Hacker News Board Hacker News, browsed like a classic message board
Live data from Hacker News

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

eclecticlight.co

Post reply on HN
Print view View original

51–60 of 419 posts

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

#51

Post by deagle50 »

Detrytus wrote: post #25
dewote wrote: post #4

Earlier quoted context omitted.

For context my M1 Air feels as good as the day I bought it and it's just about to hit 3 years old. I kept my previous 2014 MBP for 6 years and I can't see this one being any different.

I'm running Windows (ARM) virtual machine on M1 Air, and it's constantly thermally throttled. It was hitting 95 Celsius degrees sometimes. I even bought one of those cooling pads with three fans, to make up for the lack of cooling in the laptop itself, so it is now down to 65 Celsius degress. Makes my life little more bearable, but I'm switching to M3 Macbook pro (with proper cooling) ASAP.

Add a thermal pad to connect the heatsink to the back cover and the thermal throttling will go away. Tons of tutorials on Youtube and it takes 5 minutes. I did it on my M1 Air and it doesn't go above 85-90C now.

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

#52

Post by deagle50 »

cubefox wrote: post #3

This seems to be a lot of effort to rationalize the surprisingly small performance increase from M2 to M3. Initially the assumption was that M2 to M3 would be a bigger step than M1 to M2, not a smaller one. Perhaps TSMC 3nm is showing the limits of scaling?

The surface area of the M3 Pro vs M2 Pro tells you everything you need to know.

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

#53

Post by bgirard »

diffeomorphism wrote: post #34
pohl wrote: post #20

Earlier quoted context omitted.

Doesn't prioritizing efficiency and thermals make sense for any device? Even a desktop has moments when it's doing almost nothing but has tasks that can adequately be handled by an E core where powering up a P core is overkill. Doesn't being efficient mean that more of the thermal budget can be held in reserve for when it's actually needed? Don't desktop users desire for their powerful machine to also be quiet?

Not really, no. Case in point: https://www.notebookcheck.net/AMD-Ryzen-9-7940HS-analysis-Ze... Adjusting the TDP from 80W all the way down to 35W costs you relatively little in performance and gives you about the same efficiency as an M2 Pro. That is not done because it does not sell. > Don't desktop users desire for their powerful machine to also be quiet? No. That is not an "also" but an "instead". People claim the…

If I can tuck away my desktop under my desk and it acts as a large space heater in the below zero winter I don't mind if it's helping me get my compiles down faster.

There's some jobs I want a fast desktop for and other jobs, like browsing cooking recipes, I'll take my M1 macbook air for.

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

#54

Post by alpaca128 »

stuff4ben wrote: post #19
SXX wrote: post #16

Earlier quoted context omitted.

I bought M1 Air 256GB with 16GB RAM on release for light web development and professional 2D gamedev on Unity. Still using it till now and it's really really durable hardware and I've been usign it daily for 3+ years. With so little RAM had to sacrifice ability to properly run VMs, but yet till this day there just no comparable hardware on market. 8GB versions though... Certainly Apple should be damned for selling la…

8GB of RAM ought to be enough for anybody right? I'm hoping that's true for someone who spends most of their time in Google Docs or Word and the occasional light web browsing. This MBA replaced a Chromebook which really was on its last legs.

I'm using an M2 Air with 16GB and M1 Mini with 8GB. On the 8GB machine I have to restart Firefox every few days because otherwise it completely fills up RAM + Swap and the entire system starts getting sluggish. It's a very noticeable bottleneck while on the 16GB Macbook I don't really think about memory usage.

I would not recommend the 8GB variant to anyone who plans on using it for more than a couple browser tabs and some light tasks.

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

#55

Post by hajile »

jmull wrote: post #15

It seems like the M3 favors efficiency and thermals over raw power. This makes sense for the devices they've put it in. I wonder if we're going to see different processors (M3X or M4) for the next release of the pro desktops, that favor power over thermals and efficiency? Maybe there will be a kind of tick-tock, with odd numbered processors favoring efficiency and even numbered processors for power? M3, M5, M7 for th…

I think it's more accurate to say that M3 favors smaller chips because N3B has atrocious yields (not to mention cost savings).

When they switch to N3E and get better yields, I imagine that we'll see chip size grow. As we're getting close to the number of cores that the average consumer can use or would want to use in a desktop, I imagine that we'll be seeing an increase in GPU, NPU, and other specialized units instead of more CPUs though.

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

#56

Post by helf »

deagle50 wrote: post #51
Detrytus wrote: post #25

Earlier quoted context omitted.

I'm running Windows (ARM) virtual machine on M1 Air, and it's constantly thermally throttled. It was hitting 95 Celsius degrees sometimes. I even bought one of those cooling pads with three fans, to make up for the lack of cooling in the laptop itself, so it is now down to 65 Celsius degress. Makes my life little more bearable, but I'm switching to M3 Macbook pro (with proper cooling) ASAP.

Add a thermal pad to connect the heatsink to the back cover and the thermal throttling will go away. Tons of tutorials on Youtube and it takes 5 minutes. I did it on my M1 Air and it doesn't go above 85-90C now.

[dead]

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

#57

Post by hajile »

pier25 wrote: post #50
scrlk wrote: post #9

Earlier quoted context omitted.

It's a mixture of several problems: * TSMC N3B being a bit of a flop (yield issues, too expensive) * Brain drain from Apple's chip design teams over the last few years * Tim Cook trying to push the average selling price up to keep revenue growth going in the face of sales declines (e.g. hobbling memory bandwidth, reducing the number of performance cores for M3 Pro) I don't expect there to be a M1 style generational l…

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.

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

#58

Post by kalleboo »

stuff4ben wrote: post #2

This is good to know. I worry about the longevity of these new M-series processors from Apple since they're so (relatively) new. What is the usable lifespan of a base M1-MacBook Air for example? I guess I'll find out since I just bought one for my kid who is a junior in HS and I'm hoping this will last until they graduate college.

For me at least the longevity is already far better than the Intel MacBooks Pro. I bought the 2017 and 2019 models and for my use those felt obsolete from day 1, with 2 hour battery life and permanent full-speed fans and CPU throttling.

I have a M1 Pro since release day and I don't see myself wanting to replace this until probably the M5 is out.

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

#59

Post by jeffbee »

camillomiller wrote: post #29
cubefox wrote: post #3

This seems to be a lot of effort to rationalize the surprisingly small performance increase from M2 to M3. Initially the assumption was that M2 to M3 would be a bigger step than M1 to M2, not a smaller one. Perhaps TSMC 3nm is showing the limits of scaling?

I don't know where you get the idea of small performance increase. What I'm seeing until now, also in pre-review unites, is actually the opposite, especially when ray tracing and mesh shading get taken into account. For anything GPU related these new machines are a giant leap.

Are these mainstream use cases? These always feel like they are chosen to demonstrate the processor's strengths, instead of talking about the ways the processor helps in real applications.

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

#60

Post by hajile »

brigadier132 wrote: post #36
cubefox wrote: post #3

This seems to be a lot of effort to rationalize the surprisingly small performance increase from M2 to M3. Initially the assumption was that M2 to M3 would be a bigger step than M1 to M2, not a smaller one. Perhaps TSMC 3nm is showing the limits of scaling?

Isn't it a 15% increase? I really don't consider any double digit percent increases to be small.

Actual IPC increase is 1-2%. The rest is from ramping the clockspeeds. This is a problem because power consumption goes up exponentially with frequency. Go up too high and they'll be doing what AMD or Intel does where a single core is using 50+ watts to hit those peak numbers.
Post reply on HN