• Hey, guest user. Hope you're enjoying NeoGAF! Have you considered registering for an account? Come join us and add your take to the daily discourse.

Next-Gen PS5 & XSX |OT| Console tEch threaD

Status
Not open for further replies.

TLZ

Banned
They could of course just done a basic sharing of their thoughts. John himself said he was "very curious" about the PS5 in their video when hands on with Xbox. I'd be shocked if they really have nothing to add, whether insight or just opinion.

Richard did an entire 2nd video saying how the Xbox was a revolution in console design. I'm sure they could form a basic discussion or comparison of the 2 machines builds, now that we've seen both torn apart.
John did say in their 1 million subscriber video that they will leave the PS5 talk for another video, which I'm still waiting for..
 

Bo_Hazem

Banned
Spencer: Ray Tracing Support Has Been Spotty so Far; 8K Is Years Away, If Ever, from Being Standard


Yet they carved 8K on it. Hand-carved by Philly, probably, according to this shitty carving and Project Scorpio name. :lollipop_tears_of_joy:


81619272_1045367919182017_3463700805633179648_n_upscaled_image_x4.jpg


Then to erase it

1597706439_Nouveaux-details-interessants-a-linterieur-de-la-Xbox-Series-X.jpg


tenor.gif
 

TLZ

Banned
They were likely constricted to 36 CUs for BC purposes. That means the clocks had to be pushed higher than an APU with more CUs to achieve similar results. The heating solution wouldn't be as crazy if their BC solution didn't force those design choices much like Kinect forced strange Xbox One design choices. I see a lot of parallels, but obviously we'll have to wait a month or so to know for sure how it all works out.
I see what you're saying, but if that was the case, couldn't they have gone with a much bigger CU count, say 72 CUs (don't know if possible), at much lower clocks, @1.11ghz to achieve the same 10.2TF? That way they'd avoid the extra heat from higher clocks I think?
 

GreyHand23

Member
I see what you're saying, but if that was the case, couldn't they have gone with a much bigger CU count, say 72 CUs (don't know if possible), at much lower clocks, @1.11ghz to achieve the same 10.2TF? That way they'd avoid the extra heat from higher clocks I think?

That would be a much larger chip and because prices are based on how many chips you can get from each wafer, this would be a very expensive chip.
 

geordiemp

Member
They were likely constricted to 36 CUs for BC purposes. That means the clocks had to be pushed higher than an APU with more CUs to achieve similar results. The heating solution wouldn't be as crazy if their BC solution didn't force those design choices much like Kinect forced strange Xbox One design choices. I see a lot of parallels, but obviously we'll have to wait a month or so to know for sure how it all works out.

Not really, the 80 CU RDNA2 is leaked at 2.2 Ghz so...

And the 40 CU at a unbelievable 2.5 Ghz.
 
Last edited:
I think XSX GPU portion occupy 47% of the SoC. I thought I read that somewhere after Hot chip presentation

Edit: the GPU portion of XSX is 47% of the 360mm APU, so ~170mm for GPU, that leaves 190mm for everything else. Using same structure PS5 GPU would be 40CUs or around 120mm, assuming everything else is similar to XSX we end up with 311mm SoC so pretty close to this estimate.

In conclusion, I don't expect increased caches on PS5 vs XSX


I guess if they ever talk specifics on the APU we might find out. I think there is at least the added cache for the IO complex right?
 
Last edited:
T

Three Jackdaws

Unconfirmed Member
Interesting tweets from the Xbox official Twitter page, they've adopted a very friendly and pro-consumer interaction style, I like it but it's also interesting that other major companies especially fast food franchises have adopted the same technique in recent years to ensure they are seen as "innocent and friendly" rather than the profiteering billion dollar organisations they are, you can actually learn this stuff in sociology. Genuinely curious as to why Sony has not adopted the same method.

 
I wonder what custom RDNA3 features the ps5 gpu has that limit speed to 2.3Ghz. Since Cerny said the logic functions starts to break down past that point, iirc

They might have just nipped some off the top of the card they based theirs off of, same for MS. Isn't the PS5 2.2Ghz?
 

Nowcry

Member
Thanks for drawing it out - we could just look at the 4 MB L3 cache on CPU cluster, there is no way there is an additional 8 of them, 32 MB would be too much.

However, I think these calculations assume too many things.
The first thing is quite possible that the PS5 CPU is in a single Slide, that is to say that it is not divided in 2. (I think it is called that)

That will save space on the SOC infinity fabric. It also lacks 2 memory controllers since it has 8 memory controllers. The IO controller however will occupy more but not more than 1 DCU. (I still think that the IO controller is based on custom RDNA2CU), the SPUs would be somewhat less than 1 DCU.
From my point of view (I'm speculating) we should count if you want the APU to have 40CU and forget the size of the IO controller and audio. And start the accounts there.

