GFXBench: Apple M1
gfxbench.com
gfxbench.com
So its faster then a 3 generations old budget card, that doesn't run nVidia optimized drivers, over I'm assuming Thunderbolt. So?
So its faster then the last Mac book air, that was old, thermal constrained, and had a chip from Intel that has been overtaken by AMD.
Every test is single core, but guess what, modern computers have multi cores and and hyper threading and that matters.
Apples presentation was full of weasel words like "in its class" "compared to previous models". Fine thats marketing, but can we please get some real, fair benchmarks, against the best the competition has to offer before we conclude that apples new silicon is gift from god to computing?
If you are going to convince me, show me how the CPU stacks up to a top of the line Ryzen/threadripper and run Cinebench. If you want to convince me about the graphics/ML capabilities, compare it to a 3090 RTX using running Vulkan/Cuda.
It makes zero sense to compare them to high-end desktop CPUs and GPUs.
The new MacBook Air with 16GB RAM and 512GB SSD: $1399.
Geekbench: https://browser.geekbench.com/v5/cpu/compare/4679429?baselin...
MX350 is the same chip as the 1050.
And in this case, it'd make for a larger difference between Apples GPU and the nvidia. But then again, the M1 is a mobile SoC and the 1050 is a desktop GPU. So we shouldn't even be comparing them to being with.
Not sure it’s necessarily the best comparison.
So a more power-hungry notebooks battery lasts half as long, but still in the double digits. Not quite a surprise there.
Sure It makes zero sense to compare them to high-end desktop or laptops, since it is used in devices with lacking the main attribute of Personal computer - the ability to control it and install OS of your choice.
Therefore it falls into another category like phones, ipads and other toys just with attached keyboard.
You can't really compare a laptop SoC with a dedicated graphics cards like the 3090 RTX. One is using a battery on a laptop and the other is plugged in to a power source with dedicated cooling.
Yes, Apple is claiming this is a better solution but that's mostly for laptops. While they did release a Mac mini, they still haven't released an Apple SoC Mac Pro or iMac. Those would be fair game for such comparison.
So you have a platform that is (currently) embodied by entry level systems that appear to be noticeably faster than their predecessors. Apple has said publicly that they plan to finish this transition in 2 years. So more models are coming - and they'll have to be more performant again.
It seems pretty clear that the play here runs "Look here's our entry level, it's better than anyone else's entry level and could be comparable to midlevel from anyone else. But after taking crap for being underpowered while waiting for intel to deliver we cn now say that this is the new bar for performance at the entry level in these price brackets."
https://browser.geekbench.com/v5/cpu/search?q=4750U https://browser.geekbench.com/v5/cpu/search?utf8=%E2%9C%93&q...
So if they release a newer U part in the next few months it will likely best this device even on 7nm. An AMD part with a edram probably wouldn't hurt either.
It seems to me that apple hasn't proven anything yet, only that they are in the game. Lets revisit this conversation in a few years to see if they made the right decision from a technical rather than business perspective.
The business perspective seems clear, they have likely saved considerably on the processor vs paying a ransom to intel.
edit: for the downvoters, google Cezanne, because its likely due in very early 2021 and some of the parts are zen3. So apple has maybe 3 months before another set of 10-15W amd parts drop.
PS last I checked AMD was not moving Zen to 5nm until 2022 - so maybe a year plus wait is a differentiation.
Apple will have a 8+4 core out soon, and likely much larger after that. Since they're so much more power efficient, they can utilize cores better at any TDP.
The M1 has leaked scores of ~1700 single core and ~7500 multicore on Geekbench 5, versus 1200 and 6000 for the Ryzen 4750U.
I don’t know anyone who would call that an entry level car, but any Porsche-phile would.
The new Air is fast and reasonably high resolution with a ~4K resolution screen. But it is Apple’s entry level laptop.
Damn, you've got some high standards.
Brighter screen (500 nits vs. 400)
Bigger battery (58 watt-hours vs. 50)
Fan (increasing thermal headroom)
Better speakers and microphones
Touch Bar
0.2 pounds of weight (3.0 pounds vs. 2.8 — not much)
[1]: https://daringfireball.net/2020/11/one_more_thing_the_m1_mac...
Yes, but why compare it to an entry card that was released 4 years ago instead of an entry card that's been released in the past 12 months? When the 1050 Ti was released, Donald Trump was trailing Hillary Clinton in the polls. Meanwhile, the 1650 (released April 2020, retails ~$150) is significantly faster than the 1050 Ti. (released October 2016, retailed $140 but can't be purchased new anymore)
Lets see what a 3000 series nvidia mobile design does on a more recent process before declaring victory.
It still does 8 hours on battery after 3 years
Being entry level performances don't matter much, but it's still a good backup
Bonus point: Linux works great on it
For 13 hundred dollars (30% more in Euros) I can buy a pro machine, one that at least will give me the option of using more than 16GB of RAM
The new Apple Silicon looks good and I love that they are finally shipping some decent GPU, but price wise they're still not that cheap
------
The downvotes are because I said Xiaomi or for something else?
LG sells a 17 inches, 2.9 lb (same weight of the Air) 16GB of RAM (up to 40) and a discrete GPU (Nvidia GTX 1650) for 1,699
Not quite. The play here is, "Look, here is our 98% of sales laptop". That it's entry level is only an incidental factor. 98% of sales volume is at this price point, and so they get the maximum bang for their buck, the maximum impact, by attacking that one first. Not just because it's the slowest or entriest.
Had they started at the fastest possible one, sure it would have grabbed some great headlines. But wouldn't have had the same sales impact. (And it's icing that the slowest part is probably easiest to engineer.)
This article from Anandtech on the M1 is helpful in understanding why the M1 is so impressive.
https://www.anandtech.com/show/16226/apple-silicon-m1-a14-de...
Apple will be releasing something like a M1X next, which will probably have way more cores and some other differences. But this M1 is incredibly impressive for this class of device. Intel has nothing near it to compete in this space.
The bigger question is how well does Apple keep up with AMD and Nvidia for GPUs and will they allow discrete GPUs.
Actually biggest worry might be the economics - given their low volumes at the highest end (Mac Pro etc) how do they have the volumes to justify investing in building these CPUs?
So multiple superfast cores are less important for - say - audio/video if much of the processing is being handled by the GPU, or even by the ML system.
This is a difference in kind not a difference in speed, because plain old x86/64 etc isn't optimal for this.
It's a little like the next level of the multimedia PC of the mid-90s. Instead of playing video and sound the goal is to create new kinds of smart immersive experiences.
Nvidia and AMD are kinda sorta playing around the edges of the same space, but I think Apple is going to try to own it. And it's a conscious long-term goal, while the competition is still thinking of specific hardware steppings and isn't quite putting the pieces together.
Apple wouldn't go down this path if they weren't confident that their designs would scale and keep them in the performance lead for a long time.
Look, the Mac grossed $9 billion last quarter, more than the iPad ($6.7 billion) and more than Apple Watch ($7.8 billion). They've no doubt invested a lot of time and money into this; there's no way, now that they've jettisoned Intel, they haven't gamed this entire thing out. There's too much riding on this.
Yes, Apple's entry level laptops smoke much more expensive Intel-based laptops. But wait until the replacements for the 16-inch MacBook Pro and the iMac and iMac Pros are released.
By then, the geek world would have gone through all phases of grief—we seem deep into denial right now, with some anger creeping in.
