Baseless Next Generation Rumors with no Technical Merits [post E3 2019, pre GDC 2020] [XBSX, PS5]

Status
Not open for further replies.
But we don't have that in Github leak. There is no article, its pure data dump from Oberon B0 regression test, a chip we know was almost certainly designed for next gen Playstation. It contains 3 modes :

GEN0 - 0.8GHz 176GB/s 18CUs
GEN1 - 0.911GHz 217,6GB/s 36CUs
GEN2 - 2.0GHZ 530GB/s 36CUs

Now, from B0 stepping to current E0 (?), things could have changed. But I seriously, seriously doubt they added additional cluster of 20CUs, new front end and, well, redesigned entire chip. What is far more likely is that they are optimizing the chip which will have to hit high clocks at acceptable TDP and good yields. And will have to do it soon.

2.0GHz Navi in console APU certainly sounds tall order, but given that we have a legit confirmation of such chip existing, it would possibly be for best to take it at face value and assume these revisions are done in order to get most out of it.

Again, I have yet to hear of chip going through 5-6 steppings that resulted in completely different chip. I remember old Intels and Athlons going to G1 revision, but it resulted in higher clocks, lower wattage and less bugs, not additional 2 cores.

There is no way a 9.2 tflops machine (and I highly doubt a console comes with a 2 Ghz GPU), with 36 CUs can hold itself on next gen.
Although I’m speculating, I’m doing so because there is no way PS5 is 36 CU only.
Because if it is... specially if it cannot sustain 2 Ghz, it’s DOA.
 
https://www.eurogamer.net/articles/digitalfoundry-2020-amd-announces-rx-5600-xt-ryzen-4000-mobile

upload_2020-1-8_0-31-50.png

These are of course the mobile cards that makes them suitable for something like a console. Coincidentally RX5700XT is 9.75 TF and coincidentally 5600 XT and 5700 have 36 CUs.
A RRP of $399 for the big card. I have no idea what the profit margins are. But this price is expected to go down a lot near the end of 2020. The hardware most likely should already be finalized. So we can assume that the PS5 may very well be based on RX5700XT....but where is the RT support?
4K should also be a target resolution for next gen. But Sony may be aiming at 1440p with RIS, AMD's answer to NVIDIA's DLSS.
Unless AMD has another unannounced design to release near the end of 2020 or a little before that.
The 36 CU's rumor may be based on a dev kit (new or old?) that uses a variant of these cards with some tweaks.
Which to me it says that most likely we shouldnt expect too many things from Sony. The series X rumors is surprising though. A console hardware with a mobile GPU with 56 CUs which should already be almost finalized? Sounds like too much. Has MS pulled off a megaton "8GB GDDR5" that caught Sony off guard this time ?
 
Last edited:
Again, I have yet to hear of chip going through 5-6 steppings that resulted in completely different chip. I remember old Intels and Athlons going to G1 revision, but it resulted in higher clocks, lower wattage and less bugs, not additional 2 cores.
Volume CPUs tend to have multiple steppings throughout their lifespan, perhaps to make incremental improvements to some manufacturing issue or to fix errata. This makes more sense with a product that's already in volume production, as small improvements can add up over many units, but these later steppings also coincide with the overall maturation of the manufacturing process and refinement of the platform and firmware.
That's giving too much emphasis on what should be an minor update to a chip, and one where there would be diminishing returns as any readily isolated problems would be found and fixed relatively quickly.

A chip climbing into high stepping counts before its release-viable is not considered a good sign, versus a chip that has multiple acceptable steppings over a healthy production run.
Many steppings prior to finding a viable chip is usually a sign of flaws that are being patched as best as possible with layer changes, which is what respins are best used for.
If it's because they're trying to get even higher than what's considered generally acceptable clocks, steppings are usually not as effective a remedy, and generally a sign that there were early design decisions made that have not meshed with the eventual reality.
An example of a substantial improvement would be with the Thoroughbred Athlon cores, which had different CPU revisions in order to get fundamental layout and implementation changes needed for improved clocks.
 
Well, yea, we have to assume by October AMD will release big Navi (rumored to be 80CU part), probably 60CU one as well. I doubt its gonna be 64CU part, think its more likely to be 3 x 10WGPs for Scarlett, as well as big Navis. I am just saying that, compared to last time around, these consoles will be much closer to what AMD provides in high PC market.

