It’s no surprise that NVIDIA is gradually dropping support for older videocards, with the Pascal (GTX 10xx) GPUs most recently getting axed. What’s more surprising is the terrible way t…
I’ve had so many problems with Nvidia GPUs on Linux over the years that I now refuse to buy anything Nvidia. AMD cards work flawlessly and get very long-term support.
Even now, CUDA is gold standard for data science / ML / AI related research and development. AMD is slowly brining around their ROCm platform, and Vulcan is gaining steam in that area. I’d love to ditch my nvidia cards and go exclusively AMD but nvidia supporting CUDA on consumer cards was a seriously smart move that AMD needs to catch up with.
CUDA is an Nvidia technology and they’ve gone out of their way to make it difficult for a competitor to come up with a compatible implementation. With cross-vendor alternatives like OpenCL and compute shaders, they’ve not put resources into achieving performance parity, so if you write something in both CUDA and OpenCL, and run them both on an Nvidia card, the CUDA-based implementation will go way faster. Most projects prioritise the need to go fast above the need to work on hardware from more than one vendor. Fifteen years ago, an OpenCL-based compute application would run faster on an AMD card than a CUDA-based one would run on an Nvidia card, even if the Nvidia card was a chunk faster in gaming, so it’s not that CUDA’s inherently loads faster. That didn’t give AMD a huge advantage in market share as not very much was going on that cared significantly about GPU compute.
Also, Nvidia have put a lot of resources over the last fifteen years into adding CUDA support to other people’s projects, so when things did start springing up that needed GPU compute, a lot of them already worked on Nvidia cards.
Yeah, I stopped using Nvidia like 20 years ago. I think my last Nvidia card may have been a GeForce MX, then I switched to a Matrox card for a time before landing on ATI/AMD.
Back then AMD was only just starting their open source driver efforts so the “good” driver was still proprietary, but I stuck with them to support their efforts with my wallet. I’m glad I did because it’s been well over a decade since I had any GPU issues, and I no longer stress about whether the hardware I buy is going to work or not (so long as the Kernel is up to date).
I had an old NVidia gtx 970 on my previous machine when I switched to Linux and it was the source of 95% of my problems.
It died earlier this year so I finally upgraded to a new machine and put an Intel Arc B580 in it as a stop gap in hopes that video cards prices would regain some sanity eventually in a year or two. No problems whatsoever with it since then.
Now that AI is about to ruin the GPU market again I decided to bite the bullet and get myself an AMD RX 9070 XT before the prices go through the roof. I ain’t touching NVidia’s cards with a 10 foot pole. I might be able to sell my B580 for the same price I originally bought it for in a few months.
I’ve had so many problems with Nvidia GPUs on Linux over the years that I now refuse to buy anything Nvidia. AMD cards work flawlessly and get very long-term support.
I’m with you, I know we’ve had a lot of recent Linux converts, but I don’t get why so many who’ve used Linux for years still buy Nvidia.
Like yeah, there’s going to be some cool stuff, but it’s going to be clunky and temporary.
Even now, CUDA is gold standard for data science / ML / AI related research and development. AMD is slowly brining around their ROCm platform, and Vulcan is gaining steam in that area. I’d love to ditch my nvidia cards and go exclusively AMD but nvidia supporting CUDA on consumer cards was a seriously smart move that AMD needs to catch up with.
Sorry for prying for details, but why exactly do you need NVIDIA?
CUDA is an Nvidia technology and they’ve gone out of their way to make it difficult for a competitor to come up with a compatible implementation. With cross-vendor alternatives like OpenCL and compute shaders, they’ve not put resources into achieving performance parity, so if you write something in both CUDA and OpenCL, and run them both on an Nvidia card, the CUDA-based implementation will go way faster. Most projects prioritise the need to go fast above the need to work on hardware from more than one vendor. Fifteen years ago, an OpenCL-based compute application would run faster on an AMD card than a CUDA-based one would run on an Nvidia card, even if the Nvidia card was a chunk faster in gaming, so it’s not that CUDA’s inherently loads faster. That didn’t give AMD a huge advantage in market share as not very much was going on that cared significantly about GPU compute.
Also, Nvidia have put a lot of resources over the last fifteen years into adding CUDA support to other people’s projects, so when things did start springing up that needed GPU compute, a lot of them already worked on Nvidia cards.
Same. Refuse to use NVIDIA going forward for anything.
Yeah, I stopped using Nvidia like 20 years ago. I think my last Nvidia card may have been a GeForce MX, then I switched to a Matrox card for a time before landing on ATI/AMD.
Back then AMD was only just starting their open source driver efforts so the “good” driver was still proprietary, but I stuck with them to support their efforts with my wallet. I’m glad I did because it’s been well over a decade since I had any GPU issues, and I no longer stress about whether the hardware I buy is going to work or not (so long as the Kernel is up to date).
I had an old NVidia gtx 970 on my previous machine when I switched to Linux and it was the source of 95% of my problems.
It died earlier this year so I finally upgraded to a new machine and put an Intel Arc B580 in it as a stop gap in hopes that video cards prices would regain some sanity eventually in a year or two. No problems whatsoever with it since then.
Now that AI is about to ruin the GPU market again I decided to bite the bullet and get myself an AMD RX 9070 XT before the prices go through the roof. I ain’t touching NVidia’s cards with a 10 foot pole. I might be able to sell my B580 for the same price I originally bought it for in a few months.
Sadly GPU passthrough only worked on Nvidia cards when I was setting up my server, so I had to get one of them :(