AMD's TrueAudio uses Tenselica DSP cores, similar to that of Xbone's SHAPE

#1
First off, this doesnt mean that the Xbone GPU is based on the 280X. Just no. It's based on the Bonaire - 260X, just clocked much lower with 2 CUs disabled.

If you go back and watch AMD VP's interview on the three console wins, he explicitly says that AMD has this IP portfolio that nobody else has, CPU, GPU etc. Looks like it doesnt just stop there, the SHAPE audio block in the Xbone uses the same Tensilica DSP cores that are present in the Bonaire GPU.

Previously people might have assumed that the SHAPE was a custom design by MS but after this, it looks like they just re-used the TrueAudio block in Bonaire with possibly some additional customizations (speculation on my part).

From the Xbox One Hot Chips;





From AMD's TrueAudio details;









Interesting that AMD could be using their partnership with MS to bring some of those benefits over to the PC space; Mantle (as per Anandtech) & TrueAudio (?)

edit: excuse the typo in the title. :/
 
#4
I thought this would end up being the case when Dave Baumann confirmed that True Audio was using DSP cores on Beyond3D.

I think this is starting to make more sense now. Mantle and now this seems to make a lot of sense, but who knows.

http://www.gamesindustry.biz/articles/2013-05-28-xbox-one-deal-worth-more-than-USD3-billion-to-amd

AMD's deal to provide the system-on-a-chip for Microsoft's Xbox One is worth more than $3 billion. That information comes from the LinkedIn profile of Bob Feldstein, former AMD vice president of strategic development and current Nvidia vice president of technology licensing. Feldstein left AMD in July of 2012, after handling the deals that found the company's silicon in the Xbox One, PlayStation 4, and Wii U.

"My involvement was focused on business management and supply agreement negotiations. This required the coordination of multiple functional teams within AMD, as well as regular customer meetings with leadership teams responsible for handling the challenges of complex, multi-year deals. This project is valued at [$3 billion]."
 
#8
so basically amd used the money from sony and ms to develop their own chips. Smart guys. They even beat last gen IBM, those have only managed to sell the same technology twice, now AMD will sell, and sell and sell. Together with mantle they can be almost unstoppable, if not for the steambox and apparent strong nvidia involvement there.
 
#13
Is Sony using this on the PS4 or are they using their in-house audio technology from Sony Electronics division?
Not sure where it emanates from, but vgleaks described it.

http://www.vgleaks.com/playstation-4-audio-processor-acp/

Sony put their resources in other areas. A more powerful GPU, 8GB of unified GDDR5, ACE enhancements (8 instead of 2, with a total of 64 Queues), and they're going big on GPGPU. I don't think they put something like this in the PS4, but that doesn't mean the PS4 won't have awesome audio. It will. This is what bkilian had to say about the tensilica and shape work back in august.

http://beyond3d.com/showpost.php?p=1777420&postcount=414

Tensilica is a company that you contract with to design custom silicon. They have pre-purposed blocks you can just drop in, and then modify to your needs. the two scalar and two vector processors in the audio block are modified tensilica processors, where MS added their own instructions to accelerate things like speech and other audio tasks. SHAPE is custom designed by MS.
Strikes me like either MS went to Tensilica personally and AMD maybe adopted that work into their own stuff, or it was work that both AMD and MS looked into together, and each secured some of the chips for their respective products. SHAPE, on the other hand, seems entirely custom designed by MS.
 
#14
I'm really out of the hardware scene. Although I can easily build computers and troubleshoot them, I don't know how developers are supposed to use this tech in PCs.

For consoles, I guess it's just part of the APU.

But, for PCs that already have their own audio chips.

Does this tech basically allow more advanced audio processing, such as stuff like occlusion and bounce in the old Aureal tech?

Or does it just replace the need for a sound card in systems with the video card?
 
#16
I'm really out of the hardware scene. Although I can easily build computers and troubleshoot them, I don't know how developers are supposed to use this tech in PCs.

For consoles, I guess it's just part of the APU.

