ATP

00:00:00 ◼   ► He didn't follow my go-to statement that even Casey missed.

00:00:02 ◼   ► I'll move it up.

00:00:03 ◼   ► There you go.

00:00:04 ◼   ► Go-to fail.

00:00:05 ◼   ► Nice.

00:00:06 ◼   ► Because it didn't want you to just running off the end, like you got to end up going

00:00:09 ◼   ► back to Ask ATP and then I guess you just run off the end again and it loops.

00:00:13 ◼   ► What can you do?

00:00:14 ◼   ► It's basic.

00:00:16 ◼   ► Last we spoke, which feels like 13 years ago, or maybe it was just last year, I guess.

00:00:20 ◼   ► It was exactly seven days ago.

00:00:21 ◼   ► Was it really?

00:00:22 ◼   ► Oh, God.

00:00:23 ◼   ► I'm so out of whack.

00:00:24 ◼   ► You forgot the joke already?

00:00:25 ◼   ► It hasn't been that long.

00:00:26 ◼   ► Yeah, I know.

00:00:27 ◼   ► - It hasn't been that long.

00:00:28 ◼   ► - Yeah, I know.

00:00:29 ◼   ► As soon as I said, "Was it real?"

00:00:30 ◼   ► I was like, "No, wait, he's making the joke again."

00:00:33 ◼   ► No, I've been off for two weeks, so I haven't had a job,

00:00:36 ◼   ► and it's totally screwed up my schedule.

00:00:38 ◼   ► - We're gonna talk about that, by the way.

00:00:40 ◼   ► I know it's not in the list,

00:00:40 ◼   ► but we're gonna talk about that.

00:00:42 ◼   ► - Oh, God, okay, anyway.

00:00:43 ◼   ► So exactly seven days ago is what I said.

00:00:45 ◼   ► It was before Apple issued their December 28th memo

00:00:51 ◼   ► entitled "A Message to Our Customers

00:00:52 ◼   ► "About iPhone Batteries and Performance."

00:00:54 ◼   ► So this is Apple trying to do some PR cleanup about the battery gate, if you will.

00:01:02 ◼   ► The last time we spoke, this hadn't been shared yet.

00:01:06 ◼   ► A lot of people seemed semi-angry with us that we didn't take Apple to task more about

00:01:12 ◼   ► this.

00:01:13 ◼   ► A lot of people deeply believe that the only reasonable explanation for this battery stuff

00:01:17 ◼   ► isn't science, it isn't chemistry, it's greed.

00:01:21 ◼   ► The only logical explanation for this is that Apple's greedy and wants us to buy new phones.

00:01:24 ◼   ► So anyway, so Apple's been doing a lot of cleanup.

00:01:27 ◼   ► I'm satisfied with this, but I was obviously the least punchy about this issue in the first

00:01:33 ◼   ► place.

00:01:34 ◼   ► So I don't know.

00:01:35 ◼   ► Jon, you haven't had much to say.

00:01:36 ◼   ► Let's start with you.

00:01:37 ◼   ► How do you feel about this PR push?

00:01:39 ◼   ► Does this make you feel better, worse, different, don't care?

00:01:42 ◼   ► Tell me.

00:01:43 ◼   ► I thought the PR message was good.

00:01:44 ◼   ► Apple's usually pretty good with PR.

00:01:45 ◼   ► You can read it and it uses regular English sentences and explains things in a straightforward

00:01:51 ◼   ► way without tons of weasel words in it. I mean it's not going to convince anybody who's still

00:01:55 ◼   ► convinced that Apple is, you know, doing evil things on purpose, whatever. But, you know,

00:01:59 ◼   ► it's straightforward and I think it's worth linking to for people who haven't read it because

00:02:03 ◼   ► like very often companies will put out a press release and you'll kind of have to read your,

00:02:10 ◼   ► you know, the, you know, whatever the the tech side of your choice to get an explanation of what

00:02:16 ◼   ► what the press release says, you know, right?

00:02:19 ◼   ► But in the case of Apple's things,

00:02:21 ◼   ► I think you can just read the press release.

00:02:23 ◼   ► Like you don't need someone to interpret

00:02:26 ◼   ► and to translate from corporate ease into regular language.

00:02:30 ◼   ► And again, it doesn't mean you're gonna find it convincing

00:02:31 ◼   ► or like, oh, now I'm convinced before I didn't believe it,

00:02:34 ◼   ► but Apple said, Apple told me

00:02:35 ◼   ► they're not doing anything wrong

00:02:36 ◼   ► and now I believe, whatever.

00:02:37 ◼   ► Like, but anyway, you can read it and hear an explanation.

00:02:40 ◼   ► And it is typically vague

00:02:42 ◼   ► 'cause they don't wanna go into super techie details.

00:02:44 ◼   ► And I still have some curiosity about the super techie details, mostly because I want

00:02:51 ◼   ► to have something else I can tell people who have questions about it.

00:02:58 ◼   ► I know this is just about the press release and we'll get to the little thing that Renee

00:03:02 ◼   ► Ritchie tweeted about it in a second.

00:03:04 ◼   ► But the question I have, and maybe you guys know the answer, maybe you read something

00:03:08 ◼   ► more about it, is that because a lot of people are wondering, "Hey, is my iPhone, is this

00:03:13 ◼   ► happening to my iPhone. Right? So there's lots of people say, oh, you can run this thing,

00:03:16 ◼   ► it'll check your battery health and if it's below a certain percentage, then it's probably

00:03:19 ◼   ► doing it and blah, blah, blah. But the other way people have to test it is, uh, you know,

00:03:25 ◼   ► I'll run a benchmark and you know, my wife has a brand new version of the same phone

00:03:32 ◼   ► as me and she runs a benchmark and it gets this number and I run on my phone with an

00:03:36 ◼   ► old battery and I get that number or I run the benchmark and then I get my battery replaced

00:03:40 ◼   ► then I run the benchmark again.

00:03:42 ◼   ► Like the idea that running benchmarks

00:03:43 ◼   ► and getting some numbers as a way to sort of gauge

00:03:47 ◼   ► whether you're subject to the throttling.

00:03:48 ◼   ► And the question I have about this,

00:03:50 ◼   ► and I don't know the answer to it,

00:03:51 ◼   ► but like I think might be sort of a self-fulfilling prophecy

00:03:55 ◼   ► is if Apple's throttling thing is there to,

00:04:00 ◼   ► as Apple says, and as most people seem to agree,

00:04:03 ◼   ► like mechanically speaking, like how does it actually work?

00:04:05 ◼   ► Like that it clocks things down

00:04:09 ◼   ► to prevent sort of the maximum power draw like these, these things that they run at

00:04:13 ◼   ► their maximum speed and they draw the most power that they can draw. Sometimes it's just

00:04:17 ◼   ► too much and the battery old batteries can't keep up. And so what they're trying to do

00:04:21 ◼   ► by downclocking is to rent you from ever drawing that much, right? But during normal use, like

00:04:29 ◼   ► my question is, is it downclock all the time? Like from boot time on, it's like we'd normally

00:04:34 ◼   ► we would run your thing at, you know, 1 GHz, but now we're running it at 800 MHz, so

00:04:39 ◼   ► if I just make it up numbers, these aren't real. And we just run it that way from the

00:04:43 ◼   ► boot time on, and we never change the clock speed, it's 800 instead of 1 GHz the whole

00:04:48 ◼   ► time. Or does it downclock in response to demand? In other words, it's running at full

00:04:54 ◼   ► speed until you go and try to run something extremely intensive like a benchmark, and

00:04:59 ◼   ► then it downclocks you because if it didn't downclock you, you would see that you're about

00:05:03 ◼   ► to draw too much power. I don't know the answer to that, but I think if that is the case,

00:05:07 ◼   ► if it's dynamically downclocking, then by running a benchmark, you are forcing your

00:05:14 ◼   ► phone into the downclock mode, and for all you know, when you're sitting there reading

00:05:19 ◼   ► Twitter or sending text messages, it is actually running at full speed all the time, and it

00:05:22 ◼   ► only gets downclocked when you do something demanding. So again, I don't know the answer,

00:05:26 ◼   ► and that's a place where I wish Apple would be more forthcoming with the technical details,

00:05:31 ◼   ► But obviously that's not gonna be in a press release.

00:05:34 ◼   ► And so far I haven't seen any official word from Apple

00:05:36 ◼   ► explaining in nitty gritty detail exactly what it does

00:05:39 ◼   ► to your clock speed.

00:05:41 ◼   ► - I don't wanna get too far into the weeds on this.

00:05:44 ◼   ► I did wanna mention there was a great discussion

00:05:47 ◼   ► about lots of different sides of this issue,

00:05:49 ◼   ► not just how Apple has done the PR and everything,

00:05:51 ◼   ► but a lot of very good questions being brought up

00:05:54 ◼   ► on last week's episode of the talk show

00:05:56 ◼   ► with Jon Gruber and Jason Snell.

00:05:58 ◼   ► It's the talk show episode number 210 for the show notes.

00:06:01 ◼   ► And it was a really good conversation

00:06:04 ◼   ► because there's a lot of sides to this

00:06:06 ◼   ► that either we didn't cover or we only briefly covered.

00:06:09 ◼   ► And a lot of questions that have come up

00:06:11 ◼   ► kind of since then with more time to think about it

00:06:13 ◼   ► from exactly seven days ago.

00:06:15 ◼   ► And I think one of the biggest things is like

00:06:19 ◼   ► there were so many factors that made this a problem

00:06:21 ◼   ► that are totally Apple's fault.

00:06:23 ◼   ► Like, you know, things like the OS's

00:06:26 ◼   ► really do run too slowly on old hardware.

00:06:28 ◼   ► And it really does seem like they don't care that much.

00:06:32 ◼   ► I'm sure it's somebody's job,

00:06:34 ◼   ► and maybe a lot of somebody's jobs,

00:06:36 ◼   ► to make the OS's work on the older hardware,

00:06:39 ◼   ► but the actions speak louder than words,

00:06:41 ◼   ► and it clearly is not as high of a priority

00:06:43 ◼   ► as it should be, and that's been true for years.

00:06:45 ◼   ► So that's one of the many problems

00:06:48 ◼   ► that really exacerbated this for Apple.

00:06:52 ◼   ► The reason why people have been thinking for years

00:06:55 ◼   ► that Apple secretly slows down their phones

00:06:56 ◼   ► them by new ones is because new OS's so often run

00:07:00 ◼   ► so much slower than old ones when you install them

00:07:02 ◼   ► on like a one or two generation old phone.

00:07:05 ◼   ► So that's a big question.

00:07:07 ◼   ► I would also, you know, one of the bigger questions

00:07:10 ◼   ► that I think people have been asking over the last

00:07:13 ◼   ► couple of weeks is why did this only happen

00:07:17 ◼   ► from the iPhone 6 forward?

00:07:19 ◼   ► And is it happening faster than it should?

00:07:23 ◼   ► We've heard all sorts of crazy reports,

00:07:26 ◼   ► all over Twitter and email and everything,

00:07:29 ◼   ► but a lot of people are saying that they're seeing

00:07:32 ◼   ► this battery throttling happening on a phone

00:07:34 ◼   ► within its first year or its second year,

00:07:37 ◼   ► and I don't think that's reasonable.

00:07:40 ◼   ► And somebody, honestly, I haven't had a lot of time

00:07:44 ◼   ► to do research on this.

00:07:45 ◼   ► Somebody said Samsung makes some kind of guarantee

00:07:47 ◼   ► about a certain battery health percentage after two years.

00:07:51 ◼   ► I don't know enough about that, but it certainly seems like,

00:07:55 ◼   ► yes, batteries do degrade over time,

00:07:59 ◼   ► and yes, they have problems,

00:08:01 ◼   ► and this is actually a pretty clever engineering solution

00:08:03 ◼   ► to try to make phones with severely degraded batteries

00:08:06 ◼   ► still usable, but I would like to hear maybe from people

00:08:11 ◼   ► who know more about this stuff,

00:08:13 ◼   ► who know more about battery hardware design.

00:08:16 ◼   ► Is this actually inevitable at the scale and timeline

00:08:20 ◼   ► that it's happening or is there possibly some kind of bigger design problem here?

00:08:24 ◼   ► So just to get to what Renee communicated, like Apple in this press release had said

00:08:28 ◼   ► that they were going to start this like $30 battery replacement thing in mid-January but

00:08:33 ◼   ► actually they're starting it now. Like their, you know, their supplies may be limited but

00:08:37 ◼   ► they're not waiting. So if you want to go to an Apple store right now because you think

00:08:41 ◼   ► you have a bad battery and you want to get the $30 deal, you can do that. Like you don't

00:08:45 ◼   ► have to wait. So that was the minor announcement. As to what you were getting, Adam Marko, I

00:08:49 ◼   ► I think a lot of it is explained by some of the things we've experienced, or you've

00:08:55 ◼   ► experienced multiple times, with the new laptops.

00:08:58 ◼   ► It's a combination of processors getting more powerful and being more power efficient

00:09:11 ◼   ► by turning off or downclocking or momentarily increasing and decreasing their clock speed.

00:09:22 ◼   ► Like basically a larger delta between how much power does the CPU draw when it's not

00:09:26 ◼   ► really doing anything and how much does it draw when it's doing all the things.

00:09:31 ◼   ► That gap has gotten wider and at the same time Apple has been aggressively making their

00:09:37 ◼   ► things thinner and their batteries smaller.

00:09:39 ◼   ► maybe around the iPhone 6 is when it started to cross that threshold of, "Hey, this system

00:09:44 ◼   ► on a chip has a really big range between I'm doing all the things playing a game and I'm

00:09:50 ◼   ► just scrolling a text message screen."

00:09:55 ◼   ► And at the same time, the batteries are getting bigger to power the bigger screens, but also

00:10:00 ◼   ► there was sort of the end of the thinness race.

00:10:03 ◼   ► They don't think they really got much thinner after the 6, model 6 and 7 and 8 are kind

00:10:07 ◼   ► of the same and actually the 10 is even a little bit thicker than those.

00:10:12 ◼   ► So those phones could sort of be the phone equivalent of the 2016-2017 MacBook Pros where

00:10:20 ◼   ► the battery can last a long time if you're scrolling web pages in Safari or using TextEdit

00:10:24 ◼   ► or something.

00:10:26 ◼   ► But if you're doing anything aggressive on them, your battery life goes in the toilet

00:10:30 ◼   ► because the battery is scaled to sort of a low average usage, but if you use all the

00:10:37 ◼   ► CPU and GPU all the time, it destroys the battery.

00:10:41 ◼   ► And that's just, that's, you know, obviously it's not shutting off your computer or whatever,

00:10:45 ◼   ► but we discussed before the throttling presumably for thermal reasons on the Macbook Pros, but

00:10:53 ◼   ► I think this, as to whether this is inevitable, I think it's a natural consequence of essentially

00:11:00 ◼   ► chips getting better about power efficiency because they have all these things that if

00:11:04 ◼   ► they use all of them at once is tremendously fast and uses a lot of power, but it's relying

00:11:09 ◼   ► on the fact that most of the time you're not doing that.

00:11:13 ◼   ► And so they can get away with putting a very small battery in there.

00:11:17 ◼   ► If they had to treat the system on a chip on the iPhone 6, 7, 8, and 10 as if they're

00:11:23 ◼   ► running at full blast all the time, the batteries would be three times as big.

00:11:27 ◼   ► they wouldn't be able to get away with a battery that small in there.

00:11:30 ◼   ► And the other possible thing is heat.

00:11:32 ◼   ► I don't know what kind of factor that is.

00:11:34 ◼   ► I don't feel like the phones get particularly hot, but that could just be because the heat

00:11:37 ◼   ► is being trapped on the inside.

00:11:39 ◼   ► But any kind of excessive heat, or I suppose cold, really hoses your battery.

00:11:45 ◼   ► And I always, I don't know if this is founded in battery chemistry, some battery expert

00:11:49 ◼   ► can tell me or not, but I always get the feeling that like the margin of error on a really

00:11:53 ◼   ► small battery is just so much lower, like that it's thinner, that the temperature gradients

00:11:58 ◼   ► can get to the middle of it, you know what I mean, like faster. That it's just so wafer-thin

00:12:04 ◼   ► that if it's hosed in any way, if it's stressed or hosed in any way, there's so little battery

00:12:10 ◼   ► there. Like if it's made hot or cold, the whole thing heats through or gets cold straight

00:12:15 ◼   ► through immediately because it's like wafer-thin, right? As opposed to if it was like a battery

00:12:20 ◼   ► like in an old power book that was just huge brick, subjecting that to external heat for

00:12:25 ◼   ► a short period of time.

00:12:26 ◼   ► It's like, well, the outside might get hot from whatever warm thing it's next to, but

00:12:30 ◼   ► it's going to take a while for that heat to penetrate all the way through the entire cell

00:12:34 ◼   ► if the heat is an external source.

00:12:36 ◼   ► I don't know.

00:12:39 ◼   ► The solution to all these things is make processes that take even less energy even when they're

00:12:46 ◼   ► They're going full blast and put in bigger batteries because bigger batteries, it's the

00:12:51 ◼   ► cure-all, it's all battery-related things.

00:12:53 ◼   ► Yeah, everything gets heavier and it takes longer to charge.

00:12:56 ◼   ► But if you're having problems delivering enough power or your battery is getting damaged in

00:13:01 ◼   ► some way or whatever, if you just have tons of headroom and lots of excess battery capacity,

00:13:06 ◼   ► then even if the thing degrades by 20% in the first year, if you are over-provisioned

00:13:11 ◼   ► on battery by 50%, you're still good.

00:13:14 ◼   ► Yeah, I think to hopefully put a period on this whole topic, I think the thing that's

00:13:18 ◼   ► most unfortunate and distressing about this is that a lot of it, if not all of it, could

00:13:23 ◼   ► have been avoided with better messaging from Apple.