There are two reasons to think this might not be the case. The first is that they could justify continuing to do this to their shareholders based solely on the cost savings from not paying margins to Intel, even if the performance is only the same and not better. Their customers might not appreciate having the transition dumped on them in that case, but Apple has a specific relationship with their customers.
And the second is that these things have a long lead time. They made the call to do this at a time when Intel was at once stagnant and the holder of the performance crown. Intel is still stagnant but now they have to contend with AMD. And with whatever Intel's response to AMD is going to be now that they've finally got an existential fire lit under them again.
So it was reasonable for them to expect to beat Intel's 14nm++++++ with TSMC's 5nm, but what happens now that AMD is no longer dead and is using TSMC too?
You know Apple’s market capitalization is a little over $2 trillion dollars, right? And Apple's gross margins have been in the 30-35% range for many years. This isn't a shareholder issue. They are by far the most profitable computer/gadget manufacturer around.
So it was reasonable for them to expect to beat Intel's 14nm++++++ with TSMC's 5nm, but what happens now that AMD is no longer dead and is using TSMC too?
No matter what AMD does in the short term, they're not going to beat the performance per watt of the M1, let alone the graphics, the Neural Engine and the rest of components of Apple's SoC. It's not just 14nm vs. 5nm; it's also ARM’s architecture vs. x86-64.
Apple has scaled A series production for more than a decade and nobody has caught them yet in iPhone/iPad performance. There were 64-bit iPhones for at least a year before Qualcomm and other ARM licensees could catch up.
There's no evidence or reason to believe it'll be any different with the M series in laptops and desktops.
> ARM’s architecture vs. x86-64
x86 will always need more power spent on instruction decode, sure, but it's not a huge amount.
Perhaps you haven't read the Anandtech article? [1]
Intel has stagnated itself out of the market, and has lost
a major customer today. AMD has shown lots of progress
lately, however it’ll be incredibly hard to catch up to
Apple’s power efficiency. If Apple’s performance trajectory
continues at this pace, the x86 performance crown might
never be regained.
That's the issue. Shareholders always want to see growth, but when you're that big, how do you do that? There isn't much uncaptured customer base left while they're charging premium prices, but offering lower-priced macOS/iOS devices would cannibalize the margins on existing sales. Solution: Increase the margins on existing sales without changing prices so that profitability increases at the same sales volume.
> No matter what AMD does in the short term, they're not going to beat the performance per watt of the M1
Zen 3 mobile APUs aren't out yet, but multiply the performance of the Zen 2-based Ryzen 7 4800U by the 20% gain from Zen 3 and the multi-threaded performance (i.e. the thing power efficiency is relevant to) is already there, and that's with Zen 3 on 7nm while Apple is using 5nm.
> it's also ARM’s architecture vs. x86-64.
The architecture is basically irrelevant. ARM architecture devices were traditionally designed to prioritize low power consumption over performance whereas x86-64 devices the opposite, but that isn't a characteristic of the ISA, it's just the design considerations of the target market.
And that distinction is disappearing now that everything is moving toward high core counts where the name of the game is performance per watt, because that's how you get more performance into the same power envelope. Epyc 7702 has a 200W TDP but that's what allows it to have 64 cores; it's only ~3W/core.
> Apple has scaled A series production for more than a decade and nobody has caught them yet in iPhone/iPad performance.
Qualcomm never caught Intel/AMD either.
Ryzen 7 4800U has eight large cores vs 4 large plus 4 small in the M1 and even with your (hypothetical) 20% uplift multicore is just about matching M1. Single core is nowhere near as good.
'Architecture is basically irrelevant' not the biggest factor but not irrelevant - x64 still has to support all those legacy modes and has more complex front end.
You're working very hard to try to deny that Apple has passed AMD and Intel in this bit of the market. We'll have to see what happens at higher TDPs but they clearly have the architecture and process access to do very well.
No idea what Qualcomm has to do with this.
We're talking about performance per watt. The little cores aren't a disadvantage there -- that's what they're designed for. They use less power than the big cores, allowing the big cores to consume more than half of the power budget and run at higher clocks, but the little cores still exist and do work at high power efficiency. It would actually be a credit to AMD to reach similar efficiency with entirely big cores and on an older process.
> Single core is nowhere near as good.
Geekbench shows Zen 3 as >25% faster than Zen 2 for single thread. Basically everything else shows it as ~20% faster. Geekbench is ridiculous.
> 'Architecture is basically irrelevant' not the biggest factor but not irrelevant - x64 still has to support all those legacy modes and has more complex front end.
This is the same argument people were making twenty years ago about why RISC architectures would overtake x86. They didn't. The transistors dedicated to those aspects of instruction decoding are a smaller percentage of the die today than they were in those days.
> No idea what Qualcomm has to do with this.
The claim was made that Apple has kept ahead of Qualcomm. But Intel and AMD have kept ahead of Qualcomm too, so that isn't saying much.
> You're working very hard to try to deny that Apple has passed AMD and Intel in this bit of the market.
People are working very hard to try to assert that Apple has passed AMD and Intel in this bit of the market. We still don't have any decent benchmarks to know one way or the other.
Half the reason I'm expecting this to be over-hyped is that we keep getting synthetic Geekbench results and not real results from real benchmarks of applications people actually use, which you would think Apple would be touting left and right if they were favorable.
So why make this move? It's a long list.
- Greater confidence in TSMC vs Intel to deliver process improvements
- Leverage in house silicon design expertise
- Leverage Apple's collaboration with TSMC
- More modern CPU architecture
- Single architecture across all products (so can run iPhone apps on Mac)
- Ability to incorporate Apple's own IP (e.g. neural engine) in SoC
- Cost savings
- Full control of the hardware design
I've not mentioned performance as this follows from the other factors.
Only one of these benefits would be delivered by moving to AMD.
So some short term transition pain but I think this would be pretty compelling from Apple's perspective when the decision was made.
If they wanted raw performance in general code, they would stay with what they already had. The switch means that their goal was different.
I guess we'll see hordes of fans defending that decision quite soon.
Perhaps you haven't been paying attention but this is Apple's third processor transition: 68K to PowerPC to Intel to ARM. Each time was to push the envelope of performance and to have a roadmap that wouldn't limit them in the future.
When the first PowerPC-based Macs shipped, they were so fast compared to what was available at the time, they couldn't be exported to North Korea, China or Iran; they were classified as a type of weapon [1].
The fact the PowerMac G4 was too fast to export at the time was even part of Apple's advertising in the 90s [2].
It's always been part of Apple's DNA to stay on the leading edge, especially with performance.
Apple's strategy has never been to settle for good enough. If that were the case, they wouldn't have spent the last 10+ years designing their own processors and ASICs. Dell, HP, Acer, etc. just put commodity parts in a case and ship lowest-commodity hardware. It shouldn't be a surprise that the M1 MacBook Air blows these guys out of the water.
Anyone paying attention saw this coming a mile away.
I have a quad-core 3.4 GHz Intel iMac and it's pretty clear the MacBook Pro with the M1 is going to be noticeably faster for some, if not all, of the common things I do as a web developer.
We know the M2 and the M3 are in development; I suspect 2021 will really be the year of shock and awe when the desktops start shipping.
[1]: https://www.wired.com/2002/01/thats-a-whole-lot-of-power-mac...
It shows the A9-A14 chips compared vs the K-series processors by Intel.
I did not choose it, they gave it to me at work, but it's definitely a workstation
It's not a lightweight laptop, but it's a laptop nonetheless
You know. The "laptops" that work about 1 hour, maybe 2 on battery power and require a power brick the size of a VHS cassette to operate.
