Yeah, it was fun, and I appreciated the talk. But that conference was marketing as much as tech discussion. Cerney talks alot of hypotheticals which to be fair was self admitted and he wasn't sure if some of the stuff would work beyond cans.
-
Ever wanted an RSS feed of all your favorite gaming news sites? Go check out our new Gaming Headlines feed! Read more about it here.
Dolby Atmos dev talk about Sony 3D audio "Tempest Engine"
- Thread starter GING-SAMA
- Start date
No your right my bad lol... The page needs to be updated, who knows what else is missing.
it's 128... it's on the Dolby Atmos very own website, and for larger systems you can many more channels say a cinema or club for example. but for xbox/windows headphones they've limited it to 32.
Apologies for insinuating you were misinformed. 3D audio is the bed rock of Audio presence in VR so it is widely used. What we currently have in Dolby atmos in gaming is merely an approximation of the experience audio in VR provides. Things are only going to get interesting in this regard nextgen.
Beat me to it.it's 128... it's on the Dolby Atmos very own website, and for larger systems you can many more channels say a cinema or club for example. but for xbox/windows headphones they've limited it to 32.
I've never used VR (on Playstation, nor any other platform) so I can't speak to how its used there. All I know is that in the games I've played that support it present an amazing experience and I was disappointed to find it wasn't a feature that was widely available on Playstation, even as a way to experience Surround Sound (which is mostly what its used for on Xbox and PC since the list of supported games is quite short).
I would imagine that's the case. There's nothing special about what comes out of the speakers, all the magic is happening inside the console.
my headset came with a redemption code for atmos......and i have to say its pretty damn impressive....works very well on headphones.
Marketing? Hypotheticals? The man is literally the lead designer and hardware architect of the PS5. Im sure he's put the hardware into real practice. I have no idea how a developer presentation is marketing. It's ok to have a favored console, XSX is a great machine, but you shouldn't obfuscate details and info on the PS5. It comes across as being insecure about a piece of technology, which frankly, is depressing.
It was clearly marketing to some degree. It was the announcement of the specs and the hypotheticals of what it could do. You can't come out with a decent response and you somehow turn that to console warring? I did not even mention another platform.
Everyone says that Windows Sonic is MS' version of this, but I thought their version of this is Project Acoustics? If you look at the XSX page it lists Project Acoustics as their audio ray tracing that seems to be doing what Sony is talking about with Tempest.
I'm excited for Sony's tech, but we already have amazing audio tech not being fully by game studios
Audio is really underrated, unfortunatelly
Wolfenstein The New Colossus audio mixing is dreadful. And so was The New Order's.
Reviewers havent even mentioned about it, and it was something that was really distracting to the experience. I was baffled.
Cerny said that while visiting the studios, he had a hard time having meetings with sound experts because there werent many out there
I'm worried about the implementation, not about the tech
Audio is really underrated, unfortunatelly
Wolfenstein The New Colossus audio mixing is dreadful. And so was The New Order's.
Reviewers havent even mentioned about it, and it was something that was really distracting to the experience. I was baffled.
Cerny said that while visiting the studios, he had a hard time having meetings with sound experts because there werent many out there
I'm worried about the implementation, not about the tech
My understanding is that Project Acoustics is a way to "render" sound within games while stuff like Dolby Atmos or Windows Sonic is how you experience that rendered sound.
Right now audio takes a back seat to everything else in terms processor and performance budget. Now audio is getting dedicated silicon way faster than they've ever had before. I think that should motivate audio engineers to do amazing things.
That's my understanding as well. Tempest appears to be more advanced but also potentially more work to implement.
While Dolby labs is quick to point out that Atmos can support hundreds of independent objects, they're less inclined to point out that the specific consumer formats that support it have much tighter restrictions. There are hard limits on audio bandwidth over HDMI, and home receivers cannot handle as many objects as theatrical installations. While those limits will go up over time as the design is quite scalable, what people have in their homes will be stuck with the limits designed in when they shipped. When everything is processed in the box there's a whole lot more bandwidth available and some really impressive results are possible.
