Recent content by virgult

Aaaaand the CEO of Cadence in that period was........ (Talking about killing two birds with one stone, much)

That would be monumentally stupid. The one ace they have up their sleeve is offering a competitor to CUDA on reasonably priced consumer GPUs. That's their entry point into the AI market. They've already developed it, all they need to do is support it and market if for what it is - then the...

AMD's D-Rex. :D

...yet there's no mention of which Radeon GPUs are supported. There's conflicting information on the website. Here this "hardware spec" list seems to imply that the stack supports most modern Radeons down to the 6600 (plus Vega, the only arch ever really supported by ROCm anyway)...

That's because it cannot compete. It's a bit worse, for a bit more power, if you're a gamer. Non-gaming workloads run abysmally bad compared to Nvidia, due to AMD's neglect of HIP, ROCm, and any effort to make pro workloads run well. This is not a competitive product, that's why it should be...

...because it's really not that great a GPU. (Well, the 8GB is straight e-waste.) A wee bit less money than Nvidia, but no CUDA, no productivity, worse efficiency?

And that is the biggest mystery. I take it that Nvidia is after AI. Fine. Let them ride the bubble until it bursts. However, I'm astonished at how slow, lazy, and dismissive Nvidia's competitors have been since a wide gaping hole in the market was opened for them with the release of the 40...

Is it me, or the config has a pretty underpowered GPU?? Loads of muscle hampered by anemic graphics (and an overly hot and old CPU, but the large form factor of the laptop might just be enough to tame it)

I think the lack of interest is shown by the fact the they were *asked*. Intel published a guide on how to use their new GPUs with PyTorch and their ML stack when they released Alchemist. They gave Alchemist and Battlemage samples to AI YouTubers, not just gaming reviewers. If AMD cared, they...

And this is exactly why AMD can't gain any market share. There's legions of CS students and pros who would ditch Nvidia the second ROCm becomes a bit less of a bad-taste parody of CUDA. Every single generation, AMD spectacularly fails to deliver. I'm amazed they sell any HPC accelerators at...

EXACTLY! This is huge. It's as though AMD finally gained the willpower to close the 15-year gap with Nvidia. Because yes, CUDA is what made Nvidia rich. Not more gaming performance, not better drivers. The Nvidia tax is a CUDA tax. CUDA was there when the world of GPGPU needed it, and AMD was...

So the price went from completely insane to just grossly overpriced? These are £7-800 specs. I could stretch to 1k for brand and supposed reliability.

A "pro" machine (with a price to match) that's designed to never support pro workloads. Not only the 8GB RAM, let's not forget the single external display, no expansion ports... How are people insane enough to pay $1400 to have the usability of a Chromebook????

...about damn time? :) Still, supporting only the halo card is not going to do anyone any favours. Why would a student or someone prototyping smaller models dish out $1k for a 7900 XTX when they can pay $300 and get an RTX 3060? This is important as what AMD should do is establish market...

Nvidia: "No blower cards!" PNY: "What if we upcharged $300, then split the difference?" Nvidia: "Oh, go on then..." >Consumer card with consumer drivers a pro price.