But, for PCs that already have their own audio chips.

Does this tech basically allow more advanced audio processing, such as stuff like occlusion and bounce in the old Aureal tech?

Or does it just replace the need for a sound card in systems with the video card?
This wont replace the sound card on your PCs. It will help compute the directional audio;

 
#17
http://www.tomshardware.com/reviews/radeon-r9-280x-r9-270x-r7-260x,3635-5.html

The point of TrueAudio is to facilitate more complex sound effects (those HRTFs aren’t computationally free) without burdening the host processor. Today, AMD says that audio gets as much as 10% of a game’s CPU utilization budget, limiting what developers can do. But with TrueAudio, AMD wants to guarantee the availability of real-time processing resources specifically for sound, and regardless of the host CPU you have installed.
Sounds remarkably similar to what bkilian said about the amount of CPU resources that were taken up by audio tasks on the Xbox 360 CPU.

Some other interesting bkilian statements.

http://beyond3d.com/showpost.php?p=1787498&postcount=587

Well, yes, that's the point. The MS cores have a bunch of customized instructions specifically to accelerate audio workloads in general, and the MS speech pipeline in particular. The scalar core has customized instructions to accelerate codec functions. Like I've said multiple times before, this is not a general purpose DSP to be used in high end mixing stations. It's a game console audio engine with a focus primarily on reducing power consumption and offloading audio workloads from the CPU. There's an entire core dedicated just to managing the other cores, doing housekeeping on the audio graph, moving memory around, and keeping the fixed function hardware well utilized, so the CPU doesn't have to (something no desktop audio card supports, as far as I know)
http://beyond3d.com/showpost.php?p=1764893&postcount=223

Any numbers I could give you would be irrelevant. Games are not about FLOPS. SHAPE was created in order to give developers more flexibility in where they want to spend development time. Do you a) spend a bunch of time converting your audio engine over to SHAPE, and then save time on optimizing your physics engine, or b) leave the audio on CPU and spend a bunch of time optimizing the hell out of your physics engine? In the end, you will optimize and fiddle with the engines exactly the amount required to get the effect the designers want.

I like to use the example of the HD DVD player on the 360, since I have personal experience with it. When we started, we were using 100% of the 360, and getting 1 fps on a 24fps H.264 stream. When we finished, we were using 100% of the 360, and getting 24fps on a 24fps H.264 stream. Could we have optimized even more and gotten 30 or 60fps on that stream? Yes. Did we? No. It performed to spec, so we stopped right there.

In theory, the audio hardware in the One can produce results that could not be replicated by the entire 360 CPU. But what about good enough? If you don't use a polyphase SRC, and just go for the standard linear SRC? Now your audio is not quite as good (but you'd be hard pressed to tell most of the time) and you've reduced your CPU requirements a ton.
http://beyond3d.com/showpost.php?p=1765802&postcount=277

Not bad. The Audio processor was originally devised to be able to offload Kinect Audio processing, and the chip designers came to the audio team and said "We have a bunch of extra transistors we can throw in for free, what would you like them to do?" or something close to that. The SHAPE block was the result of that conversation.
Ahh shit, getting called in to work. Pretty much a guarantee I'm going to sleep all day tomorrow when I get in. Adios all.


One last thing: SHAPE isn't a tensilica core. The 4 DSP processors in the other part of the Xbox One's audio block are the Tensilica cores. SHAPE is six fixed function hardware blocks that were custom designed to Microsoft's needs by the same people that designed the Tensilica Cores according to Bkilian.
 
#19
I remember how amateurishly these were presented in AMD conference, streamed in MONO, for Kratos' sake, these quite important advances diluted in awful developer presentations.. I'm looking at you, Lichdom developers.
 
#20
I remember how amateurishly these were presented in AMD conference, streamed in MONO, for Kratos' sake, these quite important advances diluted in awful developer presentations.. I'm looking at you, Lichdom developers.
That was such a terrible, awkward conference. Lichdom was the icing on the cake for me.