00:13:25 ◼   ► And if Apple had just been more upfront about, "Hey, you know, the unfortunate reality is

00:13:33 ◼   ► your phone's battery is a little older than, you know, it's designed to be or not, it's

00:13:39 ◼   ► not up to the snuff that we hoped it was."

00:13:42 ◼   ► So the bad news is it's not going to run quite at full speed like we use – the CPU

00:13:46 ◼   ► won't run quite as full speed as we want it to, but the good news is it won't randomly

00:13:49 ◼   ► shut down on you.

00:13:51 ◼   ► And if you have the time, why don't you go to the Apple store and we'll charge you

00:13:54 ◼   ► a little bit of money to replace the battery.

00:13:57 ◼   ► And had it been messaged more appropriately, I think it would have been a much smaller

00:14:03 ◼   ► story, and I think that we can all agree that.

00:14:05 ◼   ► But –

00:14:06 ◼   ► Or a thing about what you were just talking about with the messaging.

00:14:10 ◼   ► Aside from people just not believing the messaging and everything which we went over last show,

00:14:16 ◼   ► the other problem with that that makes it tricky and maybe one of the things that motivated

00:14:20 ◼   ► Apple not to do it, it's not as if there is a little thermometer gauge or a single number

00:14:27 ◼   ► that pops out of the battery that tells you how good or bad your battery is.

00:14:31 ◼   ► It's a lot of guesswork and sort of like, "Well, last time we went through a charge

00:14:34 ◼   ► cycle, here's what the curve looked like and here's what the voltages were at various times."

00:14:39 ◼   ► a lot of guessing and it's like well but it was just it was in a really cold car then

00:14:43 ◼   ► so does that counter but now it's really warm and like there are just so many variables

00:14:49 ◼   ► and because batteries like they're not like bananas where you get a good one and a bad

00:14:53 ◼   ► one or it's an overripe one or whatever but they are not as they're not quite as uniform

00:14:59 ◼   ► as other components might be because it's all just a big chemical soup and they try

00:15:03 ◼   ► to make them all in uniform and try to make them all good but it really depends on how

00:15:07 ◼   ► you use your phone. You leave your phone on the dashboard in the sun or in your car when

00:15:13 ◼   ► it's freezing cold weather because you leave it in your car and you forget it for two hours.

00:15:17 ◼   ► Those things hurt your battery. So if they were to bring up that message, the first problem

00:15:22 ◼   ► is you could be bringing up a message that says, "Based on all our heuristics, we kind

00:15:26 ◼   ► of think your battery is a little bit crappy." And maybe they're wrong because their heuristics

00:15:29 ◼   ► are a little bit off because of extenuating circumstances that weren't taken into account.

00:15:33 ◼   ► And now someone is really mad because their phone is three months old and now you get

00:15:35 ◼   ► story. My three-month-old iPhone told me that my battery's bad. Refund, refund, right? And

00:15:42 ◼   ► the other possibility is that you are an unlucky person who happened to get a battery that either

00:15:49 ◼   ► was bad from the factory or got damaged by some excessive temperature changes during shipping or

00:15:56 ◼   ► something like that. Like someone, if all of the components that you could potentially get that

00:16:01 ◼   ► that are end up being like, you know, bad, right?

00:16:05 ◼   ► Most of the other components have good, easy ways

00:16:07 ◼   ► to test them, but batteries, you can test it at the factory

00:16:11 ◼   ► and it'll be like, passes all, it tests with flying colors,

00:16:13 ◼   ► but after three weeks of charge cycle,

00:16:14 ◼   ► some internal weakness or flaw in the battery

00:16:17 ◼   ► causes it to not be performing well.

00:16:20 ◼   ► And in that case, I mean, in that case,

00:16:22 ◼   ► the messaging is good for the consumer to say like,

00:16:25 ◼   ► oh, I've got a bad battery,

00:16:26 ◼   ► and presumably they'd get it replaced under warranty.

00:16:30 ◼   ► But it's like, how do you distinguish that

00:16:33 ◼   ► from the case I described before,

00:16:34 ◼   ► where actually your battery is fine

00:16:35 ◼   ► and the thing is just freaking out for no good reason?

00:16:37 ◼   ► So I'm not sure what the solution is.

00:16:40 ◼   ► A lot of the time people have battery issues

00:16:42 ◼   ► with their phone and the easy answer

00:16:44 ◼   ► is to just get a new battery.

00:16:45 ◼   ► And one of the things I think I learned

00:16:46 ◼   ► from the feedback from last show

00:16:48 ◼   ► is that a surprising number of people

00:16:51 ◼   ► who listen to the show, either themselves

00:16:55 ◼   ► or didn't know or know people who don't know

00:16:58 ◼   ► that you can replace the battery on an iPhone.

00:17:00 ◼   ► Like that's even a thing that can be done

00:17:02 ◼   ► for any amount of money anywhere.

00:17:04 ◼   ► I assumed that it was a thing that everybody knew

00:17:07 ◼   ► and I was definitely wrong about that.

00:17:09 ◼   ► A lot of people were saying,

00:17:10 ◼   ► until this thing, I didn't even know

00:17:12 ◼   ► you could replace a battery in the phone.

00:17:13 ◼   ► If I had known I could have just replaced the battery,

00:17:15 ◼   ► I never would have got it on a new phone.

00:17:16 ◼   ► Now some people are saying

00:17:17 ◼   ► if I had known a battery replacement would fix the problem,

00:17:19 ◼   ► but other people were saying,

00:17:20 ◼   ► I didn't even know that was a thing that you would do,

00:17:22 ◼   ► or maybe they didn't think it was a thing

00:17:23 ◼   ► that Apple would do.

00:17:24 ◼   ► Like maybe you could find some sketchy vendor

00:17:27 ◼   ► who will crack open your phone and give it a go,

00:17:29 ◼   ► or you could try to do it yourself,

00:17:30 ◼   ► but that it's an official Apple thing that you can do.

00:17:33 ◼   ► And the final thing on that is,

00:17:35 ◼   ► lots of people have been reporting,

00:17:38 ◼   ► and I don't know if this is Apple policy

00:17:41 ◼   ► or informal Apple policy or not,

00:17:42 ◼   ► but before this all happened,

00:17:44 ◼   ► if you took a phone to an Apple store and say,

00:17:47 ◼   ► "Hey, I'm having problems with my battery,"

00:17:50 ◼   ► they would run your phone through like a diagnostic

00:17:52 ◼   ► to check like your battery health according to whatever,

00:17:54 ◼   ► you know, whatever rules and number pops out

00:17:57 ◼   ► says your battery is like 80% okay or whatever.

00:17:59 ◼   ► And there was some number, and if it was below that number,

00:18:02 ◼   ► they would say, okay, we'll replace your battery

00:18:05 ◼   ► for 80 bucks or whatever,

00:18:06 ◼   ► like whatever they were charging, right?

00:18:07 ◼   ► But if it was not below that number,

00:18:09 ◼   ► if it says your battery is 99% healthy,

00:18:11 ◼   ► according to our tests,

00:18:12 ◼   ► that they wouldn't do the replacement for you.

00:18:14 ◼   ► And that you would throw your money at this,

00:18:16 ◼   ► no, look here, money, $80,

00:18:18 ◼   ► take this money and replace my battery.

00:18:20 ◼   ► And they say, no, sorry, our diagnosis tell us

00:18:22 ◼   ► that your battery is still good,

00:18:24 ◼   ► so we won't replace your thing, right?

00:18:26 ◼   ► And enough people gave me that story

00:18:29 ◼   ► that it seems like a thing that's really happening.

00:18:30 ◼   ► And the reason I was interested in it is because

00:18:33 ◼   ► we had an iPhone 5S that we were,

00:18:34 ◼   ► it was gonna be a hand-me-down phone for my son.

00:18:37 ◼   ► And we thought like, oh, before I, you know,

00:18:40 ◼   ► it had been well used.

00:18:41 ◼   ► It had been used for at least two years, maybe longer.

00:18:44 ◼   ► Heavily used.

00:18:45 ◼   ► And so we're like, oh, well,

00:18:46 ◼   ► I don't wanna give him a phone with a two-year-old battery.

00:18:48 ◼   ► You know, we wanna be able to reach him

00:18:49 ◼   ► and he's probably not gonna be very good about charging it.

00:18:51 ◼   ► So before I give him this phone,

00:18:52 ◼   ► let me go to the Apple store and replace the battery.

00:18:54 ◼   ► So I brought the 5S to the Apple store to replace the battery.

00:18:57 ◼   ► I gave it to the Apple person.

00:18:58 ◼   ► I said to her, I made a genius bar appointment.

00:19:00 ◼   ► I said, I just want a new battery in this phone.

00:19:03 ◼   ► And he takes it from me and runs it through diagnostic

00:19:06 ◼   ► and says, oh, this battery actually

00:19:07 ◼   ► looks like it's pretty good.

00:19:08 ◼   ► It's like 85%.

00:19:09 ◼   ► I'm like, yeah, I know.

00:19:11 ◼   ► There's nothing wrong with the battery.

00:19:12 ◼   ► I just want to replace it because I'm giving it

00:19:13 ◼   ► to my son as a gift and I want him to have a fresh battery.

00:19:16 ◼   ► And the guy said, oh, OK.

00:19:20 ◼   ► And he did it for me.

00:19:22 ◼   ► And only now in retrospect, do I look back,

00:19:24 ◼   ► think back to the like the strange pause

00:19:26 ◼   ► and the way he kind of said, oh, to think,

00:19:29 ◼   ► maybe he was doing me a favor.

00:19:31 ◼   ► Like I wasn't like aggressive about it.

00:19:33 ◼   ► I had no expectation that I was gonna be told no.

00:19:35 ◼   ► I was just explaining, oh yeah, no, sorry.

00:19:37 ◼   ► Sorry to make you think I had a problem with my battery.

00:19:40 ◼   ► I don't have a problem with my battery.

00:19:41 ◼   ► I am just here saying just period, I want a new battery.

00:19:44 ◼   ► Didn't argue with me and didn't tell me

00:19:46 ◼   ► they have a policy they do, but he did do a little pause.

00:19:49 ◼   ► So I don't know what to think anymore.

00:19:52 ◼   ► Maybe Apple geniuses can chime in and say,

00:19:54 ◼   ► what was the policy before this whole throttling PR thing?

00:19:58 ◼   ► But either way, at $80, I feel like Apple

00:20:02 ◼   ► should have replaced the battery for anybody

00:20:04 ◼   ► who could throw that money at them.

00:20:05 ◼   ► At $30, I feel like Apple's probably losing money

00:20:07 ◼   ► in every one of these battery replacements,

00:20:08 ◼   ► so I kind of understand if they're not gonna give you

00:20:10 ◼   ► the $30 deal if your battery's like brand spanking new

00:20:13 ◼   ► and their test says you have 95% capacity or whatever.

00:20:17 ◼   ► All right, so tell me about phone throttling

00:20:21 ◼   ► and how it's not always the explanation.

00:20:23 ◼   ► - This is another big thread of feedback.

00:20:24 ◼   ► Hey, my phone takes 25 seconds

00:20:27 ◼   ► to launch the messages application.

00:20:29 ◼   ► Damn Apple and that thermal throttling.

00:20:31 ◼   ► A lot of things can be wrong with phones.

00:20:34 ◼   ► I wanna tell people that if your phone is super slow

00:20:37 ◼   ► and doing all sorts of terrible things,

00:20:39 ◼   ► possibly related to this just started

00:20:41 ◼   ► after the iOS 11 update or whatever it may be,

00:20:44 ◼   ► There are many, many things that people have experienced

00:20:49 ◼   ► and have written to us about

00:20:49 ◼   ► and that I've experienced personally

00:20:51 ◼   ► where your phone gets unreasonably slow

00:20:55 ◼   ► when it's doing nothing hard and with a brand new battery.

00:20:59 ◼   ► And I can tell you that those things

00:21:02 ◼   ► are not related to phone throttling.

00:21:05 ◼   ► Those are related to something else terrible

00:21:06 ◼   ► and I wish I could tell you what it was.

00:21:08 ◼   ► Lots of people like wipe and restore your phone

00:21:10 ◼   ► and that would fix it, which is the worst possible,

00:21:12 ◼   ► like you almost wish that doesn't work.

00:21:14 ◼   ► like please don't let this work because it's the worst kind of sort of cross my fingers

00:21:18 ◼   ► plug in and unplug it thing.

00:21:19 ◼   ► But very often that actually does work, which is terrible.

00:21:22 ◼   ► That should not work.

00:21:24 ◼   ► You would love to know what the problem is, but all I'm going to say is if your phone

00:21:27 ◼   ► is like really slow, like someone just wrote and said they press play on like a podcast

00:21:33 ◼   ► player and it takes five seconds for audio to start, that is, I'm pretty sure that's

00:21:38 ◼   ► not CPU throttling, right?

00:21:41 ◼   ► Because the throttling is not, will you slow your CPU to 1%, right?

00:21:45 ◼   ► And so if you're getting any kind of performance problems that are pervasive everywhere throughout

00:21:51 ◼   ► the operating system and really, really slow, like 100 times, 200 times slower than it should

00:21:57 ◼   ► be, like you're waiting three minutes for messages to launch, that is not this issue.

00:22:02 ◼   ► That is another issue that is probably worse, and we can't even tell you what it is.

00:22:08 ◼   ► And if you want to go get a new battery anyway, go ahead.

00:22:10 ◼   ► But for those things, like, you know, sometimes it's not a tumor and sometimes it's not phone

00:22:16 ◼   ► throttling, sometimes it's something even worse.

00:22:20 ◼   ► And yes, sometimes it does coincide with an OS upgrade and I have no explanation for what

00:22:24 ◼   ► it is, but I can tell you it is definitely a thing.

00:22:26 ◼   ► I experienced it myself, my wife experienced it on her phone.

00:22:31 ◼   ► Wiping in restore fixes it for me about like 60% of the time and I hate that it does.

00:22:36 ◼   ► I don't really recommend people wipe and restore as a matter of course, but unless

00:22:41 ◼   ► someone else knows a better way to figure out what the heck that problem is, you just

00:22:45 ◼   ► want to make the problem go away sometimes.

00:22:46 ◼   ► Unfortunately, sometimes wipe and restore doesn't fix it, and then what do you do?

00:22:49 ◼   ► I don't even know.

00:22:50 ◼   ► But anyway, all that is to say, if there's something wrong with your phone and it's

00:22:54 ◼   ► still under warranty, bring it to the Apple genius.

00:22:57 ◼   ► Let them sort it out.

00:22:58 ◼   ► If they can't figure it out, maybe they'll just give you a new phone.

00:23:01 ◼   ► But it's not always throttling.

00:23:02 ◼   ► I find in my experience it's usually lupus.

00:23:06 ◼   ► Speaking more of throttling, let's talk about the iMac Pro and its thermal throttling.

00:23:10 ◼   ► This is the show about throttling apparently.

00:23:12 ◼   ► Yeah, this is slightly different, but anyway, people have iMac Pros now, including Marco,

00:23:16 ◼   ► which we'll get to later in the show.

00:23:19 ◼   ► And some of the people who have had them are testing them out.

00:23:22 ◼   ► And one of the interesting tests that you can do with these, now that they have them

00:23:25 ◼   ► and they have something that can measure all the temperatures of all the little bits is,

00:23:29 ◼   ► "Hey, let's see if anything we can do with this iMac Pro causes any of the parts inside

00:23:35 ◼   ► it to not run at their normal clock speed."

00:23:41 ◼   ► And the answer is that yes, there are things you can do to make both the CPU and the GPU

00:23:45 ◼   ► dip below their normal clock speed, albeit briefly.

00:23:49 ◼   ► This is an Apple Insider article, and they ran it through a bunch of benchmarks on CPU

00:23:53 ◼   ► and GPU.

00:23:54 ◼   ► And for the most part, it stays pegged at whatever its, you know, speed that its rated

00:23:58 ◼   ► speed is, but there's a nice little graph, so you can see it dips for a second or two

00:24:02 ◼   ► down to a slower clock speed, and it goes back up to the top.

00:24:09 ◼   ► And the interesting part, I think, about this throttling, and this is nothing to do with

00:24:13 ◼   ► battery, obviously, because this guy's plugged in, the interesting—this is all about heat—the

00:24:16 ◼   ► interesting part of this throttling is to say, during all these tests, the fans are

00:24:21 ◼   ► not really going nuts, right?

00:24:24 ◼   ► In other words, that it seems like this throttling could be avoided if the system was tuned to

00:24:31 ◼   ► crank the fan speed up a little bit more.

00:24:35 ◼   ► So it seems like, at least as far as this current iteration of whatever firmware and

00:24:39 ◼   ► OS combination determines all this stuff, that the iMac Pro is tuned to be quiet, as

00:24:46 ◼   ► quiet as possible, while more or less maintaining maximum speed.

00:24:50 ◼   ► And again, you can't instantaneously crank up the fans and instantaneously extract heat

00:24:55 ◼   ► that's going to cause the thing to throttle itself, right?

00:24:58 ◼   ► So I think their option would be to avoid these brief, you know, one or two second dips

00:25:04 ◼   ► below normal clock speed, right?

00:25:07 ◼   ► To avoid that, you would actually have to run the fans louder and sooner, perhaps to

00:25:11 ◼   ► a significant degree.

00:25:13 ◼   ► And instead, they've chosen to tune this thing to be as quiet as possible while allowing

00:25:19 ◼   ► for occasional dips in clock speed, which is an interesting trade-off, which means that

00:25:23 ◼   ► I bet this machine is pretty quiet, and again, Marco, you will get a chance to explain all

00:25:27 ◼   ► your experiences with this eventually.