[1]: including the 16” MBP, but certainly not limited to it
https://en.m.wikipedia.org/wiki/Subnotebook
P.s. in Italian they are usually referred to as "portatili" (portable) from the 80s portable computers, even though they were not laptops
I think we should bring back the term "luggable" for those mobile workstations and gaming notebooks that are hot, heavy, have 2hr or less of battery life.
The M1 is not currently in any workstation class computer.
It is in a budget desktop computer, a throw-it-in-your-bag travel computer, and a low-end laptop.
When an M series chip can't perform in a workstation class computer, then your argument will be valid. But you're trying to compare a VW bug with a Porsche because they look similar.
"The 8-core CPU, when paired with the MacBook Pro’s active cooling system, is up to 2.8x faster than the previous generation, delivering game-changing performance when compiling code, transcoding video, editing high-resolution photos, and more"
This MBP13 is a replacement for the previous MBP13; but the previous MBP13 was not a workstation either. It was a slightly-less-thermally-constrained thin-and-light. It existed almost entirely to be “the Air, but with cooling bolted on until it achieves the performance Intel originally promised us we could achieve in the Air’s thermal envelope.”
Note that, now that Apple are mostly free of that thermal constraint, the MBA and MBP13 are near-identical. Very likely the MBP13 is going away, and this release was just to satisfy corporate-leasing upgrade paths.
I took "budget desktop computer" to be the Mac Mini, "throw-it-in-your-bag travel computer" to be the Macbook Pro, and "a low-end laptop" to be the Macbook Air.
But I agree - the 13" is billed as a workstation and used as such by a huge portion of the tech industry, to say nothing of any others.
I’m talking Silicon Graphics, not Dell.
You say something like that about a low power fanless design and every tech nerd's first reaction is "bullshit". And now they want to call you on your bullshit.
[1] https://www.apple.com/newsroom/2020/11/introducing-the-next-...
https://www.newegg.com/Gaming-Laptops/SubCategory/ID-3365
As a Joe Schmoe it's hard to get good figures, but it appears the total laptop market is about $161.952B[1] with the "gaming" laptop segment selling about $10.96B[2]. Since gaming laptops are more expensive this undercounts cheap laptops, but there are other classes of laptop that are going to outperform this mac, like business workstations.
There might be one way to massage the numbers to pull out that statistic somehow, but it is at best misleading.
[1] https://www.statista.com/outlook/15030100/100/laptops-tablet...
[2] https://www.statista.com/statistics/1027216/global-gaming-la...
I'd really suggest you look at the CPU benchmarks (single and multi core) for M1 vs Intel / AMD laptops.
So this statement is made on the basis of what data?
That's the most meaningful constraint for mobile CPUs today, after all.
Apples measurement comes pretty close to comparing "laptops that most people would actually buy". Not sure why it's meaningful that a laptop maker can put out a model that's as thick as a college book, has the very top bins of all parts, sounds like a jet engine when running at max speed, and is purchased by 1% of the most avid gamers.
Oh, and if someone puts out a second model that adds RGB backlit keyboard, but is otherwise equivalent, that should somehow count against Apples achievements, because for some reason counting by number of models is meaningful regardless of how many copies that model sold o_O
If it was Model they would be spinning it. But they said sold in the past year. I dont know how anyone else would interpret it, but in financial and analytics that is very clearly implying unit sold.
Your [1] is Laptop with Tablet, total Laptop market is about $100B, although this year we might see a sharp increase cause of pandemic.
Let say there are $10 Gaming Laptop market. So 10% of Market Value are going to Gaming Laptop. Total Laptop Market includes Chromebook, so if you do ASP averaging I would expect at least 3x ( if not 4x or higher ) difference in Gaming and Rest of Laptop Market. So your hypothesis of "All Gaming Laptop" would be faster than M1 gives you roughly 3.3% of the Marketshare. Not too far off 2%.
And all of that is before we put any performance number into comparison.
This is what I don't get.. why would you ever assume they meant counting by model? That's a nearly meaningless measurement. How do you even distinguish between models in that measurement? Where do you set the thresholds? The supercharged gaming laptops are absolutely going to dominate that statistics no matter what, because there's a huge number of models, lots of which only differ mostly by cosmetic changes. The margins are likely higher, so they don't need to sell as many of a given model to make it worthwhile to put one out. Does a laptop maker even have to actually sell a single model for it to count? How many do they have to sell for it to count? Does it make sense to count models where every part is picked from the best performing bins, so that you're guaranteed that the model couldn't count for more than a fraction of sales?
Counting by number of laptops actually sold is the only meaningful measurement, at least you have a decent chance of finding an objective way to measure that.
And I thought it was 100% obvious from Apples marketing material what they meant, so I really don't get why anyone is confused about this.
Not a single one will beat it in single core performance.
And their statement will have been through all sorts of validation before they use it so it's almost certainly not 'bullshit'.
That statistic from Apple is just as bad as those.
Not the same as "Faster than 98% of the cheapest OEM laptop available!".
Apple has not made pure fantasy claims in the past so why should they now? The trend has been clear. Their ARM chips have made really rapid performance gains.
At best we have some trite stats about single core performance but I’m interested to see whether or not this maps to reality on some much harder workloads that run a long time. Cinebench is an important one for me...
"And in MacBook Air, M1 is faster than the chips in 98 percent of PC laptops sold in the past year.1"
There is a subtle difference between "98 percent of laptops sold" and your rephrasing as "2% of laptops you can buy today".
If you doubt the meaning, check out the footnote which refers to "publicly available sales data". You only need sales data if sales volume is a factor in the calculation.
I also don't doubt that "every tech nerd's reaction is 'bullshit'", but only because supremely confident proclamations of universal truth that are soon proven wrong is pretty much the defining trait among that community (c. f. various proclamations that solar power is useless, CSI-style super-resolution image enhancement is "impossible" because "the information was lost", "512k should be enough...", "less space than a Nomad..", and everything Paul Graham has said, ever).
Here, these nerds should have been pointed at, for example, last year's iPad Pro. Which has no fans (only -boys) and was already at that performance level (https://www.tomsguide.com/us/new-ipad-pro-benchmarks,news-28...).
After the odd graphs/numbers from the event, I was worried it was going to be an awful ~2 year period of jumping to conclusions based on:
- Geekbench scores
- "It's slow because Rosetta"
- HW comparisons that compare against ancient hardware because "[the more powerful PC equivalent] uses _too_ much power" implying that "[M1] against 4 year old HW is just the right amount of power", erasing the tradeoff between powerfulness and power consumption
The people claiming this blows Intel/AMD out of the water need to have stronger evidence than comparing against parts launched years ago for budget consumers, then waving away any other alternative based on power consumption.[1]
Trading off power for power consumption is an inherent property of chip design, refusing to consider other chipsets because they have a different set of tradeoffs mean you're talking about power consumption alone, not the chip design.
[1] n.b. this is irrational because the 4 year old part is likely to use more power than the current part. So, why is it more valid to compare against the 4 year old part? Following this logic, we need to find a low power GPU, not throw out the current part and bring in a 4 year old budget part.
IMHO it's much more likely UI people will remember it kinda sucks to have 20 hour battery life with a bunch of touch apps than that we'll clear up the gish-gallop of Apple's charts and publications rushing to provide an intellectual basis for them, without having _any_ access to the chip under discussion, so they substitute iPhone/iPad chips that can completely trade off thermal concerns for long enough for a subset of benchmarks to run, making it look like the the powerfulness/power consumption tradeoff doesn't exist, though its was formerly "basic" to me.
