Baseless Next Generation Rumors with no Technical Merits [post E3 2019, pre GDC 2020] [XBSX, PS5]

Status
Not open for further replies.
Again, your scenario is nonsensical. Someone who can understand that VRMs are over designed for an application would have the knowledge to distinguish a unique identifying serial number from a product code. If they were posting that much accurate information about the other parts on the board, Sony would already have the information they needed to hunt down the source. The only thing they’d risk by posting the SoC product code is being proven false by subsequent leaks of the correct product identifier string.
You are assuming far too many things about the guy who posted rather boring details about PS5 dev kit that was sent to developers few weeks after that. If him not posting SOC ID on internet is your way to call it fake, knock your self out. If I remember correctly, back when it leaked you said its impossible that it is true duo to clamshell configuration being a no go for such high clocked RAM.

In any case, as I said few posts before, it literally doesnt even matter. Whatever he posted there, we already have now. Him guessing 256 bit bus, 18Gbps chips and posting about actual APU based dev kit so early that everybody called it fake was most likely lucky guess.
 
Odium is so obviously full of shit it's hilarious.

That said he's kvetching on about March 5, so that's something. Announcement of pre-announcement one presumes. But basically it gives us something concrete to look at/be wrong.
Ok let's wait tomorrow.

But he seems to imply 44CUs@2.06GHz 11.6TF, which can explain the thermal is the very important factor. And 44CUs active (total 48 CUs) is also in line with the observation that PS5 SOC is smaller than xsx.
 
You are assuming far too many things about the guy who posted rather boring details about PS5 dev kit that was sent to developers few weeks after that. If him not posting SOC ID on internet is your way to call it fake, knock your self out. If I remember correctly, back when it leaked you said its impossible that it is true duo to clamshell configuration being a no go for such high clocked RAM.

In any case, as I said few posts before, it literally doesnt even matter. Whatever he posted there, we already have now. Him guessing 256 bit bus, 18Gbps chips and posting about actual APU based dev kit so early that everybody called it fake was most likely lucky guess.

Again, your scenario is nonsensical. Someone who can understand that VRMs are over designed for an application would have the knowledge to distinguish a unique identifying serial number from a product code. If they were posting that much accurate information about the other parts on the board, Sony would already have the information they needed to hunt down the source. The only thing they’d risk by posting the SoC product code is being proven false by subsequent leaks of the correct product identifier string.

The reddit PCB leak is fake in my opinion.

Phison officially stated that they don't receive PS5 orders.

It's unreasonable to assume Phison offers their controller to PS5 devkits since their statement is very clear. No PS5 orders means they are not in the PS5 supply chain.
 
Ok let's wait tomorrow.

But he seems to imply 44CUs@2.06GHz 11.6TF, which can explain the thermal is the very important factor. And 44CUs active (total 48 CUs) is also in line with the observation that PS5 SOC is smaller than xsx.
Where ?
 
The reddit PCB leak is fake in my opinion.

Phison officially stated that they don't receive PS5 orders.

It's unreasonable to assume Phison offers their controller to PS5 devkits since their statement is very clear. No PS5 orders means they are not in the PS5 supply chain.
I agree its unlikely, but Phison specifically denied having a contract for PS5 supply. What they put in dev kits is anyone's guess (I think MS put Nvidias in first dev kits last gen).

Ok let's wait tomorrow.

But he seems to imply 44CUs@2.06GHz 11.6TF, which can explain the thermal is the very important factor. And 44CUs active (total 48 CUs) is also in line with the observation that PS5 SOC is smaller than xsx.
I mean, we can wait for tomorrow, but alot of these "insiders" come up with their info after something has already been hinted at. Take Klee for example, he said people should be hyped about TGA. Well, we had dozens of industry people alluding to great show and Xbox guys directly hinting on something major happening on awards (which would be the case even if XSX wasn't announced).

Result? XSX gets announced and people say Klee knew about it. But he didn't, he wasn't specific at all and Phil Spencer rehearsed his fake speech for it to keep it from leaking.

In any case, 44CUs at 2.0GHz sounds good to me. I think that chip, if it can be kept cool, would be better choice then MS. MS IMO did everything to make sure they cannot be outspecced by putting wider bus and alot of CU's that are relatively conservatively clocked, therefore having ability to counter anything Sony with presumably narrower design could deliver. If they feel confident about 12TF and are out there posting dictionary definitions of TF on official twitter, I would assume they have sizable advantage.
 
Last edited:
Ok let's wait tomorrow.

But he seems to imply 44CUs@2.06GHz 11.6TF, which can explain the thermal is the very important factor. And 44CUs active (total 48 CUs) is also in line with the observation that PS5 SOC is smaller than xsx.

