AMD Offers Mantle For OpenGL-Next, Pushes Mantle To Workstations

[asdfblah]

My only questions are: Will we EVER see Mantle in Linux (and, most importantly, in Mesa)? if Mantle gets included in some OpenGL spec... How many years will pass until we see it in Linux?

[Creak]

My only questions are: Will we EVER see Mantle in Linux (and, most importantly, in Mesa)? if Mantle gets included in some OpenGL spec... How many years will pass until we see it in Linux?

I don't think it will be that long. It will be longer than Windows, for obvious reasons, but not years I think.
As said in a previous post, what's taking so long for Mesa and Radeon developers is just to make up for lost time. As you can see, some 4.5 extensions have already been implemented although the announce have been made only few days ago. But they still have to keep up to 4.4 first.

For Mantle, first, it will take a lot of time to have a proper standard. Largely the time to finish OpenGL 4.5 implementation I think.
Then, this time, open source developers will begin almost at the same time as proprietary developers.
And, last but not least, AMD is behind Mantle and Intel likes it too, those are the two most open source friendly graphics companies.

[justmy2cents]

I'm actually quite amazed that anyone can say anything good about nvidia. Every time I touch their hardware, its a nightmare. Recently, had to deal with a laptop with an nvidia -- installed Fedora on it, display working (nouveau), but performance stank. So installed the blob, blacklisted all the modules that they say you have to blacklist, regenerated the initrd, rebooted, and..... black screen. Tried a few other things, and it ended up that their blob just doesn't bloody work. Back to nouveau and its high temperatures and extreme slowness. Better to show something than nothing, and it really wasn't worth it to waste a lot of my life trying to make that junk work. AMD means that the thing actually *works*, performance is plenty fine, temperatures are nice and low. No screwing around with crap that doesn't work.

now... why would you do that kind of install on fedora? rpmfusion is simple copy/paste away from working nvidia

And, last but not least, AMD is behind Mantle and Intel likes it too, those are the two most open source friendly graphics companies.

Intel likes it? #.#' you got that from... where?

[Creak]

Intel likes it? #.#' you got that from... where?

I added Intel because that's what I gathered reading the other posts. If it's not true, meh, no biggy, my point was more on the fact that AMD is behind Mantle.

[M@yeulC]

Intel likes it? #.#' you got that from... where?

here ("Like" may be a little extrapolated...)

I actually like the work AMD is doing with radeon and the open source community, but I can't say the same about Nvidia. It sums up my opinion quite well.
(Think about tressFX, free sync, AMD fusion, OpenCL, their paid OSS devs, etc... vs gameworks, G-sync, cuda, poor OSS driver, etc...)

[Creak]

here ("Like" may be a little extrapolated...)

Maybe "Intel is interested by the tech" would have been better indeed

[justmy2cents]

here ("Like" may be a little extrapolated...)

I actually like the work AMD is doing with radeon and the open source community, but I can't say the same about Nvidia. It sums up my opinion quite well.
(Think about tressFX, free sync, AMD fusion, OpenCL, their paid OSS devs, etc... vs gameworks, G-sync, cuda, poor OSS driver, etc...)

so, while preaching you like oss, you also love tressfx? http://www.amd.com/en-gb/innovations...logies/tressfx which is DX11 only, not open, and only works on GCN as hw option? and you hate gameworks which works on dx+opengl and works on all nvidia cards. same for mantle, 1os and 1/10 of 1 vendor supported only.

i don't hate either, each vendor has its own place in my heart where i won't reaplce it with anything else: nvidia+blob for games and 3d, amd(oss only, catalyst is... performance cataclysm) for desktop and intel on servers. only thing i might replace is nvidia blob if mesa reaches 4.4 and in that case i'd be willing to spend 2x more on graphic card just to retain perf level.

[justmy2cents]

here ("Like" may be a little extrapolated...)

