PS5 Die Shot has been revealed



bill hader popcorn GIF by Saturday Night Live
 
How am I upset exactly? You asked for a post of someone spouting the PS5 RDNA1 narrative, which you also called bullshit in the same post, and hey, turned out to be yourself. Getting yourself so twisted in these threads lately you can't see the wood for the trees
I don't even understand why he is even in PS5 threads? He just adds negative comments about PS5 and how much better his precious is.
 
i feel sony was referring to their cache scrubbers that might appear in future amd gpu.

not the GE, which feels less advance as SX already
Actually, cache scrubbers looked from Cerny's tsk about a feature that might not be picked up for the desktop cards (possibly there are other cheaper ways to sort this problem for Big Navi and other cards with outrageous bandwidth). :LOL: at your GE comment...
 
i don't uunderstand why people think cache scrubber will come to pc when cerny use it as an example of something specific enough to their customization to be only on ps5.
 
Last edited:
as clueless as i am reading this thread for the last couple of days, i can only come to one conclusion
the ps5 is weak asf and the xbox sX is op but no games to show its true power yet
 
This sounds interesting: (thanks to @LiquidRex )



I forget to added the frontend....

Render Frontend
PS5: RDNA 2?
Series: RDNA

Render Backends
PS5: RDNA
Series: RDNA 2

Compute Units
PS5: RDNA 2
Series: RDNA

Ray-tracing / TMUs
PS5: RDNA 2
Series: RDNA 2


People remember that? Guy got a lot of shit from Xbox fans but...



Yes, was waiting for this, lmao... One thing Cerny was explicit about mentioning was RDNA 2 cu's. We know for a fact thats in the PS5, if nothing else.
 
Last edited:
i don't uunderstand why people think cache scrubber will come to pc when cerny use it as an example of something specific enough in their customization to be only on ps5.
It may be in the next series of RDNA cards. I think that was his point when he said stuff in the PS5 may come to RDNA3 if it works well.
 
Is basically what Locuza and the other are saying
I'm the one saying not this shit again
pls . not this fanboy shit again to mudd the water.
no one saying that ps5 is rdna2 after the dieshot. faster you accept it and faster you will live better
Nothing is changed , people before the dieshot was just not accepting lots of facts
He's back again trying to spread fud. Never learns
 
After further meditation and enlightenment i decided to call PC discrete version 'Vanilla RDNA 2' and PS5's 'Modded RDNA 2'. I think i finally found peace.
 
so basically like the ps5 soft engineer said and that all sony fanboys didn't accept? ...if we had a bet i would have won and also mods after they soft/banned/silenced me from that thread and not someone else .. should revisit the decision ;)
So can you provide the specification for RDNA1.5???
Also provide which other AMD products are RDNA1.5?
Don't post anything else until you provide YOUR PROOF.
You know it's nonsense when rabbid Xbox fangirls like Timdog and Dealer start posting stuff like RDNA1.1
 
I personaly don't find Locuza's twitter posts to be very constructive with all that fixation and speculation with "RDNA1.", "true RDNA2", "Cut down FFU", "True Zen 2", "128-bit Zen 1 FPU" etc. It reminds me of earlier Microsoft PR statements.
Yup, the whole version numbering thing is a mess and kind of useless (and it seems some did not catch that my post trying to give "exact" numbers to the PS5's GPU was a joke).

These machines are mix and match of features that AMD is offering to their clients as base IP.
RDNA 2.0 FeatureRay tracing VRSMesh ShadersMachine LearningPerformance Per WattInfinity Cache
Playstation 5YesNoNoNoYesNo
Xbox Series XYesYesYesYesYes but using RDNA 1.0 clocksNo

Using this metric, I would say the PS5 is RDNA 1.33 and XSX is RDNA 1.66

The XSX might take the lead once devs start utilizing mesh shaders and VRS, but the PS5 is probably doing better at the moment because it's hitting those higher RDNA 2.0 clocks where as MS used those perf per watt gains to get more CUs on the GPU.

Both are lacking Infinity Cache which is the secret sauce and the key to the 6800xt keeping pace with the 3080.

Then there is this: The PS5 higher clocks give it several advantages over the xsx which is probably why we have seen it perform better or the same in several games in the cross gen period.

s0n39Hi.png
The charts and tables with precise features are pretty useful to help us get a proper picture, the one with actual performance numbers is even more important as it tells us comparable graphics task performance on each machines (but we have to keep in mind that each frame is a mix of all of them + game logic and other things, which is why the machines trade blows in most situations).
If that RDNA 1.1 is true then ML is the same.

EufJ0crXYAEvsIg
Interesting, maybe some people assume that ML support is in the "old" version then who knows. None of it seems like what the RTX cards have, but I never looked into the details of what's needed to do ML well.
Since we're talking DMA, Amiga had DMA from '85 already which predates the Sega Genesis by a good margin.
The C64 had DMA access features as well, apparently people use it to stream video from carts these days.
 
info we have is rdna3 will be a refresh focusing on perf/watt
it doesn't ring like a shift to me but maybe it will come with more detail later
AMD probably changed the comercial/marketing name just like GCN.

