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

Apple M3 Ultra

apple.com

Post reply on HN
Print view View original

931–940 of 1001 posts

Re: Apple M3 Ultra

#931

Post by SV_BubbleTime »

happyopossum wrote: post #926
lynguist wrote: post #718

Earlier quoted context omitted.

I actually think it’s not a coincidence and they specifically built this M3 Ultra for DeepSeek R1 4-bit. They also highlight in their press release that they tested it with 600B class LLMs (DeepSeek R1 without referring to it by name). And they specifically did not stop at 256 GB RAM to make this happen. Maybe I’m reading too much into it.

> they specifically built this M3 Ultra for DeepSeek R1 4-bit Which came out in what, mid January? Yeah, there's no chance Apple (or anyone) has built a new chip in the last 45 days.

No one is saying they built a new chip.

But the decision to come to market with a 512GB sku may have changed from not making sense to “people will buy this”.

Re: Apple M3 Ultra

#932

Post by adgjlsfhk1 »

imatrix wrote: post #895
adgjlsfhk1 wrote: post #776

Earlier quoted context omitted.

This chip has 0GB vram. It has 8 channel lpddr5.

This is not correct. It is like VRAM, not like normal PC RAM. Lowest latency of DDR5-6400 on normal PC starting at 60ns+ Lowest latency of VRAM on GeForce RTX 4090 starting at 14 ns Lowest latency of Apple M1 Memory starting at 5 ns, its more like L3 cache And on Apple M chip, this ultrafast memory is available for CPU, GPU and NPU. https://www.anandtech.com/show/17024/apple-m1-max-performanc... https://chipsandchees…

You are misreading that chart. 5ns is the L2 latency. The ram latency from your source is 111 ms (once the size gets bigger than the size of the cache). The ram on a 4090 is 240 ns https://chipsandcheese.com/p/microbenchmarking-nvidias-rtx-4...

Re: Apple M3 Ultra

#933

Post by leetharris »

tobyhinloopen wrote: post #868
MR4D wrote: post #819

Earlier quoted context omitted.

Remember, that RAM is also VRAM, so 1/2 terabyte of VRAM ain’t cheap. By comparison, Apple is a downright bargain!

It doesn't have the bandwidth of dedicated GPU VRAM.

Yes it does. It is just short of a 4090's memory bandwidth.

It's still far away from an H100 though.

Re: Apple M3 Ultra

#934

Post by hot_gril »

pjmlp wrote: post #837
hot_gril wrote: post #694

Earlier quoted context omitted.

It's new for mainstream PCs to have it.

Nope, it was common in 8 and 16 bit home computers, and in respect to PCs themselves graphics memory was mapped into the main memory until the arrival of 3D dedicated cards. And even with 3D, integrated GPUs have existed for years.

The CPUs with iGPUs didn't also have the memory on-chip. The Nintendo 64 did. Not sure about the old home computers, but I thought those had separate memory usually.

Re: Apple M3 Ultra

#935

Post by bigyabai »

scarface_74 wrote: post #915
BobAliceInATree wrote: post #897

Earlier quoted context omitted.

> (just the OS updating?) "Just the OS updating" is not insignificant. Software developers, in general, are not known for making sure latest versions of their software run smoothly on older hardware. Also, performance on iPhones is throttled when your battery is very old. There was a whole class-action lawsuit about it. https://support.apple.com/en-us/106348

You realize the alternative is the iPhone shutting down completely?

No, I don't. There are multiple alternatives including (though not limited to) opt-in battery protection, removing battery DRM to enable hardware repairs, or declaring a first-party recall on the faulty units to replace components that are damaging to the hardware.

Conspicuously, Apple just so happened to pick the one that encouraged people to upgrade the entire phone. You know, an entire phone that is otherwise functional without arbitrary restrictions by the OEM.

Re: Apple M3 Ultra

#936

Post by pjmlp »

hot_gril wrote: post #934
pjmlp wrote: post #837

Earlier quoted context omitted.

Nope, it was common in 8 and 16 bit home computers, and in respect to PCs themselves graphics memory was mapped into the main memory until the arrival of 3D dedicated cards. And even with 3D, integrated GPUs have existed for years.

The CPUs with iGPUs didn't also have the memory on-chip. The Nintendo 64 did. Not sure about the old home computers, but I thought those had separate memory usually.

Of course not, because they are not designed as SOCs, the only memory on chip is cache, it doesn't change the fact the memory is one whole block shared between CPU and iGPU.

Re: Apple M3 Ultra

#937

Post by astrange »

ngneer wrote: post #924
astrange wrote: post #762

Earlier quoted context omitted.

The interesting difference between x86 and ARM is security, not performance, btw.

Hardly. x86 and ARM follow similar security underpinnings and outcomes. ARM has TrustZone, x86 has TEEs. I cannot think of a single attack demonstrated on x86 and not on ARM or vice versa. Could you please cite one?

Who cares about TrustZone? x86 doesn't have PAC or MTE.

Re: Apple M3 Ultra

#938

Post by scarface_74 »

bigyabai wrote: post #935
scarface_74 wrote: post #915

Earlier quoted context omitted.

You realize the alternative is the iPhone shutting down completely?

No, I don't. There are multiple alternatives including (though not limited to) opt-in battery protection, removing battery DRM to enable hardware repairs, or declaring a first-party recall on the faulty units to replace components that are damaging to the hardware. Conspicuously, Apple just so happened to pick the one that encouraged people to upgrade the entire phone. You know, an entire phone that is otherwise func…

At the time, battery replacement from Apple or an authorized dealer was $69.

https://support.apple.com/iphone/repair/battery-replacement

There is no “DRM” in their battery.

If you had the wherewithal, you could do it yourself

https://www.ifixit.com/Guide/iPhone+6s+Battery+Replacement/5...

The units weren’t “faulty” all batteries degrade over time.

Re: Apple M3 Ultra

#939

Post by bliteben »

andrewmcwatters wrote: post #691
theturtle32 wrote: post #674

Earlier quoted context omitted.

While technically true, the NEMA 5-15R receptacles are rated for use on 20A circuits, and circuits for receptacles are almost always 20A circuits, in modern construction at least. Older builds may not be, of course. That said, if your load is going to be a continuous load drawing 80% of the rated amperage, it really should be a NEMA 5-20 plug and receptacle, the one where one of the prongs is horizontal instead of ve…

> and circuits for receptacles are almost always 20A circuits, in modern construction at least. This is not true. Standard builds (a majority) still use 15-amp circuits where 20-amp is not required by NEC.

Yes, almost all post 2000 houses will have 20 amps in kitchen but in many areas the other circuits will be 15 amp.

Re: Apple M3 Ultra

#940

Post by seec »

gatienboquet wrote: post #111
behnamoh wrote: post #83

Earlier quoted context omitted.

> My guess is that Apple developed this chip for their internal AI efforts what internal AI efforts? Apple Intelligence is bunkers, and Apple MLX framework remains a hobby project for Apple

They use their own M chips for IA. They are far more advanced on AI than the majority of company. They are using OpenAI for now but in couple months they will own the full chain of value.

They were more "advanced" with the Siri release also, yet today it is barely competitive with the other assistant.

Apple is very good at marketing, a lot less at delivering actual value, especially if it's not about hardware.

Post reply on HN