I actually like the work AMD is doing with radeon and the open source community, but I can't say the same about Nvidia. It sums up my opinion quite well.
(Think about tressFX, free sync, AMD fusion, OpenCL, their paid OSS devs, etc... vs gameworks, G-sync, cuda, poor OSS driver, etc...)

just wanted to add this. mantle is actually less open than cuda
- both controlled by one vendor
- both not open implementation wise
- both only working on one vendor hw
- cuda is crossplatform, mantle is not by amd decision
- cuda at least doesn't fake open api like mantle. if intel couldn't get access before it is out of beta when they requested, how should ordinary mortal then. open api my shinny a$$

the only reason why they proposed this to next GL is to salvage dead horse. no vendor support (mainly amds own fault), not cross platform (again, fault goes to amd) and no real time to establish it self since they actually created 0 playground with their marketing spins

btw, i actually like mantle or at least what it stands for, lowering cpu overhead, better multitasking... just hate how amd succeeded to kill it in one single sweep before it even got to exist

[schmidtbag]

just wanted to add this. mantle is actually less open than cuda
- both controlled by one vendor
- both not open implementation wise
- both only working on one vendor hw
- cuda is crossplatform, mantle is not by amd decision
- cuda at least doesn't fake open api like mantle. if intel couldn't get access before it is out of beta when they requested, how should ordinary mortal then. open api my shinny a$$

You seem to forget that CUDA used to have all of those same restrictions. The only reason CUDA expanded is because it didn't get much attention. The only reason it gained early linux access is because linux is popular for servers and that's what CUDA is most invested in. As I have stated before, Mantle is a proof of concept. It doesn't need to be ported to linux and it doesn't matter if it's going to die off - there's barely enough games in Windows that take advantage of it. AMD doesn't have the resources to port mantle to linux right now and developers aren't showing the interest. If Mantle were released in catalyst right now and was fully functional, I'd bet it would take well over a year before we even hear about a linux title (worth our interest) that'd actually use it.

As I said in an earlier post - Mantle did what it was supposed to do. AMD got what they wanted out of it and they have very little obligation to keep it going. If PS4 and XB1 used Mantle, I think it'd be in AMD's best interest to keep it alive, since it'd be a GREAT way to effectively port games between platforms. But since neither console is interested in using it, they might as well let it fade away.


I'd say you could argue PhysX was a proof of concept on the behalf of CUDA. Physx was a technology by Ageia but nvidia bought them out and implemented the technology in their own GPUs, and formed CUDA out of it. Thanks to PhysX, GPGPU investment would be severely behind today, but very few games actually take advantage of it.

[justmy2cents]

You seem to forget that CUDA used to have all of those same restrictions. The only reason CUDA expanded is because it didn't get much attention. The only reason it gained early linux access is because linux is popular for servers and that's what CUDA is most invested in. As I have stated before, Mantle is a proof of concept. It doesn't need to be ported to linux and it doesn't matter if it's going to die off - there's barely enough games in Windows that take advantage of it. AMD doesn't have the resources to port mantle to linux right now and developers aren't showing the interest. If Mantle were released in catalyst right now and was fully functional, I'd bet it would take well over a year before we even hear about a linux title (worth our interest) that'd actually use it.

As I said in an earlier post - Mantle did what it was supposed to do. AMD got what they wanted out of it and they have very little obligation to keep it going. If PS4 and XB1 used Mantle, I think it'd be in AMD's best interest to keep it alive, since it'd be a GREAT way to effectively port games between platforms. But since neither console is interested in using it, they might as well let it fade away.


I'd say you could argue PhysX was a proof of concept on the behalf of CUDA. Physx was a technology by Ageia but nvidia bought them out and implemented the technology in their own GPUs, and formed CUDA out of it. Thanks to PhysX, GPGPU investment would be severely behind today, but very few games actually take advantage of it.

nah, i could argue something else. gameworks IS cuda based, so cuda somehow can't simply die off as you state. still, nvidia stated they use best tech for the problem in every part of Xworks. hairworks for example uses DirectCompute (not cuda) which is exactly the same thing amd uses in tressfx and can as such work on any amd card

i could also argue about something else "Mantle is a proof of concept". REALLY? show me one statement where they said anything remote to that. i can show you zillion claims how they can compete with dx/gl or how dx is no competition since it will only work with last windows

i could also argue about "AMD doesn't have the resources to port mantle to linux right now". there are 2 possibilities here
- give out specs... ahhh, right... not even intel can get those. they are so open
- if company like amd can't afford few developers for that, maybe it is better to sign up for bankrupcy, or don't they have resources to pull money for filling papers?