Technical investigation into PS4 and XB1 audio solutions *spawn

[Deleted member 13524]

I actually only have a 5.0 setup at the moment too, but especially when watching films, it's quite apparent in some scenes how much an LFE channel is missing. There are some frequencies a speaker can't reproduce effectively. Like just about any shockwave scene in movies where the sub would reproduce a shockwave that you can simply feel, but not necessarely hear, followed by the audible explosion that follows reproduced perhaps by the normal speakers because it's in the mid frequency range.

What speakers/receiver do you have?
If you have that option, bi-amping the front speakers could make a substantial difference. Especially if the speakers have a large, dedicated driver for low frequencies (which isn't uncommon for floorstanding models?).

 

[Phil]

What speakers/receiver do you have?
If you have that option, bi-amping the front speakers could make a substantial difference. Especially if the speakers have a large, dedicated driver for low frequencies (which isn't uncommon for floorstanding models?).

My speakers go down quite low, but they wouldn't be nearly as accurate or powerful as a dedicated subwoofer (and they don't go that low frequency wise)... I don't have a subwoofer by choice, it just hasn't been far up on my priority list just yet. But it will come, eventually. It's just not cheap.

For the record, I have Piega speakers with Rotel equipment. Power is not an issue.

 

[Arwin]

Wouldn't that really depend on the cross-over point that is calibrated? Then again, not everyone has the same set-up and would calibrate things the same. I have 4 fullsize speakers in my set-up and when I do get my sub, the sub will only be taking care of the LFE channel, while the stand-speakers will still take care of the normal bass (not LFE frequencies).

I actually only have a 5.0 setup at the moment too, but especially when watching films, it's quite apparent in some scenes how much an LFE channel is missing. There are some frequencies a speaker can't reproduce effectively. Like just about any shockwave scene in movies where the sub would reproduce a shockwave that you can simply feel, but not necessarely hear, followed by the audible explosion that follows reproduced perhaps by the normal speakers because it's in the mid frequency range.

I know that actually, but I maintain that I just haven't had anything sound acceptably before. I can hook up a .1 of course still whenever I want to, so if one day I hear something that convinces me, I'll go for it. But it seems pretty hard to get something I want.

 

[Cyan]

Yay! After watching and listening to Dead Rising 3 I wonder if Aureal 3d-like sound is coming back in the next generation. The sound was much better than anything I had heard on the Xbox 360 to date.

 

[bkilian]

Mark Cerny has already said it would do the processing for hundreds of audio streams

Mark Cerny was talking about decoding hundreds of MP3 streams. It's probably just a decoder chip, like XMA on the 360. That's the biggest bang for your buck, very cheap to do in hardware, and expensive in software. (I say "just", but it would be almost a requirement for a machine without SPEs to be able to decode hundreds of MP3 channels simultaneously.)

And I really wish people would stop throwing around the GFLOP number for SHAPE its practically meaningless. Its a fixed function device, you can't do anything else with the flops and we have no idea how the number was calculated, nearly all of its cores are integer after all.

To do the majority of the normal audio functions of SHAPE (i.e. none of the kinect stuff) would probably take less then a single jaguar core I'm led to believe.

You're led to believe by whom? And while FLOPS is not accurate, it is useful if you're tying to gauge how much CPU you would be saving by using the fixed function hardware. On a CPU, you would have to use the vector instructions to even approach it, and they're listed in FLOPS. Also, 24bit integer audio is bitwise identical to 32bit float audio, as long as you have a few extra bits available for mixing (which SHAPE does - it uses 28bit internally) So they are comparable. You write a float version of your fixed function algorithm, measure it, and that's how much CPU you're saving with the hardware.

And no, a single jaguar core could not even begin to process as much as the SHAPE block, not counting the AVPs and ASP and ACP, those cores are good, but not that powerful, it's the fixed function stuff that provides most of the systems advantages.

I've seen no one say that a 8 core CPU couldn't reach the same number of voices after all SHAPE does much more then voices, i could believe maybe that a 8 core CPU couldn't do _ALL_ of shape, but I have to see anyone provide any evidence that a CPU couldn't do all of the _NORMAL_ audio processing.

Define "normal" audio processing? Just reading streams off disc, mixing them, and outputting them is not what a normal audio engine does. If that were so, no one would need FMOD or WWISE.

