Who knows but
Plus this post by Sangreal
Makes me think it's definitely Navi10Lite. 00840F40 fits the Oberon E0 criteria.
-
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.
-
We have made minor adjustments to how the search bar works on ResetEra. You can read about the changes here.
PS5 Speculation |OT12| - Aw hell, Transistor's running this again?
- Thread starter Mecha Meister
- Start date
- Status
Plus this post by Sangreal
Makes me think it's definitely Navi10Lite. 00840F40 fits the Oberon E0 criteria.
Meme 😂
knowyourmeme.com
I Showed You My Dick Please Respond
I Showed You My Dick Please Respond refers to several images of unattractive boys with a variation of the text “I showed you my dick, please respond” over them, implying that they’ve recently sent a picture of their penis to a woman and are being ignored. After the pictures spread, the text...
knowyourmeme.com
Not watching that. 😂😂Meme 😂
I Showed You My Dick Please Respond
I Showed You My Dick Please Respond refers to several images of unattractive boys with a variation of the text “I showed you my dick, please respond” over them, implying that they’ve recently sent a picture of their penis to a woman and are being ignored. After the pictures spread, the text...
Actually it doesn't. In the same way, the PS4pro has a "boost mode".
All it suggests is that sony's BC solution is using CU matching to support both PS4 non pro patch games and PS4pro supported games. The improvements would come from the PS5s GPU being able to be clocked almost 100% higher.
Surely, the PS5 running games in boost mode with almost twice the clock speed should have some IQ improvements and framerate improvements. I think their approach, however, suggests that from their internal testing wth the PS4pro, they are aware that some games don't scale well with just throwing more clocks at the problem, hence why they have a system where clocks can be matched to the original.
Endorse.13F8 was Navi 10 Lite indeed, 13F9 is a new ID and you must have seen the question mark in the tweet of KOMACHI which indicates to me that this was speculative. Somebody told me (I never saw it myself) that it was corrected by KOMACHI to "unknown" in a later tweet.
Also thank you to link the Beyond3D post, which underlines that Oberon/Flute is PS5 and perfectly lines up with Github!
Here is something I want to add to the discussion too (reddit):
Reddit - Dive into anything
www.reddit.com
To me this little tidbit (the bolded part of the quite) is a really good case for Ariel (RDNA1) was upgraded to Oberon (RDNA2) because the improved efficiency would allow 2000 Mhz instead of 1800 Mhz and such a change of architecture would cause a new code name. I always wondered why there was that move from Ariel to Oberon and what triggered it. Now we have the highly likely answer.
This also would explain all the stories posted in here from 3 different Dev Kits and that Dev Kit V2 would further improve performance. And indeed just the change RNDA2 would further increase performance as we just learned yesterday -> another IPC (unknown value) and Perf per Watt (50%) increase:
Alleged PS5 dev kits as far as I remember the discussions about it:
This is how I see the news from yesterday.
- Silver Tower: which I assume was Vega based
- V-Shape Dev Kit V1: which I assume now was RDNA1 (Ariel) based
- V-Shape Dev Kit V2: which I assume now is RDNA2 (Oberon) based
As many would say "it all adds up now".
Btw, your avatar of shame, my condolences. Really ugly ....
Edit:
Saw your edit and the addition of some screenshots of _rogame tweets.
In response to that and because you got them from beyond3D I just want to add this from there too:
Baseless Next Generation Rumors with no Technical Merits [post E3 2019, pre GDC 2020] [XBSX, PS5]
So, The best news of the day that no one is talking about (yet) ... With all the power efficiency gains AMD talked about for their GPU and CPU, a SOC with LockHart's rumored specifications would be in the 22 Watt range. Sure would fit with the Surface product line. Does that sound about right?forum.beyond3d.com
Where does what you posted say that?
It's just a link to a site that explains memes, nothing dodgy 😅
sure.They might want to shift the message away from raw numbers. Probably depends on if they are more powerful or not.
i doubt they will do that. They have always disclosed the hardware numbers.
I think even your average consumer would find out about this somehow. That's a red flag if you ask me.
But has or is IPC gains being discussed? I thought the man takeaway from that AMDthingy with regards to RDNA2 was PPW gains. Basically he only thing being discussed here is that at any given clockRDNA requires less power to drive it.