00:25:30 ◼   ► But also, this is another way that I feel like this thing is hopefully, probably, potentially

00:25:37 ◼   ► differentiated from the Mac Pro, because again, if a machine is ever going to be tuned to

00:25:44 ◼   ► never dip its clock speed because it can't get enough heat away from it, a purpose design

00:25:49 ◼   ► modular doesn't have a screen attached to it, yada yada yada, like pro computer, would

00:25:55 ◼   ► be the one to do it. And the all-in-one sleek, slim, amazingly compact, pro components jammed

00:26:01 ◼   ► behind a screen is the one that is going to stay as quiet as possible, but occasionally

00:26:07 ◼   ► allow dips in clock speed so it doesn't have to be too noisy.

00:26:13 ◼   ► We are sponsored this week by Flight Logger. Go to flightlogger.co or search for Flight

00:26:18 ◼   ► Logger in the App Store. Flight Logger lets you easily search and save your

00:26:22 ◼   ► flights, organize them into custom trips, share your travel schedule with friends

00:26:26 ◼   ► and family, and much more. All from a very easy to use app and in my opinion a

00:26:31 ◼   ► really nice looking one. Flight Logger brings the ease back to air travel with

00:26:36 ◼   ► a visual timeline of your itinerary to keep you on schedule no matter where you

00:26:39 ◼   ► are in the world. You can use Flight Logger to keep track of all your travel

00:26:43 ◼   ► information, any updates or changes it will alert you. You will never miss a

00:26:46 ◼   ► of flight again with their real-time notifications on terminal and gate changes. You can track

00:26:50 ◼   ► and get notified of arrival and departure times, delays and baggage claim information,

00:26:54 ◼   ► which is awesome. It will tell you what claim your baggage are coming on. And you can share

00:26:58 ◼   ► your flight information with friends and family for no waiting airport pickups and peace of

00:27:01 ◼   ► mind. Flight logging and flight tracking and flight status apps, honestly, the ones I've

00:27:07 ◼   ► seen that aren't flight logger, they're kind of a dumpster fire. Like, it is not an easy-to-use

00:27:13 ◼   ► attractive category of apps, generally speaking.

00:27:16 ◼   ► And I actually used FlightLogger before they were

00:27:18 ◼   ► a sponsor here because I've heard them recommended

00:27:20 ◼   ► strongly on other podcasts.

00:27:21 ◼   ► And I gotta tell ya, the hype is real.

00:27:24 ◼   ► It's a great app and it's miles ahead of the others

00:27:28 ◼   ► in terms of having a nice appearance and making the

00:27:31 ◼   ► features easy to use and easy to understand.

00:27:33 ◼   ► It really is great and it supports all,

00:27:35 ◼   ► I mean there's so much stuff, I can't even tell you

00:27:37 ◼   ► it all in an ad.

00:27:38 ◼   ► You can locate the aircraft on a map,

00:27:40 ◼   ► you can see your flight as you go, which I love.

00:27:42 ◼   ► You can import flights via email forwarding.

00:27:44 ◼   ► You can just forward your confirmation

00:27:45 ◼   ► to a special email address,

00:27:47 ◼   ► and it'll import automatically.

00:27:49 ◼   ► There's so much to see here,

00:27:50 ◼   ► and they even have a cool little trick.

00:27:52 ◼   ► Go to the settings menu and shake your phone,

00:27:55 ◼   ► and you will get a new entry that I think ATP listeners,

00:27:58 ◼   ► especially John Siracusa, might appreciate.

00:28:01 ◼   ► So check it out.

00:28:02 ◼   ► Go to flightlogger.co.

00:28:04 ◼   ► That's flightlogger in the App Store,

00:28:06 ◼   ► or flightlogger.co for an incredibly nice, easy to use app

00:28:10 ◼   ► for easily tracking, saving,

00:28:13 ◼   ► and getting alerts for your flights.

00:28:15 ◼   ► Once again, thank you so much to Flight Logger,

00:28:18 ◼   ► a really cool flight tracking app,

00:28:19 ◼   ► for sponsoring our show.

00:28:20 ◼   ► - You did order an iMac Pro, as it turns out.

00:28:26 ◼   ► - As it turns out, yes I did.

00:28:28 ◼   ► - And?

00:28:29 ◼   ► - And I am now running High Sierra.

00:28:33 ◼   ► - Aw, welcome to months ago.

00:28:35 ◼   ► - Yeah, my fonts are now overly smooth and blurry

00:28:38 ◼   ► because my font smoothing setting is not fixed.

00:28:41 ◼   ► - Let me interrupt you right there.

00:28:42 ◼   ► I have heard you, you know,

00:28:44 ◼   ► conventioning about this for months.

00:28:47 ◼   ► I don't understand what it is that you're so angry about,

00:28:50 ◼   ► but I really kind of don't want to know

00:28:53 ◼   ► because I don't see it right now.

00:28:55 ◼   ► I don't see the arrow in the FedEx logo,

00:28:57 ◼   ► and I don't want to, so.

00:28:59 ◼   ► - You haven't changed your settings.

00:29:00 ◼   ► Marco is changing his settings from the defaults

00:29:02 ◼   ► for his font smoothing.

00:29:03 ◼   ► - I mean, I used to expect that if a setting

00:29:05 ◼   ► is going to be there, that it should work properly.

00:29:07 ◼   ► Otherwise, why have the setting?

00:29:09 ◼   ► - Wait, so what do you set it to?

00:29:11 ◼   ► - So the setting I'm talking about is,

00:29:12 ◼   ► in settings, general, turning off,

00:29:16 ◼   ► use LCD font smoothing when available.

00:29:19 ◼   ► - Yeah, mine is off.

00:29:20 ◼   ► So what am I supposed to see that's garbage?

00:29:22 ◼   ► - It's off?

00:29:23 ◼   ► - Yeah.

00:29:24 ◼   ► - Okay, open Xcode or something else

00:29:27 ◼   ► that would truncate text automatically with an ellipsis.

00:29:30 ◼   ► Look in the source list, make it nice and narrow.

00:29:33 ◼   ► - All right, hold on, hold on, hold on.

00:29:34 ◼   ► Slow down, man, slow down.

00:29:35 ◼   ► I was trying to find the right project, okay.

00:29:36 ◼   ► - Launching Xcode, bounce, bounce, bounce.

00:29:39 ◼   ► - Launching Xcode on your non-pro iMac must take ages.

00:29:43 ◼   ► - Oh, you're such a jerk.

00:29:44 ◼   ► Okay, anyway.

00:29:45 ◼   ► So I see an ellipsis in the middle of one of the folder names

00:29:49 ◼   ► in this particular project.

00:29:50 ◼   ► - So what you should see, if that setting is unchecked,

00:29:54 ◼   ► is you should see that the text with an ellipsis

00:29:56 ◼   ► should be rendered thicker than the text without the ellipsis.

00:30:00 ◼   ► - Oh, (bleep) you, Marco.

00:30:02 ◼   ► Oh, God bless America.

00:30:04 ◼   ► - You see it, right?

00:30:05 ◼   ► "Hey, I'm never gonna be able to unsee this now."

00:30:07 ◼   ► - Right, and that isn't the only place that this shows up.

00:30:11 ◼   ► Basically, if text is being rendered with certain,

00:30:15 ◼   ► I don't know the details, but with certain

00:30:17 ◼   ► core text behaviors that are being applied to it,

00:30:20 ◼   ► they basically make it ignore that setting

00:30:22 ◼   ► and make it always use LCD font smoothing.

00:30:26 ◼   ► And that's new to High Sierra, that is not yet fixed.

00:30:28 ◼   ► In whatever apps and workflows and things I use,

00:30:32 ◼   ► that shows up a lot.

00:30:34 ◼   ► So I'd rather just live with my blurry text everywhere

00:30:38 ◼   ► than have it be inconsistently blurry.

00:30:40 ◼   ► So right now I am living with that setting back on,

00:30:43 ◼   ► which is the default, and it sucks,

00:30:46 ◼   ► but it sucks less than having everything be inconsistent.

00:30:50 ◼   ► I think it also applies to finder windows and stuff.

00:30:52 ◼   ► It applies a lot of places, and it's annoying.

00:30:55 ◼   ► - So why do you think LCD font smoothing is blurrier

00:30:57 ◼   ► than plain old anti-aliasing?

00:30:58 ◼   ► To my eyes, plain old anti-aliasing with gray pixels

00:31:02 ◼   ► is blurrier than the little sub-pixel thing

00:31:05 ◼   ► that the LCD thing is doing.

00:31:07 ◼   ► - I haven't ever taken a macro lens to my screen

00:31:10 ◼   ► and really compared how they're doing it,

00:31:12 ◼   ► but it seems like the effective visual thickness

00:31:17 ◼   ► of the fonts, how thick they appear to be,

00:31:21 ◼   ► appears slightly thicker when it's using the LCD fonts,

00:31:25 ◼   ► which is their version of sub-pixel anti-aliasing.

00:31:27 ◼   ► That's what we're talking about.

00:31:28 ◼   ► - See, I would think it was the opposite.

00:31:29 ◼   ► In old non-retinal monitors, like the one I'm staring at,

00:31:31 ◼   ► I always find the opposite, that LCD font smoothing

00:31:34 ◼   ► makes them appear thinner, and plain old anti-aliasing

00:31:37 ◼   ► with a series of grayish pixels makes them look thicker.

00:31:40 ◼   ► - Yeah, and that's what I would assume,

00:31:42 ◼   ► but regardless, the way it appears visually to me,

00:31:46 ◼   ► like perceiving, the perception of how it looks,

00:31:49 ◼   ► is that when the sub-pixel anti-aliasing is being used,

00:31:54 ◼   ► which is the default, the fonts do appear thicker.

00:31:57 ◼   ► - Yeah, I would agree with that,

00:31:58 ◼   ► that when I ticked this checkbox back on,

00:32:02 ◼   ► everything appeared thicker to me than otherwise.

00:32:05 ◼   ► I don't know, weird though.

00:32:06 ◼   ► Well, thank you for ruining that.

00:32:08 ◼   ► - Yeah, you're welcome.

00:32:09 ◼   ► (laughs)

00:32:10 ◼   ► You can go back to Sierra, it won't happen anymore.

00:32:12 ◼   ► - All right, so how's your iMac Pro?

00:32:14 ◼   ► It's beautiful, I've seen one once at WWDC.

00:32:17 ◼   ► In fact, Jon was with me.

00:32:19 ◼   ► I can tell you they're beautiful,

00:32:20 ◼   ► but tell me, other than the physical appearance, how is it?

00:32:22 ◼   ► - Honestly, ask me next week,

00:32:24 ◼   ► because I've had it for literally one day,

00:32:28 ◼   ► a day during which I have been able to do very little on it

00:32:31 ◼   ► because I've been very, very busy.

00:32:33 ◼   ► The one thing I will, I mean, I'm using it right now,

00:32:34 ◼   ► it's wonderful so far, it does look incredible,

00:32:37 ◼   ► all the black things on my desk now look cool.

00:32:40 ◼   ► The few remaining silver things on my desk,

00:32:41 ◼   ► like my speaker and headphone amps,

00:32:45 ◼   ► now look really out of place

00:32:46 ◼   ► 'cause they're silver instead of black,

00:32:47 ◼   ► so God knows what I'm gonna do about that.

00:32:49 ◼   ► (laughs)

00:32:50 ◼   ► - Yeah, what about all your (bleep) boxes?

00:32:52 ◼   ► Those all are silver, right? - Exactly.

00:32:54 ◼   ► - They are, yeah.

00:32:54 ◼   ► Now, it does have black versions of some of their products,

00:32:58 ◼   ► so I might look into that, but.

00:33:00 ◼   ► - Oh, God.

00:33:01 ◼   ► - I could just move them or just deal with it.

00:33:02 ◼   ► - I can get you a can of Krylon,

00:33:04 ◼   ► you can fix all those silver boxes real quick.

00:33:06 ◼   ► - Yeah, anyway, so yeah.

00:33:09 ◼   ► So, my initial impressions, I don't really have any yet

00:33:13 ◼   ► because it hasn't been enough time.

00:33:14 ◼   ► However, one thing that is immediately striking,

00:33:17 ◼   ► besides the look, is just how incredibly quiet it is.

00:33:22 ◼   ► even quieter than the 5K, for me at least.

00:33:25 ◼   ► And maybe that's because my 5K is old

00:33:26 ◼   ► and the platters are full of bad blocks

00:33:28 ◼   ► or whatever the Apple people told me.

00:33:30 ◼   ► I don't know. (laughs)

00:33:32 ◼   ► - More likely the inside of the case

00:33:33 ◼   ► is filled with dog hair.

00:33:35 ◼   ► - My dog doesn't shed.

00:33:36 ◼   ► But yeah, it's-- - You're right,

00:33:37 ◼   ► it doesn't shed.

00:33:38 ◼   ► - That makes a difference, and he really doesn't.

00:33:41 ◼   ► He needs a haircut very badly, but he does not shed.

00:33:43 ◼   ► Anyway, so it's strikingly quiet.

00:33:47 ◼   ► It is ridiculously quiet.

00:33:49 ◼   ► So I am very much enjoying that,

00:33:51 ◼   ► And that, if nothing else, will push me to very aggressively

00:33:55 ◼   ► move this into being Tiff's computer as soon as the Mac Pro

00:34:00 ◼   ► is presumably released if I want it,

00:34:02 ◼   ► because I want her computer across the room

00:34:04 ◼   ► to be this quiet as well.

00:34:06 ◼   ► It is shocking how quiet it is.

00:34:08 ◼   ► And how quiet it seems to remain under load

00:34:11 ◼   ► by all the reviews, but I have not experienced that yet.

00:34:14 ◼   ► But ask me again next week.

00:34:16 ◼   ► - So both Jason Snell and Underscore have

00:34:19 ◼   ► have told me that FFmpeg encodes tend to be twice as quick on the iMac Pro as compared

00:34:27 ◼   ► to my 5K iMac.

00:34:29 ◼   ► And although I don't do FFmpeg encodes as often as I probably paint it, it happens often

00:34:35 ◼   ► enough that I am salivating at the thought of it.

00:34:38 ◼   ► And then I see the price tag and I'm reminded that I don't need one of these after all.

00:34:42 ◼   ► My goodness, it seems like it would be, it's just so much faster and I'm super jealous.

00:34:47 ◼   ► - I just think all my RAM errors

00:34:49 ◼   ► are being corrected and detected.

00:34:51 ◼   ► Oh man, this is so great.

00:34:52 ◼   ► - That's so harsh.

00:34:53 ◼   ► - It's so nice to be back on a Mac Pro again.

00:34:55 ◼   ► - Oh, you're so mean to Jon.

00:34:57 ◼   ► - Yeah, like I said in the follow-up,

00:34:59 ◼   ► but I do find it interesting that this machine

00:35:01 ◼   ► appears to be tuned for quiet above all else,

00:35:04 ◼   ► which is like an interesting way to go for, you know,

00:35:07 ◼   ► like what it says to me that actually,

00:35:09 ◼   ► that they do actually have cooling, more cooling,

00:35:11 ◼   ► especially since they chose parts that are clocked lower,

00:35:14 ◼   ► like whatever these special number parts

00:35:16 ◼   ► they use from Intel. They're like similar to other parts that Intel offers but a little bit lower

00:35:20 ◼   ► clocked. Like they would have the headroom but they wanted it to be quieter, right? Rather than

00:35:27 ◼   ► having it like it's almost as if the design goal was quieter than the 5k iMac. Which you could see

00:35:33 ◼   ► them then maybe saying it should be as quiet as the 5k iMac but it ends up being quieter.

00:35:37 ◼   ► Like you know dramatically quieter. And not because it can be that quiet and never throttle

00:35:45 ◼   ► but it can be that quiet and almost never throttle. Like just little blips, little blips here and

00:35:51 ◼   ► there. Like I encourage people to check out the Apple Insider article and look at the graphs. It's

00:35:54 ◼   ► mostly never throttling, little dips here and there. And so, and again, like this probably varies

00:35:59 ◼   ► on the ambient temperature of the room, how good the thermal paste is on the model that you happen

00:36:04 ◼   ► to get, all sorts of things like that. That's something else that, a couple of follow-up

00:36:08 ◼   ► YouTube links that unfortunately I don't have for the show notes, but if you just go onto YouTube

00:36:12 ◼   ► and look for like iMac or Apple thermal throttle cooling blah blah blah, PowerBook, sorry,

00:36:20 ◼   ► MacBook thermal throttling.

00:36:22 ◼   ► One of the things that the PC builders like to do is take apart Apple computers and take

00:36:28 ◼   ► off the heat sinks and the heat pipes and scrape off the thermal paste and put on like

00:36:34 ◼   ► their fancy thermal paste.

00:36:36 ◼   ► Sometimes they put on multiple kinds of fancy thermal paste which boggles my mind that they

00:36:39 ◼   ► They take the entire thing apart, scrape off the thing, clean the CPU in the heat sink,

00:36:44 ◼   ► put on thermal paste, reassemble the entire thing, test it, and then repeat it, only to

00:36:48 ◼   ► scrape off the paste they just put on and put on an even more ridiculously expensive

00:36:52 ◼   ► special thermal paste made with like, you know, ARCTIC SILVER!

00:36:56 ◼   ► Yeah, exactly, made with these adamantium elements that make the things more expensive

00:37:01 ◼   ► than printer ink.

00:37:03 ◼   ► But they get like, you know, double, you know, single or even sometimes double digit increases

00:37:08 ◼   ► in or decreases in temperature just from merely applying new thermal paste or better thermal

00:37:15 ◼   ► paste.

00:37:17 ◼   ► And this is true not just of laptops but of all computers.