Notwithstanding the fact that you do have a point here, polemically phrased as it may somewhat ironically be, I just want to point out that the Paul Graham reference is probably not the best example of the "tech nerd community" trait you're describing. At least this particular community doesn't quite believe that everything Paul Graham say is true; a couple of examples:
* "Let the Other 95% of Great Programmers In": https://news.ycombinator.com/item?id=8799572
* "The Refragmentation": https://news.ycombinator.com/item?id=10826836
I could share a lot more HN discussions, and some of his essays where he pretty much describe the trait you're taking issue with here -- but I'm already dangerously close to inadvertently becoming an example of a tech nerd that believe "everything Paul Graham has said, ever", is absolutely true ;) I don't, and I know for a fact that he doesn't think so either (there's an essay about that too).
By introducing a new source of information (the memory of the NN) it can reconstruct things it has seen before, and generalise this to new data.
In some cases this means hallucinations, true. But in other times (eg text where the NN has seen the font) it is reconstructing what that font is from memory.
So if we are reconstructing letters from a known font we essentially are extracting 8 bits of information from the image. I'm pretty certain that if you distort the image to an SNR equivalent of below 8 bits you will not be able to extract the information.
Tiny camera shakes on a static scene gives away more information about that scene, it's effectively multisampling of the static scene. (If I have to hunch it, any regular video can "easily" be upscaled 50% without resorting to interpolation or hallucination.)
Our wetware image processing does the same - look at a movie shot at the regular 24fps where people walk around. Their faces look normal. But pause any given frame, and it's likely a blur. (But our wetware image processing likely does hallucination too, so it's maybe not a fair comparison.)
"Faster than 98% of PC laptops"
[1]: https://d1abomko0vm8t1.cloudfront.net/article/images/800x800...
Your quote is just not very specific. On its face, it could mean every PC Laptop ever produced. I'm somewhat certain that every Apple computing device ever has beaten that standard. Even Airpods and the Macbook chargers might be getting close these days.
I simply can't agree with these logical gymnastics.
Then you have to add on top of that the consideration of what are the laptops people are actually buying and using. I can assure you that neither enterprise clients nor everyone outside of select, relatively spoiled, group people concentrated in just a few areas are regularly buying top of the line workstation or gaming laptops.
The parent is very much right, amongst techies there's an unhealthy tendency for self-righteous adjudication of the narrative or context.
If it ever started to be used in court, Ryan Gosling will have a lot of explaining to do. I hope that day never comes.
Exactly. It's a meaningless number.
They also conspicuously avoid posting any GHz information of any kind. My assumption is that it's a fine laptop, but a bullshit performance claim.
Is that information actually of any real use when dealing with a machine with asymmetric cores plus various other bits on the chip dedicated to specific tasks [1]?
Benchmarks are as good as it gets. Aggregate benchmarks used for marketing slides are less than ideal, but the problem with marketing is that it has to apply to everyone. Better workload focused benchmarks might come out later... In their defense most people won't look at these, because most people don't really stress their CPUs anyway.
Understanding the clock difference between the low energy and performance cores might be an interesting comparator, for example.
I find that Teams, Slack and VS Code running together stresses my 2019 Macbook Pro 16.
Comparing CPUs by GHz is like comparing cars by seeing which car can rev the hardest.
For example when you look up the 15w i5-1035G1 and see 1ghz base, 3.6ghz boost, you can figure out that no, those geekbench results are not representative of the chip running at 15w. It's pulling way, way more because it's in the turbo window, and the gap between base & turbo is huuuuuge.
So right now when Apple claims the M1 is 10w, we really have no context for that. Is it actually capped at 10w? Or is it 10w sustained like Intel's TDP measurements? How much faster is the M1 over the A14? How different is the M1's performance in the fanless Air vs. the fan MBP 13"?
Frequency would give insights into most all of that. It's not useless.
Therefore M1 succeeds to have a slightly higher single-thread performance (between 3% and 8%) than Intel Tiger Lake and AMD Zen 3 at only 2/3 of their clock frequency.
The speed advantage of M1 in single-thread is not high enough to be noticeable in practice, but what counts is that reaching the same performance at high IPC and low frequency instead of low IPC and high frequency results in having the same performance as the competitors at a much lower power consumption.
This is the real achievement of Apple and not the ridiculous claims about M1 being 3 times faster than obsolete and slow older products.
The cores and frequency can scale up and down quickly so there's not a meaningful clock speed measure. The best number you'll get is the maximum frequency and that's still not useful for comparing ARM chips to x86.
Even Intel's chips don't have frequency measures that are all that useful. Some workloads support TurboBoost but only within a certain thermal envelope. Even the max conventional clock for a chip is variable depending on thermals.
I don't think it's worthwhile faulting Apple for not harping on the M series clock speeds since they're a vector instead of a scalar value.
In this case, they make impressive sounding, but almost completely meaningless assertions about performance. If anything, they are underselling an impressive achievement.
Intel does provide performance ranges in thermally constrained packages that are meaningful.
And that's when both the chips are on the same x86 platform. Differences in architecture can never be compared or should be compared with GHz as any kind of basis.
The only place where the GHz is useful for comparison is with products running the same chip
Every true tech nerd knows this is fucking impressive ever since iPhones started being benchmarked against Intel and every true tech nerd can tell that this is only going to get better. :)
No, it's faster than all those laptops, not just "fanless" ones. It's actually faster than several fan-using i9 models. And it has already been compared to those.
The grantparent talks about comparisons for GPUs...
Their take reminds me of the Far Side cartoon, where the dog is mowing the lawn, a little irregularly, and a guy is yelling at him, "You call that mowing the lawn?"[1]
The problem here is that the obvious competition is Zen 3, but AMD has released the desktop part and not the laptop part while Apple has released the laptop part and not the desktop part. (Technically the Mini is a desktop, but you know what I mean.)
However, the extra TDP has minimal effect on single thread performance because a single thread won't use the whole desktop TDP. Compare the laptop vs. desktop Zen 2 APUs:
https://www.anandtech.com/bench/product/2633?vs=2635
Around 5% apart on single thread due to slightly higher boost clock, sometimes not even that. Much larger differences in the multi-threaded tests because that's where the higher TDP gets you a much higher base clock.
So comparing the single thread performance to desktop parts isn't that unreasonable. The real problem is that we don't have any real-world benchmarks yet, just synthetic dreck like geekbench.
You're seriously suggesting we should compare the integrated graphics in a fanless bottom-of-their-line laptop that starts at $999 to a $1,500 graphics card ?
(Let's be brutally clear - that graphics card can do precisely squat on it's own. No CPU, no RAM, no PSU, no display, no keyboard, etc. etc. etc.)
Surely that makes no sense in any universe.
I sure don't.
"Best-in-class" isn't a weasel word – it's recognition that no, this $1000 laptop is not going to be the fastest computer of all time. Just faster than other products similar to it.
Woot? So you are buying a new Fiat Punto and compare it to the latest spec of a Koenigsegg? What are you even doing?
What we need to know is how these perform against previous MacBooks and potentially Microsoft Surface and Dell XPS. those are competitors.
You're stuck with what you bought :)
(in the article: Fiat Uno - 30 years old Fiat model - with Lamborghini engine runs at 320km/h)
https://www.automotorinews.it/2020/05/27/fiat-uno-motore-lam...