Likewise, it's entertaining to assert that people are overwhelmed by too many spatially positioned sound sources. After all, that's how the real world works. I am generally not overwhelmed in rainstorms by all the individual droplets, instead I am immersed. Yes, it's quite possible to produce a muddy sound mix by having everything constantly in motion making a lot of noise, but again that's true of the real world. Tasteful application doesn't mean sticking to a small number of sound origins.
Likewise, it's entertaining to assert that people are overwhelmed by too many spatially positioned sound sources. After all, that's how the real world works. I am generally not overwhelmed in rainstorms by all the individual droplets, instead I am immersed. Yes, it's quite possible to produce a muddy sound mix by having everything constantly in motion making a lot of noise, but again that's true of the real world. Tasteful application doesn't mean sticking to a small number of sound origins.
Good points. If this was an article about a sony dev praising the tempest engine it would be filled with doubt but when an atmos dev downplays it's competition its sound information. Does anyone actually think you would get more then PR talk from this?
Last edited:
HRTF, Fast Fourier Transform, Phase... none of that which Cerny talked about is hypothetical, it's fact. You can use a generic HRTF binaural recording to get a rough idea of 3D sound is like... Barbershop on youtube is the most popular. But to get the best experience an accurate personalised measurement of your own HRTF reading is ideal, but again not necessary.
Also you have the stuff he was talking about developing with TVs and moving to larger systems, will be similar to how they adapt DTSx and Dolby Atmos to soundbars, just requires time and testing.
I was thinking the same just yesterday, how many people buy a sound blaster or whatever nowadays? Or is it just whatever you have built into your Mobo?
I think they are specifically talking about Dolby Atmos for headphones , which will fully virtualise a 7.1 setup when 7.1.4.4 data isn't available.
No idea how Atmos for games differs, but I work in Atmos Cinema mode and Home Theater mode for my livelihood. We have 128 channels of audio to play with in both modes. The first 10 are reserved for the first bed, and then the rest of them can be used for additional beds or all objects if we want. It's usually all objects after the first 10. That aspect is the same whether it is cinema or home theater.
anyone who doubts audio is importand should just mute their game while playing...
having better audio will make a difference. games comunicate with picture And sound
I love great sound more than anything in a game. And I can see the tech itself being a game changer. But honestly, not many devs are going to put much of the needed effort in to create an appropriate soundscape in their game, or are they?. And for the consumer the process to get this revolutionary experience seems too inconvenient and too much of a hassle (scan your ears, sit in the sweet spot) to convince the masses, I think.
Yeah, Project Acoustic is a set of tools for rendering ray traced type audio (project triton) for gameengines (unreal and unity are supported).
It may be that the Xbox Series X has hardware acceleration for this tool, they have talked about the SX having a newer audio block.
DolbyAtmos for Headphones and Windows Sonic are 2 HRTF algorithms/processes to place audio in a 3D space for headphones. The Xbox One and PC's both have the hardware to deliver this currently.
On Windows , Sonic currently allows 112 virtual speakers or dynamic moving objects simultaneously.
On windows Dolby Atmos for headphones is 32 (not a hard limit of the format). This is normally used to create a bed of core audio - a 7.1.4.4 speaker setup of 16 virtual speakers, which you hear as if they were physically placed around you., with 16 moving objects or speakers left to fill in gaps or represent sound beyond a typical speaker setup.
Cerny talked about the hardware of the chip (placing this filter on the audio is expensive)
And that they had made their own algorithm for placing the audio in 3D space (all the HRTF stuff)
What the Ps5 has a pimped up version of what Is available for Pc and Xbox One currently as PS5 supports "hundreds" of virtual objects or speakers.