00:37:21 ◼   ► As it ages and as you go through heating cool and heating cool cycles over and over again,

00:37:28 ◼   ► the thermal bond between whatever is pulling heat away from your chip and the top of the

00:37:32 ◼   ► chip itself degrades, right? And I've always wondered why Apple doesn't use whatever these

00:37:38 ◼   ► very fancy thermal paste things are. Maybe they, maybe these ones last even, like they're

00:37:45 ◼   ► good when they're installed new, but after a year they're crap. Maybe it's just a cost

00:37:49 ◼   ► saving thing. Maybe they can't get thermal paste of that quality in the volumes needed

00:37:54 ◼   ► to sell millions and millions of whatevers. But it's always something I suspect when I

00:37:59 ◼   ► hear someone says like, "Oh, I got a 5K iMac and it's like super loud." Or, you know, that

00:38:04 ◼   ► happened with the PlayStation 4, so a lot of the PlayStation 4 is like, "I got a brand

00:38:07 ◼   ► new PlayStation 4 and it sounds like a jet engine is taking off." And again, getting

00:38:11 ◼   ► back to the battery as being like the sort of touchy-feely squishy element that could

00:38:17 ◼   ► go, that could be like a, you get a bad banana, right? The touchy-feely squishy element of

00:38:22 ◼   ► almost all electronics these days is how, how well assembled was it? Like, how good

00:38:29 ◼   ► as the thermal bond between the hot little square or rectangle and the apparatus meant

00:38:35 ◼   ► to pull heat away. And so you may end up getting a PlayStation 4 where it's just not seated

00:38:41 ◼   ► properly or there's not enough thermal paste or there's too much thermal paste or it's

00:38:45 ◼   ► crooked or some other problem with it. And right out of the box the fans run at max speed.

00:38:50 ◼   ► You're like, "Oh, what's wrong with this thing?" It's like, yeah, that part of it. The silicon

00:38:54 ◼   ► chip is solid state and they can test it and verify it and if it's good it's probably good.

00:39:00 ◼   ► And the assembly process, if they over torque or under torque a few screws it doesn't matter,

00:39:04 ◼   ► but if one of those screws is one of the screws that keeps the heat sink properly seated on

00:39:09 ◼   ► the thing or if the little machine or person that puts the thermal paste on got it off

00:39:14 ◼   ► by a few millimeters or just so many things can go wrong that seem very analog. And if

00:39:22 ◼   ► If that goes wrong, it's like you just feel like you lost the lottery, like you're bummed

00:39:26 ◼   ► out or you won the crappy lottery.

00:39:29 ◼   ► Most of the time everything's fine, but every once in a while you get a lemon and you're

00:39:32 ◼   ► like, "Why are these fans so loud?"

00:39:33 ◼   ► It's like, if you're one of those people, you could crack it open and try to fix it,

00:39:37 ◼   ► but if you're not, you have to just try to return it or try to...

00:39:40 ◼   ► They'll say, "Oh, fan noise is normal.

00:39:42 ◼   ► When you're playing a game, the fans just spin up."

00:39:44 ◼   ► It's like, "Yeah, but when I just turn it on, listen to this.

00:39:46 ◼   ► You can tell there's something wrong with it, right?

00:39:47 ◼   ► And you have to argue with the person and beg."

00:39:49 ◼   ► It's like my worst nightmare.

00:39:52 ◼   ► I've had pretty good luck with fans so far.

00:39:55 ◼   ► But like imagine, you get a new 5K iMac.

00:39:57 ◼   ► Who was the one who did this?

00:39:58 ◼   ► Was it Stephen Hackett who was complaining

00:39:59 ◼   ► about his noisy 5K iMac?

00:40:00 ◼   ► Someone we know, I think, recently got a brand new 5K iMac

00:40:03 ◼   ► and could not believe how loud it was.

00:40:05 ◼   ► And now whenever I hear that, I'm like,

00:40:06 ◼   ► maybe they just expected it to be quieter

00:40:09 ◼   ► or maybe they lost the thermal paste lottery.

00:40:12 ◼   ► - Yeah, well, and I think that one was actually

00:40:14 ◼   ► like a defective one.

00:40:14 ◼   ► But I will say that like, you know,

00:40:17 ◼   ► now having used the 5K iMac full-time for three years,

00:40:22 ◼   ► My only complaint really was that under load, it got loud.

00:40:27 ◼   ► And it was kind of annoying to hear the fans

00:40:29 ◼   ► when it was under a load.

00:40:31 ◼   ► And so to have this brand new, awesome, high-end generation

00:40:36 ◼   ► of the iMac basically not do that at all,

00:40:40 ◼   ► or do that to such a small level

00:40:41 ◼   ► that most people don't even notice it's doing it at all,

00:40:43 ◼   ► basically to have it be quiet under most loads,

00:40:47 ◼   ► that is a huge upgrade.

00:40:51 ◼   ► And you could argue, as some people,

00:40:53 ◼   ► like on that Apple Insider comment thread did,

00:40:56 ◼   ► you can argue like it's a pro machine,

00:40:57 ◼   ► it should never throttle from thermals.

00:41:00 ◼   ► But I think there's a question of degree there.

00:41:05 ◼   ► If it can be really, really quiet under most workloads

00:41:11 ◼   ► and only give up like 1% of its total throughput on the CPU,

00:41:15 ◼   ► I would probably take that trade off,

00:41:17 ◼   ► and I bet most buyers would too.

00:41:20 ◼   ► - Yeah, especially since you would have to run the fan,

00:41:23 ◼   ► not just like, oh, like during those blips,

00:41:26 ◼   ► just make the fan a little faster.

00:41:27 ◼   ► I think to get rid of those blips,

00:41:29 ◼   ► you would have to substantially,

00:41:32 ◼   ► run the fan substantially faster,

00:41:34 ◼   ► much earlier than you think you would have to.

00:41:36 ◼   ► It's not as if you can react to them in real time

00:41:38 ◼   ► with fan speed, and so it is actually a trade off.

00:41:41 ◼   ► And so basically, like these things look like

00:41:42 ◼   ► they're pretty much never throttling

00:41:45 ◼   ► in exchange for very quiet performance.

00:41:47 ◼   ► And you could go to absolutely never throttling

00:41:50 ◼   ► in exchange for iMac 5K sounds, performance, you know?

00:41:54 ◼   ► - Right, but I don't want that.

00:41:56 ◼   ► I gladly would give up that last little tiny percentage

00:41:59 ◼   ► of performance for a computer that is quiet

00:42:02 ◼   ► pretty much all the time than one that does

00:42:06 ◼   ► what the 5K iMac does, which is just ramp up, ramp down,

00:42:08 ◼   ► ramp up, ramp down.

00:42:09 ◼   ► You can always hear those things moving.

00:42:11 ◼   ► - Or we could just wait for the Mac Pro and have it all.

00:42:13 ◼   ► - We hope, right?

00:42:15 ◼   ► - That's right.

00:42:16 ◼   ► Until it's introduced, it's the fantasy computer

00:42:18 ◼   ► that solves everyone's problems.

00:42:19 ◼   ► - Exactly.

00:42:20 ◼   ► So Maxuck in the chat asks,

00:42:22 ◼   ► "Is it possible that I would ever stick with the iMac Pro

00:42:24 ◼   ► "and not buy the Mac Pro?"

00:42:27 ◼   ► That is possible.

00:42:28 ◼   ► Again, we won't really know until the Mac Pro is released,

00:42:32 ◼   ► you know, what it is.

00:42:34 ◼   ► - Correction, it is impossible that you won't buy

00:42:37 ◼   ► the Mac Pro, but it is possible that you buy it,

00:42:39 ◼   ► return it, and then go back to the iMac Pro.

00:42:41 ◼   ► - Truth.

00:42:41 ◼   ► - It's possible, but you know, 'cause I do,

00:42:44 ◼   ► I am going to probably be incredibly satisfied

00:42:47 ◼   ► with the iMac Pro in the meantime.

00:42:49 ◼   ► And as I mentioned on previous shows,

00:42:52 ◼   ► if the iMac Pro was it,

00:42:54 ◼   ► if there was not going to be another Mac Pro,

00:42:57 ◼   ► if this was the only Mac Pro Xeon workstation

00:43:01 ◼   ► that Apple's going to make,

00:43:03 ◼   ► I would totally just buy it and be fine with it.

00:43:05 ◼   ► And I would occasionally complain,

00:43:07 ◼   ► if I had to bring in my computer for service

00:43:08 ◼   ► and I lost my whole computer

00:43:09 ◼   ► 'cause the monitor had a dead pixel,

00:43:11 ◼   ► that would make me upset.

00:43:12 ◼   ► But otherwise, I've been using a 5K iMac

00:43:16 ◼   ► for the last three years

00:43:17 ◼   ► because I decided the overall package of the iMac

00:43:20 ◼   ► with the wonderful Retina screen and everything

00:43:22 ◼   ► was better for me than the 2013 Mac Pro.

00:43:25 ◼   ► I might make that same decision again

00:43:27 ◼   ► after seeing the Mac Pro, the next Mac Pro,

00:43:30 ◼   ► but the next Mac Pro will also be apparently,

00:43:34 ◼   ► allegedly coming with a new Pro display.

00:43:37 ◼   ► I assume it's going to be at least a 5K Retina display,

00:43:41 ◼   ► if not bigger.

00:43:42 ◼   ► Man, it'd be awesome if it was 8K or something,

00:43:44 ◼   ► but I don't know how likely that is

00:43:47 ◼   ► at this year, but if they release something

00:43:50 ◼   ► that ends up being less compelling to me

00:43:55 ◼   ► than the iMac Pro, then I'll get an iMac Pro for myself

00:43:59 ◼   ► and give this one to TIFF.

00:44:00 ◼   ► Maybe I'll wait 'til the next generation, probably not.

00:44:03 ◼   ► (laughs)

00:44:04 ◼   ► Because, and I've seen a lot of people,

00:44:07 ◼   ► a lot of Mac commentators and everything saying,

00:44:11 ◼   ► "We really have to see how often Apple updates the iMac Pro.

00:44:15 ◼   ► expecting like annual updates and don't expect that because even if Apple is

00:44:21 ◼   ► serving this line responsibly and as well as they possibly can they probably

00:44:27 ◼   ► won't update the iMac Pro in a meaningful way annually. They might and

00:44:32 ◼   ► they should update the GPUs annually. I don't know that they will but the most

00:44:38 ◼   ► responsible way to update this machine would be to update the GPUs basically

00:44:42 ◼   ► whenever there's a new major update to the GPU line that it uses, which is usually about

00:44:48 ◼   ► once a year. But the Xeon CPUs that it uses are not updated every year. Usually Intel

00:44:55 ◼   ► gives meaningful Xeon updates about every 18 months to two years. So that's the kind

00:45:00 ◼   ► of upgrade cycle that I hope to see here. Anything less than that I think is optimistic,

00:45:05 ◼   ► and they would only be able to update the GPU, maybe the SSD, but probably not any meaningful

00:45:11 ◼   ► CPU updates more often than that.

00:45:14 ◼   ► - So, so far so good, not a lot to say,

00:45:16 ◼   ► 'cause it's brand new.

00:45:18 ◼   ► - An hour in.

00:45:19 ◼   ► - Yep, I was gonna be quick, remember,

00:45:21 ◼   ► are you keeping your peripherals?

00:45:24 ◼   ► - The black ones?

00:45:25 ◼   ► Hell yeah.

00:45:27 ◼   ► - But you won't be using the keyboard, I assume,

00:45:29 ◼   ► is that correct?

00:45:30 ◼   ► - No, that's true, the keyboard I am not planning on using

00:45:32 ◼   ► because I use a natural keyboard,

00:45:35 ◼   ► which is all, the keyboard I use is the Microsoft

00:45:37 ◼   ► Sculpt ergonomic keyboard, which happens to also be

00:45:39 ◼   ► black and gray, so that works out well.

00:45:42 ◼   ► It looks awesome with all the new stuff.

00:45:45 ◼   ► So I do have the black trackpad and the black mouse.

00:45:48 ◼   ► The black keyboard is still in the box,

00:45:50 ◼   ► and I don't know, I'll probably just send it

00:45:52 ◼   ► to John or somebody, I don't know, do you want it?

00:45:54 ◼   ► - I'm already buying Snails.

00:45:56 ◼   ► - Oh, I was gonna give you mine for free.

00:45:57 ◼   ► - If you wanna send me it for free, I'll also take it.

00:45:59 ◼   ► (laughing)

00:46:00 ◼   ► - Casey, do you want?

00:46:01 ◼   ► - Well, I am acquiring Underscores, coincidentally.

00:46:04 ◼   ► Underscore got a--

00:46:05 ◼   ► (laughing)

00:46:06 ◼   ► - The, what do you call it, the black market,

00:46:08 ◼   ► to speak for iMac Pro peripherals is alive and kicking.

00:46:12 ◼   ► The space gray marking.

00:46:14 ◼   ► Coincidentally underscore, this is the thing, like, if I were to order myself in iMac Pro,

00:46:21 ◼   ► whether or not I wanted any of the peripherals, there is absolutely no chance I would not

00:46:26 ◼   ► order all three of them.

00:46:28 ◼   ► I would order the keyboard, which of course you have to get, I would order a mouse and

00:46:31 ◼   ► I would order a trackpad because why would you not get the whole set?

00:46:36 ◼   ► And Underscore did that, and even though I am not a trackpad kind of guy, when he told

00:46:40 ◼   ► me he had the whole set, I was like, "Alright, just send me an Apple Pay Cash request for

00:46:45 ◼   ► whatever the sum total of that would be.

00:46:47 ◼   ► I don't even want to know what it is until you ask for it, so I can't back down, because

00:46:50 ◼   ► I'm sure it'll be like $300.

00:46:51 ◼   ► But I'll just take the whole set off your hands, and I'm very looking forward to it."

00:46:56 ◼   ► So about, I don't know, a year or two ago, I added a trackpad to my setup.

00:47:00 ◼   ► So I have left hand, left trackpad, right mouse.

00:47:03 ◼   ► And that's what I plan to do.

00:47:04 ◼   ► Keyboard in the middle.

00:47:05 ◼   ► really nice.

00:47:06 ◼   ► Like, it takes only days to get into the habit

00:47:09 ◼   ► of using the trackpad with your left hand.

00:47:11 ◼   ► It does like, so I'm not using it all the time,

00:47:13 ◼   ► I'm basically just switching off.

00:47:15 ◼   ► Like, if I need to reach something with my left hand,

00:47:17 ◼   ► I can, if I need to move with the mouse,

00:47:20 ◼   ► use my right hand, I can do that too.

00:47:21 ◼   ► Like, it's surprisingly easy to get accustomed

00:47:24 ◼   ► to regularly using more than one input device,

00:47:28 ◼   ► even with your non-dominant hand.

00:47:30 ◼   ► It's really nice.

00:47:31 ◼   ► And now, like, I'll occasionally have,

00:47:34 ◼   ► like I'll need the desk space for something else

00:47:36 ◼   ► for like an hour and I'll move the trackpad

00:47:38 ◼   ► like up to a different spot where it is

00:47:40 ◼   ► so it's not accessible to me.

00:47:42 ◼   ► And I miss it immediately.

00:47:44 ◼   ► Like I keep putting my hand where it is

00:47:46 ◼   ► expecting to use it and it's not there.

00:47:49 ◼   ► So it really has worked itself into my setup

00:47:52 ◼   ► very, very fully and it's really nice

00:47:55 ◼   ► and I miss it when it isn't there.

00:47:57 ◼   ► - Yeah, I don't know how much I'll use it in general

00:47:59 ◼   ► but in the brief amount of time I've used Final Cut Pro,

00:48:03 ◼   ► Oh boy, do I wish I had a trackpad for a lot of that.

00:48:06 ◼   ► So, I am a devout mouse user in general,

00:48:09 ◼   ► but for that, I am very looking forward

00:48:13 ◼   ► to having the trackpad.

00:48:14 ◼   ► - Yeah, it's very nice for the two finger scrolling gestures

00:48:16 ◼   ► in both directions.

00:48:17 ◼   ► - Yep.

00:48:18 ◼   ► - And so like, I started it for Logic

00:48:21 ◼   ► when editing podcasts so I could scroll the timeline

00:48:24 ◼   ► left and right.

00:48:25 ◼   ► That's probably why you want it for Final Cut Pro, right?

00:48:27 ◼   ► - Yep, yep.

00:48:28 ◼   ► - Now, I'm good enough, I'm precise enough with it

00:48:31 ◼   ► that I can pretty much use regular mouse functionality

00:48:35 ◼   ► at pretty much full speed with my left hand

00:48:39 ◼   ► whenever I need to, and it's wonderful.

00:48:42 ◼   ► - That's pretty cool.

00:48:43 ◼   ► Yeah, I wanna use it for scrolling laterally

00:48:45 ◼   ► in the timeline and also zooming, you know, so making,

00:48:48 ◼   ► you know, I don't know what the technical terms are

00:48:51 ◼   ► for this, but you know, rather than showing

00:48:53 ◼   ► the entire timeline of a maybe 15-minute video

00:48:56 ◼   ► all on screen at once, maybe I wanna blow it up

00:48:58 ◼   ► that there's only 10 seconds of video being shown

00:49:01 ◼   ► on my entire 5K screen.

00:49:03 ◼   ► And I'm doing that deliberately

00:49:04 ◼   ► because I'm like really trying to tweak timing on something.

00:49:06 ◼   ► So I'm looking forward to pinch to zoom and all that.

00:49:08 ◼   ► So yeah, I'm gonna be freaking broke,

00:49:10 ◼   ► but I am looking forward to these stealthy,

00:49:13 ◼   ► space gray, don't call it black, black peripherals.

00:49:18 ◼   ► - You guys both need 27 inch iPad Pros.

00:49:20 ◼   ► So you can do multi-touch editing

00:49:22 ◼   ► of your podcasts and video.

00:49:23 ◼   ► - True.

