You're accusing him of being a liar?!He also spoke of a difference of 15%, but now it seems that it will be 30%.
If you are, then please say it clearly.
You're accusing him of being a liar?!He also spoke of a difference of 15%, but now it seems that it will be 30%.
Why exactly? I don't see how that list changes anything. Just to be completely honest, I have all of this leaked data sitting in front of me right now (it's not exactly hard to get), so just to go over some of the points:So yes, while the data derived from the mining may lineup with actual code names used internally at AMD, their usefulness or relevance to the PS5 at this point is dubious at best.
What lol? First of all Matt never said the difference is 15%You're accusing him of being a liar?!
If you are, then please say it clearly.
Most of us aren't in here looking for "a good deal". I know I am not. I want the most powerful console I can realistically get, and I'll happily pay $500 or $600 for that. Saving $100-200 and getting a significantly weaker console is not a "deal" I care for.
Mate, can you please stop using larger fonts to make your points? It's obnoxious.And according to AMD's own whitepaper on RDNA, no, a Tflop is not just a TFLOP when compared to GCN.
Look at these pictures. Check page 21 and 22 of the rdna whitepaper
You will see this as well as a detailed comparison as to why the 9 teraflop RX 5700XT beats the pants off the 13 teraflop Vega 64
"The Radeon RX 5700 XT delivers over 2.5X higher bytes per FLOP, a tremendous improvement that is crucial for the Radeon RX 5700 XT to deliver better performance than the Radeon RX Vega 64, especially on more sophisticated and complex shaders and compute-based rendering systems. "
Even being a smaller GPU RX 5700 XT beats Vega 64 on triangle performance, triangle cull rate, fp16 texturing, L0 bandwidth, l1 bandwidth, l2 bandwidth, total cache bandwidth/flop. In other words, an RDNA teraflop is much more performant than a GCN teraflop.
No way. My comment is not going in that direction, I am just talking about the information we have outside at the moment.You're accusing him of being a liar?!
If you are, then please say it clearly.
that would basically be a Megaton
The information which speaks of a GPU without RT, which PS5 GPU is confirmed to have.No way. My comment is not going in that direction, I am just talking about the information we have outside at the moment.
Why exactly? I don't see how that list changes anything. Just to be completely honest, I have all of this leaked data sitting in front of me right now (it's not exactly hard to get), so just to go over some of the points:
Well, of course not. It doesn't need to be me. It's coming directly from an AMD employee.
- This info is not datamined from drivers.
No, but it is related (potentially) to the APUs that'll be going inside devkits.
- It is not related to devkits
That's true. By 'missing features', I assume he's talking about the lack of ray-tracing tests, but I think that's been discussed to death already.
- It is not complete so I wouldn't read into "missing features." It's not a spec sheet
This is also true, but at this point I'm don't think there can be any huge changes to the design, so I don't know how relevant the date is.
- It is dated June/July and is the same source for info "leaked" around that time
Looks weird tbh
A huge "change" is that PS5 GPU is confirmed to have RT accelerators incorporated into the GPU. The datamined GPU does not. No one knows precisely what Oberon/Ariel are, we just have educated guesses at this point.That's my point. This information is of course factual, but when comparing it to the information we were getting around that time, it doesn't add up. Which is why I believe it's not something we should put too much stock into especially with the full unveiling around the corner (if you believe Klee).
The final controller is probably not going to look like this.
No one has denied _rogame, I always keep the latest information. If I didn't keep thinking Lockhart is dead.The information which speaks of a GPU without RT, which PS5 GPU is confirmed to have.
There are missing pieces here and y'all need to chill, acting like 9TF GPU is confirmed, when multiple reliable sources have said PS5 is close and has double-digit FLOPS.
Im surprised people are now saying Oberon/Ariel isn't PS5, even DF seemed to think it was & that famous 2Ghz number gets thrown around constantly even when talking about a bigger GPU.A huge "change" is that PS5 GPU is confirmed to have RT accelerators incorporated into the GPU. The datamined GPU does not. No one knows precisely what Oberon/Ariel are, we just have educated guesses at this point.
