Intel's new discrete GPU will have a focus on Linux gaming

That's a yay from me, go Mesa, go opensource, thanks Intel :D

Best of luck Intel. The competition is needed, and the FOSS support is appreciated.

I'm pretty happy to wait and see how Intel's new GPU turns out first.

You're going to be waiting a while :P

Just go get a AMD card!

I'm pretty happy to wait and see how Intel's new GPU turns out first.

You're going to be waiting a while :P

Just go get a AMD card!

My 980ti doesn't need replacing and won't until after those are released anyway, I'm not in any rush.


Last edited by Liam Dawe on 3 December 2018 at 1:57 pm UTC

I'm pretty happy to wait and see how Intel's new GPU turns out first.

You're going to be waiting a while :P

Just go get a AMD card!

2020 is indeed far away in computer years...

However, and this almost certainly isn't the place for this, I don't see the point of an AMD card. I'm thinking of getting a new desktop and it appears on Linux even the fastest AMD card is slower than the Nvidia 1070 I've been running the past two years without issue. Except if you're very devoted to the idea of FOSS for drivers, and if you're not cost retrained, why go AMD? Again, taking the FOSS vs non-FOSS out of the equation... Honestly wondering.


Last edited by iiari on 3 December 2018 at 2:45 pm UTC

I'm pretty happy to wait and see how Intel's new GPU turns out first.

You're going to be waiting a while :P

Just go get a AMD card!

2020 is indeed far away in computer years...

However, and this almost certainly isn't the place for this, I don't see the point of an AMD card. I'm thinking of getting a new desktop and it appears on Linux even the fastest AMD card is slower than the Nvidia 1070 I've been running the past two years without issue. Except if you're very devoted to the idea of FOSS for drivers, and if you're not cost retrained, why go AMD? Again, taking the FOSS vs non-FOSS out of the equation... Honestly wondering.

Well, one thing to consider is the Wayland issue. Yeah, I know, Wayland is taking forever to dominate, but it is getting used more and more and adoption of this sort of thing tends to accelerate after a certain point, so if we're talking time horizons like 2020 . . .
My understanding of just what the problem is is fuzzy, but I hear Nvidia don't play well with Wayland.


Last edited by Purple Library Guy on 3 December 2018 at 6:02 pm UTC

AMD graphics power to power use ratio is way off. One of the primary reasons I won't go AMD even though I want to. Their APU game looks decent though.

So, I'm definitely keen on seeing this new card. Might be something that can give them something of a boost.

I distinctly remember their LAST attempt at going back into the Discrete market and how lackluster it was overall.

One of the problems was that they didn't try enough R&D experimentation before going for announcing it. They foolishly thought that someone could just field stripped down CPUs (it's doable, just not in the way THEY did it, actually) and never once found out the problem they ran into.

http://libre-riscv.org/3d_gpu/

Here's a bit of a set of musings by part of the RISC-V crowd that run similar to some of the same ones I have had and continue to do so. One of the telling things is this:

nyuzi is a modern "software shader / renderer" and is a replication of the intel larrabee architecture. it explored the concept of doing recursive software-driven rasterisation (as did larrabee) where hardware rasterisation uses brute force and often wastes time and power. jeff went to a lot of trouble to find out why intel's researchers were um "not permitted" to actually put performance numbers into their published papers. he found out why :) one of the main facts that jeff's research reveals (and there are a lot of them) is that most of the energy of a GPU is spent getting data each way past the L2/L1 cache barrier, and secondly much of the time (if doing software-only rendering) you have several instruction cycles where in a hardware design you issue one and a separate pipeline takes over (see videocore-iv below)

Hm... There's a reason they went down in flames. They got broken upon the wheel on bandwidth. Couldn't get the graphics data in and out of the pipelines fast enough with the bus design they made.

Now, if you duplicate something like the AMD Southern Islands with a decent enough rasterizer and peel part of the crap out of AMD's design, they might have something as this is quite in the space AMD and NVidia are currently living in and it might even consume less power.

If they try something Larrabee-ish or something a bit bolder like Adapteva's mesh. coupled with something more like a VideoCore IV, it might have better legs and maybe even hand you other rendering methods than the GL/Vulkan/Metal/DX ones.

There might be other paths. But unless they've been experimenting with Silicon in the form of high-end, high F-Max capable (1GHz or better) class FPGAs or fully taped out silicon, this is sadly going to be another attempt at hyping up a failed attempt to, "stay relevant," in that space for Investors, much like Larrabee before it.

MMM makes sense from intel point of view they are never going to take nvidia place from windows desktop gaming but on linux its much easier task especially if they hit the ground with as good or better performance in gaming.

MMM makes sense from intel point of view they are never going to take nvidia place from windows desktop gaming but on linux its much easier task especially if they hit the ground with as good or better performance in gaming.

Oh, it most certainly does. But the thing is...they claimed this once before.

MMM makes sense from intel point of view they are never going to take nvidia place from windows desktop gaming but on linux its much easier task especially if they hit the ground with as good or better performance in gaming.

Oh, it most certainly does. But the thing is...they claimed this once before.

True promises in interviews are easy time will tell

I'm pretty happy to wait and see how Intel's new GPU turns out first.

You're going to be waiting a while :P

Just go get a AMD card!

2020 is indeed far away in computer years...