GCN actually has six versions 1.0, 1.1, 1.2, 1.3, 1.4 and 1.5 but AMD after GCN 1.3 changed to 1, 2, 3, 4 and 5.

RDNA 1.1 could be the internal name that ended RDNA 2 and probably RDNA 3 will be something like RDNA 1.2 or 1.3.

We won't see drastic changes from now until AMD changes the RDNA name.

That is just a guess.
 
Last edited:
Me spreading nonsense? Poor Xbox fan. Since you love PS5 engineer Leonardi


Just some one less RDNA 2 feature, but it has more of them

Your conclusion : between RDNA1 and 2.
He only accepts receipts for certain quotes that help and strengthen his agenda. All other quotes by said engineer are fake news.
 
it's not my conclusion IT IS (proved) between rdna1 and 2 ...with customization added by Sony. If you don't know of what you are talking about ...stop spreading nonsense. at this point you aren't accepting reality If you continue I'll report
If anyone needs reporting it's you.
 
xsx have rdna2 front end with mesh shader and rdna2 cu's
Ok, so either you go with the info that was been posted about the PS5 die shot from the same set of tweets or you dont.

You cant believe so much in one and dismiss the other.
 
RDNA2 is a commercial acronym to simplify the market, otherwise GPUs with completely random features would come out and it would be difficult for the average user to choose.

Cerny did say in his GDC keynote that their collaboration with AMD was going both ways. ""If you see a similar discrete GPU available as a PC card at roughly the same time as we release our console, that means our collaboration with AMD succeeded in producing technology useful in both worlds. It doesn't mean that we at Sony simply incorporated the PC part into our console."

The PS5 Geometry Engine is also likely what's going to the basis for the improved geometry handling of RDNA 3 too, along with other improvements such as changes to the caches.

PS5 simply has a very custom RDNA2 GPU, some off the shelf RDNA2 PC features have been removed and some custom features have been implemented, which cannot be defined as RDNA 1,2 or 3.

You cannot talk about RDNA 1.7 or 1.8, or whatever, it doesn't make any sense. It is simply a very Custom RDNA2 GPU.
Let the games do the talk, and we must wait for the best exclusive games on both consoles to have a more reliable scenario.
 
If you take clock and perf/clock as evidence... not.
So it is a matter if you believe in Kitty or not... the others 3 parts he posted were proved true.
No the clock it would not seem so obvious the Xbox still goes at 1800 + Ghz as a game clock of a rdna2 6800 no ?. nothing said at hotchip being rdna2 if you remember ?
 
Last edited:
I personaly don't find Locuza's twitter posts to be very constructive with all that fixation and speculation with "RDNA1.", "true RDNA2", "Cut down FFU", "True Zen 2", "128-bit Zen 1 FPU" etc. It reminds me of earlier Microsoft PR statements.
I think you would personally find them very constructive if you would read them at full length, together with the context because nearly all of them are stated in a neutral and objective matter.

In regards to "RDNA1" and "true RDNA2", this was my commentary:



I never mentioned "true RDNA2" and I tried to make it clear, that the PS5 and Xbox Series are neither 100% RDNA1 nor RDNA2 technology.
However boths sides integrate huge improvements from RDNA2, so personally I would put them much closer to RDNA2 than RDNA1 but as said, it's not possible to slap a branding on them without it being largely oversimplified and subjective.


"Cut down FPU" is a simple fact, which I didn't try to downplay nor did I overdramatized it.
I do regret my initial wording though, rather than saying "that Sony likely cut down the 256-Bit FP pipes to just 128-Bit", I should have worded it as "Sony might have cut down the 256-Bit FP pipes to just 128-Bit" to include a larger uncertainty.
As "true RDNA2" I never said anything about "true Zen2" but a cut down or "vanilla" version.

In relation to the "128-Bit Zen1 FPU" I made the following tweet:


I don't think any comment reads like a Microsoft or Sony PR statement.
 
Last edited:
If you take clock and perf/clock as evidence... not.
So it is a matter if you believe in Kitty or not... the others 3 parts he posted were proved true.
Now we get into a situation I was gonna post the other day to answer that person with but decided to delete it. Might as well now...

One card has RDNA 2 cu's, but not all the RDNA 2 features, has their own custom for some, doesn't have Infinity Cache

One card has RDNA 1 cu's, but all/support for all the RDNA 2 features, doesn't have Infinity Cache

Wouldnt that make them both custom RDNA 2?

Te me, if the cu's are different between the 2 consoles it could help explain BC and next gen game comparisons.
 
Last edited:
as clueless as i am reading this thread for the last couple of days, i can only come to one conclusion
the ps5 is weak asf and the xbox sX is op but no games to show its true power yet

PS5 slightly weaker based on what the new Hitman game showed. We should see a steady stream of new titles after summer, so that's when the real hardware tests come out. Right now you just seeing next gen patches to older titles.
 
I think you would personally find them very constructive if you would read them at full length, together with the context because nearly all of them are stated in a neutral and objective matter.