Definitely going AMD again when I upgrade, Mantle and TrueAudio being so compatible with Xbone hardware makes me hopeful AMD can rustle some feathers over at Nvidia.
 
#22
I hope this tech became standard in all GPU families from both AMD and Nvidia, because we really need that tech in all games, not only ones financed by AMD. I want this in freaking next Elder Scrolls, its the best tech for this game.

The demo GenAudio was outstanding and people impressions of Lichdom demo with this tech were great too.
 
#24
I think this is why being the manufacturer/designer/whatever of the 3 console's hardware
is an advantage.

They could copy any design Sony/MS/Nintendo made with their consoles and profit from them.

I also suspect that mantle is a derivative of the drivers found in the consoles.
 
#27
I'm extremely hyped, I got no other choice. I take anything that improves audio rendering.

After all these years with bland, generic audio in many titles (both in processing and recording) this new gen might kickstart hardware audio back and possibly encourage developers to invest more into audio immersion and do less prebaked bulls***.

In order to have exact placement in the 3D space TrueAudio needs to have every single coordinate. How much info do we got on binaural rendering like CMSS-3D did with OpenAL/DS3D and creating a 3D soundfield based on the exact positonal info of every of the 128 voices? (contrary to the bland speaker virtualization we use today)
 
#28
Pretty good to see AMD giving audio attention, it's a pity it's limited to 3 GPUs but I'd imagine their next generation of cards will all have it.
 
#29
If the xbox one is so weak, and such a last gen GPU why is everything we are hearing about next gen pc cards, using technology that MS is using??
 
#30
If the xbox one is so weak, and such a last gen GPU why is everything we are hearing about next gen pc cards, using technology that MS is using??
Because it's AMD tech? Were you under the impression that AMD no longer makes GPUs for PCs? This also has zero to do with the power of specific GPUs.
 
#31
If the xbox one is so weak, and such a last gen GPU why is everything we are hearing about next gen pc cards, using technology that MS is using??
Because .. it's based on the Bonaire - 260X, just clocked much lower with 2 CUs disabled.

I havent seen anybody complain that the Xbone GPU is "last-gen", just that its weak.
 
#32
According to this article (attention: it's in German) the Radeon HD 7790 which is technically the same as the new R7-260X (Bonaire GPU) already has the TrueAudio functionality integrated. They're saying that AMD currently plans to activate this feature with an future update of Catalyst Software, but it's not 100% confirmed.

Hopefully they'll do it - would be awesome (especially because I have a 7790 :>)
 
#33
One thing to keep in mind is that Tensilica's business is to sell processors that can be heavily customized. Just because two devices use Tensilica or even the same extensions doesn't mean they have exactly the same capability.
That said, it would be a hell of a coincidence if they weren't the same.
 
#35
According to this article (attention: it's in German) the Radeon HD 7790 which is technically the same as the new R7-260X (Bonaire GPU) already has the TrueAudio functionality integrated. They're saying that AMD currently plans to activate this feature with an future update of Catalyst Software, but it's not 100% confirmed.

Hopefully they'll do it - would be awesome (especially because I have a 7790 :>)
Tech Report confirmed it will be activated with new drivers.
 
#38
http://www.eurogamer.net/articles/digitalfoundry-the-complete-xbox-one-interview

This was mentioned in that article.

Digital Foundry: You talk about having 15 processors. Can you break that down?

Nick Baker: On the SoC, there are many parallel engines - some of those are more like CPU cores or DSP cores. How we count to 15: [we have] eight inside the audio block, four move engines, one video encode, one video decode and one video compositor/resizer.

The audio block was completely unique. That was designed by us in-house. It's based on four tensilica DSP cores and several programmable processing engines. We break it up as one core running control, two cores running a lot of vector code for speech and one for general purpose DSP. We couple with that sample rate conversion, filtering, mixing, equalisation, dynamic range compensation then also the XMA audio block. The goal was to run 512 simultaneous voices for game audio as well as being able to do speech pre-processing for Kinect.

