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

Status
Not open for further replies.
I haven't been following this thread for months, what's Behind BCP textures making up for 2x the disk speed? How is that even possible?
As always a good feature of XSX but in some moment some fans of the brand take that feature out of proportion.

Is better in compression of texture (is only for textures) than Kraken which is used in PS5, but as Xbox advertised as something never seen
before well you see the consequences.
 
It is just a form of hardware texture compression. Both consoles have some type of compression.
It won't magically make up the difference between consoles.
Exactly especially when come from a company which will release all its games day 1 in PC and I am sure will part of
directx 12 ultimate in some moment as Sampler Feedback is now part of that.
 

tenor.gif
 
My point is BCPack is just a library they improved and renamed again which the last version already exists also in PC, the main change is now they hardware to support it regarding . But people
put crazy because they heard things like 'BCPack will reach and overcome and exceed the bandwidth of PS5' that is a nonsense.

With SF is a similar situation the difference is that is already compatible with Nvidia turing so is not exclusive (SFS is part of SF).
No, I think the way a xbox engineer tweeted suggests it will be in the texture filtering stage, in much the same way the s3_tc/DXT is, but BCpack (which they didn't want to call BCbc, apparently, even though it is a block compressed texture being block compressed). The main difference I would think to this being achieved in a shader, instead of a custom hw texture filter, is that you don't need to use any memory to expand the data to texture sample the inner (BC) texture. The BCpack hw filter will directly recover the RGB texel from the texture sampler using the custom filter on the BCpack data.
 
No, I think the way a xbox engineer tweeted suggests it will be in the texture filtering stage, in much the same way the s3_tc/DXT is, but BCpack (which they didn't want to call BCbc, apparently, even though it is a block compressed texture being block compressed). The main difference I would think to this being achieved in a shader, instead of a custom hw texture filter, is that you don't need to use any memory to expand the data to texture sample the inner (BC) texture. The BCpack hw filter will directly recover the RGB texel from the texture sampler using the custom filter on the BCpack data.

Uhm.. so it does make up for 2x the data speed difference?
 
So according to dealer in his today's podcast, the ps5 is 9.2tf overclocked to 10.3 with workload it will be downclocked to 9.2tf again...


I mean REALLY??
ZPeGJtK.jpg
 
No, I think the way a xbox engineer tweeted suggests it will be in the texture filtering stage, in much the same way the s3_tc/DXT is, but BCpack (which they didn't want to call BCbc, apparently, even though it is a block compressed texture being block compressed). The main difference I would think to this being achieved in a shader, instead of a custom hw texture filter, is that you don't need to use any memory to expand the data to texture sample the inner (BC) texture. The BCpack hw filter will directly recover the RGB texel from the texture sampler using the custom filter on the BCpack data.

But isn't BCPack wasteful if it's already lossy? Meaning the game size is the same, yet you use less quality of the actually available data. In a simpler example, it's like watching 4K content on a 1080p-1440p TV/monitor.

As I understand that both have decompressors, which is the most important, isn't compression also important for saving data as well? Do they both have compression functionality inside as well? Thanks.
 
Last edited:
Navi supports a wide variety of configurations, you adjust front-end size of the SEs as needed. The 5500 has 24 CUs on a single (larger) SE.
Does the size of the CUs vary as well? I mean can they make smaller CUs inside SE to put more of then for more parallel processing? Or is the size of CUs fixed for the architecture?
 
Astro Bot and Bloodborne are too low...

No Ratchet and Clank in that list is a sin. God forgive those bastards.
 
Last edited:
Wait so next week's PS event is for third-parties or am I missing something here? Page 28-29 lists all of the 38 games.
No , I am pretty sure if ps5 gonna have event next week , it will have first party titles as main show and new 3rd parties other than ones mentioned in opm issue
 
Amazing list, nothing new, games like battlefield 6, the sims 5, elder scrolls 6. They don't even know full title and when those games release but hey, let's be hyped!
 
No old bioshock info , also magazine contains nothing new regardless ps5 games , don't know why they doing that , all 38 games they talked about are 3rd parties , and already announced ,

My disappointment is immeasurable, and my day is ruined


Does the size of the CUs vary as well? I mean can they make smaller CUs inside SE to put more of then for more parallel processing? Or is the size of CUs fixed for the architecture?

CU size should be liked to nm process, 7nm for RDNA 2
 
Status
Not open for further replies.
Top Bottom