00:49:24 ◼   ► - Like you're like one degree,

00:49:25 ◼   ► like you're doing this indirect thing

00:49:26 ◼   ► where you're swiping around while staring at a screen.

00:49:29 ◼   ► It's like, just bring them together, you're so close.

00:49:31 ◼   ► - Yeah. - Use all your fingers

00:49:32 ◼   ► and your hands and your elbows.

00:49:33 ◼   ► - We are sponsored this week by Casper.

00:49:37 ◼   ► Start sleeping ahead of the curve

00:49:39 ◼   ► and get $50 toward select mattresses

00:49:41 ◼   ► by visiting casper.com/ATP

00:49:44 ◼   ► and using code ATP at checkout.

00:49:46 ◼   ► Terms and conditions apply.

00:49:48 ◼   ► Casper mattresses are wonderful.

00:49:51 ◼   ► They are perfectly designed for humans,

00:49:53 ◼   ► engineered to soothe and cradle your natural geometry.

00:49:57 ◼   ► You spent a third of your life sleeping,

00:49:59 ◼   ► so you should be comfortable,

00:50:00 ◼   ► and Casper helps you do that

00:50:02 ◼   ► by providing all the support the human body needs

00:50:04 ◼   ► in all the right places, 'cause it's fairly complex.

00:50:07 ◼   ► You can't just have the same foam layering

00:50:10 ◼   ► or the same density at all parts of the mattress,

00:50:12 ◼   ► because our bodies have these shapes

00:50:14 ◼   ► that need to be supported in different ways,

00:50:15 ◼   ► have different weights to them,

00:50:17 ◼   ► and Casper has engineered this.

00:50:19 ◼   ► They've designed and developed

00:50:20 ◼   ► and engineered these mattresses,

00:50:22 ◼   ► All in the US, by the way, with these wonderful,

00:50:24 ◼   ► breathable designs to help you sleep cool

00:50:27 ◼   ► and regulate your temperature throughout the night

00:50:29 ◼   ► with multiple supportive memory foams

00:50:31 ◼   ► for a quality sleep with just the right sink

00:50:34 ◼   ► and just the right bounce.

00:50:36 ◼   ► With over 20,000 reviews and an average of 4.8 stars

00:50:39 ◼   ► across Casper's online store, Amazon, and Google,

00:50:42 ◼   ► it is very quickly becoming the internet's favorite mattress.

00:50:45 ◼   ► They now have three mattress models,

00:50:46 ◼   ► the original Casper, the Wave, and the Essential,

00:50:49 ◼   ► and they also have a wide variety of other products

00:50:51 ◼   ► to ensure a great sleep like sheets and pillows.

00:50:54 ◼   ► And all this is available at very affordable prices

00:50:57 ◼   ► because they cut out the middleman

00:50:58 ◼   ► and sell directly to you.

00:51:00 ◼   ► All this is very easy, it kinda sounds crazy

00:51:03 ◼   ► to buy mattresses online,

00:51:04 ◼   ► but they have wonderful easy shipping,

00:51:06 ◼   ► it comes in a box, you can even get it upstairs

00:51:08 ◼   ► if you need to, then it expands out and it's wonderful,

00:51:11 ◼   ► and they have no hassle returns after 100 nights

00:51:15 ◼   ► if you are not completely satisfied.

00:51:17 ◼   ► Free shipping and returns in the US and Canada,

00:51:19 ◼   ► they couldn't be easier.

00:51:20 ◼   ► Start sleeping ahead of the curve with the Casper mattress.

00:51:23 ◼   ► Get $50 towards select mattress models

00:51:26 ◼   ► by visiting casper.com/atp and using code ATP at checkout.

00:51:31 ◼   ► That's casper.com/atp with code ATP

00:51:34 ◼   ► for $50 towards select mattresses.

00:51:36 ◼   ► Terms and conditions apply.

00:51:38 ◼   ► Thank you so much to Casper for sponsoring our show.

00:51:40 ◼   ► - All right, anything else about the iMac Pro

00:51:45 ◼   ► before we move on to this Intel thing?

00:51:47 ◼   ► - I don't know, do we have time for the Intel thing?

00:51:48 ◼   ► I think we have to get to Ask ATP

00:51:50 ◼   ► and we can push the Intel thing off till next week,

00:51:52 ◼   ► by which time planes will be falling from the sky.

00:51:55 ◼   ► (laughs)

00:51:56 ◼   ► - Come on.

00:51:57 ◼   ► We have time for at least the opening of the Intel thing.

00:51:59 ◼   ► We'll talk about it for a little bit,

00:52:01 ◼   ► and then I think Ask ATP will be quick, famous last words.

00:52:04 ◼   ► - Well, the thing is, like the Intel thing, it's huge news,

00:52:08 ◼   ► but we don't really know a lot about it yet.

00:52:10 ◼   ► So we need to talk about it,

00:52:12 ◼   ► but we're gonna be talking about it more next week.

00:52:14 ◼   ► - I know enough about it to give a reasonable summary.

00:52:17 ◼   ► - All right, so let me take a stab

00:52:19 ◼   ► as Chief Summarizer-in-Chief, and John, whenever you get fed up with it, just interrupt me

00:52:23 ◼   ► and we'll move along.

00:52:25 ◼   ► It appears that on Intel CPUs for sure, but probably on other CPU architectures, there

00:52:34 ◼   ► is an exploit wherein—and I have to back up just a half-step—you know, as the CPU

00:52:40 ◼   ► is executing instructions, as it's executing code, one of the things that modern CPUs will

00:52:45 ◼   ► do is they'll start looking ahead to what code is coming, and they'll start actually

00:52:49 ◼   ► pre-executing some of that code in some circumstances to get kind of ahead of the game.

00:52:55 ◼   ► This is like when you grab a book and jump to the last page and see, you know, who killed

00:53:00 ◼   ► who or, you know, who survives at the end or whatever the case may be.

00:53:03 ◼   ► So these CPUs will look a little bit ahead and start to execute stuff that they think

00:53:09 ◼   ► is coming.

00:53:11 ◼   ► And that's true of most modern CPU architectures,

00:53:15 ◼   ► but what's a little bit different between CPU architectures

00:53:17 ◼   ► is that some of them, or I understand,

00:53:20 ◼   ► that some of them have some protections

00:53:22 ◼   ► around what's considered like operating system

00:53:25 ◼   ► or really kernel code and what's considered

00:53:28 ◼   ► kind of quote unquote the user's code.

00:53:32 ◼   ► - You're straying, you're straying.

00:53:33 ◼   ► - All right, sorry.

00:53:34 ◼   ► - Warning, warning.

00:53:35 ◼   ► - All right, here we go.

00:53:35 ◼   ► - Abort, abort.

00:53:36 ◼   ► - Just cut in now, cut in now.

00:53:38 ◼   ► - All right, so everything you said up to that point

00:53:41 ◼   ► is good about speculative execution. And again, I'm not an expert on this either, and it just came

00:53:46 ◼   ► out today, and I'm reading about it after work, so forgive me if I'm getting some parts of it wrong.

00:53:50 ◼   ► But the exploit, as it has been explained to me in various articles and over Twitter,

00:53:59 ◼   ► so take that for what it is, is that during the speculative execution of instructions,

00:54:09 ◼   ► it will start executing instructions before it's sure whether they are valid things to do. And

00:54:13 ◼   ► depending on what mode your CPU is in and what address it's trying to reference and so on and

00:54:18 ◼   ► so forth, there are certain things that you could be trying to do that are invalid. Reading from a

00:54:22 ◼   ► memory location that you're not allowed to read from because it's not mapped into your process

00:54:26 ◼   ► or something like that or whatever. Or reading from kernel memory, which is a distinction that

00:54:32 ◼   ► exists in pretty much all modern processors of having memory space that belongs to the kernel,

00:54:38 ◼   ► that can't be read by usual processes.

00:54:41 ◼   ► And it will do the speculative execution

00:54:43 ◼   ► and then eventually figure out, oh, you're

00:54:46 ◼   ► not supposed to do that.

00:54:47 ◼   ► I wasn't supposed to do it.

00:54:48 ◼   ► And it'll be like, oh, well, OK, I'm

00:54:49 ◼   ► not allowing that to proceed.

00:54:50 ◼   ► You're not going to actually be able to do whatever

00:54:52 ◼   ► it is you were trying to do, even though I kind of sort

00:54:55 ◼   ► of already executed part of it.

00:54:56 ◼   ► But never mind, let me undo that.

00:54:59 ◼   ► And so it undoes it, right?

00:55:00 ◼   ► But given how far ahead of itself

00:55:04 ◼   ► it does speculative execution, not just like one or two

00:55:06 ◼   ► things ahead but several things ahead.

00:55:10 ◼   ► If the instructions are dependent on each other, so I'm going to speculatively execute

00:55:15 ◼   ► this operation here, and then I'm also going to speculatively execute the next one, which

00:55:21 ◼   ► depends on the result of the first one, and then it turns out like you weren't allowed

00:55:24 ◼   ► to do that one and so it rolls everything back, right?

00:55:28 ◼   ► So it's not like you actually successfully read something you weren't supposed to.

00:55:32 ◼   ► But the side effects of doing those operations can leave a bunch of crap in the cache, and

00:55:39 ◼   ► you're like, "Well, so what?

00:55:40 ◼   ► The stuff that's in the cache is not the stuff that I wasn't supposed to read, because that

00:55:43 ◼   ► never got there, because I wasn't even supposed to do that read."

00:55:45 ◼   ► But because the thing you weren't supposed to read influenced what got put in the cache

00:55:54 ◼   ► from the other instruction, you can back solve and figure out what the place you weren't

00:56:01 ◼   ► supposed to read from was based on how the result of that was used to get, I know this

00:56:06 ◼   ► is very confusing, like it's basically like looking at the side effects of an operation

00:56:11 ◼   ► that you weren't supposed to do and that was actually rolled back just to figure out what

00:56:16 ◼   ► that number must have been, what that memory address must have been, and in that way you

00:56:21 ◼   ► can figure out, someone was saying even like just a bit at a time, like you can figure

00:56:25 ◼   ► out what the most significant or least significant bit of that address used to be. You can slowly

00:56:29 ◼   ► but surely reveal the entire contents of physical memory, much of which belongs to processes

00:56:35 ◼   ► that you're never supposed to see.

00:56:37 ◼   ► And that is a pretty terrible bug, which basically means all the protections of saying, "You

00:56:41 ◼   ► just can't read arbitrary memory on the system," because all sorts of stuff is in memory, like

00:56:45 ◼   ► encryption keys and the password that someone just typed and all sorts of stuff is in memory,

00:56:51 ◼   ► like in physical memory on your computer.

00:56:53 ◼   ► All the protections of the operating system and the CPU and all that stuff are supposed

00:56:57 ◼   ► prevent you from reading memory from processes that belong to the operating system itself,

00:57:03 ◼   ► memory from processes that belong to other users. For example, if you're using AWS, Amazon

00:57:08 ◼   ► Web Services, or other shared hosting where your stuff may be running on the same real

00:57:15 ◼   ► or virtual CPU as other people's stuff, and you have no visibility to them, you don't

00:57:19 ◼   ► see their other processes or whatever, but physically speaking, the machine that you're

00:57:22 ◼   ► running on in RAM next to your information is somebody else's information. If you can

00:57:27 ◼   ► dump all RAM by using this X point to slowly reveal every single address in memory, that's

00:57:33 ◼   ► really, really, really bad. And this is not just like, oh, someone, like there's a bug

00:57:39 ◼   ► in this processor, like the fdiv bug or something, and they'll just fix it by making a new version

00:57:42 ◼   ► of the processor. This is more or less inherent in the way speculative execution works. Like

00:57:47 ◼   ► their whole cleanup process actually does the cleanup, but they didn't realize that

00:57:52 ◼   ► And the thing that was valid to do, if it was influenced by the thing that was invalid

00:57:56 ◼   ► to do, can leave information hanging around such that you can back solve and figure out

00:58:01 ◼   ► what the invalid information you were supposed to be getting out was.

00:58:04 ◼   ► That's my attempt to explain this and why it's not just like, "Oh, there's a mistake

00:58:07 ◼   ► in Intel CPU and they'll fix it."

00:58:09 ◼   ► It's why it affects all Intel CPUs for the past many, many years, not just one thing.

00:58:15 ◼   ► It's why it also affects ARM CPUs.

00:58:17 ◼   ► ARM came out and said, "Yes, this affects our CPUs as well."

00:58:20 ◼   ► AMD says it doesn't affect their CPUs because they do speculative execution in a different

00:58:25 ◼   ► way.

00:58:26 ◼   ► And it's not like this is unfixable, but you will have to, to fix it in hardware, you'll

00:58:31 ◼   ► have to change the way the hardware works.

00:58:33 ◼   ► And the general trade-off they're making here is the hardware wants to be fast, so it's

00:58:37 ◼   ► like, "We'll clean up after ourselves and try to be secure, but if we don't have to

00:58:42 ◼   ► clean something up like that information that was pulled into that cache line or whatever,

00:58:46 ◼   ► oh, you don't need to clean it up because that was from the valid operation, so that

00:58:48 ◼   ► one is fine."

00:58:49 ◼   ► It's garbage data, but it's not insecure.

00:58:51 ◼   ► It's not like we're leaving information in there

00:58:52 ◼   ► that will let someone get something.

00:58:54 ◼   ► So just leave it there because it would be more costly

00:58:57 ◼   ► to erase that, right?

00:58:58 ◼   ► And they're gonna have to make different trade-offs

00:59:03 ◼   ► to get rid of this, 'cause it's not like something

00:59:05 ◼   ► they can, you know, first of all, they can't fix it

00:59:07 ◼   ► in the CPUs that are already out there,

00:59:08 ◼   ► 'cause it's out there, it's out there.

00:59:10 ◼   ► And they can fix it in a future architecture

00:59:12 ◼   ► so they take it into account,

00:59:13 ◼   ► but it's going to make a big difference.

00:59:15 ◼   ► Now the fix for it in software

00:59:16 ◼   ► that people are slowly rolling out

00:59:18 ◼   ► and that actually is in a Mac OS 10.3 point,

00:59:22 ◼   ► what is it, 10.13.2?

00:59:23 ◼   ► - Yep.

00:59:24 ◼   ► - Has this fix according to one random person on Twitter.

00:59:27 ◼   ► The fix they're doing is basically to say that

00:59:30 ◼   ► when you're in a process, don't map the kernel

00:59:32 ◼   ► into the same address space as the process.

00:59:36 ◼   ► Put it in an entirely different address space entirely,

00:59:39 ◼   ► like entirely separate address space

00:59:41 ◼   ► so you can't use this exploit

00:59:42 ◼   ► so that you would have to switch modes

00:59:44 ◼   ► and go into, switch memory images

00:59:46 ◼   ► and go into an entirely different address space,

00:59:47 ◼   ► which is slower, and people are doing benchmarks and saying,

00:59:50 ◼   ► okay, with this software fix where we shove the kernel

00:59:52 ◼   ► into an entirely different place,

00:59:53 ◼   ► instead of allowing the kernel to virtually live

00:59:55 ◼   ► inside the same virtual address space

00:59:57 ◼   ► as every user land processor,

00:59:58 ◼   ► which makes it really easy to,

01:00:00 ◼   ► when you switch into kernel mode to do kernel stuff,

01:00:02 ◼   ► then user stuff, right?

01:00:03 ◼   ► It makes it slower to do that

01:00:04 ◼   ► if you more widely separate them,

01:00:06 ◼   ► but it fixes this vulnerability.

01:00:08 ◼   ► So everyone now is benchmarking how much slower is it?

01:00:10 ◼   ► Is it 30% slower?

01:00:12 ◼   ► Is it 1% slower?

01:00:14 ◼   ► Is it only slower when you do heavy IO?

01:00:15 ◼   ► Is it only slower in our synthetic benchmarks,

01:00:18 ◼   ► but it's just the same and normal?

01:00:19 ◼   ► Does the frame rate of my game go down?

01:00:21 ◼   ► So now everyone's applying patches to their Linux kernels

01:00:25 ◼   ► or trying the Windows patch version or whatever,

01:00:27 ◼   ► and trying to see just how bad the performance hit is.

01:00:32 ◼   ► But this is a really critical, serious bug

01:00:35 ◼   ► because the only way it's really, truly gonna be fixed

01:00:39 ◼   ► at the hardware level is for them

01:00:41 ◼   ► to make different design decisions,

01:00:44 ◼   ► different fundamental design decisions about speculative execution in their CPUs. And they're

01:00:51 ◼   ► going to have to figure out a way to do that, so it's fast too. And everyone's freaking

01:00:54 ◼   ► out about it, especially for things like shared hosting, because you can exploit this from

01:01:00 ◼   ► anywhere, basically. If you can just get code to run on the CPU at a low enough level, then

01:01:08 ◼   ► you can just dump the entire contents of memory of that machine. No matter your processes,

01:01:12 ◼   ► people's processes, the operating system, everything. It's like heart bleed times a

01:01:16 ◼   ► thousand. Like, not just get information about the process that you're exploiting, get information

01:01:22 ◼   ► about anything else that is running on the hardware, which is slightly terrifying. So,

01:01:27 ◼   ► understandably, everyone who does shared hosting and everyone who has an operating system is

01:01:31 ◼   ► all busily trying to update their kernels to work around this thing at the cost of a

01:01:36 ◼   ► potential speed hit and then dealing with the fallout of that speed hit. But yeah, so

01:01:41 ◼   ► So Apple's already got the fix in the latest version of High Sierra.

01:01:46 ◼   ► There was some tweet about the upcoming version of High Sierra having something else in regard

01:01:51 ◼   ► to this.

01:01:52 ◼   ► Someone in the chat just pointed out that there are JavaScript proof of concepts.