As for how feasible 36CUs at 2.0GHz will be next year, I think inability to get those clocks in console are a bit overblown tbh.

5700XT with 40CUs and average frequency of ~1.85GHz is at 204W. 5700 with 36CUs and average frequency of 1750MHz is well down at 162W.

For comparison sake, a GPU found in Xbox One X is just 15W down on full Navi 10 at 191W.

Can AMD deliver such a chip in console releasing in November next year, on improved node (which gives you straight ~10% lower TDP at same clocks without changes to hardware design)? I absolutely think so. Think next years AMDs cards on 7nm will clock well, well above 2.0GHz.
The news from CES and navi chips being redesigned for mobile suggests that basing things on desktop parts would be inaccurate.The fact they got a 56% uplift in those vega chips in the 4800 series chips would apply to navi as well.

AnandTech: The rearchitect of Vega for 7nm has been given a +56% performance increase. Does this mean that there was a lot left on the table with the design for 14/12nm? I’m trying to understand how you were able to pull so much extra performance from a simple process node change.

LS: When we put Vega into a mobile form factor with Ryzen 4000, we learned a lot about power optimization. 7nm was a part of it sure, but it was also a very power optimized design of that architecture. The really good thing about that is that what we learned is all applicable to Navi as well. David’s team put a huge focus on performance per watt, and that really comes out of the mobile form factor, and so I’m pleased with what they are doing. You will see a lot of that technology will also impact when you see Navi in a mobile form factor as well.
 
Volume CPUs tend to have multiple steppings throughout their lifespan, perhaps to make incremental improvements to some manufacturing issue or to fix errata. This makes more sense with a product that's already in volume production, as small improvements can add up over many units, but these later steppings also coincide with the overall maturation of the manufacturing process and refinement of the platform and firmware.
That's giving too much emphasis on what should be an minor update to a chip, and one where there would be diminishing returns as any readily isolated problems would be found and fixed relatively quickly.

A chip climbing into high stepping counts before its release-viable is not considered a good sign, versus a chip that has multiple acceptable steppings over a healthy production run.
Many steppings prior to finding a viable chip is usually a sign of flaws that are being patched as best as possible with layer changes, which is what respins are best used for.
If it's because they're trying to get even higher than what's considered generally acceptable clocks, steppings are usually not as effective a remedy, and generally a sign that there were early design decisions made that have not meshed with the eventual reality.
An example of a substantial improvement would be with the Thoroughbred Athlon cores, which had different CPU revisions in order to get fundamental layout and implementation changes needed for improved clocks.
Informative as ever. Thanks 3dilettante!

The news from CES and navi chips being redesigned for mobile suggests that basing things on desktop parts would be inaccurate.The fact they got a 56% uplift in those vega chips in the 4800 series chips would apply to navi as well.
Yap, thats the point. We can have basic ballpark, but we cannot for sure know if 2.0GHZ is viable or not 18 month after PC cards have already came out.

I think it Navi 10 cards were a bit rushed, but perf per watt of 5700 chip is still very high. If they can deliver improvements in design, as well as N7P ones, it should be fantastic chip in console environment.
 
The news from CES and navi chips being redesigned for mobile suggests that basing things on desktop parts would be inaccurate.The fact they got a 56% uplift in those vega chips in the 4800 series chips would apply to navi as well.

The response was in the context of mobile optimization, including a mobile Navi.
While I'm sure this can have improvements that come back to the desktop, many such optimizations make trade-offs in terms of sustained performance or the upper range of the performance envelope. The improvements tend to become more modest the more the design is pushed.
Or at least I think that's part of the story, since the way that answer reads I'd be asking what AMD was doing for the last decade where power efficiency has been a consistent bugbear for them to now look into optimizing for it.
 
Volume CPUs tend to have multiple steppings throughout their lifespan, perhaps to make incremental improvements to some manufacturing issue or to fix errata. This makes more sense with a product that's already in volume production, as small improvements can add up over many units, but these later steppings also coincide with the overall maturation of the manufacturing process and refinement of the platform and firmware.
That's giving too much emphasis on what should be an minor update to a chip, and one where there would be diminishing returns as any readily isolated problems would be found and fixed relatively quickly.