Audio is cheap, the voice + kinect stuff isn't.

Actually, in an average 360 game using voice commands, the "normal" audio processing takes up to 10 times more CPU than the kinect voice pipeline.

I don't expect very many people to make their decision on which version to purchase based off an audio spec and if someone truly thinks that is relevant in the mind of consumers should show where audio has successfully marketed before as key differentiator.

Excellent point, and probably why the audio hardware has not been mentioned publicly except by the FMOD guys.

Anyone got any idea how this shape chip compares to a EMU20K1 dsp (in the xfi)

It is significantly more powerful, by about an order of magnitude.

 

[Silent_Buddha]

It is significantly more powerful, by about an order of magnitude.

Color me giddy with potential excitement. Although I should not get my hopes up too high. It's still up to the developers to actually make use of it.

But if they do, and do it properly, this could actually get me to buy the console.

God how I have dearly missed environmental sound modeling in PC games and console games have never had it in the first place.

Regards,
SB

 

[scently]

Among developers, I think DICE and most racing games dev houses will be happy about the SHAPE. While I am not an audio aficionado, I do enjoy an increase in fidelity in all aspect of gaming when moving from a gen to another.

 

[pjbliverpool]

Color me giddy with potential excitement.

I'd expect nothing less than for it to be significantly more powerful. The X-Fi was launched 8 years ago now at a time when the original Xbox was the most powerful console on the market.

I'm sure there'll be some PC equivilent if this kind of sound takes off in the console world.

 

[Davros]

One thing that has me worried about shape is it only supports the same api as the 360 which would suggest the same features, its all well and good having a ton of power but if its just used for mixing, compression/decompression and sample rate conversion its not really giving us anything new in the area of more lifelife audio

 

[bkilian]

One thing that has me worried about shape is it only supports the same api as the 360 which would suggest the same features, its all well and good having a ton of power but if its just used for mixing, compression/decompression and sample rate conversion its not really giving us anything new in the area of more lifelife audio

No, there is a shape API, independent of the other audio apis. Xaudio 2 will be using shape internally now, but developers can interact with the shape hardware directly if they so choose, which is what FMOD's announcement was about.

 

[TheChosenOne]

http://www.youtube.com/watch?feature=player_detailpage&v=H9x2Ut_R2oU

At about 1:45, in this interview with Brian Flemming of Sucker Punch, he is asked a question of what next gen hardware brings to the table and he mentions the use of "HDR sound" and simulating the way your ears perceive audio in Infamous: Second Son (if I understood correctly).

Would something like that be done with the PS4 audio chip or with GPU compute? Or the CPU? Just wondering if this is a clue as to whether the audio chip in PS4 really is just a decoder chip or perhaps a bit more capable.

(Mod: Put YouTube vid code H9x2Ut_R2oU between

 

[Silent_Buddha]

Among developers, I think DICE and most racing games dev houses will be happy about the SHAPE. While I am not an audio aficionado, I do enjoy an increase in fidelity in all aspect of gaming when moving from a gen to another.

It's not about the fidelity in this case. It's potentially about acoustic modeling. Have sound potentially react to your environment and objects in your environment in ways that you expect they would in real life. As I've said before, it's like the difference between pre-baked lighting and real time lighting. Canned animations or physics based animations.

As I've mentioned before. I'd be happy if something as simple as the Doppler effect on sound was at least somewhat correctly modeled in games so that every time I hear a siren from a police car, ambulance, fire truck, or any other sound that is approaching me or going away from me in game has an accurate sound to it. This one simple thing annoys me every single fracking time I hear it in game.

I'd expect nothing less than for it to be significantly more powerful. The X-Fi was launched 8 years ago now at a time when the original Xbox was the most powerful console on the market.

I'm sure there'll be some PC equivilent if this kind of sound takes off in the console world.

Yup, I'm still really angry with Creative Labs with making EAX closed license starting with EAX 3.0. I can only imagine how good sound could be in games now days if they hadn't done that.

So, yes, I'm hoping that this takes off on Xbox One and revives acoustic modeling in PC games again. But I'm not holding my breath on that unless Microsoft introduces a standard for it in DirectX and then gets hardware manufacturer's to implement it.

