bfgoodrich

There are loads of amazing devices out there. My home security system runs on a Jetson Nano and a bespoke solution, with image processing using Tensorflow, fed via GStreamer. I'm going to replace it at some point with a Xavier NX.

Amazon has Gravitron 2. Microsoft is doing something with "their" SQ2, though at this point they're mostly faking it by putting a badge on an uncompetitive Qualcomm part, but I'm sure they'll get there.

The M1 is a wonderful chip, and it has done loads to push the industry to really adopt ARM -- possibly the greatest fat-computing transition to ARM yet -- and it should help across every market, and loads of hardware from competitors.

Some Geekbench results have appeared and seem to show the 888 still at a pretty big disadvantage compared to the A14. e.g. the big cores in the A14 offering 40% greater performance, so much so that the 2 big+4 little of the A14 has higher MT performance than the 4 big+4 little 888.

These results could very well be fake, however early in a product release OEMs do often intentionally release these teaser numbers. This would make it the fastest Android chipset by a good margin, and bring it closer to equalling the A12 of 2018.

Intel's 10nm is approximately equal to TSMC's 7nm (even having advantages in some areas). The A-chips at 7nm (the A13) were already significantly ahead of Intel's offerings in efficiency and instructions per cycle. Intel went all in on a design that relegates performance to power gobbling turbo boost and here we are.

Just generally too much is made about node size. It isn't a magic pixie dust that you can sprinkle on a chip.

I certainly don't want to engage in chip advocacy wars, but it has been interesting seeing the goalposts moving. Last year it was just that Apple had some easily replicated, lazy big cache that was their key to performance, that's if you accept that it's actually high performance because Something Something Geekbench is a big lie and they're actual miserable performers. Now it's just node size. Super simple, trivial things that somehow discount what they're delivering.

Pulling other cores into this discussion does nothing but muddy the waters.

The Firestorm cores in the M1 are 4.5X faster at SPECfp than the Icestorm cores. The Firestorm cores in the M1 are 3.1X faster at SPECint than the Icestorm cores.

In my world, the 3.1 - 4.5X faster sibling makes them very low performance, relatively. In a fully working together scenario, the "8" core M1 is equal in performance to a conceptual 4.5 core Firestorm part (see the Geekbench scaling, for instance). This, again, has positively nothing to do with any other core or part.

Bizarre that anyone thinks this is somehow negative or dismissive of the M1. It is an extraordinary chip. But they intentionally, purposefully put 4 very low performance -- compared to Firestorm -- cores because it matches the computing model of users average workload.

To be fair, 4 of the cores on the M1 are efficiency cores and are very low performance (but sufficient performance, at incredible efficiency, for a lot of casual ongoing background things). An 8 performance core M1 would be a monster.

Not sure if it maxes out at 24 watts. Someone measured wall power of a Mac Mini hitting 21 watts at full load, but that's including power supply inefficiency, all of the supporting circuitry and IO, a fan, flash storage, etc.

Add that chess had already seen a remarkable resurgence this year, before the Queen's Gambit.

https://thenextweb.com/gaming/2020/09/13/the-classic-game-of...

https://www.insider.com/online-chess-has-surged-during-quara...

I absolutely loved the Queen's Gambit, but the root comment has it a bit backwards. It really seems like the Queen's Gambit came at a remarkably opportune time, just as chess was re-emerging into the popular mindspace.

I really doubt that had anything to do with arrogance or thinking they were smart. They knew that everyone would just turn that annoying "feature" off, but presumably some lawyer thought it was a good idea to ensure people didn't blame car crashes on Apple, particularly in a court of law.

e.g.

https://www.cnet.com/news/apple-blamed-for-accident-sued-by-...

I am not the person you replied to, however spores are readily available online in many nations (they are absent psilocybin/psilocin, and thus are often unregulated and lawfully sold as syringes or prints), and techniques like "PF Tek" are simple, well documented, and speedy.

One of the most common stats cited on HN to demonstrate that Apple has a supposed monopoly is StatCounter. StatCounter uses web browsing to measure the prevalence of given clients, in which we find that iPhone users make up more than 50% of US mobile devices.

Only in actual sales data, iPhone devices are far below 50%. It just turns out that users on iPhones browse a lot more than users on other platforms. In just about every country the representation on statcounter significantly exceeds the actual sales metrics of the device.

No, Apple doesn't "force" people into app store apps, and by actual empirical data, facilitates browsing the web even more than alternatives.