It would be necessary to see how much the ESRAM, ON-RAM CHIP, and other parts of the custom hardware would occupy. I consider it necessary to continue with the RDNA2 cache structure but with the custom hardware and all of this concluding on an L3 of at least 32 Mb to move the 4k textures with a fluency like UE5.
 
Last edited:

Zoro7

Banned
Interesting tweets from the Xbox official Twitter page, they've adopted a very friendly and pro-consumer interaction style, I like it but it's also interesting that other major companies especially fast food franchises have adopted the same technique in recent years to ensure they are seen as "innocent and friendly" rather than the profiteering billion dollar organisations they are, you can actually learn this stuff in sociology. Genuinely curious as to why Sony has not adopted the same method.


In between all the subtle digs everytime Sony shows something new.
 
T

Three Jackdaws

Unconfirmed Member
I wonder what custom RDNA3 features the ps5 gpu has that limit speed to 2.3Ghz. Since Cerny said the logic functions starts to break down past that point, iirc

👇this may help
I think RDNA 2 has been in development for a while now, at the time of the PS5 development it was leaked that RDNA 2 was struggling to go over 2.23 Ghz because there was a breakdown in logic, the same reason Mark Cerny cited.

However (allegedly) in recent months the issue was resolved and now the RDNA 2 cards can clock over 2.23 Ghz, that being said I'm guessing Cerny and his team were still satisfied and happy with the 2.23 Ghz target they achieved.
 
I wonder what custom RDNA3 features the ps5 gpu has that limit speed to 2.3Ghz. Since Cerny said the logic functions starts to break down past that point, iirc
The cause for the limit is power, not some hidden RDNA3 IP. They limited the power to keep the total system power consumption under 200-250W.

Despite the bigness of the PS5, its still faaaar smaller than most PC cases. Not only this, its an APU, not a dGPU. Lots of engineering considerations.
A 40CU dGPU can go as fast as it likes, because it will have much better cooling and airflow as its not compacted into a (relatively) small box.
 

HAL-01

Member
Interesting tweets from the Xbox official Twitter page, they've adopted a very friendly and pro-consumer interaction style, I like it but it's also interesting that other major companies especially fast food franchises have adopted the same technique in recent years to ensure they are seen as "innocent and friendly" rather than the profiteering billion dollar organisations they are, you can actually learn this stuff in sociology. Genuinely curious as to why Sony has not adopted the same method.


Sony went to the Apple school of “shut the hell up and let your product speak”
Microsoft went to the Wendy’s school of “aren’t we so relatable and cool you guys please buy our product”
 

Darklor01

Might need to stop sniffing glue
Funny you say this



Can't help but feel he's talking about MS. They're the only ones who haven't shown any next gen games and don't have any first party launch games.


In his case, that could certainly be.

I just know I've been reading this back and forth for what feels like an eternity now, and even now, knowing what's in these machine's to whatever degree I can understand it, it's just so damn exhausting.
I think, I'm a the point where I want to pick up a controller, sit back, and think..

Yeah, the waiting, everything in between then and now, and the money, yeah.. it was worth it. When I booted up PS4 and XBOX One, I felt that. I felt that with every launch so far. This is the first launch where there has been this fog and cloud over the whole thing until pretty recently.
 

user1337

Member
Interesting tweets from the Xbox official Twitter page, they've adopted a very friendly and pro-consumer interaction style, I like it but it's also interesting that other major companies especially fast food franchises have adopted the same technique in recent years to ensure they are seen as "innocent and friendly" rather than the profiteering billion dollar organisations they are, you can actually learn this stuff in sociology. Genuinely curious as to why Sony has not adopted the same method.



Usually takes a couple of years for Sony social media marketing to get with the new trend. By then it's no longer the new trend.. lol jk.

I guess they are just more controlled and probably have stricter rules to abide by.
 
The relative size difference is because stuff like the GDDR6 memory PHYs are quite large. The G6 PHYs in say for example Navi 10 (5700/XT) take up almost 100mm2 of the 250mm2 total die area. The actual CU's are quite small:
EPJuORtUUAAhx-j
Doing some napkin math on the die shot of Navi 14 gets me about 7mm2 for a DCU, which are 2 CU's conjoined together into a workgroup. Feel free to correct me on those numbers, its just an estimate. Navi 14 is 158mm2 for reference. We know that the RT on RDNA2 has a negligible area cost, so the CU's/DCUs won't be much bigger than Navi 10/14's.

Oberon has 20 DCU's - 7*20mm2 = 140 mm2
Arden has 28 DCU's - 7*28 = 196 mm2.

There's the 40% difference.