What we can say for sure is that the Playstation 5 console being released in holiday 2020 will differ significantly, in at least the one way listed above.
Probably because it lacks RT on chip? That was in the post you quoted.Im surprised people are now saying Oberon/Ariel isn't PS5, even DF seemed to think it was & that famous 2Ghz number gets thrown around constantly even when talking about a bigger GPU.
That might not be completely accurate though. This leaked data is just a small sample of the (presumably) huge amount of testing being done at the time. It's perfectly possible that any RT testing just wasn't part of this particular leak.A huge "change" is that PS5 GPU is confirmed to have RT accelerators incorporated into the GPU. The datamined GPU does not. No one knows precisely what Oberon/Ariel are, we just have educated guesses at this point.
What we can say for sure is that the Playstation 5 console being released in holiday 2020 will differ significantly, in at least the one way listed above.
Probably just gonna wait until the ps5 pro comes which is going to be the real ps5 for me lol. it seems that maybe klee has outdated spec sheet who knows
Im surprised people are now saying Oberon/Ariel isn't PS5, even DF seemed to think it was & that famous 2Ghz number gets thrown around constantly even when talking about a bigger GPU.
It's funny because people were bragging how amazing that PS5 leak was just a few months ago, 2070 performance for $400-500, which is admittedly great, now there is damage control to say it's actually under powered & can't be the PS5.
Im aware, but Sony could have their own RT/VSR solution, because MS are implying, if not out right stating they have their own solutions too, RDNA 2.0 on Desktop might be unique.Probably because it lacks RT on chip? That was in the post you quoted.
Well it's true, how did we go from Oberon being a monster to being under powered?
Not saying that it isn't related to PS5. What I'm saying is that the datamined GPU we've been discussing lacks a key feature consistent with the PS5 GPU Sony has confirmed to be present IN the PS5 GPU.Im surprised people are now saying Oberon/Ariel isn't PS5, even DF seemed to think it was & that famous 2Ghz number gets thrown around constantly even when talking about a bigger GPU.
Why not wait for the PS6?Probably just gonna wait until the ps5 pro comes which is going to be the real ps5 for me lol. it seems that maybe klee has outdated spec sheet who knows
And yet there's no reason to believe they do.Im aware, but Sony could have their own RT/VSR solution, because MS are implying, if not out right stating they have their own solutions too, RDNA 2.0 on Desktop might be unique.
Mate, can you please stop using larger fonts to make your points? It's obnoxious.
That might not be completely accurate though. This leaked data is just a small sample of the huge amount of testing being done at the time. It's perfectly possible that any RT testing just wasn't part of this particular leak.
PS5 is confirmed to have RT IN the GPU, not as a discrete co-processor (check 2nd Cerny interview from wired). That change would render an APU as non-Navi10. The datamined chip is shown to be Navi10, which does not have RT hardware IN the GPU.Im aware, but Sony could have their own RT/VSR solution, because MS are implying, if not out right stating they have their own solutions too, RDNA 2.0 on Desktop might be unique.
Yea, thanks. He said that he'd consider a 15% difference to be very close, not that there is a 15% difference.
Wait, really? Do you think that's because Visual Concepts now has to split their time between developing the WWE games and NBA games?Unfortunately with 2k, that gif is from nba 2K14 and the visuals have only gotten worse with each subsequent entry.
You're not above anyone on this forum, we are equals. Please refrain from using text in such a way as to infer that your points are more important.Mate, can you please stop trying to tell me how to make my points on a forum? It's obnoxious.
Yep, people need to stop putting words into people's mouths.
kind of.
oberon is too small to be a 40 cu part with a zen 2 cpu, hardware rt and i/o stuff. there is an amd patent out there that suggests a separate chip that contains the i/o stuff and hardware rt cores. so even if oberon is a lame 36 cu part, the hardware rt having dedicated cores could be good news.
maybe.
most likely not.
having the actual specs that i wanted that actually made excited for this gen, double digit tflops, high amounts of ram etc. What got me hyped this gen is to see both ms and sony go all out and not compromise thanks to the available to tech. I mainly play on playstation, so it sad to see sony compromise and go for the afforable route again.What is a PS5 Pro going to do to make it a "real PS5" for you?
