HN Theater @HNTheaterMonth

It's not just Adobe or games, takes 3D rendering for example, both Redshift and Cycle are now using Metal (the Metal backend for Cycle was even submitted by Apple themselves), Nvidia's GPUs absolutely destroy their (supposed) M1 equivalents, with or without using the ray tracing hardware.

https://youtu.be/HF_rMjjPoPs?t=118 https://www.youtube.com/watch?v=iS_uPukPuo4 https://youtu.be/i4CuZqXA4BM?t=359

⬐ dagmx

In Blenders case, there are still several patches left to merge in and even the Blender page strictly mentions that it's early integration without much performance work done yet.

⬐ brrrrrm

Mentioned in the video is the claim that there simply aren't many products that use Apple's Metal API. It's a similar situation with AMD's GPUs - theoretical performance that can't be achieved without commercial adoption of the software needed to actually achieve it.

I also find it interesting that they benchmarked a basically broken OpenGL driver in this video. What does that show?

⬐ mrguyorama

This title is editorialized (or possibly LTT is A/B testing titles)

The "not as fast" specifically references the 3D gpu performance. The CPU performance is pretty good, and the "media engine" it uses to render video is very powerful in very specific use cases, but the raw 3D performance is a good bit less than Apple claimed

⬐ spacedcowboy

It's not entirely clear to me if the PC laptop was running on power or not - you don't get any difference on the Mac, and it's still quiet. For me, that's a sufficiently good differentiator, but he makes some good points if that's quiet and portability aren't your thang.

⬐ SpelingBeeChamp

That's not true re: no difference on a Mac. The M1 Max has an optional "High Power" mode that can only be enabled when the computer is plugged in.

Edit: I'm an idiot. It is still available, even when unplugged. Sorry :|

⬐ musicale

Hard to get anything useful out of a video like this.

How did the numbers compare to Apple's numbers for the same benchmarks?

What are the most plausible reasons for the discrepancy?

⬐ johnthuss

I would love a transcript or if someone who watched this could summarize it.

⬐ technobabbler

Here's a transcript for you without the timestamps: https://pastebin.com/66WHkRAH

On YouTube, just click the 3 dots and go to Open transcript to see it.

⬐ InTheArena

Open tab…

LTT

….

Close tab….

I love a lot of his content, but linus’s obsession with trashing apple when they don’t say or do exactly what he want them to do got annoying a long time ago.

He mocked apple when they announced the M1, because their marketing was vague Meant that they had something to hide and it wasn’t all that.

Rather then grudgingly roll that back when it turned out it was - he moved on to the next thing.

Skip for now.