The 50% overall gains were "AMD internal estimates."
RDNA 2 is not only more efficient, but they specifically said it can clock higher due to optimization.
We got new and exciting information yesterday and this thread went to absolute madness. Let's come back to the bets and polls please.
I thought it was interesting that during the Q&A portion of the conference Mark Papermaster said TSMC has been just identifying the process node simply as 7nm and not distinguishing as 7nm+.
With that kind of efficiency gains I'm surprised we still need these form factors for the new consoles. Maybe the SSDs are also adding to the heat output.
With that kind of efficiency gains I'm surprised we still need these form factors for the new consoles. Maybe the SSDs are also adding to the heat output.
Last edited:
Komachi and IDs have said it's Navi 10. Just showing that it is also in GitHub as A0.
That chip referenced is almost certainly RDNA1.
Now the rationale is that different steppings of Oberon (B0 etc) are a different chip with a different architecture. Not sure I buy that.
If the games look the biz, would they care?
I think the 'average consumer' might care more about performance in multiplat games rather than have much awareness of actual raw specs or their (non-)disclosure. But then in come factors like price also. The 'average consumer' probably won't care so much if those multiplats already look great, and if the alternative is notably more expensive to buy for example.
So I'd say the calculus on that - on the effect of an initial non-disclosure of spec - may be a bit complicated. I'm a firm believer in 'the more the merrier', at a given pricepoint, wrt power, but market perception is a trickier puzzle.
(PS: Sony has been a bit stingy about raw specs since the PS3. Vita had hardly any despite being clearly more powerful than its direct competitiors. We got GPU Teraflops and RAM count on the PS4 - but not the same comprehensive spec sheet as on PS3. PS5...? We will see. But there may not be much rhyme or reason to the choice. If Sony has a less powerful machine I don't think they'll be too embarrassed to confirm that, if their price/value position is on the other hand attractive... but maybe they won't know that until later.)
Can you had CU by a stepping? And why change so much time of stepping if this is a final chip. I think there is something more than this.
I like to think people care about transparency. That's why I said they would find out one way or another. It doesn't necessarily have to be about the specs themselves but instead the mere idea that they refuse to release them. Red Flago
And you still got people saying we really don't know it's based off RDNA2. Like, they literally need Sony to spell it out for them, so whatever.
Actually due to the power efficiency of RDNA 2, 36 CUs at 2 GHz is more likely than ever. I think we can clearly assume that 9.2 TFLOPs is the target compute power Sony wants to achieve with the PS5.
Whats on March 18th?
Did we know what Arden was regarding Navi 10 / ... etc. ?
Sony please for our sanity throw us some spec bones...
Regarding the XSX discussion before the spec reveal. 2 times the power was many times doubted that it really means easy math and 12 tf. Regarding next generation RDNA was also said to be RDNA 1.5 at most.
Sony please for our sanity throw us some spec bones...
Regarding the XSX discussion before the spec reveal. 2 times the power was many times doubted that it really means easy math and 12 tf. Regarding next generation RDNA was also said to be RDNA 1.5 at most.
I think we are fairly sure between Komachi and Beyond3D that Oberon A0 is Navi 10. Oberon A0 is also referenced in GitHub.
The assumption now is that the different steppings of Oberon (B0, E0 etc) are different chips rather than simply steppings and have a different RDNA2 architecture.
I am not quite sure if there is any evidence or reason to assume this however.
I think that would resonate with the 'core' - clearly important for the launch year for example, and they can have an impact on word of mouth in general around a system later on with a wider base of customers - but the 'average consumer' otherwise? I think it depends what their eyes are showing them. If it's sufficiently, and even relatively, impressive vs other options, then I don't think that cohort would be concerned much more about it. At least I don't think it would be an overriding concern next to what they're seeing, and things like price.
Now multiplat 'reputation' is a different thing... assuming that typically falls in Xbox's favour, for that not to trip Sony up among a wider cohort, it will depend more on pricing than things like Sony's spec transparency. I agree though that the latter could fuel the perception of a 'confirmation' of what multiplat comparisons are showing.
We've reach the point where TF predictions are changing by the hour! A day of madness on Era!
Sony pls...
Now i have to think of the xbox as Spencer's dong.
