AMD Begins Polaris and Vega GPU Retirement Process, Reduces Ongoing Driver Support
by Ryan Smith on November 8, 2023 9:30 AM EST
As AMD is now well into their third generation of RDNA architecture GPUs, the sun has been slowly setting on AMD’s remaining Graphics Core Next (GCN) designs, better known by the architecture names of Polaris and Vega. In recent weeks the company dropped support for those GPU architectures in their open source Vulkan Linux driver, AMDVLK, and now we have confirmation that the company is slowly winding down support for these architectures in their Windows drivers as well.
Under AMD’s extended driver support schedule for Polaris and Vega, the drivers for these architectures will no longer be kept at feature parity with the RDNA architectures. And while AMD will continue to support Polaris and Vega for some time to come, that support is being reduced to security updates and “functionality updates as available.”
For AMD users keeping a close eye on their driver releases, they’ll likely recognize that AMD already began this process back in September – though AMD hasn’t officially documented the change until now. As of AMD’s September Adrenaline 23.9 driver series, AMD split up the RDNA and GCN driver packages, and with that they have also split the driver branches between the two architectures. As a result, only RDNA cards are receiving new features and updates as part of AMD’s mainline driver branch (currently 23.20), while the GCN cards have been parked on a maintenance driver branch – 23.19.
At present, AMD has not published anything about this change in driver support to their website. But, responding to a request for a comment on Windows driver support from AnandTech, the company provided the following statement:
The AMD Polaris and Vega graphics architectures are mature, stable and performant and don’t benefit as much from regular software tuning. Going forward, AMD is providing critical updates for Polaris- and Vega-based products via a separate driver package, including important security and functionality updates as available. The committed support is greater than for products AMD categorizes as legacy, and gamers can still enjoy their favorite games on Polaris and Vega-based products.
Notably, AMD is actively asserting that this is not “legacy” status for Vega and Polaris, which is an important distinction because of what “legacy” means within AMD’s ecosystem. For AMD, legacy products are effectively considered end-of-life, and ongoing driver support is retired. Which in the case of previous generation GPU architectures going legacy, AMD did not have any further driver releases of any kind planned – though in practice AMD did release a couple of drivers to fix critical security issues.
AMD’s support plans for Vega and Polaris, by contrast, still call for regular driver releases, albeit without major feature updates or performance optimizations. That means receiving bug fixes and other occasional updates as AMD sees fit to backport them to the older driver branch, but not the full scope of updates that AMD’s RDNA products are now receiving via their up-to-date mainline driver. In practice, this is much closer to how NVIDIA has handled their legacy GPU products, which have traditionally received security fixes for a minimum length of time – making for a more welcome offramp for going from fully supported to unsupported.
But regardless of what AMD is calling their driver support policy for Polaris and Vega, the end result is that major driver development for these GPU architectures has effectively come to an end, and these parts have now entered an extended support phase. AMD considers the drivers for the hardware mature, and with games increasingly taking advantage of features not supported by the hardware (e.g. Alan Wake II and mesh shaders), there’s clearly less of a need to add support/optimizations for new games to drivers for old hardware.
More broadly speaking, with the current high-end game consoles using a RDNA2-ish (DX feature level 12_2) architecture, we’re finally approaching a rather hard switch that will be leaving pre-12_2 GPUs behind. Consequently, I’m not surprised to see AMD semi-retire both Polaris and Vega at the same time – there is a clear gulf in archtiecture between GCN and the greatly reworked RDNA that underpins AMD’s more recent cards.
| AMD Recent GPU Driver Releases | |||
| Adrenaline | RDNA | GCN 4 & 5 | |
| 23.11.1 (November) | 23.20.23.01 | 23.19.05.01 | |
| 23.10.2 (October) | 23.20.17.03 | N/A | |
| 23.9.3 (September) | 23.20.11.04 | 23.19.02 | |
| 23.8.2 (August) | 23.10.29.06 | ||
As AMD enacted this driver split back in September, we can already see some of the ramifications of this with AMD’s latest drivers. Polaris and Vega did not receive an October driver release (Adrenaline 23.10), and the November release (Adrenaline 23.11.1) contains only a handful of fixes for GCN cards, as opposed to the much more extensive list of fixes and new game support for RDNA cards.