In regards to "RDNA1" and "true RDNA2", this was my commentary:



I never mentioned "true RDNA2" and I tried to make it clear, that the PS5 and Xbox Series are neither 100% RDNA1 nor RDNA2 technology.
However boths sides integrate huge improvements from RDNA2, so personally I would put them much closer to RDNA2 than RDNA1 but as said, it's not possible to slap a branding on them without it being largely oversimplified and subjective.


"Cut down FPU" is a simple fact, which I didn't try to downplay nor did I overdramatized it.
I do regret my initial wording though, rather than saying "that Sony likely cut down the 256-Bit FP pipes to just 128-Bit", I should have worded it as "Sony might have cut down the 256-Bit FP pipes to just 128-Bit" to include a larger uncertainty.
As "true RDNA2" I never said anything about "true Zen2" but a cut down or "vanilla" version.

In relation to the "128-Bit Zen1 FPU" I made the following tweet:


I don't think any comment reads like a Microsoft or Sony PR statement.

Thanks for sharing here.
I posted your tweets before and I believe you were being very fair with them... it is because the RDNA 1.1 article put people on fire here.

I want to congrats your dissection on Series X SoC... I hope you do some with PS5 when you have time.
Seems like theses APUs are all mixed techs.

BTW what do you thing about what Kitty leaked in October?

 
Last edited:
It runs about the same as a 6800 game clock, which is RDNA2.
More like PS5 clocks... unless you are using some underclocked RX 6800 as example.

clocks-and-thermals.png


In their test it never go lower than 1974Mhz.

clock-vs-voltage.png


 
Last edited:
"highly modified rdna1 gpu" with their own flavour of rdna2 feature. The ps5 is this guy's
and as we seeing from the soc i been all the time right ?

i didn't say is a rdna1 gpu the difference between rdna1 and a customized gpu that start his customization from rdna1 is enormous and if people don understand i can't help them. The gpu in ps5 is much more closer to rdna2 than rdna1

You literally said in first quoted post it is a RDNA 1 GPU. You were not right all the time. But anyway :

4) We know directly from the ps5 lead graphic engineer mouth that the ps5 gpu is a mix between rdna1 and rdna2 (based on rdna1 with customization to make it more like rdna2)

Absolutely yes. the gpu is very likely based on rdna1 to which they have added a whole series of customizations to make it similar and very close to rdna2.

.till then all facts point out at rdna1 heady modified gpu with most rdna2 features.


xsx have rdna2 front end with mesh shader and rdna2 cu's

You don't like what Yuko Yoshida was right?
Now the backtracking will start.......

Oh yes. LOL
It runs about the same as a 6800 game clock, which is RDNA2.

So, PS5 is RDNA 4 then :/
 
I think you would personally find them very constructive if you would read them at full length, together with the context because nearly all of them are stated in a neutral and objective matter.

In regards to "RDNA1" and "true RDNA2", this was my commentary:



I never mentioned "true RDNA2" and I tried to make it clear, that the PS5 and Xbox Series are neither 100% RDNA1 nor RDNA2 technology.
However boths sides integrate huge improvements from RDNA2, so personally I would put them much closer to RDNA2 than RDNA1 but as said, it's not possible to slap a branding on them without it being largely oversimplified and subjective.


"Cut down FPU" is a simple fact, which I didn't try to downplay nor did I overdramatized it.
I do regret my initial wording though, rather than saying "that Sony likely cut down the 256-Bit FP pipes to just 128-Bit", I should have worded it as "Sony might have cut down the 256-Bit FP pipes to just 128-Bit" to include a larger uncertainty.
As "true RDNA2" I never said anything about "true Zen2" but a cut down or "vanilla" version.

In relation to the "128-Bit Zen1 FPU" I made the following tweet:


I don't think any comment reads like a Microsoft or Sony PR statement.

I apologize, my post was indeed a bit crude and hasty. It was just based on my superficial impression, perhaps not helped by the way your posts were presented (in picemeal) here. Thanks for taking time to clarify your analyse.
 
You literally said in first quoted post it is a RDNA 1 GPU. You were not right all the time. But anyway :










You don't like what Yuko Yoshida was right?


Oh yes. LOL


So, PS5 is RDNA 4 then :/

you posted me saying 10 times the same the ps5 is a rdna1 highly modified gpu....to which many rdna2 features have been added
and you know what? it seem that i was right .and you with John wick and others wasn't ...ask Locuza he know more than we all here
 
Last edited:
More like PS5 clocks... unless you are using some underclocked RX 6800 as example.

clocks-and-thermals.png


In their test it never go lower than 1974Mhz.

clock-vs-voltage.png



Very interesting. I did read about AMD underreporting their clocks. Nice to see some proof.
 
I really wish ignoring a user would also hide other posts replying to that user. This thread would be so much better without a certain person shitting it up.
 
More like PS5 clocks... unless you are using some underclocked RX 6800 as example.

clocks-and-thermals.png


In their test it never go lower than 1974Mhz.

clock-vs-voltage.png


1c96455.png
 
Top Bottom