Switching to Apple has a cost, switching to Apple with Apple silicon has an even higher cost.
It all depends what you use your computer for, if you buy a Tesla you probably don't depend on your car, people buying entry level hw are people who don't need something fancy, they need a tool and good enough it's enough.
To reverse your analogy, if they have the same price, I take a computer that I can upgrade and actually own over an Apple
Whe dont know yet.
>To reverse your analogy, if they have the same price,
I was thinking that Apple is the Punto, Apple never had a Koenigsberg.
So no
> I was thinking that Apple is the Punto, Apple never had a Koenigsberg.
The Punto is reliable, durable, and easy to modify
And Koenigsberg is a burg in Germany, maybe you meant Koenigsegg?
You have a hell of a Fiat Punto
>The Punto is reliable, durable, and easy to modify
If you don't drive on salted streets..the frame oxidize under your butt.
>And Koenigsberg is a burg in Germany, maybe you meant Koenigsegg?
Yeah, but honestly i don't give a damn.
under any other circumstance i’d agree with you. i think we can agree this is not the assertion apple are trying to push in their marketing of the M1.
if Fiat are going to claim their entry level punto is, in real world terms, faster that 98% of all cars sold in the last year, they’re inviting a lot of (fair) comparison.
the 1050 is a budget chip from 3 generations ago. even in the GPU space, nvidia are claiming that their mid range GPU (RTX3080) is outpacing their previous generation top-end GPU (RTX2080ti).
the parent comment has a fair point.
It seems pretty ridiculous to put the 10 watt M1 CPU up against big time GPUs that require several hundred watts.
Regardless, all the benches anyone desires will be out next week.
And I'm not scared, I'm sad. I would love to have 3 or more open competitive desktop/laptop platforms (Windows, Linux and MacOS) but my view is that the release of the M1 makes that not very likely.
The M1 is almost identical to the A14, a great chip that was designed for mobile devices. It is designed to run one application at a time, don't do much compute or graphics, have a fixed set of memory and no hardware attachments. It does that brilliantly.
The problem is that a chip designed for a computer have different priorities. The need for low power consumption is less important. Even in a laptop, after 10h, Id rather take performance then more battery time. In a Computer you want more threads, expandable memory, PCI Gen 4.0, discreet graphics, GPU Compute, Raytracing, support for fast networking (the Mac Mini used to have 10Gig lan), multiple display support, expandable storage, and in a stationary computer preferably you want to be able to upgrade parts.
If we look at the M1 compare it to the A14, we can see that Apple has made almost no modifications to the architecture to make it fit in a computer rather then a mobile device. They didn't add more cores, they didn't add support for PCI, GPUs, Ethernet, or anything else that makes it more like a Computer CPU.
Maybe this is besides the point, but to me it was very telling that they didn't update the physical design of any of the 3 computers. The thermal envelope of the M1 Should make it possible to make some much sexier designs, then the old designs we where given. It tells me a lot about how Apple allocates its resources.
This paired with Apples messaging, tells me that Apple is not interested in making Macs that are competitive with PCs on the things that make you choose to use a computer over a mobile device. It feels to me like Apple has moved to Apple Silicon on the mac because its convenient for them to have a single platform, rather than it being the right design for the products.
Apple have neglected the mac and especially the pro segment for a long time. There was always a chance they would get their act together and build reasonably priced, top speced machines, that could compete with Linux and Windows boxes, but with this move, I cant see them put the resources behind their silicon to compete with AMD, nVidia, and Intel when it comes to Performance and features.
So, no I'm not scared, I'm sad that Apple has taken themselves out of the running.
This doesn't exclude you from being a mac hater. In fact judging from the comments you read on every post on HN that is remotely related to Apple, that seems more likely than the typical blind hatred you see on the internet broadly.
> And I'm not scared, I'm sad. I would love to have 3 or more open competitive desktop/laptop platforms (Windows, Linux and MacOS) but my view is that the release of the M1 makes that not very likely.
Based on what, exactly? Very few people have their hands on an M1 mac and there hasn't been any reviews of the systems from third parties yet.
> If we look at the M1 compare it to the A14, we can see that Apple has made almost no modifications to the architecture to make it fit in a computer rather then a mobile device. They didn't add more cores, they didn't add support for PCI, GPUs, Ethernet, or anything else that makes it more like a Computer CPU.
You don't know what they did or didn't do to adapt the M1 for computers. There has to be some degree of PCI-E support since they are using thunderbolt ports. There is ethernet on the mac mini with the M1 chip. It remains to be seen whether or not supporting discrete GPUs will be a thing and that'll largely depend on how their GPU scales. The computers that use discrete GPUs in macs right now haven't released an Apple silicon variant yet.
> Maybe this is besides the point, but to me it was very telling that they didn't update the physical design of any of the 3 computers. The thermal envelope of the M1 Should make it possible to make some much sexier designs, then the old designs we where given. It tells me a lot about how Apple allocates its resources.
I think it is beside the point. There's no reason to delay new chips just for the sake of being in sync with a new redesign of the chassis. It may or may not happen next year with the iMac or Macbook Pro 16, we don't know that yet.
> Apple have neglected the mac and especially the pro segment for a long time. There was always a chance they would get their act together and build reasonably priced, top speced machines, that could compete with Linux and Windows boxes, but with this move, I cant see them put the resources behind their silicon to compete with AMD, nVidia, and Intel when it comes to Performance and features.
They only released their entry level computers. You have nothing to base this argument on. There have only been leaked benchmarks all of which seem favorable towards Apple. We won't know what it means until people start getting them and testing them.
Also, from the Docker issue that was posted here yesterday, we know that the A14 does not have support for virtualization and M1 does (I presume the ARM equivalent of VT-X, etc.).
And yes, they did add more cores. There's PCIe 4. There's support for Ethernet. There's also Thunderbolt and HDMI.
You're extrapolating the three lowest-end Macs up through the entire product line. They aren't going to keep the higher-end Intel Mac mini available if they don't think it (and its specs) serves a purpose. They aren't going to invest everything they did in the new Mac Pro just to drop it after a year.
And if the performance of Apple's chips over the last decade makes you think they won't put the resources behind their silicon to compete, I don't know what to tell you.
We haven't seen what the desktop versions with much higher TDPs and discrete graphics will look like yet but you've written them off?
None of the computers getting the M1 today previously let you put in PCI cards or do any of the stuff you claim is needed other than expand the memory.
A14 only made to do one task at a time is just not true. iOS itself runs multiple tasks and writing software utilizing multiple threads is something you do all the time in iOS. I have written more multithreaded code on iOS than on a desktop. Why? Because using grand central dispatch is a must when dealing with all sorts of low latency web APIs you access. I don’t have that issue when building desktop apps.
These Apple laptops have always had few ports. Don’t blame M1 for that.
I do however wish the Mac Mini had more ports.
This is a key point. Nvidia GPUs have not been supported in macOS since Mojave, so this seems like an apples-oranges comparison. Unless the benchmark for M1 was also run on Mojave (unlikely), then there's 3 years worth of software optimization potentially unaccounted for.
Possibly a more realistic comparison is between the M1 and the AMD Radeon 5300M. Shows between a 10-40% deficit in performance: https://gfxbench.com/compare.jsp?benchmark=gfx50&did1=798775...
That said still an impressive showing given the TDP and the fact it's an integrated GPU vs. a dedicated GPU. It seems to hint that with enough GPU cores and a better cooling solution, it's not unreasonable to see these replacing the AMD Radeon 5500M/5600M next year in the MBP 16 and iMac lineups.