Regards,
SB

 

[Cyan]

SHAPE is starting to show how advanced it is. The audio revolution will start in Forza 5, since the sound will be procedural.

http://www.oxm.co.uk/57555/forza-5-gameplay-video-and-interview-turn-10s-audio-revolution

[YT]-zj9O56hLTY[/YT]​

I can't center and show the Youtube video like I usually do -and don't know how to fix it- with screenshots and videos, so I will only give the link.

EDIT: Fixed!

 

[Tap In]

SHAPE is starting to show how advanced it is. The audio revolution will start in Forza 5, since the sound will be procedural.

http://www.oxm.co.uk/57555/forza-5-gameplay-video-and-interview-turn-10s-audio-revolution

[YT]-zj9O56hLTY[/YT]​

I can't center and show the Youtube video like I usually do -and don't know how to fix it- with screenshots and videos, so I will only give the link.

EDIT: Fixed!

wow... can not wait.. the music soundtrack sounds brilliant! they already had the best sounds in race games and this will take it to the next level

 

[Silent_Buddha]

SHAPE is starting to show how advanced it is. The audio revolution will start in Forza 5, since the sound will be procedural.

http://www.oxm.co.uk/57555/forza-5-gameplay-video-and-interview-turn-10s-audio-revolution

Cool, so they mentioned Doppler shift, yay! Now hopefully it's modifying the sound in realtime rather than just being a canned recording of Doppler. Now if only I was still into racing games. Still not enough to convince me to get the game or an Xbox One.

Regards,
SB

 

[Shifty Geezer]

Cool, so they mentioned Doppler shift, yay! Now hopefully it's modifying the sound in realtime rather than just being a canned recording of Doppler.

You just change the playback rate of the sample for Doppler. I can't see anything hard about it.

 

[Shifty Geezer]

SHAPE is starting to show how advanced it is. The audio revolution will start in Forza 5, since the sound will be procedural.

That's not what he said in the video. The audio is same-old sample blending (even using foley tricks like using a recorded lion roar as part of the engine sound). The remark about procedural content was for the music, which is a pretty conventional technique in games. What isn't happening is real-time engine modelling and realistic sound physics. It's just more of the same, as described here, and no audio revolution.

 

[Silent_Buddha]

You just change the playback rate of the sample for Doppler. I can't see anything hard about it.

If that's the case, then it'll be crap sound as usual. Proper Doppler shift has to take into account not only the sound itself but the relative speeds (Doppler shift) and distances (time until sound wave reaches the listener) between the source and listener to properly modify the frequency of the sound wave in a believable manner.

Relatively simple in the case of a siren approaching or receding from a listener. A lot more complex if the listener is in a fast moving vehicle as then all sounds require some amount of Doppler shift for proper audio reproduction. And that isn't even accounting for any potential environmental effects on the audio (not in relation to Forza, just in general).

And even in the relatively simple case, it's still not done in games even though it'd go a long ways towards increasing the immersion of the environment.

Regards,
SB

 

[Shifty Geezer]

That's still easy to do. Base the playback sample rate on the total relative velocities between sound source and listener. As doppler is literally a stretching of the wave over time, and that's what a shift in audio playback is (whenever one sample is played at different pitches, it's just played faster/slower than the original sample rate), it's not difficult to reproduce in game. Okay, for a really fancy effect you should stretch the audio from all the reflected audio as well, but some form of chorus is probably good for simulating that. As i said in my reply though, the actual description of Forza 5's audio in that vid is nothing new, although with more RAM, a much larger sample set can be used to reconstruct audio.

 

[bkilian]

That's still easy to do. Base the playback sample rate on the total relative velocities between sound source and listener. As doppler is literally a stretching of the wave over time, and that's what a shift in audio playback is (whenever one sample is played at different pitches, it's just played faster/slower than the original sample rate), it's not difficult to reproduce in game. Okay, for a really fancy effect you should stretch the audio from all the reflected audio as well, but some form of chorus is probably good for simulating that. As i said in my reply though, the actual description of Forza 5's audio in that vid is nothing new, although with more RAM, a much larger sample set can be used to reconstruct audio.

The SRC provided in most games is normally linear interpolation, which does not give the best result. Possibly why they don't do it often. The SRC the hardware provides is doing a high quality conversion, with much better quality than devs would normally be able to afford in an in-software engine. So, yes, they could get high quality doppler for essentially free.