Ultimately, while the remaining GCN GPUs haven’t been put out to pasture quite yet, this is clearly the beginning of the end for a line of GPU architectures that stretches back to AMD’s 2011 GPU architecture modernization. AMD has been selling Polaris (GCN 4) cards since mid-2016 – starting over seven years ago – and in practice the core compute and graphics architecture of GCN 4 is virtually identical to the even older GCN 3 architecture. Consequently, AMD has essentially been supporting that core GPU architecture for almost 9 years at this point.
| GCN 4 & 5 Products | |||
| Desktop | Mobile | ||
| Radeon VII | Radeon 600 Mobile Series | ||
| Radeon RX Vega Series | Radeon 500 Mobile Series | ||
| Radeon Pro Duo | Radeon 400 Mobile Series | ||
| Radeon 600 Series | Ryzen Mobile 7030U Series | ||
| Radeon RX 500 Series | Ryzen Mobile 5000 Series | ||
| Radeon RX 400 Series | Ryzen Mobile 4000 Series | ||
| Ryzen 5000G Series APUs | Ryzen Mobile 3000 Series | ||
| Ryzen 4000G Series APUs | Ryzen Mobile 2000 Series | ||
| Ryzen 3000G Series APUs | |||
| Ryzen 2000G Series APUs | |||
Meanwhile, things are a little more short-lived for the newer Vega GPU architecture (GCN 5). While AMD introduced the first discrete Vega GPUs and video cards in mid-2017 – and replaced the whole lot of them in mid-2019 – that GPU architecture remained in use in current-generation products for much longer as an integrated graphics solution. AMD’s current desktop APUs, the Ryzen 5000G series, integrate a Vega GPU. And the same silicon is still sold in the mobile space as the budget-minded Ryzen Mobile 7030 series. So although the Vega architecture is only a year younger than Polaris, it has stuck around for much longer overall.
Unfortunately, that does also mean that these Vega-based APUs are also getting something of the short end of the stick when it comes to driver support, receiving only a few years of mainstream driver support before being deprioritized. Though as these are also the weakest Vega GPUs, they’re admittedly also the least likely to be used with new games. More critical here will be how long AMD supplies security fixes for the Vega GPU architecture, especially since GPU drivers are popular targets for privilege escalation attacks.
In any case, while this isn’t a eulogy for the final members of the Graphics Core Next GPU architecture – at least, not quite yet – it’s clear that, 12 years later, GCN’s time is finally approaching its end.
Facebook
Twitter
RSS
32 Comments
View All Comments
kn00tcn - Wednesday, November 8, 2023 - link
and the AFMF (FSR3) preview branch is 23.30.x.xprevious legacy switches were unacceptably without warning and often had no real updates since
it would be wise to do better this time especially since vega is still in stores
interestingly, on polaris i've run ai upscaling over vulkan on windows 7, and unreal engine 5 lumen also works on win7 (screenspace GI and reflections), even fortnite ran last year upon its ue5 switch during the first week of chapter 4
NikosD - Thursday, November 9, 2023 - link
@RyanAlan Wake II that you mentioned in the article, is running at 2 FPS on average using Polaris cards (but without rendering issues) while it's completely broken on Vega, although it achieves decent framerates close or better to RDNA1 performance (Navi series 5000)
This situation could clearly be the case of Primitive Shaders support of Vega in hardware (like Navi cards) that need to be leveraged via drivers for Mesh Shaders support.
Is AMD going to fix Alan Wake II on Vega hardware ?
It's not a matter of performance. The game is completely broken.
Take a look of these videos.
https://community.amd.com/t5/drivers-software/due-...
scineram - Thursday, November 9, 2023 - link
What is there to fix??? Game uses unsupported technologies, RDNA2 minimum listed.NikosD - Thursday, November 9, 2023 - link
No, of course not.As I wrote above, RDNA1 RX 5000 series Navi cards play the game just fine with Primitive Shaders.
Polaris renders the game perfectly and extremely slowly.
Only for Vega is utterly broken.
tipoo - Wednesday, November 8, 2023 - link
Feels like last year that Vega card was released. Who knows where the time goes...Threska - Wednesday, November 8, 2023 - link
It's still a capable card for those without an extreme setup.PurposelyCryptic - Thursday, November 9, 2023 - link
In my head, I still think of my RX480 8GB as that "new" card I got fairly recently. "Fairly recently" being 2016. I had specifically upgraded to it so my PC would be "VR Ready", and I could use the Rift CV.That same Rift CV has been sitting in its box for years now, together with its four cameras, ever since I got the Quest 2 (Which now also has its own successor).
