Live data from Hacker News

M3 Macs: there's more to performance than counting cores

eclecticlight.co

191–200 of 419 posts

Re: M3 Macs: there's more to performance than counting cores

#191

Earlier quoted context omitted.

For 16 core CPU, a dedicated GPU equivalent, 400GB/s LPDDR5 memory, 3024x1964 Mini-LED, and 18 hours of battery? The closest I can configure a Dell XPS 15 is $3099, and that's a 14 core CPU and likely lower memory bandwidth and lower performance SSD. They claim 18 hours of battery but only with the base screen, the upgraded screen is presumably less. And from personal experience using an XPS 15 is a significantly wor…

Yes I guess if you absolutely need those specs in a small laptop and you don’t care about value it makes sense. I think it’s more that the laptop is not a good deal compared to other Apple laptops.

Oh for sure. The value return per dollar gets worse the higher you go up Apple's options list. Especially their RAM and SSD upgrade prices.

Re: M3 Macs: there's more to performance than counting cores

#192
post #165
post #157

Earlier quoted context omitted.

Zsh has been the default shell on MacOS for ages now. If you need a more modern bash then you can easily install it using brew or other common package managers. This is a total non-issue.

Of course shipping a very outdated tool with your OS is an issue. And the fact that you need to use 3rd-party package managers to update/replace it is also an issue. But if you want to contend that anything that is solveable via customization/3rd party packages is a 'total non-issue', then I fail to see how you could argue that Linux isn't superior to MacOS in every single way.

It's not an issue even if you say it is. Takes 1 minute to setup brew on a fresh Mac, third party or not who cares when it's open source, apt is also open-source and in the same sense a third party someone develops, you just get that with the base Debian like systems, if we start counting the minute wasted to setup brew then you waste more time to install Debian in the first place, Macs come pre-installed.

Re: M3 Macs: there's more to performance than counting cores

#193
post #184

Earlier quoted context omitted.

I’ve been using macOS in a multimonitor setup for years without much issue. A good bit of it boils down to macOS expecting monitors to be well-behaved, e.g. each having unique EDIDs (many don’t, instead sharing one across all units of a particular model) and initializing in a timely fashion. > Why can't I move a window to a monitor/workspace that has a maximized window? Because it’s fullscreened, not maximized. macOS…

> A good bit of boils down to macOS expecting monitors to be well-behaved, e.g. each having unique EDIDs (many don’t, instead sharing one across all units of a particular model) and initializing in a timely fashion. I don't really buy this explanation. I'm talking about using a single external monitor here, a monitor which I regularly also use with linux (intel/amd) laptops without a single issue. > Because it’s full…

> I don't really buy this explanation. I'm talking about using a single external monitor here, a monitor which I regularly also use with linux (intel/amd) laptops without a single issue.

Might be model-specific then. It’s not something I’ve seen with displays from Asus, Alienware, and Apple. I briefly owned a Dell monitor that would periodically flicker but it got returned.

> That's not really a good reason though. It still doesn't explain why the restriction exists. Why can't a non-maximized/fullscreen window be displayed on top of a fullscreen/maximized window? What problem does that solve?

Probably because it’d be easy for windows to get “lost” if the fullscreen window were focused and non-fullscreened windows fell behind it, with there being no obvious indicator that those windows exist.

Re: M3 Macs: there's more to performance than counting cores

#194

It’s ridiculous that the base and Pro only support 1 and 2 external displays, respectively. Want 3 monitors? Have to pay for the Max. https://www.macrumors.com/2023/11/02/m3-chip-still-supports-...

Apple may be setting up for this to be a killer feature with Apple Vision headsets.

They are saying a single 4K feed, but if you could break that into segments it would change the game. That would be 4 x 1080p screens + the laptop.

If you could slice the rectangles anyway you wanted and rotate them individually then you have the perfect environment. Which you would be able to change instantly, presumably, with Mission Control.

Personally I would pay lots of dollars for that.

Re: M3 Macs: there's more to performance than counting cores

#195

Earlier quoted context omitted.

