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

Apple's M4 Max chip is the fastest single-core performer in consumer computing

twitter.com

Post reply on HN
Print view View original

391–400 of 551 posts

Re: Apple's M4 Max chip is the fastest single-core performer in consumer computing

#391

Post by MBCook »

rbanffy wrote: post #226
trynumber9 wrote: post #173

Earlier quoted context omitted.

Apple is slightly pulling away. AMD's top desktop chips were on par with M1/M2/M3 1T but now they cannot match even M4 despite releasing a new design (Zen 5) this year. It's partially because AMD is on a two year cadence while Apple is on approximately a yearly cadence. And AMD has no plans to increase the cadence of their Zen releases. 2020 - M1, Zen 3 2021 - ... 2022 - M2, Zen 4 2023 - M3 2024 - M4, Zen 5 Edit: I a…

> Edit: I am looking at peak 1T performance, not efficiency. In that regard I don't think anyone has been close. Indeed. Anything that approaches Apple performance does so at a much higher power consumption. Which is no biggie for a large-ish desktop (I often recommend getting middle-of-the-road tower servers for workstations).

Don’t thermals basically explode non-linearly with speed?

It’s possible Apple’s chips could be dramatically faster if they were willing to use up 300W.

I remember seeing an anecdote where Johny Srouji, the chief Apple Silicon designer, said something like the efficiency cores get 90% of the performance of the performance cores at like 10% of the power.

I don’t remember the exact numbers but it was staggering. While the single core wouldn’t be as high, it sounded as if they could (theoretically) make a chip of only 32 efficiency cores and just sip power.

Re: Apple's M4 Max chip is the fastest single-core performer in consumer computing

#392

Post by MBCook »

spockz wrote: post #254
carlgreene wrote: post #5

I have been out of the PC world for a long time, but in terms of performance efficiency, is Apple running away from the competition? Or are AMD and Intel producing similar performing chips at the same wattage?

For many workloads I think they are pulling definitely ahead. However, I think there is still much to gain in software. For example, my Linux/Fedora desktop with 5900X is many times more responsive than my 16” M1 Pro. Java runs faster. GraalVM native generated native images run way waster. Golang runs faster. X86_64 has seen more love from optimalisations than aarch64 has. One of the things I hit was different GC/mem…

I suspect that’s an OS issue. Linux is simply more optimized and faster at equivalent OS stuff.

Which isn’t too surprising given a lot of the biggest companies in the world have been optimizing the hell out of it for their servers for the last 25+ years.

On the flipside of the coin though Apple also clearly optimizes their OS for power efficiency. Which is likely paying some good dividends.

Re: Apple's M4 Max chip is the fastest single-core performer in consumer computing

#393

Post by MBCook »

LorenDB wrote: post #25
carlgreene wrote: post #5

I have been out of the PC world for a long time, but in terms of performance efficiency, is Apple running away from the competition? Or are AMD and Intel producing similar performing chips at the same wattage?

AMD's latest Strix Point mobile chips are on par with M3 silicon: https://youtube.com/watch?v=Z8WKR0VHfJw

Is that in performance, or performance while matching thermals?

If a competing laptop has to run significantly hotter/louder than in my mind that’s not on par.

Re: Apple's M4 Max chip is the fastest single-core performer in consumer computing

#394

Post by MBCook »

api wrote: post #23
carlgreene wrote: post #5

I have been out of the PC world for a long time, but in terms of performance efficiency, is Apple running away from the competition? Or are AMD and Intel producing similar performing chips at the same wattage?

There have always been higher performing x64 chips than the M series but they use several times more power to get that. Apple seems to be reliably annihilating everyone on performance per watt at the high end of the performance curve. It makes sense since the M series are mobile CPUs on ‘roids.

Irony being that’s the same thing Intel learned from the P4.

They gave it up for Core, grown from their low power chips, which took them far further with far less power than the P4 would have used.

Re: Apple's M4 Max chip is the fastest single-core performer in consumer computing

#395

Post by Reason077 »

Affric wrote: post #253
grahamj wrote: post #188

Earlier quoted context omitted.

Agreed. It might share the title with the M1 Air which was incredible for an ultraportable, but the M1MBP was just incredible period. Three generations later it's still more machine than most people need. M2/3/4 sped things up but the M1 set the bar.

My only regret is getting base RAM. It’s not a server so it’s not a crime to not always be using all of it and it’s not upgradable so it needs to be right the first time. I should have got 32GB to just be sure.

Apple's sky-high RAM prices and strong resale values make this a tough call, though. It might just about be better to buy only the RAM you need and upgrade earlier, considering you can often get 50% or more of the price of a new one back by selling your old one.

Thankfully, Apple recently made 16GB the base RAM in all Macs (including the M2/M3 MacBook Airs) anyway. 8GB was becoming a bad joke and it could add 40% to the price of some models to upgrade it!

Re: Apple's M4 Max chip is the fastest single-core performer in consumer computing

#396

Post by MBCook »

tacker2000 wrote: post #191
rubslopes wrote: post #139

I have a M1 MacBook Air that I use for Docker, VSCode, etc. And it still runs very smoothly. Interestingly, the only times it slows down is when opening Microsoft Excel.

Excel performance on Mac is a disaster, and I dont understand why. Everytime I paste something it lags for 1-2 seconds… so infuriating!!

It was originally a Mac app before Microsoft bought it too, wasn’t it?

Re: Apple's M4 Max chip is the fastest single-core performer in consumer computing

#397

Post by MBCook »

FactKnower69 wrote: post #259
miohtama wrote: post #193

Earlier quoted context omitted.

Game performance is often GPU bound, not CPU bound.

1. parent is talking about Unreal Editor, not playing games 2. yes, different pieces of software have different bottlenecks under different configurations... what is the point of a comment like this?

Even if it’s a relatively niche tool, if that’s the tool of your job depends on then that’s the thing that gates if you can use the computer.

Re: Apple's M4 Max chip is the fastest single-core performer in consumer computing

#398

Post by MBCook »

nicce wrote: post #72
HumblyTossed wrote: post #59

Earlier quoted context omitted.

People who browse the web and want that fastest javascript performance they can get.

Or users of Slack, Spotify, Teams.. you name it. But I don't want to make an excuse that Electron-like frameworks should be encouraged to be used even more if we have super single core computers available.

Even if we ignore them, most tasks people do on a computer end up being heavily influenced by single threaded performance.

Amdahl's law is still in control. For a great mini users single threaded performance is extremely important.

Re: Apple's M4 Max chip is the fastest single-core performer in consumer computing

#400

Post by MBCook »

iJohnDoe wrote: post #55
LorenDB wrote: post #40

How many people does this actually affect? Gamers are better off with AMD X3D chips, and most productivity workloads need good multicore performance. Obviously MR is great silicon and I don't want to downplay that, but I'm not sure that best singlecore performance is an overly useful metric for the people who need performance.

Based on my limited knowledge. Most applications aren’t great at using all cores, so single-core performance is really important most of the time.

Yep. And even if they can use multiple threads, faster single threaded performance means each of those multiple threads gets done faster.

There’s a reason consumer CPUs aren’t slower with 1024 cores instead.

Post reply on HN