A chip climbing into high stepping counts before its release-viable is not considered a good sign, versus a chip that has multiple acceptable steppings over a healthy production run.
Many steppings prior to finding a viable chip is usually a sign of flaws that are being patched as best as possible with layer changes, which is what respins are best used for.
If it's because they're trying to get even higher than what's considered generally acceptable clocks, steppings are usually not as effective a remedy, and generally a sign that there were early design decisions made that have not meshed with the eventual reality.
An example of a substantial improvement would be with the Thoroughbred Athlon cores, which had different CPU revisions in order to get fundamental layout and implementation changes needed for improved clocks.
@3dilettante you are a national treasure to this board. The amount of information you know is just unreal.
 
It was already in base consoles, i think. Just marketing 'we have 3d audio (again)'
Will be what AMD's GPU's have to offer, just like then.

I think the audio solution will be different than last gen. PS4 and Xbox One had hardware to do things like convolution reverb etc to alleviate the CPU. I'm expecting PS5 to ship with something similar to Steam Audio and fully accelerated by the GPU. https://valvesoftware.github.io/steam-audio/

Probably the next evolution of AMD's True Audio Next

https://gpuopen.com/gaming-product/true-audio-next/
https://github.com/GPUOpen-LibrariesAndSDKs/TAN
 
Last edited:
It's the majority of games period, regardless of digital or physical medium.

Yup in the majority of games as we move more and more to digital, more and more games will have a stegosaurus tail in sales will numerous spikes in unit sales years after release.

Basically the people that used to buy used, will move more and more to buying retail games on sale after months and years. You can, of course, close your eyes and cover your ears and pretend it isn't happening, but it doesn't stop the trend. It already happened on PC and it's now happening on console.

Regards,
SB
 
@3dilettante you are a national treasure to this board. The amount of information you know is just unreal.

Thank you for the kind words.
I suppose much of it is doing my best to remember what is said by subject experts and participants in many places, especially here over the years.
I hope to do those lessons they've taught me justice.

I think the audio solution will be different than last gen. PS4 and Xbox One had hardware to do things like convolution reverb etc to alleviate the CPU. I'm expecting PS5 to ship with something similar to Steam Audio and fully accelerated by the GPU. https://valvesoftware.github.io/steam-audio/
Sony had an audio engineer discuss the various processors in the PS4 in the context of using them in an audio engine, and he noted that at the time the most he'd accept for GPU audio was the longest-latency reverb effects. In the absence of prioritization and with spiky graphics loads, there was too much risk to count on anything more timely.
The DSP block that was roughly analogous to the original True Audio had a more specific use case for Sony, primarily for decode work and secure services for the platform. The limited performance and operations supported meant there was generally not enough for other uses, and there was apparently some additive latency due to the way programmers needed to go through a secure API. Better latency than the GPU, but still not good enough for music-type games and too limited for an ideal audio pipeline.
That left the Jaguar cores, which seemed decent enough.

There was a video presentation that went with the following slides:
https://www.slideshare.net/DevCentralAMD/mm-4085-laurentbetbeder
In the talk itself, he actually gave some examples of worst-case timings not in the slides themselves. I think it was something like ~30ms for launching an audio wavefront when the GPU was under load. The ACP could get below 10ms, but for the most sensitive games (dance/rhythm games were something like his example) he wanted latency in the single digit ms or sub-ms. Fast response helped with perceived latency, and very fast latency combated accumulated latency--which could make using more than one device type a problem if you were hopping queues.
Unfortunately, I've not seen quantitative numbers for launch latencies since, so what True Audio, then True Audio Next, and the high-priority queues offer today is something of a mystery to me.
One other non-performance limitation that known GPU hardware wouldn't solve was their wide hardware would require batching objects to better utilize them, which ran counter to the ideal programmable audio engine where the developer could put together effects as one wished without needing to learn the low-level batch behavior of the hardware and lose flexibility.
 
The discussion was around whether the 360 gamers that moved to PS4 would return to Xbox or stick with PS and loosing your digital library was mentioned. Was just pointing out that you can get an instant back library as well as the new gen exclusives with game pass and this might soften the blow of losing your PS library.

What would be neat is for MS to do some special offers in Game Pass for new XSX customers where they would have a special catalog of titles they might have missed by skipping XB1. They could also give them extra discounts by linking their competitor account to unlock deals on 3rd party titles they already own on the other system. Remember Game Pass has shown that subscribers still purchase games they play on the service, especially when they can get up to a 20% discount while it's on there.