01:01:55 ◼   ► Like when I say you can get something to run on the CPU, it doesn't matter what it is.

01:02:01 ◼   ► Imagine running JavaScript that accidentally dumps the entire contents of your machine's

01:02:07 ◼   ► RAM.

01:02:08 ◼   ► That's bad.

01:02:09 ◼   ► That's pretty bad.

01:02:11 ◼   ► - It's terrifying.

01:02:11 ◼   ► - Again, this is like one of those exploits

01:02:13 ◼   ► like you'd see in a movie and like, that's so stupid.

01:02:15 ◼   ► You can't run something on a webpage

01:02:16 ◼   ► that dumps the contents of RAM.

01:02:18 ◼   ► There's a little thing called memory protection dummy.

01:02:20 ◼   ► - Now wait a minute.

01:02:21 ◼   ► So I think you might be,

01:02:24 ◼   ► either you're misunderstanding these or I am, basically.

01:02:27 ◼   ► So there are two big bugs or two big exploits

01:02:31 ◼   ► that we're talking about here.

01:02:34 ◼   ► The names I think finally came out tonight.

01:02:36 ◼   ► One of them is called Meltdown

01:02:37 ◼   ► and one of them is called Spectre.

01:02:39 ◼   ► And the Meltdown is the one that basically allows

01:02:44 ◼   ► the cache priming effects that you were talking about earlier

01:02:49 ◼   ► like when you try to access an invalid address,

01:02:52 ◼   ► you can basically run timing,

01:02:56 ◼   ► like run very precise timing benchmarks

01:02:58 ◼   ► to figure out like is something cached or not,

01:03:01 ◼   ► and then you can figure out based on that

01:03:03 ◼   ► like roughly what addresses certain kernel functions map to

01:03:09 ◼   ► that have been randomized using address space

01:03:11 ◼   ► layout randomization.

01:03:13 ◼   ► And so that's how you can,

01:03:15 ◼   ► but you can't read other processes' memory space.

01:03:19 ◼   ► You can only read the kernel memory

01:03:21 ◼   ► because the kernel memory has been shadow mapped

01:03:23 ◼   ► into your process space.

01:03:25 ◼   ► And by the way, I am so sorry to anybody listening

01:03:28 ◼   ► who this is over your head.

01:03:30 ◼   ► We can try to explain it, but it would take a long time

01:03:33 ◼   ► 'cause we'd have to explain virtual memory

01:03:35 ◼   ► and addresses of functions and interrupts

01:03:39 ◼   ► and everything and it would be kind of a problem.

01:03:41 ◼   ► So I'm so sorry.

01:03:43 ◼   ► Feel free to skip the rest of the chapter if you want.

01:03:45 ◼   ► But anyway, so that's the Meltdown attack.

01:03:48 ◼   ► And then the Spectre attack is not about revealing

01:03:53 ◼   ► kernel addresses of things.

01:03:55 ◼   ► It is about actually dumping kernel memory

01:03:58 ◼   ► and being able to access the contents of kernel memory.

01:04:01 ◼   ► But it's only kernel memory not,

01:04:05 ◼   ► because it is mapped into your process,

01:04:08 ◼   ► not the memory of other processes on the system.

01:04:11 ◼   ► - I'm pretty sure, from the explanation that I read,

01:04:13 ◼   ► I'm pretty sure that either a combination

01:04:15 ◼   ► of these exploits or one or the other

01:04:18 ◼   ► can actually eventually iteratively go through

01:04:20 ◼   ► the entire address space and eventually get

01:04:24 ◼   ► all the contents of RAM by, presumably by

01:04:27 ◼   ► walking your whole giant address space

01:04:30 ◼   ► and causing it to page in to physical space

01:04:32 ◼   ► every one of the things that you read,

01:04:34 ◼   ► pushing out everything else and turning RAM.

01:04:37 ◼   ► It may just be kernel space.

01:04:38 ◼   ► Even if it's just kernel space, that's still pretty bad, because tons of interesting information

01:04:42 ◼   ► isn't there.

01:04:43 ◼   ► But it seemed to me that the reason people are scared about shared hosting is that—and

01:04:48 ◼   ► how you could read processes in a different virtual machine is one of the other proof

01:04:53 ◼   ► of concepts in the Google paper.

01:04:54 ◼   ► That's bad.

01:04:55 ◼   ► Yeah.

01:04:56 ◼   ► That's not just looking at the kernel space that's in your process, but that you're

01:05:00 ◼   ► able to look into the memory inside other virtual machines.

01:05:02 ◼   ► Makes me think that you can actually get just dumps of the contents of RAM.

01:05:07 ◼   ► But these are, you know, so we'll put a bunch of links in the show notes, all of which I

01:05:10 ◼   ► have not had a chance to read, because again, this is all coming out just, you know, today,

01:05:13 ◼   ► now as I'm leaving work, these things are coming out.

01:05:16 ◼   ► There are abstracts, there are white papers, there's proof of concepts on these things.

01:05:21 ◼   ► There's a nice Google site.

01:05:23 ◼   ► Google has one of the projects that they call Google Project Zero, found either one or both

01:05:29 ◼   ► of these exploits originally.

01:05:33 ◼   ► Google also has a good page that explains in sort of much simpler language than we're

01:05:37 ◼   ► doing here, a very broad vague kind of thing about speculative execution exists.

01:05:42 ◼   ► It's supposed to clean up, but some stuff gets left around.

01:05:44 ◼   ► Like that's the, you know, that's all you really need to know to understand the nature

01:05:47 ◼   ► of this thing.

01:05:48 ◼   ► We're trying to get down to the nitty gritty details.

01:05:50 ◼   ► But at Google also lists, okay, here's all the Google, the things Google has.

01:05:54 ◼   ► We have all of our different services and Gmail and Google Cloud things and, you know,

01:05:59 ◼   ► And they go through them all one by one and say,

01:06:02 ◼   ► here's how this is with respect to this bug.

01:06:04 ◼   ► We've patched these, we haven't patched these,

01:06:06 ◼   ► this will be patched soon, which is nice.

01:06:08 ◼   ► I was looking for a similar page from Apple,

01:06:10 ◼   ► but I suppose such a thing is not forthcoming,

01:06:12 ◼   ► or maybe they'll just come out with it tomorrow or whatever.

01:06:15 ◼   ► But every person out there,

01:06:17 ◼   ► I'm assuming Amazon is doing something similar,

01:06:20 ◼   ► has to now explain to you

01:06:23 ◼   ► how many of their systems are vulnerable,

01:06:26 ◼   ► How many have had their kernel patch to fix this and what the timelines are and what you

01:06:34 ◼   ► can expect as the customer.

01:06:36 ◼   ► Like if you have an Android phone, is that vulnerable?

01:06:38 ◼   ► How old is your Android phone?

01:06:39 ◼   ► Who makes it, so on and so forth.

01:06:41 ◼   ► Because this type of vulnerability is really scary.

01:06:45 ◼   ► Basically if you cause untrusted code to run on your CPU, and again, trying to do it from

01:06:53 ◼   ► JavaScript that shows you like code, I wouldn't run untrusted code. Oh, do you load web pages?

01:06:57 ◼   ► Because you're running essentially untrusted code then. You know, bad things can happen.

01:07:03 ◼   ► Now, as for the performance issues, when I first thought this was a performance bug,

01:07:10 ◼   ► I wasn't that worried about it because I was like, well, no matter how big the performance problem

01:07:14 ◼   ► is, if you're not worried about being exploited, you know what I mean? Like I can imagine like

01:07:21 ◼   ► having a toggle switch that says switch into this mode and I'm just going to run a local

01:07:25 ◼   ► game that I trust, right?

01:07:27 ◼   ► And I'll get full performance from my game.

01:07:30 ◼   ► But the more I hear about this bug, the more I think that whatever the performance hit

01:07:35 ◼   ► is, we're all just going to have to live with it until Intel and ARM and everybody else

01:07:41 ◼   ► who is susceptible to this update, like produce new versions of their microarchitecture, produce

01:07:46 ◼   ► new chips essentially.

01:07:49 ◼   ► There's nothing you can do about the iMac Pro that Marker got now.

01:07:52 ◼   ► He's just going to have to run it in the slow mode to be protected from this.

01:07:55 ◼   ► I mean, he already is, basically, because he's got the latest High Sierra.

01:07:58 ◼   ► So guess what?

01:07:59 ◼   ► You're already doing it.

01:08:00 ◼   ► But there's nothing you can do until you wait for Intel to come out with a new chip that

01:08:04 ◼   ► is not susceptible to this, because it's not—like I said, it's not a bug or a problem or someone

01:08:09 ◼   ► made a boo-boo and they meant it to work one way and it works the other way.

01:08:12 ◼   ► This is working as designed.

01:08:13 ◼   ► It's just an attack that they didn't account for.

01:08:16 ◼   ► And so there's really nothing to do except go back to the drawing board and say, "How

01:08:20 ◼   ► can we design a CPU that is fast but that is not susceptible to this category of attack?"

01:08:28 ◼   ► It's not just one clever thing with one silly value that trips up something.

01:08:32 ◼   ► It's an entire category of attacks that exploit the predictability of what's supposed to be

01:08:37 ◼   ► unpredictable.

01:08:38 ◼   ► Like one part of one of these exploits, I think, was like they had completely figured

01:08:43 ◼   ► out exactly how the branch predictor works on a particular Intel CPU that helps them

01:08:49 ◼   ► predict what it's going to do in various situations.

01:08:51 ◼   ► Because you do have to manipulate the speculative execution to get it to do what you want.

01:08:55 ◼   ► You need to know which things it's going to speculatively execute in which order so that

01:09:00 ◼   ► you can set things up to produce a side effect that gives you actionable information about

01:09:04 ◼   ► the thing that wasn't supposed to be read.

01:09:07 ◼   ► Anyway, if your eyes are glazing over, we apologize.

01:09:11 ◼   ► look at all the various PDFs and web pages we will link in the show notes and then I

01:09:16 ◼   ► guess we'll come back next week and if we're all still alive and the entire world hasn't

01:09:20 ◼   ► burned down in a year 2000 style apocalypse then we can talk about it more.

01:09:26 ◼   ► In the year 2000.

01:09:28 ◼   ► It seems that it's probably going to be, it's probably going to end up being a really interesting

01:09:35 ◼   ► thing to extreme nerds and most people are probably not going to notice, at least on

01:09:40 ◼   ► On other platforms, I don't know how big the differences are

01:09:43 ◼   ► because of how they manage memory and stuff,

01:09:45 ◼   ► but on Macs, people are saying,

01:09:47 ◼   ► who know more about this than we do,

01:09:49 ◼   ► that any Mac that has the PCID feature on the Intel chip,

01:09:54 ◼   ► which pretty much all modern chips do

01:09:58 ◼   ► from the last five or so years,

01:10:00 ◼   ► it's not a major performance hit

01:10:03 ◼   ► if your Mac is one of these.

01:10:05 ◼   ► And again, the problem with us trying to talk about this

01:10:09 ◼   ► this week, it's so big of a story, probably,

01:10:13 ◼   ► that we really can't ignore it.

01:10:14 ◼   ► We really can't say, "Oh, we'll talk about this next week

01:10:16 ◼   ► "because we don't know enough about it yet."

01:10:18 ◼   ► The reality is we're gonna talk about it

01:10:19 ◼   ► probably both this week and next week

01:10:21 ◼   ► because we're gonna understand it better next week

01:10:22 ◼   ► and more news will have happened,

01:10:24 ◼   ► more of the patches will be out,

01:10:26 ◼   ► we'll be able to see some of the performance impacts

01:10:28 ◼   ► that it has.

01:10:29 ◼   ► So this is probably a really big story,

01:10:33 ◼   ► but we don't really know yet.

01:10:35 ◼   ► It could just be fixed and nobody notices

01:10:38 ◼   ► the impact and we all move on.

01:10:41 ◼   ► So we apologize, we'll do our best to cover it next week

01:10:44 ◼   ► also if there's more news to cover,

01:10:46 ◼   ► and I guess we'll just see what happens.

01:10:48 ◼   ► But if you are a user who has somehow made it through

01:10:51 ◼   ► this explanation and has not skipped the chapter yet,

01:10:54 ◼   ► and you don't really understand it,

01:10:55 ◼   ► and you're worried about it, I would say

01:10:59 ◼   ► there's probably not reason to be significantly concerned

01:11:03 ◼   ► as an end user from what we know so far.

01:11:05 ◼   ► - It would just make, you know, update,

01:11:07 ◼   ► update your operating systems, though.

01:11:09 ◼   ► - Yeah, yeah, definitely, like, install the updates

01:11:10 ◼   ► and everything, but like, it's not going to all of a sudden

01:11:13 ◼   ► make your computer super slow.

01:11:14 ◼   ► It's not gonna throttle your battery or whatever,

01:11:16 ◼   ► like, it's not gonna--

01:11:17 ◼   ► - That should be the new thing, we should tell people

01:11:18 ◼   ► that Microsoft is throttling your Windows computer

01:11:21 ◼   ► and Linux is, if only Linux was a company,

01:11:24 ◼   ► tell people it's a company, they'll believe it.

01:11:25 ◼   ► The Linux company is throttling your Linux computer

01:11:28 ◼   ► by introducing a kernel patch that can do,

01:11:30 ◼   ► produce up to a 30% speed loss.

01:11:32 ◼   ► Look at this select one benchmark in Postgres.

01:11:35 ◼   ► Look how much slower it is.

01:11:38 ◼   ► Linux is slowing down your computer.

01:11:40 ◼   ► - To make you buy a new Linux.

01:11:42 ◼   ► - Which is true, they literally are.

01:11:43 ◼   ► They're literally patching the kernel to make it slower.

01:11:46 ◼   ► - Oh my God.

01:11:47 ◼   ► (laughing)

01:11:49 ◼   ► - And so did Microsoft and so did Apple,

01:11:50 ◼   ► but no one is benchmarking Apple to,

01:11:53 ◼   ► no one is running servers on,

01:11:55 ◼   ► Mac servers and mission critical things like.

01:11:57 ◼   ► The AWS thing, I heard a bunch of people like that.

01:11:59 ◼   ► If you had a bunch of systems that were near the edge

01:12:01 ◼   ► their performance envelope on the EC2 instances that you've provisioned, and they all take

01:12:06 ◼   ► like a 2% performance hit, and suddenly you go that next few percentage, and now you kind

01:12:13 ◼   ► of have a problem and have to get some more hardware.

01:12:15 ◼   ► This can actually affect, even though it's a small effect, anything you do in cloud computing

01:12:21 ◼   ► can have a ripple effect if you're close to the edge of your capacity.

01:12:25 ◼   ► So I feel kind of bad that they have to update.

01:12:28 ◼   ► If this has a big effect on you, besides the fact that

01:12:32 ◼   ► pretty much every server and cloud instance

01:12:34 ◼   ► is probably gonna have to be rebooted in the next few days,

01:12:37 ◼   ► that's gonna be a bigger problem.

01:12:40 ◼   ► Rebooting all the servers is gonna be

01:12:41 ◼   ► a more disruptive effect of this than I think anything else.

01:12:44 ◼   ► - Well, it could disrupt your monthly bill.

01:12:47 ◼   ► - Yeah, but if it disrupts your bill by a meaningful amount,

01:12:50 ◼   ► then either this problem is way worse

01:12:53 ◼   ► than we thought it was on Linux,

01:12:54 ◼   ► or you were already running too close to capacity anyway.

01:12:59 ◼   ► - Yeah, that's the thing is it depends on workload.

01:13:01 ◼   ► So if anything's gonna have a weird workload

01:13:03 ◼   ► that happens to be like super IO intensive

01:13:05 ◼   ► or something like that, it's some sort of,

01:13:06 ◼   ► you know, like running a database server

01:13:08 ◼   ► that it just all it does all day

01:13:09 ◼   ► is just tremendous amounts of IO

01:13:11 ◼   ► and a 5% hit to your million dollar bill.

01:13:15 ◼   ► That 5% is a big increase in the amount of hardware.

01:13:19 ◼   ► Anyway, we'll see how it is.

01:13:21 ◼   ► I bet most people will just patch and update everything,

01:13:23 ◼   ► But like any one of these problems, you're like, "But not everybody's going to patch."

01:13:27 ◼   ► And so this is like Heartbleed even.

01:13:30 ◼   ► You never know, like, for how long are there going to be systems with, like, the known

01:13:36 ◼   ► vulnerable hardware spans many years, and probably many vendors, at the very least Intel

01:13:40 ◼   ► and ARM.

01:13:43 ◼   ► How long is that bum hardware going to be out there, and how much of the hardware will

01:13:46 ◼   ► never be updated to run a kernel that is patched to avoid this vulnerability?

01:13:52 ◼   ► We are sponsored this week by Squarespace.

01:13:54 ◼   ► Start building your website today at squarespace.com

01:13:57 ◼   ► and enter offer code ATP at checkout to get 10% off.

01:14:00 ◼   ► Make your next move with Squarespace.

01:14:02 ◼   ► Squarespace websites are so incredibly easy to make

01:14:06 ◼   ► because these days, no matter what project you're tackling

01:14:09 ◼   ► or what new hobby you wanna start

01:14:11 ◼   ► or what new creative work you wanna try to do,

01:14:14 ◼   ► chances are you need some kind of website for that,

01:14:16 ◼   ► as a home base or promotion

01:14:18 ◼   ► or as part of the thing itself, like a podcast.

01:14:21 ◼   ► And with Squarespace you can make websites to do so many different things.

01:14:25 ◼   ► Blogs, podcasts, portfolios, corporate sites, hobby sites, so many different things.

01:14:31 ◼   ► Even digital storefronts to sell digital or physical goods.

01:14:35 ◼   ► And all of those are incredibly easy to make.

01:14:38 ◼   ► You don't have to worry about installing software, configuring things, security updates, version

01:14:44 ◼   ► upgrades.

