AMD GPU14 Tech Day Event - Sept 25'th

[hoom]

Theoretically NV coming onboard for Mantle is just as good as NV saying at CUDA launch that ATI could theoretically come onboard to CUDA.

Was never going to happen then & if NV doesn't start immediately making positive noises about Mantle then its not going to happen now.

I really do not want to even have to contemplate a return to the Glide/not Glide days where you had to actually check whether a game would run on your GPU.

 

[Kaotik]

Theoretically NV coming onboard for Mantle is just as good as NV saying at CUDA launch that ATI could theoretically come onboard to CUDA.

Was never going to happen then & if NV doesn't start immediately making positive noises about Mantle then its not going to happen now.

I really do not want to even have to contemplate a return to the Glide/not Glide days where you had to actually check whether a game would run on your GPU.

The difference is that AMD would have to pay NVIDIA to support CUDA, NVIDIA doesn't have to pay AMD to support Mantle (if I haven't understood anything wrong)

 

[3dcgi]

So far I see no reason that all the arguments against CUDA don't apply to Mantle.

What's different between CUDA and Mantle is CUDA is effectively a language while Mantle is an API. It's easier to support multiple APIs than to rewrite your entire code base. I don't expect to see any game support only Mantle. Many engines today already support DX and OpenGL.

The difference between Mantle and Glide is in the Glide days there were many graphics vendors and now it's a duopoly for discrete graphics. The other difference is many games license graphics engines so the engine teams can do the extra work and these game teams will get support for free.

 

[Dave Baumann]

Audio stuff, yeah cool but not really a game changer, I'm no audiophile.

This isn't an "audiophile" thing, its an audio immersion thing. Techniques are available to bring more realistic sound within (gaming) environments, but they generally suck up CPU horsepower, to the point where game devs don't do it because it pushes the bar too high. ACP lowers the bar (because it is CPU overhead free and will ultimately be supported on a breadth of devices) for audio processing enabling devs to put audio features in that will increase the level of immersion within the environment. Consider it "Eyefinity for your ears" (and, yes, "Earfinity" was tossed around when considering the brand!).

 

[Dave Baumann]

So AMD organized a broken live stream to announce that at some point in the future there will be a faster GPU. Do I have this right? Did they hire the PR firm that handled the first XBO announcement? What's the point of doing this?

I think, all told the GPU portion was the smaller parts of what was actually announced and discussed today. The simple fact is that if you have a bunch of press in front of you for an NDA session, why not do it?

 

[Dave Baumann]

Interesting comment, though not what we're aiming for in the latter part: https://twitter.com/ID_AA_Carmack/status/383051142755663872

 

[Dave Baumann]

Will devs bother to support multiple API's for PC gaming?

Development has changed. Previously a development house would create a new engine with a new game - maybe some follow ups, but there was a lot of starting from scratch. Now 50%-75% of all games on the PC are covered by a relatively small number of engines; the render engine now is effectively engine middleware. And guess what, the developer of that middleware makes money by supporting multiple different platforms and API's - Frostbite supports PC DX11 (and probably other DX revisions), XBOX 360, PS3, PS4, XBOX One; Unreal supports all of these and iOS, HTML5, etc., etc. Adding another optimized back end can actually be a selling feature of engine developers and the reuse is high - as Johan points out, look at the number of EA titles using Frostbite 3 in the coming year or so.

 

[ninelven]

Interesting comment, though not what we're aiming for in the latter part: https://twitter.com/ID_AA_Carmack/st...51142755663872

I was wondering about that myself...

MS probably doesn't care as much since they already work both sides, but I imagine Sony isn't exactly thrilled...

 

[willardjuice]

Interesting comment, though not what we're aiming for in the latter part: https://twitter.com/ID_AA_Carmack/status/383051142755663872

That's exactly what I was thinking! SteamOS combined with mantle could lead to easy (but "optimized") console ports. You guys should give Gabe a ring. Although they probably will not like being tied to an ihv.

 

[Dave Baumann]

Slide are available here, unfortunately Johan's slides are not in there though.

 

[Raqia]

Will Mantle require a reboot? Will standard multi-tasking via the windows GUI still work? Will it introduce any stability or incompatibility issues? Is the long rumored memory management driver rewrite the same as Mantle?

 

[Dave Baumann]

OpenGL is fully out of the domain of Windows and wholly in the hands of the IHV's, Mantle will be no different in this respect. Memory re-write was already implemented many releases ago.

 

[Scott_Arm]

I'm curious to know if Mantle will see its way onto operating systems other than Windows.

 

[itsmydamnation]

OpenGL is fully out of the domain of Windows and wholly in the hands of the IHV's, Mantle will be no different in this respect. Memory re-write was already implemented many releases ago.

how long has it been in development for?

 

[Davros]

Why cant I see any headphone and speaker outputs on these cards ?
and uncle phil from the fresh prince of bel air runs astound sound - I never expected that

 

[HMBR]

I was wondering about that myself...

MS probably doesn't care as much since they already work both sides, but I imagine Sony isn't exactly thrilled...

how can MS not care!? it means D3D and Windows is loosing importance and something outside their control competing with the Xbox/DX

 

[silent_guy]

I think, all told the GPU portion was the smaller parts of what was actually announced and discussed today.

No kidding!

And that was exactly what was wrong with it. If you're going to make a big hoopla about a live streaming and you're teasing a new GPU, then don't be surprised it people all over will disappointed when a sleep therapist goes on stage to say nothing more than "there will be a new GPU at some undefined point in the future for an undefined price with undefined performance and undefined technical specs and it's going to be great" and then starts meandering for 2 hours about tangential technologies.

Have you actually read the user commentaries on various websites?

 

[Dave Baumann]

Why cant I see any headphone and speaker outputs on these cards ?

TrueAudio is an audio processor, not the delivery mechanism - if you route your sound off the motherboard audio jacks you'll still do that with TrueAudio.

 

[Davros]

Thats fantastic I was worried it was only for people with an av receiver with a hdmi input
does it have to be via mboard jacks ? can it be via soundcard jacks ?
now can we get a guarantee amd will support it and not drop it after a generation or two aka tru-form

 

[3dilettante]

Are there more details forthcoming as to the nature of the TrueAudio DSP?
Is the scenario here similar to the makeup of Xilleon/UVD?

As far as announcements go, kudos to AMD for making a show that it cares about software.
Perhaps that was more the point, and given the mention of financials and the desire to show momentum and corporate ties, more indicative of whom the audience actually was.
I'm sure embedding itself with specific developers and creating an API bypass like Mantle can bring significant dividends, although I wonder what the implications will be from the line in the sand being drawn by manufacturer-specific features or the tight embrace of manufacturer and specific devs/publishers.

On a technical note, perhaps the expanded IOMMU capability is something of an enabler here, now that memory can be shared with a more fully vetted memory protection system in place.
The security and system integrity issues must be more acute if the low-level details are now being exposed.
There's no hiding little architectural oopsies behind the driver.
I hope exposing things to this level doesn't start to add the backwards compatibility drag that happens when you start needing to keep around the low-level quirks exposed in the GPU subsystem and its ISA. In other words, I hope AMD hasn't decided GCN as it stands is a good spot to rest on its laurels.


I guess we'll need to wait until the developer-oriented events shed light on some of this.