Im aware, but Sony could have their own RT/VSR solution, because MS are implying, if not out right stating they have their own solutions too, RDNA 2.0 on Desktop might be unique.
Well it's true, how did we go from Oberon being a monster to being under powered?
In that interview Cerny talked about audio ray tracing, different ballpark.PS5 is confirmed to have RT IN the GPU, not as a discrete co-processor (check 2nd Cerny interview from wired). That change would render an APU as non-Navi10. The datamined chip is shown to be Navi10, which does not have RT hardware IN the GPU.
Even if the memory bus changes (see Navi 12 with HBM), the gpu id changes.
That's not what im talking about, people have been sharing that 20k+ fire strike score from "PS5" for a while & DF even did a video on it, i just find it a bit funny some people think that is now under powered when if Sony can release that for $399, it would be insane value.Your assumption is that people believed Oberon was the actual PS5 devkit name when we had Prospero confirmed for months, but I digress. The real question you should ask is how did we go from PS5 can barely hit 8TF and can't do 2.0Ghz in speed without a heavy heating issue to now 9TF and 2.0Ghz is fine on 36CUs?
That was during the first interview.In that interview Cerny talked about audio ray tracing, different ballpark.
Before they do, Cerny wants to clarify something. When we last discussed the forthcoming console, he spoke about its ability to support ray-tracing, a technique that can enable complex lighting and sound effects in 3D environments. Given the many questions he's received since, he fears he may have been ambiguous about how the PS5 would accomplish this—and confirms that it's not a software-level fix, which some had feared. "There is ray-tracing acceleration in the GPU hardware," he says, "which I believe is the statement that people were looking for."
Holy balls, people need to READ PROPERLY. Read the threadsmarks, read up...In that interview Cerny talked about audio ray tracing, different ballpark.
Given the many questions he's received since, he fears he may have been ambiguous about how the PS5 would accomplish this—and confirms that it's not a software-level fix, which some had feared. "There is ray-tracing acceleration in the GPU hardware," he says, "which I believe is the statement that people were looking for."
And yet people here and on Twitter swear he was lying.That was during the first interview.
Next playstation console is called PlayStation 5, releasing holiday 2020 and supports ray tracing at hardware level
About UI and game installation: On the controller: From PS Blog: Bluepoint are working on a PS5 title: Laura Miele, chief studio officer for EA comment: https://www.wired.com/story/exclusive-playstation-5/www.resetera.com
That was the first interview. The second interview he clearly states that RT is hardware accelerated and is more focused on the visual implementations of it.In that interview Cerny talked about audio ray tracing, different ballpark.
The context of him saying that was answering this question " I trust what you say, so I want to ask if when you said the consoles are "very close" in power did you mean like a 2 teraflop difference or less? " His response " I would consider ~15% difference or less to be very close, depending on other factors of course."Yea, thanks. He said that he'd consider a 15% difference to be very close, not that there is a 15% difference.
In that interview Cerny talked about audio ray tracing, different ballpark.
That was during the first interview.
Next playstation console is called PlayStation 5, releasing holiday 2020 and supports ray tracing at hardware level
About UI and game installation: On the controller: From PS Blog: Bluepoint are working on a PS5 title: Laura Miele, chief studio officer for EA comment: https://www.wired.com/story/exclusive-playstation-5/www.resetera.com
Ah i see, still Ariel/Oberon are suspiciously close to what PS5 needs for BC. Why it has no RT or possible RDNA 2.0 features is odd.That was the first interview. The second interview he clearly states that RT is hardware accelerated and is more focused on the visual implementations of it.
Who?I mean I have them on ignore now, but senjutsusage was going around earlier basically saying that Cerny was lying and that the info from Rogame was more trustworthy than him at this point. That's almost delusional.
Because it's probably not the PS5 chip. He clearly says in the interview. " . "There is ray-tracing acceleration in the GPU hardware,"Ah i see, still Ariel/Oberon are suspiciously close to what PS5 needs for BC. Why it has no RT or possible RDNA 2.0 features is odd.