As to Safari being the "worst web browser in existence", after chuckling at what I assumed was some sort of parody, I realized some people actually believe this noise. Usually because someone, at some point, stomped their feet and had a fit because Apple didn't immediately support whatever their pet proposal was, and this got spun into a fantastic conspiracy about invented motives.

Yet in the actual real world, Safari is the most performant browser by a country mile. It's the most privacy focused. It supports everything and its brother. For all the pissing and moaning about edge conditions of standalone web applications, Safari was far and away the earliest mobile browser to actually support them! Long before Chrome on Android did.

Remarkable how cheap the comments on stories like this tend to be. The regurgitated "Apple is bad, mmmkay" low effort posts.

To be clear, guy started a sync with his phone (people still do this?). In the middle of a sync he decides he wants to restart his computer, which tells him, reasonably, to wait until its done. If you were running a long-running terminal script, in the middle of a backup or large file move, or countless other tasks, any reasonable OS does exactly the same thing -- it asks the various subsystems if now is a good time, if not it delays the reboot.

He could forcefully stop the sync (which would make it not an issue, while forfeiting the work it had done thus far), and he could forcefully reboot his computer, but this has positively nothing to do with "you have to reboot your phone first".

There needs to be a moratorium on these sorts of low-info, low-effort posts. It's just pandering to a base.

I do consume around three large coffees an average day. I would get a headache from caffeine withdrawal (not a migraine though) so it's been a pretty long time since I've really gone without caffeine.

Couldn't say whether caffeine is negative or positive as it's pretty much always been a component.

I do recreationally partake of psilocybin (for "spiritual" and simply entertainment reasons), and this was something that I personally have noticed: I no longer have migraines. Previously I suffered extreme migraines that would easily knock me out for a day, I couldn't take any medicine or I would vomit it back up, etc. I haven't since my first "shroom" experience.

The weird thing is that I get the pre-migraine experience at the same rate, it just isn't followed by a migraine. Previously I'd get the vision distortions that would get worse to the point that I could barely see, and then an extreme migraine would start. I still get those vision distortions (every week or two) just as frequently, but they're followed by just returning to normal.

This is obviously single case anecdotal, could be purely coincidental, and I don't encourage anyone to base any action on it, but it was something I literally had a discussion about this weekend.

Small doses of psilocybin are very difficult to detect. Personally I barely find it barely noticeable at 1.5g (in the dried shroom measure). Microdose levels are 1/10th of that, and often much lower.

Give someone a so-called microdose and it is unlikely that a placebo vs actual study would actually find that people could detect it.

A disclaimer on your battery life: that is doing literally nothing with the 2080. Effectively turning it off and carrying it around as an extra cost/weight for nothing.

Because the moment you use that 2080, your battery life will be South of 3 hours with a 100% charge and a factory new battery. You'll also have a jet engine fan and a leg burning device.

Which is the point.

"The 2080 Max-Q in my laptop is fairly underpowered"

The 2080 Max-Q only exists in ~$3000 gaming laptops. Poor battery life, 4lb+ laptops. There is no universe where that is "fairly underpowered" unless we've moved the goal posts to pretend that everyone is rolling with a GTX 3080 and anything less is unbearable. In the real world you have to go to the seriously compromised "gaming" tier to get better graphics performance.

"but in terms of raw power it's just not there"

That depends upon what we define as the destination. It in no way competes with dedicated gaming rigs, as I made clear. Not by a huge margin. It isn't going to be something a guy reviewing 300W, $1000 GPUs will care about. That isn't the target market.

But the vast majority of people don't having those gaming rigs. Laptops that have better GPUs are a _tiny_, minuscule component of the market. And if you actually use those dedicated GPUs, you'd better be plugged in.

So when we talk about "gaming" we get into a No True Scotsman thing (to repeat myself) where it isn't gaming unless it's someone playing Warzone at 120Hz at 4K. But a pretty heady number of users are doing things like Civilization, The Sims, Roblox, Minecraft, and similar gaming. The M1 can host that style of gaming with ease.

"thanks mostly to TSMC's 5nm process"

It's kind of interesting that we're at this point. Apple has made a pretty capable GPU, stellar CPU cores, stellar inference cores, among other remarkable hardware. It was pretty stellar at 7nm too. It's all being dismissed as the 5nm advantage. I don't think it's remotely so simple.

The power-sipping, integrated GPU has performance 30% better than a GTX 1050 Ti [1].

The GTX 1050 Ti is still a current, sold product [2], selling for $150 or so USD [3], and consumes about 75-100W in your PC.