> expect 2010s Intel style yearly performance gains from here on out Intel saw very little gain this year at all in return for a 400 watt power draw under load. The plain old M3 saw a 20% performance gain along side efficiency gains. Having a 22 hour battery life is insane and you certainly aren't going to manage that with a 400 watt power draw.

400 watts is on a desktop chip where there is no concept of battery life. 20% increase on performance is compared to M1 not, M2 - which also had 20% increase in performance on M1.

> 20% increase on performance is compared to M1 not, M2

Nope.

> The M3 chip has single-core and multi-core scores of about 3,000 and 11,700, respectively, in the Geekbench 6 database. When you compare these scores to those of the M2's single-core and multi-core scores (around 2,600 and 9,700, respectively), the M3 chip is indeed up to 20% faster like Apple claims.

https://www.laptopmag.com/laptops/macbooks/apple-m3-benchmar...

> 400 watts is on a desktop chip where there is no concept of battery life.

Yes, and in exchange for that ridiculous 400 watt power draw, Intel saw negligible performance gains.

> In some areas, the extra clock speeds available on the Core i9-14900K show some benefit, but generally speaking, it won't make much difference in most areas.

https://www.anandtech.com/show/21084/intel-core-i9-14900k-co...

Intel only wishes they could hit a 20% gain in exchange for all that increased power draw and heat. As that review noted the best improvement they saw in any of the common benchmarks was just 6%.

Re: M3 Macs: there's more to performance than counting cores

#196
post #187

Earlier quoted context omitted.

Interesting - it's one of the killer features of macos for me. I use it to compartmentalize stuff I'm working on, and 4 finger swipe between them. I have a couple instances of VSCode, some app/debugging browser windows, some chrome profile windows etc... all running full screen and I spend my day brush swiping between them. It may be different for me because I work on a 14" macbook pro, mostly at coffee shops or unus…

But you can have workspaces/maximized windows without having the restriction of preventing moving windows to another monitor/workspace that has a maximized window. I use workspaces on my linux laptops all the time - this 'feature' has existed for ~20 years, and is not hard to implement. The difference being that macos seems to force the restriction of having only 1 window per workspace and not being able to drag a wi…

You can have multiple monitors and multiple workspaces with multiple windows that can be freely moved between monitors and spaces. The only restriction is that when you make a window full screen (not the same as maximizing), it becomes its own new single-occupancy space. You seem to think that making a window fullscreen is the only way to make a new space, but it's just a special case of a larger system that already has the functionality you're asking for.

Re: M3 Macs: there's more to performance than counting cores

#197

I don't see apple silicon in macs having been the right decision a decade from now because competing toe to toe with the entire semis industry on architecture has never worked before except in iphone where they had a huge first mover advantage.

I certainly thought this back in 2012 when Apple started designing their own mobile CPU cores. How could Apple compete with ARM's designs which would have much greater economies of scale or with Qualcomm who would be shipping way more units than Apple? More than a decade later and it's proven to be a durable competitive advantage.

I would point out that Apple had no first mover advantage with the iPhone. Apple was using standard ARM cores for the first 5 years. It's not like ARM didn't have a multi-decade head start on Apple designing ARM cores.

There are certainly risks, but so far it looks like Apple is making the right decision. Intel and AMD are mostly still targeting a higher power/thermal level than Apple is looking for. Apple likes being able to run things without relying on an external GPU. Having the ability to design things themselves means they're able to make what is most important to their users.

With Intel, Apple was always at the whims of what Intel wanted to produce. It took forever to get the 1038NG7 28W part from Intel which meant that Apple couldn't offer a 13" laptop with the speed that a lot of users wanted. It also meant that they were beholden to crappy Intel graphics - even as they paid up for the Iris graphics.

With their own CPUs, their destiny is in their hands. It's worked extremely well for the iPhone and it's working well for their Macs too.