Tommy McClain
 
It was already in base consoles, i think. Just marketing 'we have 3d audio (again)'
Will be what AMD's GPU's have to offer, just like then.

I disagree with this but it's just speculation on my part based on the first wired interview with Cerny who made a big deal about it and that it is critical for VR and the break out box for PSVR had dedicated audio hardware I think.
 
I disagree with this but it's just speculation on my part based on the first wired interview with Cerny who made a big deal about it and that it is critical for VR and the break out box for PSVR had dedicated audio hardware I think.

That's true, other than 3D audio, the box was a glorified pass through to the TV.
 
Btw wasn't 3D audio already inside Xbox One X?
No. If it was, we'd have games with immersive audio through stereo headphones. Both consoles had 'TrueAudio' DSPs, but neither implemented the HRTF processing needed to produce 3 dimensional audio. Some games may have done something in software - Hellblade supposedly did a good job although I've not found the 3D audio in YT videos at all spatial.

Recap of last gen including the audio we were hoping for (examples of 3D audio)...https://forum.beyond3d.com/threads/...into-ps4-and-xb1-audio-solutions-spawn.54290/

There's a clear generational advance possible for audio where it's been stuck at rudimentary for the past 30 years unless you have a surround system. The pinnacle will be per-user calibration taking in the shape of the ear etc.
 
It was already in base consoles, i think. Just marketing 'we have 3d audio (again)'
Will be what AMD's GPU's have to offer, just like then.
Mark Cerny identified it as a weak point being addressed as explicitly as he was addressing loading times...

The AMD chip also includes a custom unit for 3D audio that Cerny thinks will redefine what sound can do in a videogame. “As a gamer,” he says, “it's been a little bit of a frustration that audio did not change too much between PlayStation 3 and PlayStation 4. With the next console the dream is to show how dramatically different the audio experience can be when we apply significant amounts of hardware horsepower to it.”

The result, Cerny says, will make you feel more immersed in the game as sounds come at you from above, from behind, and from the side. While the effect will require no external hardware—it will work through TV speakers and virtual surround sound—he allows that the “gold standard” will be headphone audio.

There's been a promise of true 3D audio for many years now, so it's high time it finally emerged as a core feature instead of a few rare test-cases and half-way solutions!
 
No. If it was, we'd have games with immersive audio through stereo headphones. Both consoles had 'TrueAudio' DSPs, but neither implemented the HRTF processing needed to produce 3 dimensional audio. Some games may have done something in software - Hellblade supposedly did a good job although I've not found the 3D audio in YT videos at all spatial.

Recap of last gen including the audio we were hoping for (examples of 3D audio)...https://forum.beyond3d.com/threads/...into-ps4-and-xb1-audio-solutions-spawn.54290/

There's a clear generational advance possible for audio where it's been stuck at rudimentary for the past 30 years unless you have a surround system. The pinnacle will be per-user calibration taking in the shape of the ear etc.

I'm hoping some of the advances on the PC gaming side WRT audio makes it to consoles. And I think to an extent there's more interest from AAA developers to investigate more robust audio solutions. One of the key points that Infinity Ward focused on with the last COD was the audio in the game, and they rightly won an award for that work.

But even that pales in comparison to some of the indie work that is done with audio. Post Scriptum is excellent in spots. Escape from Tarkov's audio is sublime although it still has some problems with presenting verticality with stereo headphones. You can close your eyes in that game and basically pinpoint where any audio source is coming from, except WRT the above mentioned problem with verticality (sometimes you can tell, sometimes you can't).

While Dolby Atmos and Windows Sonic for headphones are impressive, they still rely on the audio generated by the games themselves. Hopefully we'll see an increased focus on that by developers next gen, but I've been hoping for that for years and I'm being constantly disappointed.

That said, an excess in CPU power will likely aleve some (but likely not all) of the problems that developers had with focusing more fully on audio. Regardless, we'll have to hope that the game developers take the opportunity to evolve game audio beyond what it has been for over 2 decades now.

Regards,
SB
 
hs0bzyc4fg941.jpg


Another baseless rumor coming from reddit
 
Status
Not open for further replies.
Back
Top