Quoting: mylka

Quoting: svartalf

Quoting: loggfreakdoes this mean AMD can implement GPU physx?

Yes. Someone could also extend OpenCL, etc. versions of it and make it even MORE generic.

doesnt nvidia has a processor thingy in the gpu, which is only for physx? so it isnt just software, but hardware?

Nope. They bought the company which had dedicated silicon before they got bought- which is part of WHY they got bought. Another $150-200 card for physics accel? It didn't sell well. Once NVidia bought them, they supported the old, now utterly deprecated, hardware for a while, but moved it all over to CUDA compute kernels. All but the lowest end cards make the other hardware utterly irrelevant- because it had more moxie than the cards ever imagined of having.

View this comment - View article - View full comments

Quoting: Stebs

Quoting: soulsourceNevertheless, I just recently installed it via the unofficial Gentoo ebuilds [External Link], and I must say, that it indeed ROCks.

Just got me a RX 580 and was sad to learn that I can't use ROCm because ... they think my i7 2600k Sandy Bridge is too old!
So OpenCL on my GPU does not work because of my CPU :huh:
That leaves me with Clover and OpenCL 1.1, worse than with my old Nvidia Card before, not exactly what I was expecting...

Uhm... WHAT? The only real differences there is overall speed of the CPU and the Intel GPU (which isn't getting used...or, rather, I hope not..). There's no architectural or ISA differences that should matter or prevent it from being usable on that CPU.

View this comment - View article - View full comments

Quoting: GuestI think Bullet was going for that, but last I heard it was still experimental. And I had trouble with Mesa's OpenCL implementation, which is a shame.

Yeah, it's a bit of a pain. Maybe someone can get it sorted out, or maybe get an impetus to get someone ON that. This being there as a largely full and complete framework would be a start. Sometimes it's better to work with something more full than "Open Source" to get a goad applied to get something done.

Of course, compute is available through OpenGL and Vulkan, but I don't think they're as complete (for lack of a better term) for a purely compute based solution like OpenCL or CUDA. If the physics was just for gaming, not sure that would matter.

"Complete," isn't a word I'd use for that. OpenCL makes it easier to code for a given GPGPU compute kernel, but saying it's incomplete is not strictly correct. You're doing odd, counterintuitive operations within textures, surfaces, etc. to get compute operations in OpenGL or Vulkan.

But, I'm sure someone will start converting from cuda (if this release includes the gpu acceleration - I've not delved into it, but maybe it's just the CPU side of physx).

Heh. Might want to go over to the linked Blog page.

"Free, Open-Source, GPU-Accelerated"

I think they're giving it ALL out.

View this comment - View article - View full comments

Quoting: orochi_kyoWell, I guess theyll make Raytracking opensource when its not relevant(a gimmick to sell videocards) anymore.

Meh... One could get there if you followed the papers. Patents MIGHT be of concern, but they'd need to have filed those patents within a year of their disclosure of the basics of the whole process.

View this comment - View article - View full comments

Quoting: loggfreakdoes this mean AMD can implement GPU physx?

Yes. Someone could also extend OpenCL, etc. versions of it and make it even MORE generic.

View this comment - View article - View full comments

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/ [External Link]

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.

View this comment - View article - View full comments

Quoting: ShmerlIf you can't substantiate your claims, don't make them in the first place, that way you won't need this condescending tone of "I'm the vet, and see no point to explain things to you user peasants". It looks more like you have no arguments really. But if you make claims, better substantiate them properly, that's the way discussion should be held.

I just did, NUMEROUS times. YOU don't want to listen because it doesn't validate what you feel to be the truth- you dismiss actual facts in evidence like what happened with Quake3:Arena as not being even remotely valid. In a store situation, you're buying a PHYSICAL artifact, so it's **NOT** even remotely analogous like you keep trying to do.

Like I said... I can't help you. Keep telling yourself you're in the right...nobody else will convince you otherwise anyhow.

View this comment - View article - View full comments

Quoting: thykrI really doubt this.
Those few adventurous Windows "gamers" who might have been seduced by the idea of using GNU/Linux instead of Windows, will very quickly realize that their games run like crap in Wine / Proton. They will quickly move back to Windows with their tail between their legs and probably never think about using GNU ever again.

Well, if you're just running the ones they certified, they work in the same quality domain as they do on Windows. It's only when you turn off the "bar" to anything that they've not checked out/fixed that you run into any "problems"- it actually works better than I'd thought it would, all things considered. The quality is faintly better in some situations than the official WINE, in fact- which would be what people would run if they weren't using Proton.

Exactly.
This is also shifting a lot of work onto the Valve's Linux team, which as far as I know is relatively small.

Heh. What if you get an update, while running in Windows that breaks the game? Happens ALL the time, folks. ALL THE TIME. NVidia pushes an update...breaks something...hell, even on Oculus, guys... Windows pushes an update, breaks the game. Game pushes an update, but now you have to wait for NVidia, AMD, Windows, etc. to fix their f-ups. WINE doesn't just magically make it easier to get broken or not. It just IS.

YES, native code is "better", but only sort-of. I can't count how many times a distro broke some of my game ports over the years or, or, or... C'mon. If this is the best you've got...

View this comment - View article - View full comments

Quoting: Sir_DiealotAre you using Proton though? My guess is no. It's probably not that easy to actually use it outside of steam.

Proton's basically a version of WINE with a bit of hooks to more readily use it within Steam and all. Much of it's been upstreamed all the same, so a fresh pull of the latest will get you most of the fun as long as you're skilled at things like many of the current "regular" WINE users are.

View this comment - View article - View full comments

Quoting: Shmerl@svartalf: You didn't elaborate on why such stats method is a problem besides the "user agent can be spoofed" or "what if it's downloaded from Windows" (why would Linux users even do that?), but I assume that you mean, this position is why Feral don't release DRM-free games. I don't find it convincing in the least, since other porters release DRM-free fine. Including Icculus whom you mentioned above.

Why should I- I already had elaborated NUMEROUS examples of why it doesn't work, to whit you said, "Well, don't do that," which is BS and rubbish. I told you but you seem to know it all over an Industry Vet that's gone through all of this. I wandered off, not caring about you. The only reason you're getting the time of day from me right now is to put paid to your remarks once and for all. I've little time for what just came my way- and most everyone WANTS me to be that busy to be blunt.

Your answer? Browser ID downloads didn't work for iD on Quake3:Arena, where you could buy any SKU and "patch" it with a downloaded engine install for the other Two OSes. What in the HECK posseses you to think that it will be better "now"? Carmack still swears that he lost a quarter of a million in royalties off of the Linux side of things. That should be a telling thing...but apparently not.

It's better than nothing- but it doesn't tell you much of anything realistically USEFUL in the industry's eyes. How many times was <x> downloaded for <y> OS? How do you know it's not the same person re-downloading, etc.? Like I said. Your thinking is **WHY** it's been this long in the making. It's why iD's coming of age party for Linux **NEVER HAPPENED** and it doesn't change with the "now"- because you're **NOT** measuring anything USEFUL. Three decades of this sort of thinking and you're not getting any further. It. Doesn't. Work.

View this comment - View article - View full comments