One of the things to remember: Apple already needs to design these cores for their iPhones and iPads. Once you're doing that, it makes a certain amount of sense to re-use those core designs for the Macs. It isn't zero effort, but they aren't starting from scratch.

I'd also note that Apple tends to do less than a lot of companies and the same applies to their processors. Intel, AMD, and Qualcomm are trying to make parts for a huge range of devices. Apple has a much narrower field of products. While Qualcomm is trying to create processors for $100 cheap phones, Apple just goes with a single CPU for its mobile devices (sometimes using last year's CPU for some devices). With the Mac, Apple has a little more variety, but it's still pretty constrained. An M Ultra is basically just two M Maxes put together. Most of the time, the design is basically the same and Apple has just changed the core count or something like that.

Will Apple compete with Nvidia's GPUs? Probably not, but an M2 Ultra is competing with an Nvidia RTX 3070 desktop GPU. Sure, that's not the current generation or Nvidia's highest spec, but it's still good - and OpenCL isn't great on Mac so it isn't even a fair test. The new M3s have much upgraded graphics and it'll be interesting to see how well they do.

You can say it has never worked before, but I think that might ignore a few things. First, one of the reasons that Intel came to dominate things is because everyone who tried to vertically integrate their processors tried to charge too much. Regardless of what you think of Apple's pricing, they aren't charging more for the privilege of their M processors than they were for their Intel ones. I think it also ignores the fact that Apple has so much money. Finally, in terms of cores shipped in the segments that Apple is shipping them, they're huge. ARM isn't shipping many X-series cores (their top performance cores). They're mostly shipping lower spec'd cores. There aren't a ton of flagship Android phones being sold. Intel is mostly shipping lower-end laptop CPUs destined for machines in the $400-800 range, server CPUs, etc. Apple has a lot of scale for the cores it is designing.

There is risk, but Apple took that risk a decade ago with their iPhone CPUs and they've had a great advantage there. While Intel and others are looking to revitalize their CPU game, it seems like they're doing it at a higher thermal/power level than Apple is looking for - and trying to benchmark themselves against Apple parts at a fraction of the power. Apple is getting to design parts that do what they need and they've proven that they can beat the industry for more than a decade. I'd say it was the right decision to move to Apple Silicon for Macs.

Re: M3 Macs: there's more to performance than counting cores

#198
post #22

Feels like we’ve gone from “it’s just better, it’s obviously better you can see with your eyes and feel it as you use it” when M1/2 dropped. To trying to justify why the M3 is 20% faster than the M1 and the M2 was also 20% faster than the M1 and weirdly Apple is only comparing it to their older processor. Like people, maybe it’s just an underwhelming update… no need to pretend it’s not.

I bought my M1 MacBook Pro not long after they were released. It's approaching 3 years old which is a pretty standard age for people to start looking at a replacement - I guess that's why they're comparing it to the M1. Slight tangent - even though the MBP has been struggling to handle my dev work (partly because the codebase I'm working on has grown significantly in the last two years) I won't be upgrading. Work bou…

This is the way, portable machine should have a good battery life, good screen and good keyboard for me, everything else gets done on a remote, network connected machine with gobs of RAM and Linux.

Re: M3 Macs: there's more to performance than counting cores

#199
post #57
post #50

Earlier quoted context omitted.

The GPU is increasing a lot though. Another point is the CPU improvements are constant. In about two years the base M3 is now on par with the M1 Max.

The GPU is critical to future profits though. Apple really wants your monthly subscription to Apple Arcade and they want to expand in other game areas which is why they've been paying AAA companies to optimize for Mac. This also ties into their VR headset where gaming will be one of the core features.

Hardly any AAA companies are optimising for the Apple GPU. MoltenVK is where all the interest is.

Even if they did, there is very little in Apple Arcade which taxes the GPU, most target the lowest common denominator in terms of supported iOS/phone combinations.

The original Apple Arcade strategy was for AAA titles, but for whatever reason that wasn’t pursued, so now we have a tonne of casual games and re-releases of old titles.

Apple just seems to run hot and cold on gaming.

Post reply on HN