01:14:45 ◼   ► There are so many parts of running a website that are kind of a pain if you do it really

01:14:49 ◼   ► pretty much any other way and Squarespace takes care of all that for you.

01:14:53 ◼   ► In addition, it's all super easy to actually use. So regardless of your

01:14:57 ◼   ► skill level, whether you are a novice to a full-blown web programmer, Squarespace

01:15:04 ◼   ► can be for you. It is very, very easy to use. You never have to see or touch any

01:15:08 ◼   ► code. Everything is what you see is what you get and if you want to get super in

01:15:12 ◼   ► there and get really involved and actually see the code you can do that

01:15:15 ◼   ► too and everything in between. It is so easy to make great websites with Squarespace.

01:15:19 ◼   ► you will wonder why anybody does it any other way.

01:15:22 ◼   ► Try it there first.

01:15:24 ◼   ► Whenever you need to do something, just try it there first.

01:15:26 ◼   ► See how far you get.

01:15:27 ◼   ► I say give it like an hour, and you'll see as I have

01:15:30 ◼   ► just how awesome it is to spend a very small amount of time

01:15:33 ◼   ► on Squarespace to get a whole site done,

01:15:35 ◼   ► and then move on and do the thing

01:15:36 ◼   ► you actually made the website for.

01:15:38 ◼   ► Check it out today at squarespace.com.

01:15:40 ◼   ► Start building your website right now with a free trial,

01:15:43 ◼   ► no credit card required.

01:15:44 ◼   ► Enter offer code ATP at checkout to get 10% off.

01:15:47 ◼   ► Squarespace, make your next move.

01:15:49 ◼   ► - So Erin Bushnell writes,

01:15:54 ◼   ► "Suppose you had to give up your Mac and solely use iOS,

01:15:59 ◼   ► "but Apple agreed to add one iOS feature of your choice.

01:16:03 ◼   ► "What would it be?"

01:16:05 ◼   ► And let's start with Marco, please.

01:16:07 ◼   ► - I had a really hard time with this question

01:16:10 ◼   ► because if I were to solely use iOS,

01:16:14 ◼   ► I mean, that presumes a lot of things,

01:16:17 ◼   ► like can I still make iOS apps on iOS?

01:16:22 ◼   ► Like is there a way for me to do that?

01:16:24 ◼   ► Because otherwise my career has a problem.

01:16:28 ◼   ► - iOS has a problem if there's no way to make iOS apps.

01:16:31 ◼   ► - Well that's true.

01:16:33 ◼   ► Yeah, that's, yeah.

01:16:34 ◼   ► So like, I mean technically I suppose I should say

01:16:38 ◼   ► my answer is the iOS feature that would make me use iOS

01:16:42 ◼   ► would be Xcode, but it's so much more complicated than that.

01:16:46 ◼   ► So let's assume that somehow the tools to do

01:16:50 ◼   ► software development exist as part of this new

01:16:53 ◼   ► theoretical world where I'd be forced to use iOS.

01:16:57 ◼   ► Other than that, I think I would have to say

01:17:00 ◼   ► just larger screen devices, because,

01:17:03 ◼   ► like John's wish for the 27-inch iPad,

01:17:07 ◼   ► honestly, I don't love that concept for myself

01:17:11 ◼   ► the way I work, but whenever I try to do anything on iOS

01:17:16 ◼   ► that is non-trivial, I run into screen size limitations

01:17:20 ◼   ► really fast, and I've tried the 12.9 inch iPad briefly,

01:17:24 ◼   ► that I didn't find that that was for me,

01:17:26 ◼   ► just physically it felt weird to have like a handheld device

01:17:29 ◼   ► of that size for me, so I don't know how this would be

01:17:33 ◼   ► in practice, but just some way to get larger screen space.

01:17:37 ◼   ► I mean, if only there was a way that the physical size

01:17:40 ◼   ► of the input devices that I was interacting with

01:17:42 ◼   ► was not tied to the screen size of the device I was using.

01:17:46 ◼   ► Maybe we could have like a little proxy touch device

01:17:50 ◼   ► that I could touch that was smaller,

01:17:52 ◼   ► maybe like the size of like a large iPhone maybe.

01:17:55 ◼   ► I could touch that, maybe put on the desk.

01:17:58 ◼   ► And then I could have a nice big screen

01:18:01 ◼   ► like a little bit back, maybe vertical,

01:18:02 ◼   ► so I could see it better,

01:18:03 ◼   ► where I could arrange a whole bunch of small apps

01:18:07 ◼   ► in little rounded rectangles

01:18:09 ◼   ► that maybe I can move around and rearrange as necessary.

01:18:11 ◼   ► Something like that.

01:18:12 ◼   ► - I see what you did there.

01:18:15 ◼   ► John?

01:18:16 ◼   ► - Marco cheated, 'cause this was supposed to be

01:18:19 ◼   ► an iOS feature and he started describing hardware,

01:18:21 ◼   ► but we'll let it slide in just a time.

01:18:23 ◼   ► My iOS feature, I think I would probably go

01:18:27 ◼   ► with external storage support,

01:18:29 ◼   ► 'cause it doesn't sound fancy,

01:18:32 ◼   ► but it's a really annoying limitation

01:18:35 ◼   ► that on an iOS device, you can't just plug something in

01:18:38 ◼   ► and then like go to the files app and just like drag things around in it and like go

01:18:42 ◼   ► from dropbox and drag things like mounting storage so that it's available to every single

01:18:47 ◼   ► application sort of natively everywhere.

01:18:50 ◼   ► It's just an unnecessary thing that impedes, you know, like it seems like the battle world

01:18:57 ◼   ► where instead of there being like sort of an equivalent of the finder or whatever, that

01:19:02 ◼   ► everything is application specific.

01:19:04 ◼   ► So yeah the photos application can read from your SD card or whatever but SD card is not

01:19:07 ◼   ► universally available and mounted as a volume to any application that wants to look at it

01:19:12 ◼   ► because that's just not the way iOS rolls with the file system.

01:19:14 ◼   ► So if I had to use it all the time, I would want it to act in a slightly more Mac-like

01:19:21 ◼   ► manner with regards to storage.

01:19:24 ◼   ► For me, I was going to say like a real true-to-form terminal, which I know you can get from like

01:19:30 ◼   ► Jailbreaking and I know you can get, you know, wonderful apps like, what is it, Prompt?

01:19:35 ◼   ► Is that what I'm thinking of?

01:19:36 ◼   ► - Yeah, but that doesn't open up a local terminal,

01:19:40 ◼   ► that's for connecting to other machines.

01:19:41 ◼   ► - Exactly.

01:19:42 ◼   ► But I think if I had to choose just one,

01:19:44 ◼   ► what I'd actually choose is just being able to side load apps

01:19:48 ◼   ► without going through the App Store

01:19:50 ◼   ► and understanding their risks involved and blah, blah, blah.

01:19:52 ◼   ► But I think I would just wanna be able to side load apps

01:19:55 ◼   ► and I think that would make a tremendous difference.

01:19:56 ◼   ► - That's a good one.

01:19:57 ◼   ► - And I mean like, you know, being able to download a binary

01:20:01 ◼   ► in the same way that you can on a Mac

01:20:03 ◼   ► or, you know, the same way you would get a binary

01:20:05 ◼   ► the App Store, but have it not come from the App Store.

01:20:09 ◼   ► So that's what I would say.

01:20:11 ◼   ► Brandon Butler asks, and I'm going to slightly tweak the

01:20:14 ◼   ► verbiage here, could all of you give some specific details

01:20:18 ◼   ► about why you don't like Windows?

01:20:20 ◼   ► I'm genuinely curious.

01:20:23 ◼   ► As the potential most recent Windows user, I'll kick this

01:20:26 ◼   ► off and say, I haven't used Windows for more than a few

01:20:30 ◼   ► minutes in about a year and a half.

01:20:32 ◼   ► But the last time I used Windows, which was Windows 10, I believe, the things that bothered

01:20:39 ◼   ► me about it the most were high DPI was not pervasive throughout the entire operating

01:20:46 ◼   ► system, so some things worked at what we would call retina resolutions in the Mac world and

01:20:51 ◼   ► some would not.

01:20:53 ◼   ► They're clearly in the midst of a...

01:20:55 ◼   ► They were at the time in the midst of a transition, so there was a control panel, but then there

01:21:00 ◼   ► were like other places to tweak a lot of the same settings or sometimes some settings were

01:21:06 ◼   ► only in the control panel or only in other places and it's really really really frustrating especially

01:21:12 ◼   ► if you're not deep in the windows world to figure out okay which one of these two or three places

01:21:17 ◼   ► do i need to go to in order to say change your desktop resolution or something like that and

01:21:21 ◼   ► a third of all the third-party apps while there is an unbelievable amount of third-party apps for

01:21:30 ◼   ► Windows. Most of them, in fact, I would almost say the overwhelming majority of them are utter garbage.

01:21:36 ◼   ► And look at something like, the last time I used Audacity, it was very, very powerful, but it was

01:21:43 ◼   ► visually, I'm not going to say offensive, but hideous. And Audacity is available on the Mac,

01:21:50 ◼   ► as far as I'm aware, but it is like the quintessential GarageBand equivalent for Windows.

01:21:55 ◼   ► and it is just hideous to look at.

01:21:59 ◼   ► And I just don't think that third-party developers

01:22:01 ◼   ► on the Windows platform really take design

01:22:04 ◼   ► nearly as seriously as your average third-party developer

01:22:08 ◼   ► on the Mac.

01:22:09 ◼   ► I believe we started with Marco last time,

01:22:11 ◼   ► so John, let's move to you.

01:22:12 ◼   ► - So luckily this question just says,

01:22:15 ◼   ► "Some specific details of why I don't like Windows,

01:22:17 ◼   ► "so I don't have to be exhaustive."

01:22:18 ◼   ► You already touched on a bunch of them.

01:22:20 ◼   ► I'll throw out some that are near and dear to my heart.

01:22:25 ◼   ► The basic interface paradigm of not having a menu bar at the top of the screen, but instead

01:22:29 ◼   ► having the menu bar embedded in the window, for someone who likes to have a lot of windows

01:22:33 ◼   ► and arrange them, repeating the menu bar in every single one of those windows is incredibly

01:22:37 ◼   ► wasteful and infuriating and it makes it hard to know where the menus are and you know,

01:22:40 ◼   ► you've got the whole infinite height target, yadda yadda.

01:22:42 ◼   ► Not that I had used the menu bar that much, but just the space savings alone to say, look,

01:22:46 ◼   ► we're going to burn the top part of the monitor for the menu bar, but that's it.

01:22:49 ◼   ► No matter how many windows you get open, no more space is taken up by the menu bar.

01:22:53 ◼   ► the obsession with full screen and the tiling features that unfortunately the Mac has copied

01:22:57 ◼   ► where oh, slam the window against the side of the screen, now it's taking up half or

01:23:01 ◼   ► the top third or the bottom third.

01:23:02 ◼   ► That's not how I use Windows.

01:23:03 ◼   ► I don't want that to be that way.

01:23:04 ◼   ► I hate full screen.

01:23:06 ◼   ► I don't like the taskbar or whatever the hell they're calling it these days in the bottom

01:23:11 ◼   ► that's mutated a lot.

01:23:12 ◼   ► I've never liked it.

01:23:13 ◼   ► I don't really like the dock that much either, but I like the dock better than that stupid

01:23:15 ◼   ► thing.

01:23:17 ◼   ► The aesthetics of its DOS origins with backslashes instead of forward slashes and all caps file

01:23:23 ◼   ► names that are shouting at me and dot three extensions and many of those limitations have

01:23:29 ◼   ► been slowly winnowed down over the years, but tons of them are still there, including

01:23:33 ◼   ► stupid things like drive letters, which just aesthetically just give me the willies and

01:23:37 ◼   ► I do not like.

01:23:40 ◼   ► The mouse cursor is, the pointer is white instead of black, which is wrong.

01:23:46 ◼   ► I mean, just, I don't use Windows a lot, but as part of my work, I use it in a virtual

01:23:51 ◼   ► machine enough to be, enough to know that it just doesn't appeal to me in so many ways.

01:23:59 ◼   ► Like, it's like fundamental ways, like how window management works and how applications

01:24:03 ◼   ► work, and sort of the connection between, like, applications and Windows and how they

01:24:08 ◼   ► appear in the taskbar to just little tiny picky things, so just plain old aesthetics

01:24:12 ◼   ► of what icons and menus look like,

01:24:14 ◼   ► and what font rendering looks like,

01:24:15 ◼   ► and drive letters, I mean, just drive letters,

01:24:19 ◼   ► that should have been all I had to say.

01:24:21 ◼   ► (laughing)

01:24:23 ◼   ► - Marco.

01:24:24 ◼   ► - You guys basically covered it,

01:24:25 ◼   ► I mean, it's like death by a thousand cuts,

01:24:28 ◼   ► I mean, it's hard to nail down specifics

01:24:31 ◼   ► without having used Windows in a very long time.

01:24:34 ◼   ► A lot of it comes down to what John said

01:24:37 ◼   ► about just design preferences,

01:24:39 ◼   ► a lot of it comes down to, Casey,

01:24:40 ◼   ► what you said about the quality of most software available

01:24:44 ◼   ► and the community around it just being pretty lacking.

01:24:48 ◼   ► It's just kind of a wasteland of mediocrity

01:24:52 ◼   ► and bad decision making.

01:24:54 ◼   ► If I really had to use something that was not Mac OS,

01:24:59 ◼   ► I honestly think I would be more inclined

01:25:02 ◼   ► to try to use Linux than Windows at this point.

01:25:05 ◼   ► And I used Windows for a very long time.

01:25:06 ◼   ► It's not that I didn't know,

01:25:10 ◼   ► I know what I'm missing, basically.

01:25:11 ◼   ► And while, granted, I haven't used the versions

01:25:15 ◼   ► from the last probably literally 10 years or so,

01:25:19 ◼   ► from what people say, they aren't that much better

01:25:21 ◼   ► in a lot of pretty critical ways

01:25:23 ◼   ► or just kind of in like ecosystem conditions.

01:25:27 ◼   ► So I do think I'm speaking from some position of experience,

01:25:31 ◼   ► you know, probably not as much as Casey,

01:25:33 ◼   ► but more than Jon.

01:25:35 ◼   ► And it just does not fit for me.

01:25:38 ◼   ► which it really seems incredibly mediocre.

01:25:42 ◼   ► And if it's the only thing that I ever knew,

01:25:46 ◼   ► and it's the only thing that was available,

01:25:48 ◼   ► I could make it work.

01:25:50 ◼   ► I would just be annoyed with it pretty frequently,

01:25:52 ◼   ► like I was when I was using it full-time.

01:25:55 ◼   ► - And finally, Mark W. asks,

01:25:56 ◼   ► "If podcast advertising went away overnight,

01:25:59 ◼   ► "would you still do the show?"

01:26:01 ◼   ► I haven't talked to you guys about this,

01:26:03 ◼   ► but we used to, semi-regularly, ask each other,

01:26:06 ◼   ► "Hey, is everyone still cool?"

01:26:08 ◼   ► Now we just assume it, I guess.

01:26:10 ◼   ► But when we started Neutral,

01:26:13 ◼   ► when we first recorded, to my recollection,

01:26:16 ◼   ► we didn't think we had any sort of advertiser at the time.

01:26:20 ◼   ► And then Marco was able to get, was it Squarespace?

01:26:22 ◼   ► Is that right?

01:26:24 ◼   ► - Yeah, for Neutral, we got Squarespace

01:26:25 ◼   ► for the entire run of it.

01:26:27 ◼   ► Like, kind of at the last minute,

01:26:28 ◼   ► like right before we were gonna record

01:26:30 ◼   ► and release the episodes,

01:26:31 ◼   ► my contact there said, "Yeah, we'll take 'em all."

01:26:34 ◼   ► - So it was before we recorded?

01:26:35 ◼   ► I thought it was after, but you would remember better.

01:26:37 ◼   ► at least before we released them.

01:26:39 ◼   ► That, yeah, that I agree on. I can't, it doesn't really matter in the grand scheme of things. What I was driving at though

01:26:44 ◼   ► is that one way or another, at least for me, when I started doing this, I was doing it just for fun.

01:26:48 ◼   ► I just wanted to talk to my buddies about cars, and it's turned out to be lucrative, which is a tremendous blessing,

01:26:55 ◼   ► and I'm super duper thankful for it.

01:26:57 ◼   ► But yeah, I think, I certainly hope it doesn't go away tomorrow, but if podcast advertising went away tomorrow,

01:27:05 ◼   ► tomorrow, I would absolutely still talk to my two good friends for a couple hours about

01:27:08 ◼   ► nerdy crap each week. I would suffer through. Marco?

01:27:12 ◼   ► - I would do this podcast even if it made a lot less money than it does. And the fact

01:27:16 ◼   ► is if the advertising went away and we had to switch to a direct payment model, I think

01:27:22 ◼   ► it probably would make a lot less money than it does. But I would still love to do it.

01:27:27 ◼   ► And what's nice is that having the money coming in from the ads more strongly justifies putting

01:27:34 ◼   ► putting in a lot of time and doing it on a regular schedule.

01:27:38 ◼   ► If we didn't have the ad money coming in,

01:27:41 ◼   ► it would be a lot easier to say like one week

01:27:43 ◼   ► when it's really busy, "Oh, you know,

01:27:44 ◼   ► "I can't do it this week.

01:27:45 ◼   ► "We'll skip this week and we'll just do it next week."

01:27:48 ◼   ► Or, "We'll skip the next two weeks

01:27:49 ◼   ► "as some of us are traveling

01:27:51 ◼   ► "and it's hard to schedule it," or something.

01:27:53 ◼   ► - Oh, that's a very good point I didn't consider.