MS will stream their GDC content
I think it's too soon for the next gen systems to be using PCIe 4.0 spec nvme's. The ssd's getting that hot seems out of the qurstion6
In terms of Sony not having announced yet I only just remembered that MS announced the XO in May..
Here also is Tom's Hardware listing everything as NAVI10 Lite.
AMD Flute SoC Rumored for Xbox Project Scarlett Sees First Leak
The mysterious AMD Flute system-on-a-chip (SoC) could be powering Microsoft's upcoming Xbox Scarlett console.
www.tomshardware.com
Not only _rogame, but also Komachi and APISAK assumed it was RDNA 1.0 based on the name.
To me, the jury is still out. I need to re-review the AMD naming scheme.
Last edited:
Your core is what's going to carry your platform in the first year and if you fail at enticing your core, your casual market isn't going to follow. No one is going to go out and buy the $400 console that no one wants to play.
Ah that's neat. I thought it was cancelled altogether.
Hopefully we will hear more about DXR 1.1. Seems like it can deliver quite a performance boost according to AMD, as they mentioned low level API performance improvements in their slide.
Is there some proof for this? + we don't know how big rdna2 scale ;)
I don't think Sony will have that problem, at 9Tf, 10Tf, 11Tf, etc. If the price is right and the games impress. Sony's biggest year-one/year-two marketing challenge ever was when they overshot on tech and price, it's hard to think of anything more challenging than that. A slightly weaker PS5, with the right marketing mix (i.e. price, games etc), will be fine.
MS is doing them a favour with Lockhart in this scenario by the way. By definition they're lowering the level of 'acceptability' well below where PS5 will stand. MS , for the sake of their own offerings, can't really argue that XSX is the only desirable next-gen option. If PS5 comes in between the two Xboxs in price, and closer to XSX in perceived 'impressiveness', Sony will be very happy with that positioning. That's an if, but my point is, the calculus in terms of perception - wider perception - is more than the tech in the box.
Last edited:
P
The thing is there would have to be for this to make sense.
if I'm remembering correctly, AMD reported a 50% power efficiency gain from GCN to RDNA 1 with a 25% gain in IPC. Most of that I would have attributed to a node jump down to 7nm, but here we are with RDNA 2 reporting another 50% jump in power efficiency on the same 7nm. There would have to be some kind of IPC to make that large of a jump which is why I'm willing to guess its around the same IPC jump from GCN to RDNA1 which would be 25%
The thing is there would have to be for this to make sense.
if I'm remembering correctly, AMD reported a 50% power efficiency gain from GCN to RDNA 1 with a 25% gain in IPC. Most of that I would have attributed to a node jump down to 7nm, but here we are with RDNA 2 reporting another 50% jump in power efficiency on the same 7nm. There would have to be some kind of IPC to make that large of a jump which is why I'm willing to guess its around the same IPC jump from GCN to RDNA1 which would be 25%
See III-V post below. To me, there have been too many retractions and not enough posting of direct code snippets to conclusively say Oberon is Navi 10.
It would be uncommon, but it's also hard to imagine three different developmental GPUs.
It's also hard to reconcile going all the way to E0 stepping. If this is an AMD issue and a forum poster is worried they can make it in time for launch, then why isn't MS facing the same issue? If it's not an AMD issue, what the hell is Sony changing so much, and why does it jeopardize their launch? A reallocation of the GPU is one way to do that.
I am probably sounding like a broken record now... but here goes (again).
Everyone is entitled to their opinions and theories. At the end of the day, it doesn't really matter. 36CU, 54CU....etc
I say we should all assume we are getting a 36CU GPU for the sake of peace. The worst that happens is that we end up with something better than that.
So for me, we are getting a 36CU GPU in a $399 console. And that's fine by me. The only time I would have a problem is if its anything above $399. Unless it's more than 10TF though.
Everyone is entitled to their opinions and theories. At the end of the day, it doesn't really matter. 36CU, 54CU....etc
I say we should all assume we are getting a 36CU GPU for the sake of peace. The worst that happens is that we end up with something better than that.
So for me, we are getting a 36CU GPU in a $399 console. And that's fine by me. The only time I would have a problem is if its anything above $399. Unless it's more than 10TF though.
- Status