Let us not forget the other aspects of a GPU outside of the compute units. There is the render backends, i.e the ROPs. The ROPs we know are configured based on the number of Shader Engines (each contain 2 Shader Arrays) there are. Both Arden and Oberon have 2 Shader Engines (4 Shader Arrays), so both contain 64 ROPs. These will be pretty much exactly the same size for both consoles. Then there is the central GPU command processor, which will likely be pretty much exactly the same for both consoles. Given the slightly different memory config, Arden likely has 25% more L2 cache than Oberon 5 MB vs 4MB. This is an assumption based on the fact that Navi 10 has 4MB L2 cache and Microsoft has told us Arden has 5MB. The L1 Caches are likely to be pretty much the same for Arden and Oberon and thus take up the same amount of space.

Of course both APU's also contain 8 Renoir Zen 2 cores and their associated caches. As both are likely to be virtually identical, they're both the same size, at around 30-40mm2. I must reiterate, this will take up the same die area for both chips.

Then there are the massive GDDR6 PHYs, of which Oberon has 8*32-bit PHYs and Arden has 10*32-bit Phys, and their associated memory controllers. 10 PHYS will only be 25% larger than 8 PHYs. Memory controllers will be marginally larger for Arden vs Oberon.

Then there will be other miscellaneous stuff on each die - the display engine, encoders etc. Each of Oberon and Arden may also contain additional fixed function/custom units which will take up an unspecified amount of die area.

Here's a die diagram of Arden graciously provided by Microsoft:

Series_X_die_breakdown.png


Each black box I've drawn here is a DCU (someone smarter than I can probably make much more accurate annotations). You have 28 boxes, which amounts to 56CU.
We know Oberon (probably) has 40CUs. We know for certainty that the main differences between Arden and Oberon are the number of CU's and the number of Memory PHYs. So in essence, you just need to remove what I've highlighted in red, and boom, you have an approximation of the PS5's APU.

I shouldn't need to say it, but this is a gross oversimplification. In reality there will be additional differences, as I've briefly touched upon above. This is just intended to be a visual aid.
This image should be a good enough representation of what a 56CU vs 40CU actually looks like on a die. The differences is ultimately quite small, because there is a hell of a lot of uncore stuff that is present on the die.
40% more shaders does not at all imply 40% larger. There areas of the APU that will be pretty similar between each console.
The PS5 has a Navi 22 GPU,it won't be long before you can buy the Navi 22 card that has 40 CU's running above 2200 Mhz.
 
Interesting tweets from the Xbox official Twitter page, they've adopted a very friendly and pro-consumer interaction style, I like it but it's also interesting that other major companies especially fast food franchises have adopted the same technique in recent years to ensure they are seen as "innocent and friendly" rather than the profiteering billion dollar organisations they are, you can actually learn this stuff in sociology. Genuinely curious as to why Sony has not adopted the same method.



Sometimes saying nothing speaks volumes.
 

Rea

Member
I just dont get Alex to him testing a last gen 5 year old game in BC mode is more intersting than talking about the tear down of a new console that is just weird.
Screenshot-2020-10-10-143118.png

I understand they have already committed to releasing these videos, but his response is just baffling comes across a bit snobbish. Like this video was 100% fluff, anyway i am sorry for the rant.
This Alex guy is So unprofessional, whole world is waiting for Ps5 tear down video and he called it uninteresting!! Wtf. That video should be interesting for a channel like DF and called themselves a tech analysis. Oh my god!
 

reksveks

Member


Hate the fact that they haven't shown xbox series x games yet but I think you can see that Microsoft are having weekly/bi-weekly marketing drops which might help them. I do wonder what the next embargo is going to reveal:
- xbox series x enhanced games?
- specific cross gen games like yakuza and Dirt 5?

I am personally waiting for the cyberpunk performance reveal between the one x and the series x.
 

HAL-01

Member


Hate the fact that they haven't shown xbox series x games yet but I think you can see that Microsoft are having weekly/bi-weekly marketing drops which might help them. I do wonder what the next embargo is going to reveal:
- xbox series x enhanced games?
- specific cross gen games like yakuza and Dirt 5?

I am personally waiting for the cyberpunk performance reveal between the one x and the series x.

IGN already leaked some yakuza XSX gameplay accidentally, so that’s next on the list along with some other cross gen games

As for Sony, they still have the UI, bloodborne remastered, horizon gameplay and 1 secret unannounced first party to reveal, so we’ll probably see that in a last big event before release
 

reksveks

Member
IGN already leaked some yakuza XSX gameplay accidentally, so that’s next on the list along with some other cross gen games

As for Sony, they still have the UI, bloodborne remastered, horizon gameplay and 1 secret unannounced first party to reveal, so we’ll probably see that in a last big event before release

I think both Sony and Microsoft have one event left (xbox fanfest is the day before launch I think so that's their one not sure it's going to have an immediate sales impact as I think it will be hard to find one out anyways)

But yeah got that Bloodbourne remastered and first party game from Sony still to reveal and then the Initiative's game at some point to reveal from Microsoft.
 
Status
Not open for further replies.
Top Bottom