EDIT: pasted wrong compare link
If it helps, take into consideration performance:power ratios. None of your scenarios are fair otherwise, and I personally haven't seen anyone here claim the M1 will outperform everything. Hence, "in its class."
Maybe you saw some errant comments on PCMag.com claiming the M1 was the be all end all of computing?
Good luck.
You think a fair comparison is a passively heated MacBook Air compared to a top spec PC cpu?
It absolutely destroys it’s competition in performance (low end light notebooks) - why would you think comparing it to a 3090rtx is anything like fair.
Like calling a dell XPS 13 a piece of shit because it can’t keep up with a thread-ripper Titan pc desktop.
The fact you’re even in a place (where the benchmarks destroy in class competition) to complain about how it’s not being put up against top spec pc hardware is testament to how powerful it is.
The Macbook Air isn't low end in price.
If you’re looking for a passive thin laptop that’s powerful enough to run pro apps it’s a good option if you have the money to spend.
It's a passively cooled laptop that's super thin and can crunch through 4k video editing and serious audio workflows.
We'd have shit our pants about this kind of leap 10 years ago. It really is impressive stuff.
It seems amazing that Apple can make its own desktop chips now - imagine if the origin Apple had an apple chip instead of a 6502! OTOH, everyone used to design their own CPUs, like Sun and Acorn. Wozniak too.
These are chips for their low end, entry level products. Why on earth would you think that would be an appropriate comparison? That's absolutely absurd. They aren't putting these in iMacs or Mac Pros.
They aren't putting these in their high end MacBook Pros (they put it in their low end 13" pro, which has always been only a minor step up from the Air - the higher end 13" pro was more powerful, which they haven't replaced yet.)
Apple's "faster than 98% of laptops sold" line is obviously nonsense, like I'm sure it's technically correct - most laptops sold are cheap, low end things - but that's not a particularly huge achievement. I'm not sure why this line in particular is inviting people to make such ridiculous statements in response though.
Apple could nearly buy Nvidia with their cash in hand alone, is it that surprising? I'd be more surprised if they fucked it up completely.
That they have actually chose to do it is impressive, although highly worrying to me from a software freedom perspective. What are the odds of Apple releasing drivers for any of their hardware?
[1] https://github.com/TechEmpower/FrameworkBenchmarks/wiki/Tech...
Like.. what the hell were you thinking typing this comment?
It seems odd to be so cynical about a drastic performance increase yoy and a breakthrough in performance per watt. I bet if it wasn’t Apple you would be a little more excited.
I think that’s the important point to consider. Nvidia is non grata with Apple, and that rift happened way way before Metal. These drivers have no chance of performing representatively.
Compared to the AMD Radeon Pro 5600M, the +$700 upgrade in the top of the line MacBook Pro 16": https://gfxbench.com/compare.jsp?benchmark=gfx50&did1=907542...
Note that the Onscreen numbers are capped at 60fps on OS X, so ignore any Onscreen results at 59-60fps.
https://gpu.userbenchmark.com/Compare/Nvidia-GTX-1050-Ti-vs-...
1: Userbenchmark.com is terrible and nobody should use it for anything. At least their CPU side of things is hopelessly bought & paid for by Intel (and even within the Intel lineup they give terrible & wrong advice), maybe the GPU side is better but I wouldn't count on it.
2: The real question there isn't "why is the 1050 Ti not faster?" it's "how did you run a 1050 Ti on MacOS in the first place, since Nvidia doesn't make MacOS drivers anymore and hasn't for a long time?"
2. That’s a good question and I don’t have an answer for that.
As for a better source, I don't know with the M1 being so new, but that's no reason to accept bad data, if this benchmark actually is as bad as others here are saying.
To provide some elaboration on this: Their overall CPU score used to be 30% single, 60% quad, 10% multicore. Last year around the launch of zen 2 they gave it an update. Which makes sense; the increasing ability of programs to actually scale beyond four cores means that multicore should get more importance. And so they changed the influence of multicore numbers from 10% to... 2%. Not only was it a blatant and ridiculous move to hurt the scores of AMD chips, you got results like this, an i3 beating an i9 https://cdn.mos.cms.futurecdn.net/jDJP8prZywSyLPesLtrak4-970...
And there was some suspicious dropping of zen 3 scores a week ago, too, it looks like.
Most applications can only make use of a few CPU-heavy threads at a time, and these systems with with 18 cores will not make any difference for the average user. In fact, the 18 core behemoth might actually feel slower for regular desktop usage since it’s clocked lower.
If you are a pro with a CPU-heavy workflow that scales well with more threads, then you probably don’t need some consumer benchmark website to tell you that you need a CPU with more cores.
> In fact, the 18 core behemoth might actually feel slower for regular desktop usage since it’s clocked lower.
It has a similar turbo, it won't.
A quick inspection of the details tab in my task manager shows around 200 processes, only half of which are browser.
And starting a web browser with one page results in around half a dozen processes
Every user is now a multi-core user.
And they added some "subjective" metric so even when an AMD CPU wins at every single test, the Intel one can still be ranked higher.
There's a reason they've been banned from most major subreddits. Including /r/Intel.
https://gfxbench.com/compare.jsp?benchmark=gfx50&did1=907542...
That’s the most relevant CPI compare, given it is the entry level apple endorsed way to boost Mac gpu.
It also helps understand the value of the 580 and Vega on pre-m1 macs.
Whatever bus video runs over is wired through the dedicated GPU, so integrated is not an option with external monitors connected. That by itself would be fine, except for whatever reason, driving monitors with mismatched resolutions maxes the memory clock on the 5300M and 5500M. This causes a persistent 20W power draw from the GPU, which results in a lot of heat and near-constant fan noise, even while idle. As there isn’t a monitor in the world that matches the resolution of the built-in screen, this means the clocks are always maxed (unless you close the laptop and run in clamshell mode).
The 5600M uses HBM2 memory and doesn’t suffer from the issue, but a £600 upgrade to work around the issue is lame, especially when I don’t actually need the GPU, you just can’t buy a 16” without one.
Disabling turbo boost does help a little, but it doesn’t come close to solving it.
Your best bet is to run external displays at the MBPs resolutions and because that is not possible/ideal you are left with choices of running at 1080p/4k/5k. macOS no longer renders crisp on 1080p so 3840x2160 is the last remaining widely available and affordable choice while 5K is still very expensive.
Part of the reason for 5K’s low popularity was its limited compatibility: they required 2xDP1.2 connections in MST mode. Apple’s single-cable 5K and 6K monitors both depend on Thunderbolt’s support for 2x DP streams to work. I’m not aware of them being PC-compatible monitors at native resolution yet.
I love 5K - but given a bandwidth boost I’d prefer 5K @ 120Hz instead of 8K @ 60Hz.
The AMD Radeon Pro 5600M is built on TSMC 7nm.
https://gfxbench.com/compare.jsp?benchmark=gfx50&did1=907542...
It's not the 4-6x in raw graphics improvement advertised but at 10W vs Xe Max's 25W for just the GPU, the M1 getting 50% more fps, that's still 3.75x in perf/watt.
Let alone - let's really not forget that this is an HBM type architecture. As a complete package it seems awesome, but we can argue for ages about the performance of GPU cores with no end result.
You'd think so, but seems most people[1] think it's just LPDDR wired to the SoC using the standard protocol, inside the same package. (Though it might use > 1 channel I guess?)