The psychological perception of the passage of time for humans is a very weird subject, especially for those who don't spend much time thinking about it too deeply:
When you were a child, events from one year ago felt more distant than events that happened a decade ago do now. That is because of a combination of factors, but the two big ones are
a) The longer you live, the smaller each additional unit of time becomes in relation to the overall time you've lived. When you were 5 years old, one year represented one fifth of your entire life, and functionally, even less than that, as your first year or two you are two busy acclimating to existing to really retain any compatible memories of that time, afterwards. Even in your twenties, that same year already represents less than 5% of your total existence. So, every new year you experience becomes perceptually shorter than the last, until they just seem to fly right by. If we were immortal, eventually even decades and centuries would feel pretty short. This is also directly related to
b) Your brain is a finite, continuously self-optimizing neural network. As you get older, certain processes are made more efficient, but less complex and flexible, in order to optimize resources, making it harder to learn new things and adapt to new concepts and ways of thinking. The same thing happens with memory - to use computing terminology, your brain continuously refines the lossy compression algorithm it uses to store experiences, keeping only the data judged as most relevant, and culling the rest, much of it before it even has a chance to be stored in the first place. The end result is that the more memories you have, the less detailed they are likely to be. To make up for the missing information when you actively try to recall it, your brain uses heuristic processes to generate approximations on the fly based on related data to fill in the gaps - these approximations are basically no more than informed guesses, and, if they are accurate, that is pretty much just a coincidence. So, you literally remember less details of new events experienced and information learned with every passing year, and older memories are passively optimized as the specific neuronal connections that form them are less frequently innervated, making the links between the individual neurons weaker, and those memories harder to access,
I could go on for literal days on the subject, as it has always fascinated me, and I've been studying it on and off since I started back in college, but no one wants to actually read that, so
TL;DR:
The more years you live, the shorter each additional year feels, and, as far as memories are concerned, actually is, as you store less and less nonessential details to optimize the use of the fundamentally limited resources your brain has.
For example, I know what I had for breakfast today and yesterday, but if you asked me what exactly I ate for each meal this past week, I truthfully wouldn't be able to tell you. It gets worse, too: if you asked me what I did over the last month, I could tell you the general activities and a few keys points, but actual specific details would be extremely limited. And I'm only just shy of 39.
It's no wonder time feels like it flies by when 90% of it has been compressed down to a vague outline with some bullet points.
...and with that, my ramblings over the perception of time and its neural correlates that no one was remotely interested in is officially over. For now.
GeoffreyA - Friday, November 17, 2023 - link
I enjoyed the rambling, and it's a subject that fascinates me too!Taking this one step further, science has yet to explain the passage of time properly, and some, like Carlo Rovelli, have suggested it has got to do with the mind.
FWhitTrampoline - Wednesday, November 8, 2023 - link
I'm on Linux so Vega and Polaris driver support comes from the MESA drivers and that's supported a lot longer for older hardware on Linux than on Windows. But I have a Laptop with Vega 8CU integrated graphics and that laptop also has a Polaris dGPU and really AMD dropping support for Polaris GPUs for it's ROCm/HIP translation layer(longer ago) has put Blender 3.0/Later editions out of reach there for that Polaris RX560X GPU in the laptop.And that Laptop was from the 2019 Model Year and AMD featured that at CES 2019. But for dGPUs and Laptops in 2019 any Vega Discrete Mobile GPUs had to come with expensive HBM memory and Apple was the only laptop maker using Vega Discrete Mobile so AMD had to use Polaris dGPUs in laptops as late as 2019/2020 until the first RDNA1 discrete mobile GPUs became available.
The Blender Foundation deciding to only support CUDA natively for Blender 3.0 later editions for any non Apple hardware means that Polaris GPUs do not have any current ROCm/HIP support to take that CUDA and convert that to a form that can be run on the Polaris Hardware. And only the Older Blender 2.93/earlier editions support OpenCL as the GPU compute language for Blender's GPU Accelerated Cycles rendering as Blender 3.0/Later have to utilize ROCm/HIP to translate CUDA to a form that can be run on the AMD hardware.
Blender 3D will Default to Cycles rendering on the CPU cores if Blender can not detect any Software/API support for GPU compute on the GPU hardware but that's slower to render the Ray Traced scene and ties all the CPU cores/threads up at 100% until the render is completed so GPUs are much faster for that than CPUs.
kn00tcn - Wednesday, November 8, 2023 - link
maybe i didnt test a serious scene a couple years ago, but my cycles opencl 580 didnt have any useful gain over cycles 2600xHIP says it supports radeon7 on linux, that is vega, and if it's open source then there may still be a chance the same way mesa can enable compute based raytracing on polaris
but if you're rendering on a laptop, isnt that concerning? higher temps, harder if not impossible to replace components