01:27:55 ◼   ► And I completely agree with you.

01:27:57 ◼   ► I mean, I don't remember exactly when it was

01:28:00 ◼   ► that we decided this was really a thing,

01:28:02 ◼   ► but it was somewhere around the middle of 2013, which is almost five years ago. And

01:28:09 ◼   ► in almost five years, we have not missed a week. And sometimes, particularly in summertime,

01:28:14 ◼   ► and particularly around Christmas time, we have to do some really ridiculous scheduling

01:28:19 ◼   ► in order to get everything to line up such that all three of us will be here every single

01:28:24 ◼   ► week. And with one exception, when we deliberately traded Jon for Christina Warren, we have always

01:28:30 ◼   ► had all three of us every week for like four and a half years and that is normally very easy but

01:28:36 ◼   ► occasionally extremely difficult and I agree with you Marco we would definitely punt and I wouldn't

01:28:42 ◼   ► say frequently but a lot more than never if there was no if there was no real money involved.

01:28:47 ◼   ► I think you guys covered it I mean practically speaking the show is big enough now that we just

01:28:53 ◼   ► do like Patreon or some sort of direct payment and it would and you know we would just continue

01:28:56 ◼   ► like and I'm not saying that would be the same as advertising Marco said it would almost

01:29:00 ◼   ► certainly be much much less but you know it would be enough to to make it so that

01:29:05 ◼   ► we I think probably maintain the exact same commitment that we have now.

01:29:08 ◼   ► Alright thanks to our sponsors this week Casper Squarespace and Flight Logger and

01:29:13 ◼   ► we will see you next week.

01:29:16 ◼   ► Now the show is over they didn't even mean to begin because it was accidental

01:29:24 ◼   ► Oh it was accidental.

01:29:28 ◼   ► John didn't do any research.

01:29:30 ◼   ► Margo and Casey wouldn't let him.

01:29:33 ◼   ► Cause it was accidental.

01:29:36 ◼   ► It was accidental.

01:29:39 ◼   ► And you can find the show notes at ATP.fm.

01:29:44 ◼   ► And if you're into Twitter, you can follow them at C-A-S-E-Y-L-I-S-S.

01:29:52 ◼   ► So that's Casey List, M-A-R-C-O-A-R-M-N-T-Marco-R-M-N-S-I-R-A-C-U-S-A-C-R-A-C-U-S-A-C-R-A-C-U-S-A-C-R-A-C-U-S-A-C-R-A-C-U-S-A-C-R-A-C-R-A-C-R-A-C-R-A-C-R-A-C-R-A-C-R-A-C-R-A-C-R-A-C-R-A-C-R-A-C-R-A-C-R-A-C-R-A-C-R-A-C-R-A-C-R-A-C-R-A-C-R-A

01:30:22 ◼   ► episode of Analog, right?

01:30:23 ◼   ► - I'm not, no, I'm not.

01:30:25 ◼   ► I worked from home today, but I worked today.

01:30:29 ◼   ► So the agreement was once the baby is born, I'm taking--

01:30:32 ◼   ► - Which is any day now.

01:30:34 ◼   ► - Which is-- - Any second.

01:30:35 ◼   ► - Literally, it could be happening--

01:30:36 ◼   ► - Any day is Mac Pro Day and also Baby Day.

01:30:38 ◼   ► - Yeah, no seriously.

01:30:39 ◼   ► - It is possible that the baby might come out

01:30:42 ◼   ► between the time that we record this

01:30:44 ◼   ► and when I release it tomorrow.

01:30:45 ◼   ► - That's true.

01:30:46 ◼   ► - That is possible.

01:30:47 ◼   ► Unlikely, but possible.

01:30:49 ◼   ► So anyway, so yeah, so the agreement with work,

01:30:51 ◼   ► Which, when we recorded Analog, I was going on the assumption that Baby would be born

01:30:56 ◼   ► before the new year.

01:30:57 ◼   ► And there's reasons why I assume that.

01:31:00 ◼   ► It's not really worth getting into.

01:31:01 ◼   ► But suffice it to say, Smart Money said it was already going to be here at this point,

01:31:05 ◼   ► and it isn't here.

01:31:06 ◼   ► So once the Baby is born, I will be off of work for roundabouts of eight weeks.

01:31:13 ◼   ► Something like that.

01:31:14 ◼   ► I'm going, oh, my company changed from 2017 in 2018 to giving us three weeks of paternity

01:31:23 ◼   ► leave instead of two.

01:31:24 ◼   ► So if the baby was born before the new year, I would have only gotten two weeks of paternity

01:31:28 ◼   ► leave.

01:31:29 ◼   ► And since it's been, since the policy has changed and it will be born sometime in 2018,

01:31:33 ◼   ► I'm getting three weeks of paternity leave, which is excellent.

01:31:37 ◼   ► And if you're from another country and you think that's barbaric, hey, welcome to America.

01:31:42 ◼   ► This is how it works here.

01:31:43 ◼   ► three weeks of paternity leave is unheard of around these parts. Anyway, so after those three weeks of

01:31:49 ◼   ► paternity leave, I've told work, "Hey, I'm just going to take some unpaid leave. I'm going to

01:31:55 ◼   ► leverage FMLA," which is a law in America that says basically they can't fire you or lay you off

01:32:02 ◼   ► when you have to deal with like a big medical issue like a birth. And so I'm going to take

01:32:08 ◼   ► some FMLA and they don't have to pay me, and they won't be paying me, but I'm just going to

01:32:13 ◼   ► take a lot of time off. So the thought was that once I was supposed to come back in January,

01:32:20 ◼   ► I would just basically blow off January and February. In reality, what's going to happen is

01:32:24 ◼   ► I'm going to blow off all of January, probably all of February, and maybe even a little bit of March,

01:32:30 ◼   ► depending on when this baby comes. And that's because I want to give Aaron some time to adjust

01:32:36 ◼   ► to having two kids. I want to be there, of course, and I want to adjust to having two kids. And

01:32:43 ◼   ► it would just be nice to have some time with the family. And to bring this back to the SKTP that

01:32:47 ◼   ► we just finished, really the express reason that I am able to do this, other than just being

01:32:54 ◼   ► reasonably financially savvy and not spending eight times what I make, in no small part,

01:33:01 ◼   ► the reason that this is possible is because of every person listening to the show. And to that,

01:33:06 ◼   ► I owe all of you and especially my two co-hosts my deepest gratitude because were it not for the

01:33:12 ◼   ► extra money that this show provides my family, I don't think I would be able to do this and

01:33:16 ◼   ► take some unpaid leave to be with my new baby and my family.

01:33:20 ◼   ► So I'm really excited by it.

01:33:23 ◼   ► I have all these visions of all these outside of work endeavors that I want to work on,

01:33:29 ◼   ► like a video of Aaron's car, which I've filmed a teeny bit of, but very little of.

01:33:34 ◼   ► I want to do an iOS app that's like half cooked, but I have a lot more I want to do for it.

01:33:41 ◼   ► And there's all sorts of other things I wanna do,

01:33:42 ◼   ► but the reality of the situation is

01:33:44 ◼   ► your life is destroyed in the best possible way

01:33:46 ◼   ► for like two months once a, well, at least two months,

01:33:50 ◼   ► once a baby is born.

01:33:52 ◼   ► And so because of that, we'll see if I accomplish

01:33:55 ◼   ► getting changed in the mornings,

01:33:57 ◼   ► or if I'm just stuck in the same PJs for like two months.

01:34:01 ◼   ► But we'll see.

01:34:01 ◼   ► Anyway, why do you bring it up, Marco?

01:34:03 ◼   ► - I would like to make you

01:34:04 ◼   ► a little bit more uncomfortable right now.

01:34:06 ◼   ► - Okay.

01:34:08 ◼   ► - So I apologize in advance.

01:34:09 ◼   ► You're gonna hate this.

01:34:10 ◼   ► - No, no, no, that's fine.

01:34:11 ◼   ► - You need to hear this.

01:34:13 ◼   ► This is your chance to quit your job.

01:34:17 ◼   ► (laughing)

01:34:19 ◼   ► Listen, I'm serious.

01:34:20 ◼   ► Okay, what you're gonna have is two or maybe part,

01:34:24 ◼   ► almost three months of not going to your job every day,

01:34:29 ◼   ► basically not having a job.

01:34:31 ◼   ► So you're gonna have a nice little preview

01:34:33 ◼   ► of what it is like to be independent.

01:34:35 ◼   ► - Oh, it's gonna ruin me.

01:34:36 ◼   ► Oh, it's gonna ruin me.

01:34:37 ◼   ► - So A, yes it will.

01:34:40 ◼   ► B, your job can't fire you for taking this time,

01:34:44 ◼   ► but you're certainly gonna probably be knocked down

01:34:49 ◼   ► the favorites list.

01:34:51 ◼   ► I wouldn't assume that you're gonna be long

01:34:54 ◼   ► for that job anyway.

01:34:55 ◼   ► And C, you've been talking for years about wanting,

01:34:59 ◼   ► how great it would be to work for yourself

01:35:05 ◼   ► full time at home, basically.

01:35:07 ◼   ► To be there with your family,

01:35:09 ◼   ► Basically, I had to do the kind of life

01:35:10 ◼   ► that I hope that I do, which is like,

01:35:13 ◼   ► I'm here when the family needs me,

01:35:14 ◼   ► I can be present for moments,

01:35:15 ◼   ► I can help out around the house,

01:35:17 ◼   ► I can also do some work sometimes and make some money.

01:35:21 ◼   ► The reason I'm able to do it is because

01:35:24 ◼   ► I have good income from a podcast

01:35:27 ◼   ► and I have software development income

01:35:30 ◼   ► making that even better.

01:35:33 ◼   ► Well, I know how much you make from your podcast

01:35:35 ◼   ► because we split this evenly between the three of us,

01:35:38 ◼   ► so you make the same amount that I make

01:35:39 ◼   ► and you make the same amount that John makes.

01:35:41 ◼   ► (laughing)

01:35:42 ◼   ► So I know exactly how much you make from this show

01:35:45 ◼   ► 'cause it's the same that I do.

01:35:47 ◼   ► And so I know that's pretty good

01:35:49 ◼   ► and I know of course it's irregular

01:35:53 ◼   ► with when the hell advertisers pay and everything

01:35:55 ◼   ► and we get like bursts of money here and there

01:35:57 ◼   ► and then nothing for months and then big bursts of money.

01:35:59 ◼   ► So it isn't like a normal income.

01:36:01 ◼   ► It doesn't provide health insurance and stuff like that.

01:36:04 ◼   ► So I augment this with software development.

01:36:08 ◼   ► That's basically what you do too.

01:36:10 ◼   ► Your software development is from a job.

01:36:12 ◼   ► Like you're working for somebody else,

01:36:14 ◼   ► they're paying you, they're handling a bunch of crap for you,

01:36:16 ◼   ► it's pretty consistent, et cetera.

01:36:17 ◼   ► So, this is a chance for you to try

01:36:20 ◼   ► not to do YouTube full time,

01:36:25 ◼   ► because that has a very, very slow ramp up.

01:36:28 ◼   ► You're not gonna get to a YouTube career in two months.

01:36:31 ◼   ► That's not gonna happen.

01:36:33 ◼   ► This gives you a chance to become an iOS consultant

01:36:37 ◼   ► the other part of your income.

01:36:40 ◼   ► So you too will have software development income

01:36:44 ◼   ► as well as podcast income.

01:36:47 ◼   ► That is enough to have a pretty nice life.

01:36:50 ◼   ► And so I'm telling you right now,

01:36:54 ◼   ► I think you should go into this break seriously

01:36:58 ◼   ► with the expectation that you might not go back.

01:37:01 ◼   ► And now I'm telling the listeners,

01:37:05 ◼   ► because I know the kind of listeners that we have

01:37:08 ◼   ► make Casey do this, send, no I'm serious,

01:37:13 ◼   ► make Casey do this because by sending him work,

01:37:18 ◼   ► if you have consulting work, Casey did not ask me

01:37:21 ◼   ► to say this and he's probably really mortified

01:37:22 ◼   ► right now that I'm saying this,

01:37:24 ◼   ► if you have consulting work that Casey could do

01:37:28 ◼   ► for your company, hire him now so that he can get this going

01:37:34 ◼   ► and then in two months, he won't have to go back to his job.

01:37:37 ◼   ► This is your time to do this.

01:37:40 ◼   ► This is an opportunity.

01:37:43 ◼   ► 'Cause look, some people are naturals

01:37:46 ◼   ► at going into independent life.

01:37:48 ◼   ► Some people, they jump right in,

01:37:51 ◼   ► or they make a big plan and execute that plan,

01:37:52 ◼   ► and they're able to do it right.

01:37:56 ◼   ► That's not what I did.

01:37:57 ◼   ► I was pushed into it.

01:37:59 ◼   ► I was quit-fired.

01:38:03 ◼   ► I had been kind of wanting to maybe do it

01:38:06 ◼   ► like you say sometimes, but I didn't choose

01:38:08 ◼   ► when that was happened.

01:38:10 ◼   ► I was pushed into the deep end,

01:38:12 ◼   ► and I was so incredibly thankful for that.

01:38:15 ◼   ► It was like the best thing that could have happened to me.

01:38:19 ◼   ► It was an external force forcing me to do something

01:38:23 ◼   ► that I should have done, because I needed that push.

01:38:26 ◼   ► You have that push now, sort of, in a way,

01:38:31 ◼   ► because a baby's coming into your life again,

01:38:33 ◼   ► you know how much work that is,

01:38:35 ◼   ► so you're taking this nice sabbatical from work

01:38:37 ◼   ► to have paternity leave to get things going.

01:38:40 ◼   ► This is your push, make it happen.

01:38:44 ◼   ► And I'm telling you, audience, please,

01:38:48 ◼   ► for the love of God, hire Casey, make him do this.

01:38:53 ◼   ► We will all be so thankful, including Casey,

01:38:58 ◼   ► 'Cause look, you will be so much happier.

01:39:01 ◼   ► - You know someone who might not be thankful

01:39:04 ◼   ► and might not be happy?

01:39:06 ◼   ► - No, no, there isn't.

01:39:08 ◼   ► - No, there is this one person that I think

01:39:10 ◼   ► might not be so happy about this

01:39:11 ◼   ► and might not be so thankful, and that's Aaron,

01:39:14 ◼   ► who expects Casey to be on paternity leave

01:39:17 ◼   ► because he's gonna help with the new baby and Declan,

01:39:20 ◼   ► where if instead he's on paternity leave

01:39:22 ◼   ► working on himself, starting his consulting business

01:39:25 ◼   ► while she juggles two screaming children,

01:39:28 ◼   ► I'm thinking maybe she's not gonna be so thankful

01:39:31 ◼   ► for this advice.

01:39:32 ◼   ► Like this is the flaw in this plan.

01:39:33 ◼   ► Like I'm with Marco, we're at the point where,

01:39:36 ◼   ► oh, and by the way, Casey has a new baby.

01:39:37 ◼   ► Like the whole point of this vacation is not

01:39:39 ◼   ► let's launch Casey's independent career.

01:39:41 ◼   ► Like he's taking his time off for a reason.

01:39:45 ◼   ► And everything you're describing

01:39:47 ◼   ► takes time away from the reason

01:39:48 ◼   ► he's supposedly taking off for.

01:39:50 ◼   ► And so I feel like this is maybe not the right time.

01:39:52 ◼   ► And secondarily, having a screaming baby at home

01:39:55 ◼   ► is going to make Casey long for the office on some days.

01:39:58 ◼   ► It's going to be like, "I wish I could leave this house and go to a place with adults and

01:40:03 ◼   ► just do work in quiet peace."

01:40:06 ◼   ► Like, everyone has those moments with babies.

01:40:07 ◼   ► It's just a fact of life.

01:40:08 ◼   ► And so, unfortunately, when he's at the end of this time, he may find himself begging

01:40:13 ◼   ► to go back to the office, which is exactly the opposite of what you want to get him,

01:40:17 ◼   ► like, booted out into the independent lifestyle.

01:40:20 ◼   ► So I'm thinking maybe the timing isn't exactly right.

01:40:23 ◼   ► Obviously, when is the time right?

01:40:24 ◼   ► If you really wanted to make this happen, Marco, you need to get him fired from his

01:40:27 ◼   ► job after he goes back.

01:40:28 ◼   ► >> Marco: Oh, thanks, John!

01:40:29 ◼   ► [laughs]

01:40:30 ◼   ► >> John: But I think in that case, Casey and Aaron both wouldn't be very thankful to you.

01:40:35 ◼   ► So I don't know what the solution is here.

01:40:36 ◼   ► >> Aaron: No, you know, so I appreciate it, first of all.

01:40:40 ◼   ► Let me say I appreciate it.

01:40:42 ◼   ► It's funny because I really don't have any particular interest in going back to consulting.

01:40:51 ◼   ► Let me do that like provide a verbal glossary for a minute when I say consulting I mean

01:40:58 ◼   ► Both flavors of consulting because there's the consulting that I did in years past

01:41:04 ◼   ► Which is I was part of a company and a group of us would be given no poor choice words

01:41:12 ◼   ► But given to another company to do work for that other company and then that work would cease and I would

01:41:18 ◼   ► And I would go back to my original company

01:41:20 ◼   ► So I'm always an employee of my original company, but I bring this up to say I didn't have to pay for health care

01:41:26 ◼   ► Well, I did but you know what? I mean like there was a company health care

01:41:29 ◼   ► I didn't have to drum up work because there was a sales team that did that for me

01:41:34 ◼   ► so I basically just showed up build my hours and left and

01:41:37 ◼   ► So it was still consulting in the sense that I had little to no control over my destiny

01:41:43 ◼   ► But it was not consulting in the same way that you're talking about Marco because the work just fell in my lap