Digital Foundry: There's concern that custom hardware may not be utilised in multi-platform games but I'm assuming that hardware-accelerated functions would be integrated into middlewares and would see wide utilisation.

Nick Baker: Yeah, Andrew can talk about the middleware point but some of these things are just reserved for the system to do things like Kinect processing. These are system services we provide. Part of that processing is dedicated to the Kinect.

Andrew Goossen: So a lot of what we've designed for the system and the system reservation is to offload a lot of the work from the title and onto the system. You have to keep in mind that this is doing a bunch of work that is actually on behalf of the title. We're taking on the voice recognition mode in our system reservations whereas other platforms will have that as code that developers will have to link in and pay out of from their budget. Same thing with Kinect and most of our NUI [Natural User Interface] features are provided free for the games - also the Game DVR.
They use the cores but they say it's mostly reserved for Kinect so I don't know if you will get the same benefits of TrueAudio.
 

Raistlin

Post Count: 9999
#39
Curiously, DailyTech states that the TrueAudio tech is in both consoles.

http://www.dailytech.com/AMD+Soft+L...Programmable+Audio+in+Hawaii/article33449.htm

Versus competitive digital signal processing (DSP) engines, AMD brags that its TrueAudio solution is the most effective for game audio as it works hand in hand with graphical position data, so that it's far simpler for developers to write code to simulate 3D sound.

AMD is rolling out TrueAudio to its higher end R9 family, and to its GPUs in the Xbox One/PS4. This means that most of the console market and a significant chunk of the PC enthusiast markets will have game-optimized 3D audio DSP on tap for the first time.
 
#41
Curiously, DailyTech states that the TrueAudio tech is in both consoles.

http://www.dailytech.com/AMD+Soft+L...Programmable+Audio+in+Hawaii/article33449.htm
I'd guess that True Audio tech is not about the hardware. It's about positional values calculated by shaders and applied to 200 voices of non-player characters (NPCs) processed in real-time (think Skyrim) to the Ultra HD audio standard (22.2) 24 channel audio then squeezed into 7.1 channel audio via a computational approximation. (lifted from your cite with minor editing)

Shape and the audio hardware in the AMD GPUs have at it's base DSPs from Tenselica but according to Microsoft, Shape is a custom design. Likely the same is true for the Sony PS4. I supect that 4k Ultra HD TV standards for audio are supported by AMD, Microsoft and Sony. This would be the common API that they might share.

PS4 Audio Processor (ACP)

Custom Audio Block
Access mediated by OS
Not programmable by game teams
Performs encoding and decoding
Game: 200+ MP3 or equivalent streams
System: Encoding and decoding for voice chat, etc...
Filtering and effects performed by game code
Presumably CPU, but GPGPU is also a possibility
 
#44
http://www.tomshardware.com/reviews/radeon-r9-280x-r9-270x-r7-260x,3635-4.html

Tomshardware's article touches on this some also. OP is def correct then. Xbox One's GPU is almost certainly, from the looks of it anyway, based on Bonaire, Radeon 260X. I'm guessing this means the 7790 that was already released and out there also secretly had this tensilica hardware inside without people knowing? Tomshardware seems to imply just that.
i dont get that impression AT ALL...

isn't SHAPE off-die? there is no correlation between the GPU used in the Xbone's GPU and the existence of SHAPE...
 
#46
I hope this means true multichannel for stereo headsets for every game:


Listen on HEADPHONES !
that video is a clone of this :

http://www.youtube.com/watch?v=IUDTlvagjJA
(virtual barbershop)

Do they suggest that you get this effect in games or?
I wasn't aware that this wasn't in games for the past 15 years already, I had perfect stereo echolocation from footsteps (exact direction as well as distance, they get the effect in this video by moving the object around the microphones, ingame you get the exact same effect by rotating your character - just like you rotate your head irl to hear where exactly a sound is coming from) in counter strike back in 2000

Genuine questions, sounds like a load of fluff to me.

Listen to the virtual barber video guys, the plastic bag over head effect is far more effective than the ruffling of some paper