However, and this almost certainly isn't the place for this, I don't see the point of an AMD card. I'm thinking of getting a new desktop and it appears on Linux even the fastest AMD card is slower than the Nvidia 1070 I've been running the past two years without issue. Except if you're very devoted to the idea of FOSS for drivers, and if you're not cost retrained, why go AMD? Again, taking the FOSS vs non-FOSS out of the equation... Honestly wondering.

Well, one thing to consider is the Wayland issue. Yeah, I know, Wayland is taking forever to dominate, but it is getting used more and more and adoption of this sort of thing tends to accelerate after a certain point, so if we're talking time horizons like 2020 . . .
My understanding of just what the problem is is fuzzy, but I hear Nvidia don't play well with Wayland.

I know wayland has progressed by still the best way to play games on linux with with xorg yes i know its old as dirt but it works. Wayland needs about another 3 to 5 years of development to reduce its impact and to intergrate better into steam etc. Just my opinion.

While it would be nice to have another player in the discrete GPU market. I really want AMD to make a big leap forward like they've managed on the desktop first. I'd hate to see Intel come along and trounce AMD and Nvidia and then do the same with CPUs. They've hired the architect behind Zen and AMDs GPU guru, so its not out of the realm they could do really really well.

My only concern is that Intel is now going to have to go up against nvidia & amd, and those two have quite the portfolio of patents between them for anything high performance in video cards.

The Patent Cross Licence between AMD and Intel is available on the SEC website. Looks to me like it includes GPUs (the "Processor" definition includes "graphics coprocessor"). Maybe the part of the definition of "AMD Processor" which says "or any AMD Subsidiary (provided such purchased or acquired design was not previously generally publicly available)" excludes things from ATI but as that acquisition was in 2006 and the first GCN cards came out in 2012 there should still be quite a bit that is covered. Some parts of the agreement are omitted from that and filed separately with the SEC to be kept confidential so perhaps something there excludes GPUs.

I'm pretty happy to wait and see how Intel's new GPU turns out first.

You're going to be waiting a while :P

Just go get a AMD card!

2020 is indeed far away in computer years...

However, and this almost certainly isn't the place for this, I don't see the point of an AMD card. I'm thinking of getting a new desktop and it appears on Linux even the fastest AMD card is slower than the Nvidia 1070 I've been running the past two years without issue. Except if you're very devoted to the idea of FOSS for drivers, and if you're not cost retrained, why go AMD? Again, taking the FOSS vs non-FOSS out of the equation... Honestly wondering.

Well, one thing to consider is the Wayland issue. Yeah, I know, Wayland is taking forever to dominate, but it is getting used more and more and adoption of this sort of thing tends to accelerate after a certain point, so if we're talking time horizons like 2020 . . .
My understanding of just what the problem is is fuzzy, but I hear Nvidia don't play well with Wayland.

Wayland use a MESA API to allocate memory, NVIDIA want them to add EGLStreams as well.

Neither want to do what the other asks. *shrug*.

I assume the open source driver, Nouveau, does work with Wayland, but then you lose a lot of the performance advantage from buying NVIDIA in the first place.

I assume the open source driver, Nouveau, does work with Wayland, but then you lose a lot of the performance advantage from buying NVIDIA in the first place.

Can confirm. I'm actually running Arch with GNOME on Wayland using Nouveau on an old laptop of mine. The proprietary driver has been problematic on that machine.

I know wayland has progressed by still the best way to play games on linux with with xorg yes i know its old as dirt but it works. Wayland needs about another 3 to 5 years of development to reduce its impact and to intergrate better into steam etc. Just my opinion.

x.org is still preferable but Wayland is progressing and eventually even old games running through xwayland will probably run just as well on (x)Wayland as on native x.org. This recent merge request is a great step in the right direction for GNOME on Wayland but currently it doesn't help much in games because xwayland applications still seem to be stuck at 60fps but it's something that's being worked on.

I know wayland has progressed by still the best way to play games on linux with with xorg yes i know its old as dirt but it works. Wayland needs about another 3 to 5 years of development to reduce its impact and to intergrate better into steam etc. Just my opinion.

x.org is still preferable but Wayland is progressing and eventually even old games running through xwayland will probably run just as well on (x)Wayland as on native x.org. This recent merge request is a great step in the right direction for GNOME on Wayland but currently it doesn't help much in games because xwayland applications still seem to be stuck at 60fps but it's something that's being worked on.

Yes, once XWayland is in a good state, the issue between the project team and NVidia probably will become more important, unless the Nouveau team can get close in performance. I am led to believe ( no proof, of course) that some of the NVidia performance advantage is due to recognising the executable and adapting the driver code paths to suit it, which will be difficult to match with Nouveau, especially without full documentation.

Do you have any idea how big the performance gap is for your setup? (assuming you are prepared to use the proprietary driver for testing)

Do you have any idea how big the performance gap is for your setup? (assuming you are prepared to use the proprietary driver for testing)

I have no idea. I don't use the laptop for gaming and it's too old to run games anyway. I'm happy as long as it browses the web and plays video, which it does fine with Nouveau. I don't think theres too much performance disparity on old hardware though. The main problem for Nouveau is lack of re-clocking on modern hardware due to Nvidia's refusal to release firmware and documentation to Nouveau developers.