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

Ask HN: Why hasn't x86 caught up with Apple M series?

news.ycombinator.com

Post reply on HN
Print view View original

261–270 of 640 posts

Re: Ask HN: Why hasn't x86 caught up with Apple M series?

#261

Post by skeezyboy »

jonwinstanley wrote: post #185
ben-schaaf wrote: post #101

Battery efficiency comes from a million little optimizations in the technology stack, most of which comes down to using the CPU as little as possible. As such the instruction set architecture and process node aren't usually that important when it comes to your battery life. If you fully load the CPU and calculate how much energy a AI340 needs to perform a fixed workload and compare that to a M1 you'll probably find s…

A huge reason for the low power usage is the iPhone. Apple spent years incrementally improving efficiency and performance of their chips for phones. Intel and AMD were more desktop based so power efficiency wasnt the goal. When Apple's chips got so good they could transition into laptops, x86 wasn't in the same ballpark. Also the iPhone is the most lucrative product of all time (I think) and Apple poured a tonne of t…

> and Apple poured a tonne of that money into R&D and taking the top engineers from Intel, AMD, and ARM, building one of the best silicon teams.

how much silicon did Apple actually create? I thought they outsourced all the components?

Re: Ask HN: Why hasn't x86 caught up with Apple M series?

#262

Post by yaro330 »

nicolaslem wrote: post #256
yaro330 wrote: post #218

Earlier quoted context omitted.

> Adding to that, it is very picky about which power brick it accepts (not every 140W PD compliant works) and the one that comes with the laptop is bulky and heavy. I am used to plugging my laptop into whatever USB-C PD adapter is around, down to 20W phone chargers. Having the zbook refuse to charge on them is a big downgrade for It's Dell, they are probably not actually using PD3.1 to achieve the 140w mark, instead…

It is HP, and the output of the provided adapter does 28V 5A, so in spec. With some of the other 28V 5A adapters I have, it charges until triggering a compute heavy task and then stops. I have seen reports online of people seeing this behavior with the official adapter. My theory is that the laptop itself does not accept any ripple at all.

Ah my bad. Are you sure your cable can do 140w? That was the source of most of my pains trying to push 100w to my work laptop. Baseus and Anker have some good PD3.1 chipped cables that worked for me. What kind of charger are you using?

Re: Ask HN: Why hasn't x86 caught up with Apple M series?

#263

Post by akho »

gettingoverit wrote: post #245
akho wrote: post #238

Earlier quoted context omitted.

x12 and x6 do not seem plausible. Something is very wrong.

My CPU is at over 5GHz, 1% load and 70C at the moment. That's in a "power-saving mode". If nothing would be wrong, it'd be at something like 1.5GHz with most of the cores unpowered.

What is the laptop, and what's it doing?

Re: Ask HN: Why hasn't x86 caught up with Apple M series?

#264

Post by fullstop »

NitpickLawyer wrote: post #187
socalgal2 wrote: post #146

Earlier quoted context omitted.

AFAICT the comparisons to safari are no longer true https://birchtree.me/blog/everyone-says-chrome-devastates-ma... That might be different on other platforms

I think the GP is talking about linux specifically. On a Mac I can see that Chrome disables unused tabs (mouse over says "Inactive tab, xxx MB freed up")

I have inactive tabs on Linux, and it shows the same thing.

Re: Ask HN: Why hasn't x86 caught up with Apple M series?

#265

Post by aurareturn »

yaro330 wrote: post #234
aurareturn wrote: post #222

Earlier quoted context omitted.

This is false, in cross platform tasks it's on par if not worse than latest X86 arches. This is Cinebench 2024, a cross platform application: https://imgur.com/a/yvpEpKF They are willing due to lower idle and low load consumption, which they achieve by integrating everything as much as possible - something that's basically impossible for AMD and Intel. Weird because LNL achieved similar idle wattage as Apple Silicon.…

> Weird because LNL achieved similar idle wattage as Apple Silicon.[0] Why do you say it's impossible? And where is LNL now? How's the company that produced it? Even under Pat Gelsinger they said that LNL is a one off and they're not gonna make any more of them. It's commercially infeasible. > Honestly not sure how your statement is relevant. How is you bringing up synthetics relevant to race to idle? Regardless, a n…

  And where is LNL now? How's the company that produced it? Even under Pat Gelsinger they said that LNL is a one off and they're not gonna make any more of them. It's commercially infeasible.
Why does it matter that LNL is bad economically? LNL shows that it's definitely possible to achieve same idle or even better idle wattage than Apple Silicon.

  How is you bringing up synthetics relevant to race to idle?
I truly don't understand what you mean.

Re: Ask HN: Why hasn't x86 caught up with Apple M series?

#266

Post by Fluorescence »

aurareturn wrote: post #251
Fluorescence wrote: post #233

Earlier quoted context omitted.