[1] eg in the spec table here https://www.anandtech.com/show/16235/apple-intros-first-thre... - an interesing thinb in the mini spec table is also the 10x downgrade of the ethernet speed.
Which would be the same width as dual channel DDR4 – 8x16 == 2x64 :)
Also it's completely ridiculous that some people think it might be HBM. The presentation slide (shown in both anandtech articles) very obviously shows DDR style chips, with their own plastic packaging. That is not how HBM looks :) Also HBM would noticeably raise the price.
This in turn means that every PC and console game created up until now, and probably for the next year or so, will comfortably run on a 1050Ti, even with compromises (“medium” settings, middling resolution like 1080p).
For the first generation of Apple’s Mac GPU, running at 10W, this is a great result. This means, Apples idiosyncrasies aside, there will be no technical problems in porting games to macOS. (Metal is less of a problem than you’d think - if you use Unreal, Unity, DirectX 11/12 or Vulkan for graphics, the work needed will be on the order of a few man/months, which is small change in game development budgets.)
Not at all - I'm using it now. It's 5 years old and not adequate for most games & renders.
I use an eGPU for the sole purpose of running triple monitors. It's a docking station, and the whole setup works brilliantly with my 13" Macbook. Productive workstation setup at home, super nimble laptop on the go.
Apparantly these M1 chips don't support eGPUs and don't even support two external monitors, let alone three. I know it's asked a lot, but what's "Pro" about that?
Completely negates any of the benefits of M1, which is a shame because I really wanted one.
edit: You know, if they don't resolve this then the final round of Intel Macbooks with the non-butterfly keyboards are going to be the holy grail of laptops to a lot of people for a long time, just like the pre-butterfly ones with the ports still are.
Multiple external monitors, eGPUs and bootcamp... that's a lot to lose, man.
Might be better to go for a System76 or Dell if you can find all the apps you need for them.
I actually bought one of the last generation of MacBook Pros so I can continue to build an x86 based server application. Maybe these machines will command a premium on the second hand market in due course?
Not sure what the 2-port M1 MBP13 is really for, other than price descrimination.
I do wish that Apple would go ahead and release a 27” 5K display for the current machines. The 6K is way overkill and there are no good options for monitors >4K. They are probably waiting for the new iMacs with a new design language to put out a matching monitor.
>The power consumption of the GeForce GTX 1050 is roughly on par with the old GTX 960M, which would mean around 40-50 Watts
https://www.notebookcheck.net/NVIDIA-GeForce-GTX-1050-Notebo...
In comparison, the mobile version of the 1080 drew between 150 and 200 Watts, with the lower clocked Max Q version getting the power draw down to 90 to 110 Watts.
https://www.notebookcheck.net/NVIDIA-GeForce-GTX-1080-Laptop...
Ther 1050Ti was replaced by the 1650, 1650S and 1650Ti
The ARM Mac hardware is looking fantastic, but OTOH macOS is getting worse every year...
Phoning home on every executable launch. Both because it's bad for privacy, and because the implementation of it is absolutely horrific such that when Apple's servers went down it basically locked up everyones computers at the same time.
It's on the front page still, even: https://news.ycombinator.com/item?id=25074959
In MacOS, we've recently seen: pushy siri, sending search results to the cloud, and yesterday the OCSP failure made it obvious they were sending logs of every app launch to the cloud :/ . It's the same direction, even if they aren't yet quite as lost as Microsoft.
You would’ve seen it at least a dozen times by now.
System Preferences -> Keyboard
Click on "Customize Control strip"
You then drag and drop items on and off the touchbar. It is a totally inane, unintuitive interface and it took me forever to find it. Also, I couldn't figure out how to change it because the option DISAPPEARS if you're trying to customize in clamshell mode. The touch bar has to be open
I can't tell if it's deliberately bad UX, but I spent months being asked if I wanted to turn on Siri typing on this keyboard...
I mean, if you enable a completely optional feature that requires giving up a bit of privacy for its literal intended functionality, how is it Apple's fault? And unlike Cortana on Windows 10, you can disable Siri feature just with a click of a button, or you can just click a button to not enable it in the first place. When you start your new Mac for the first time, it asks you very explicitly if you want it enabled or not.
Really off-putting and unexpected from Apple.
Most people are not as technical as the people on this site, and we know that blanket access to all executables is not a great idea. MacOS is not immune to malware.
Having executables be signed means the signature has to be checked and the certificate has to be checked to see if it hasn't been revoked. I don't see how to do that without "phoning home" and checking, tbh.
I mean yes Apple could enable some expert setting or something for developer/expert types where you can say "trust me I know what I'm doing" but a lot of people who don't know what they're doing will enable that and then malware will run rampant again.
It's a shitty situation. But I'm not going to go down the paranoia rabbit-hole of assuming this is done to spy on me. At least not yet.
Speaking of Windows, they also moved to a "we know what's best for you" model with Windows 10.
Definitely a "damned if you do, damned if you don't" situation.
I understand why "power users" feel frustrated but also understand the company's POV. A story like "Macs are invulnerable to this latest ransomware attack" looks pretty good to investors; the random complaints of nerds and power users go mostly unnoticed.
I bet $100 this latest scandal will not affect Apple's bottom line nor will anyone care within a month - there'll be other reasons to be outraged over on Twitter.
I don't even remember Siri also exists on Mac OS most times.
I guess I'm in the wrong leg of the A/B test. Fing fantastic.
You could try to delete all of the preference files:
find ~/Library/Preferences/ -iname com.apple.siri\* -delete 0.0.0.0 ocsp.apple.com
sudo dscacheutil -flushcache; sudo killall -HUP mDNSResponder # refresh hosts
defaults write /Library/Preferences/com.apple.security.revocation.plist CRLStyle None
defaults write /Library/Preferences/com.apple.security.revocation.plist OCSPStyle None
defaults write com.apple.security.revocation.plist CRLStyle None
defaults write com.apple.security.revocation.plist OCSPStyle NoneThey are bringing the best bits of iOS to macOS
Why ? Because once everyone starts using iOS apps on their Mac eg. Netflix, Outlook Mac-only apps will slowly disappear. Hence you will need a look and feel that works on touch.
You can run android apps on chrome(/chromium) and other than for novelty I don't know anyone who does so.
As I mentioned, chrome and chromium have enabled this for android for awhile now.
I imagine that eventually devs will target pro software for the ipad and have it come to macos for free.
See also this marketing asset that shows finger interaction with Mac OS: https://twitter.com/mantia/status/1327055883025539072?s=21
You can turn off the menu bar, the rounded corners just look nicer, and I’d rather have simple looking icons than a bunch of words in the UI.
Apple asks you every time you upgrade OSX whether you want to send anonymous data to Apple and third parties. You just need to click no.
By the standards of modern disk and network, couldn't they download revocation caches the way they do with malware?
The whole point is to check if a cert has been revoked. If you have an out of date cache, you'll falsely approve a cert that should be revoked. I'm not defending the system as a whole, but if you care about revoking authentication – which they clearly do – then a cache directly undermines that goal.
A malware hash doesn't get revoked, new ones just get added.
Or every time it feels the need to check a program, instead of asking about that program, it could ask for all revocations from the last day.
I think I've blocked 20 different processes from talking to apple.
Little Snitch helps - but we've seen where that's heading.
Win32 ui just looks 90s, and some looks broken on high dpi screens. While microsoft's new ui is just broken for the desktop use. Text size is different between uwp apps like mail, and win32 apps like word. They are even from the same company and same product team.