We can no true Scotsman this, but that GPU performance is a literal magnitude+ greater than what the majority of users have in their computers (and I am including full desktops in this discussion). The overwhelming majority of users are not paying $500-$1000 for a GPU (now add the 700W+ PSU, and what in effect becomes a space heater), which is the current going rate for the mythical gaming market.

Is it intended for enthusiast gamers who have giant $1000 graphics cards sucking 300W? No, of course not. There is no surprise that Apple didn't send it to a guy who targets that crowd. But for the vast majority of people that chip would represent a pretty significant upgrade over what they currently have.

[1] - Even in Shadow of the Tomb Raider, demonstrated purely given that it was running under Rosetta 2, it did quite decent.

[2] - Current products from EVGA, Gigabyte, ASUS and others feature the GTX 1050 Ti, making it a current product.

[3] - Which is a range, and above, that similar cards sell for. If you can time travel you may be able to get an equal performance level for less, but I lack time travel abilities.

Non tech people didn't care. Tech people could read the many details that Apple provided on specific performance claims.

And to be clear, the presentation claimed "this is a fast mobile processor that's very efficient". Both claims have been very soundly proven. So I'm not sure why they would attempt the deception you seem to be implying. There was simply no need.

The M1 release is fascinating because never have I seen such a frantic effort to dismiss something in the face of overwhelming evidence that it's a pretty great thing. Eh.

When a die shrink is correlated with higher performance, generally that means higher clock speed for a given heat profile. The clock speed of the 4800U is already 50% higher than the M1 when running a single-thread task. And despite that, it is 25-50% slower. And it's significantly less energy efficient.

So put that to 5nm. Either you've partly closed the large efficiency gap, or you've closed the significant performance gap, but in neither case will you come close to closing it entirely. So you're either a little less slower, but still a lot less efficient, or a lot slower, but a little less efficient.

Yeah, maybe they'll do some amazing things on the core that'll overcome all of this. But right now the Firestorm core has a massive advantage. AMD can go to Zen4 and 5nm and maybe they'll close the gap, but Apple won't be sitting still.

Apple entered the desktop space and brought the most efficient core (by _far_), hilariously offering the best per core performance without going to magnitudes higher thermal profiles. It's pretty amazing. So now we're into comparing it to hypothetic, mythical alternatives from competitors in the future.

Sidenote: Apple has had blazing cores for a few years now, and every denier will claim it's just some big cache or some other absurd simplification. If that were the case, everyone would just copy them.

To be clear, I said that a 4800u @ 5nm (if one could simply scale a design like that) would only be marginally better. That the 5nm boogeyman is more incremental than the big advantage it is held as.

You replied that if you take the 4800u, switched it to 5nm, switch it to Zen 3...no actually switch it to to Zen 4 and a completely different chip, it would be lots better so what I said is "definitely not true".

I'm not sure this logic follows.

AMD definitely is the best of the rest, but it doesn't seem quite as close as it may be held to.

In single core tests the 4800u is running that core at 4.2Ghz. Yet it gets soundly bested by the M1 @ 3 - 3.2Ghz (running at a 50%+ advantage, 65%+ clock per clock). The M1 has an enormous IPC advantage.

In a multicore test the 4800u has 8 performance cores with HT. It only marginally beats an M1 with 4 performance cores and 4 efficiency cores (by the scaling the efficiency cores look like they're 1/4 performance or worse -- these are very lightweight cores).

Again, it's the best of the rest, but Apple clearly holds an enormous lead here. Somehow everyone is focused on 5nm, but the A13 on 7nm was still in a substantial lead. The 4800u on 5nm is only going to be marginally better.

Apple clearly sandbagged this first entrant because they're packing it into their "entry level" devices. In six months or whatever they'll unveil the 6+2 core device in the mid range, the 12+2 in the high range, etc, and we'll be back at these discussions.

(Speaking generally) - This whole discussion about Apple Silicon is fascinating because the goal posts have moved so much. Looking back to HN discussions a year ago and everyone was talking about some pathetically weak entrant that would be a joke, etc. Now people are celebrating that it doesn't beat a 24-core, 300W Threadripper. Now the narrative is that it isn't impressive because the v1 didn't overwhelming destroy everything else in the market.

I have a Yoga 720 with a touchscreen and honestly the only time I use the touch is inadvertently when I'm wiping a piece of dust or the like off the screen. Every time it is a bit of a shock.

The form just doesn't conduce itself to touch. As someone earlier said, even if Apple does eventually add touch, they need to figure out the collision with iOS to allow functional use without requiring it.