> Despite being ~1.5x bigger than the M4 Pro Where are you getting M4 die sizes from? It would hardly be surprising given the Max+ 395 has more, and on average, better cores fabbed with 5nm unlike the M4's 3nm. Die size is mostly GPU though. Looking at some benchmarks: > slightly more MT. AMD's multicore passmark score is more than 40% higher. https://www.cpubenchmark.net/compare/6345vs6403/Apple-M4-Pro... > worse ef…

Where are you getting M4 die sizes from? M1 Pro is ~250mm2. M4 Pro likely increased in size a bit. So I estimated 300mm2. There are no official measurements but should be directionally correct. AMD's multicore passmark score is more than 40% higher. It's an out of date benchmark that not even AMD endorses and the industry does not use. Meanwhile, AMD officially endorses Cinebench 2024 and Geekbench. Let's use those.…

What a waste of time.

"directionally correct"... so you don't know and made up some numbers? Great.

AMD doesn't "endorse benchmarks" especially not fucking Geekbench for multi-core. No-one could because it's famously nonsense for higher core counts. AMD's decade old beef with Sysmark was about pro-Intel bias.

Re: Ask HN: Why hasn't x86 caught up with Apple M series?

#267

Post by jcranmer »

BirAdam wrote: post #250

First, Apple did an excellent job optimizing their software stack for their hardware. This is something that few companies have the ability to do as they target a wide array of hardware. This is even more impressive given the scale of Apple's hardware. The same kernel runs on a Watch and a Mac Studio. Second, the x86 platform has a lot of legacy, and each operation on x86 is translated from an x86 instruction into RI…

> Second, the x86 platform has a lot of legacy, and each operation on x86 is translated from an x86 instruction into RISC-like micro-ops. This is an inherent penalty that Apple doesn't have pay, and it is also why Rosetta 2 can achieve "near native" x86 performance; both platform translate the x86 instructions.

Can we please stop with this myth? Every superscalar processor is doing the exact same thing, converting the ISA into the µops (which may involve fission or fusion) that are actually serviced by the execution units. It doesn't matter if the ISA is x86 or ARM or RISC-V--it's a feature of the superscalar architecture, not the ISA itself.

The only reason that this canard keeps coming out is because the RISC advocates thought that superscalar was impossible to implement for a CISC architecture and x86 proved them wrong, and so instead they pretend that it's only because x86 somehow cheats and converts itself to RISC internally.

Re: Ask HN: Why hasn't x86 caught up with Apple M series?

#268

Post by rollcat »

ben-schaaf wrote: post #101

Battery efficiency comes from a million little optimizations in the technology stack, most of which comes down to using the CPU as little as possible. As such the instruction set architecture and process node aren't usually that important when it comes to your battery life. If you fully load the CPU and calculate how much energy a AI340 needs to perform a fixed workload and compare that to a M1 you'll probably find s…

> It's a fairly common issue on Linux to be missing hardware acceleration, especially for video decoding. I've had to enable gpu video decoding on my fw16 and haven't noticed the fans on youtube.

I've worked in video delivery for quite a while.

If I were to write the law, decision-makers wilfully forcing software video decoding where hardware is available would be made to sit on these CPUs with their bare buttocks. If that sounds inhumane, then yes, this is the harm they're bringing upon their users, and maybe it's time to stop turning the other cheek.

Re: Ask HN: Why hasn't x86 caught up with Apple M series?

#269

Post by aurareturn »

KingOfCoders wrote: post #255
aurareturn wrote: post #253

Earlier quoted context omitted.

Great. Here's what you're getting between an M4 Max vs an AMD AI 395+: https://imgur.com/a/yvpEpKF And of course, the Mac Studio itself is a much more capable box with things like Thunderbolt5, more ports, quieter, etc. I can see why some people would choose the AMD solution. It runs x86, works well with Linux, can play DirectX games natively, and is much cheaper. Meanwhile, the M4 Max performs significantly better,…

AMD 395+/ Cachyos MT Geekbench 6 25334 - not sure where you get your Geekbench number from for the 395+ You: "If it helps you focus on what the actual discussion, we are comparing maximum CPU and GPU speeds for the dollar." You: "Mac Studio itself is a much more capable box with things like Thunderbolt5, more ports, quieter"

You: AMD 395+/ Cachyos MT Geekbench 6 25334

Me: https://imgur.com/a/yvpEpKF

Re: Ask HN: Why hasn't x86 caught up with Apple M series?

#270

Post by loudmax »

akho wrote: post #238
gettingoverit wrote: post #220

> might be my Linux setup being inefficient Given that videos spin up those coolers, there is actually a problem with your GPU setup on Linux, and I expect there'd be an improvement if you managed to fix it. Another thing is that Chrome on Linux tends to consume exorbitant amount of power with all the background processes, inefficient rendering and disk IO, so updating it to one of the latest versions and enabling "m…

x12 and x6 do not seem plausible. Something is very wrong.

These figures are very plausible. Most Linux distros are terribly inefficient by default.

Linux can actually meet or even exceed Window's power efficiently, at least at some tasks, but it takes a lot of work to get there. I'd start with powertop and TLP.

As usual, the Arch wiki is a good place to find more information: https://wiki.archlinux.org/title/Power_management

Post reply on HN