It's exactly the same, there is just a max object limit of 32 and the bed can be speakers underneath the listener (I'm guessing that isn't used in the real speaker setups)
When Mark Cerny revealed the PS4, with its garbage-tier, mobile-ass CPU/GPU, he called it a super-charged PC.
Cerny isn't just a hardware designer, he's also trying to sell you a product. Keep that in mind.
OK yeah that makes sense. The dolby app is pretty good value for what you get on Xbox and PC, and it's better than buying a gaming headset, since you can use your own headphones with it.
They do:
"Developers will be able to easily leverage Project Acoustics with Xbox Series X through the addition of a new custom audio hardware block."
Ninja Theory
"We're going to have a dedicated chip to work with audio, which means we finally won't have to fight with programmers and artists for memory and CPU power. We take for granted that graphics are powered by their own video cards. But in audio, we haven't had anything like that. Now we have some power dedicated to us."
Xbox Series X Confirmed to Have Dedicated Audio Chip; Sound Developers "Won't Have to Fight Programmers and Artists for Memory and CPU Power.”
According to Ninja Theory senior sound designer Daniele Galante, the Xbox Series X will have a dedicated audio chip that will help sound designers considerably
wccftech.com
I think that's the case as long as you don't turn on any other processing like dolby or windows sonic. It's how it worked on hellblade, just regular 2 channel audio gave you the 3d audio when you used headphones. YouTube has lots of videos with 3d audio, and the effect is there if you listen with headphones.
Dolby Atmos for headphones also has simulated processing for any game that doesn't official support it, same as Windows Sonic and DTS:X.
The thing about Tempest is we don't know if they plan on doing something similar and if not we don't know whether this is something 3rd party devs will need to implement or if it just works, and if they do have to implement, will they do that instead of the other broader use solutions that work on multiple platforms over this propietary Tempest.
That 32 object limit makes it very much not the same. Home Theater and Cinema modes do up to 118 objects.
Yeah I've experienced some of those. Unfortunately I've usually found them underwhelming, especially when they attempt to make it sound like there's something behind me. Maybe it's something with my hearing.
The best example I've experienced is still the old "virtual barber shop" demo using binaural recording. Still impressive so many years later.
It supports multiple hundreds of objects on any output instead of being capped at 128 on only the most expensive installs. It has more than 100 dedicated GFlops of processing power compared to what is probably dozens.
and if i told you XSX audio processor chip has more than 5x as much processing power as that, what would you say?
p.s. besides purchasing largest audio gaming middleware company Audiokinetic, Sony has also launched their 360 Music format as well, which sounds pretty similar in the end to this and it has been getting rave views:
www.techhive.com
So yes, it is something new and different.
Sony 360 Reality Audio review: This headphone-virtualization system expands music beyond your head
Sony's new immersive audio format sounds fantastic with headphones.
www.techhive.com
So yes, it is something new and different.
From the way cerny worded things it sounds like they are using amd gpu cores stripped of their caches as well as a few things. it is amd tech but very custom. He seemed to want something that works similar to cell from my understanding.
I was under the impression that that's kinda what TrueAudio Next already is.
Dolby is to sound what Sony is to videogames.
They understand acoustics. It also helps that they got history.
No, True Audio was Tensilica DSP but AMD removed then and started using regular CUs.
Yeah, I edited, I meant TrueAudio Next. Wouldn't this just mean that Sony dedicated cut down CUs for it? Essentially replicating the function of the ASICs?
I'd love to know where you're getting detailed info on the XSX audio chip, I can't find any info anywhere.
According to the presentation they customized CU's to more closely resemble PS3's SPU's so it's definitely not the standard AMD solution.
You should relax a tad. Cerny is also trying to sell you on the PS5. He is a very smart man. But he doesnt know more about acoustics than the experts at Dolby. His words are not sacrosanct.
Same applies to Phil Spencer, Kamiya, and whoever else that this place loves deitfy.