I don't think ever mac will be touch based. They share design style with ios. But it's going to be optimized for mouse and keyboard. And I love how you can search for commands in a app on the Mac in the help menu. It allow you to access app features all by typing what you want. You can use gui like a cmd.
As late as today I believe you can still turn off most of it.
I'm running Big Sur and have been since the public beta. I also run iTerm2 and Macports and Firefox and Thunderbird. My editor is Neovim.
MacOS is still a BSD Unix underneath. I can't think of anything I want to run on Linux that I can't run on MacOS.
I also get a nice UI, native first class apps for all my various WFH chat/video clients, MS Office for work etc.
Why exactly would I want to dual boot into Linux? What runs on Linux that doesn't on MacOS?
MacOS just doesn't run great on my 2015 MBP (dual-core) either. I have a 4k monitor, and launching 5 chrome/safari tabs, zoom, and VSCode creates noticeable lag. Window dragging being 24fps, choppy scrolling, etc. I'm sure linux would run better on the same hardware. Its not what doesn't run, its how it runs.
Oh yes. Apple Hardware are still in great shape from a high level overview. But Software and Services.....
Despite both having its own sets of problems, they are still industry leading. Which always make me unsure what to make of it, are Apple really that good? Or Microsoft and Google just doesn't know how to compete? Or more likely a bit of both.
OnScreen Aztec High Tier gives you 54fps on the M1, vs 106fps on the 1050Ti. Normal tier is capped at 60fps for OSX while Windows goes up to 156fps.
That's a better comparison since it represents the best optimised platforms for this hardware. Comparing with NVidia on OSX isn't very meaningful.
Let's see how the reviews turn out, where people will use at real-world tasks.
It annoys me any my coworkers (we use 2017/2018 15" MBP's) every work day to no end. I hate it when when I'm accidentally logged out by feathering the 'lock' icon.
I'm aware I could move it - but ideally where are times that I'd want to use the feature as intended.
The lack of haptic feedback for the Touch Bar is a bizarre and unusual oversight for a company usually so detail-obsessed as Apple. We are not allowed to install anything but specifically approved software, so any third-party solutions are a no-go here. It should be built-in, it's a no-brainer.
I've personally taken to using a bluetooth Apple keyboard half the time with my work laptop just to avoid the Touch Bar as it honestly feels infinitely more comfortable for coding to not have it there.
> Items purchased at the Apple Online Store that are received between 10 November and 25 December 2020 may be returned up to 8 January 2021.
I hadn't realized how much faster computers had gotten since 2015, this new machine runs circles around the old one. The keyboard is great and I actually like the touch bar. 0 regrets, it's an upgrade in every single way.
I don't know your workflow but maybe some dependency in your pipeline has issues and then you are in a world of pain trying to figure out why.
I'm sure the M1 is going to light a fire in getting ARM support on most projects, but it's not there yet.
The idea is that this can do the same as what was historically in a thick and heavy gaming laptop that got 4 hours of battery life in use..is very impressive. It firmly destroys all SKUs of the 15 inch MBP
edit - ignore that I'm an idiot :)
https://news.ycombinator.com/item?id=25065026 ("Apple Silicon M1 chip in MacBook Air outperforms high-end 16-inch MacBook Pro")
https://news.ycombinator.com/item?id=25049079 ("Apple unveils M1, its first system-on-a-chip for portable Mac computers")
https://gfxbench.com/compare.jsp?benchmark=gfx50&did1=651421...
About 60% of Radeon Pro 5500M (in the 16" 2019 Pro)
https://browser.geekbench.com/v5/compute/search?utf8=&q=GTX+...
https://browser.geekbench.com/v5/compute/search?utf8=&q=1050...
my skeptic hat has gotten smaller, still there, just smaller
It's plenty for everything but modern gaming, and since those games aren't likely to be ported to ARM on Mac anytime soon it's not a huge problem. Apple has always had something of a rocky relationship with game publishers, at least on the Mac. Lots of older games will probably work fine, assuming the driver situation isn't a nightmare. Apple is somewhat notorious for neglecting graphics card drivers unfortunately.
The bigger question to be answered is whether this is a baseline that will be surpassed handily by the higher end released coming later, or that this is about as good as it gets now.
I'm assuming that the reason these were released separately was because the later arriving devices have significantly differing SoC's with even better performance, and maybe even discrete GPUs with variable, scaling performance.
1050 Ti is not current gen. Geforce 10 series have been out for more than 4 years now. the current gen is RTX 30 series
1050 is just one generation removed from most current discrete GPUs, I don't believe the 3000 series is out yet on laptops, the 2080 just came out last year.
The 1050 Ti (4 years old) has about the same performance as a GTX 680. A card from 2012.
This comparison makes absolutely no sense. You'd want to compare the M1 against either the current generation integrated, such as the Vega 8 or 11 in the Ryzen 4xxx mobile CPUs or the Intel Xe-LP in the current tiger lake CPUs, or you'd want to compare it against last gen integrated.
Comparing it against a discreet card from 4 years ago with the performance of a card from 8 years ago is just... weird?
If they can't deliver that, it probably should have been held back.
Let's see if iFixit can get us some nice high-res pictures. Heck, someone might leak schematics and any normal users can probably show a IOTree for the system. Always nice to explore those on new machines.
Including Macs.
So this seems pretty useless.
Too bad that with the architecture change all you have to play is iOS games...
On the other hand, there are probably more PC game ports to iOS than there ever were to macOS, and they may run better than ever on the M1.
So, probably!
https://gpu.userbenchmark.com/Compare/Nvidia-GTX-970-vs-Nvid...
[edit: to be clear I mean modern feature set type stuff, not just raw framerate]
Is it a synthetic benchmark or not?
I believe we simply assume that the reputation and trust built over time allows us to take the numbers as 'real enough' from that program & website.
Anyway, answering my own question: multiple results have been uploaded in the past two days and uploaded by anonymous users. Maybe tech reviewers running it with devices they received early?
https://gfxbench.com/allresults.jsp?order=date&page=1&D=Appl...
Also large developers like Adobe, Microsoft would have had retail units months ago to help with final testing.
I was trying to point out that disputing things is fine, but the whole basis of a website where benchmarks are uploaded is trust-and-reputation-over-time to the point where enough other people can re-run the same tests on their machines (one they get them) to validate the results. Heck, you might almost call that science!
Right now, there aren't additional results and you can't easily reproduce them because the machines aren't wide spread or available. But we can take the track record and reputation of the site and application and use that to value the integrity of the published benchmarks to be 'likely correct'.
I'm not an Apple fan at all but that is some mighty impressive performance in such a small package and considering the thermal- and power envelope.
Microsoft has faced this same problem themselves. Releasing the Surface Pro X is a great example of a machine that is limited by software.
As others have noted in other threads, Apple's ability to run iOS apps natively on the M1 chip seems like a great mechanism to lower the switching costs, though I maintain the chasm left to cross is software.
This is all of course, notwithstanding the "locked down OS" concerns from the front page for the past day or so. Does an M1 Macbook Air with BigSur make a competent development machine?
Rosetta 2 is emulation in once sense, but it precompiles applications in to M1 code for speed.
Running all x86-64 mac apps plus iOS apps plus any updated / universal apps is pretty far from a software shortage.
There is absolutely nothing even remotely dubious about universal binaries. They were used during the PPC->Intel transition and again for x86->amd64. You can create them right now.
They’ve done universal bins before. They work. What’s dubious?
Rosetta is translation not emulation.