Do you really believe in those speeds?
Yield could be a problem and make that solution as expensive as a larger chip.
Besides, what about the thermal efficiency loss over the years with dust clogged fans? Sony could face a lot of problems and insatisfaction from the fans.
People are speculating wildly, and using just their ass to do it. Nobody knows for shure the PS5 APU size. The spoken values, if true, may had been from an old oberon version.
I’m not saying that Sony didnt f****d up, and that now, she is trying to minimize damage. It’s a possibilty as good as any other.
 
Last edited:
Source?

I've been following his posts for a while and I haven't seen a single post with the claims you mention here.
His claims are: first party Sony developers told him 10.5 to 11.5, and obviously they'll go as high as they can.

I wonder...


Hmmm...Yup. Those goalposts keep on moving all right. :)

Of course since he likely doesn't know any of the specs he leaves himself an out by claiming that Sony might adjust things if heat is an issue. :p Of course then he goes on to claim that high 12+ is possible. :D

Basically make claims so that whatever the speed ends up being he can claim he was correct. 10 TF? He was correct as they had to adjust the speed due to heat. 12.5 TF? He was correct because he mentioned it was possible. Pick a number, he'll be correct regardless. :D Someone could do just as well throwing a dart at a dart board.

Now, it's certainly possible he knows something, but I'm going to remain skeptical. Just like I'm still skeptical of the GitHub numbers.

Regards,
SB
 
Ok let's wait tomorrow.

But he seems to imply 44CUs@2.06GHz 11.6TF, which can explain the thermal is the very important factor. And 44CUs active (total 48 CUs) is also in line with the observation that PS5 SOC is smaller than xsx.
52CU@1.75Ghz = 11.6TF or 56CU@1.62Ghz = 11.6TF, take your pick. Sony would be choosing the one that translates the best balance of cost, heat, and power over 100 million pieces. Going beyond the thermal limit of the chip would not be one of the choices.
 
52CU@1.75Ghz = 11.6TF or 56CU@1.62Ghz = 11.6TF, take your pick. Sony would be choosing the one that translates the best balance of cost, heat, and power over 100 million pieces. Going beyond the thermal limit of the chip would not be one of the choices.
I think that 56CU chip would be to big for 256-bit bus to feed it.

In addition to that, if you are going with single SKU, you will need to keep die size in check. As per AMD, cost per yielded mm2 for 250mm2 chip on 7nm node is almost 2x that of 16nm, so even by releasing ~325mm2 chip (PS4 Pro die size), Sony would be paying at least +50% more per chip then with PS4Pro (thats in case that number got down and didnt' stay in ballpark). To put that in perspective, that chip would still be significantly more expensive then Xbox One X's.

This is why its of utmost importance for MS to release both SKUs, because Arden die size makes it economically unfeasible to release just one. In case of dual SKU strategy, they will have bunch of ~200mm2 chips coming out of fab, therefore considerably higher chip per waffer then Sony, which will give them ability to have premium console with gigantic and expensive die. On average, price should be similar, but they will cover both bases this way.

Another point I was thinking about was console's BW per TF and how it compares to similar spec'ced PC GPUs.

PS4, PS4 Pro and XBX all had at least 24-29% higher BW per TF in comparison to their PC GPU equivalents, so I couldn't see how 512-528GB/s would be enough for total system BW if PS5 was 12-13TF console.

Example :

5700XT - 48.7 GB/s per TF (448GB/s)

Oberon - 57.3 GB/s (528GB/s total BW)

XT + 25% - 60.8 GB/s (565 GB/s total BW)


By that same calculation, XSX will need 730GB/s of BW if same 25% is applied (and 12TF as final GPU spec).

I don't see either happening, but I can see MS having at very least 16Gbps chips in there for 640GB/s, if not downclocked 18Gbps similar to Flute to achieve this. Obviously, there can be slight difference in TF per GB/s for consoles, but I personally don't think known BW is sufficient for rumored chips (528GB/s or 560GB/s).

In case above, I even took XT version (which is best case scenario) instead of 5700, because 5700 has considerably higher BW per TF then XT.
 
Last edited:
Considering better cache, but higher utilization, how much bandwidth needs an 8 core zen2 compared to an 8 core frankenjaguar?
 
I've always wondered about what the CPU needs - is the worst case scenario for CPU when fully utilizing the FP/INT units? (I don't know if simultaneously for either Jaguar or Zen)

@3dilettante help :p
 
So this BGs insider is using this gif to describe how he envision's Sony's situation:oops:
giphy.gif

I..I just can't.
 
Basically these "insiders" think all of us have gold fish-tier memory.
I think a fair few of them mean well but just don’t have the knowledge or sense to vet their sources like a journalist would. It’s easy to get caught up in the hype and have that color your judgement.
 
I wonder if GAF moderators and/or Evilore are actually participating in next gen hyping, especially for PS5, because amount of traffic that site has got duo to Osiris/Tommy Fischer/Heisenberg/BG/O'dium/etc. is pretty big. Certain gaming sites have even written articles about these insiders originating from GAF, therefore they actively spread good PR for Evilore and co.

I know last gen mods/Evilore created CBOAT character and role played secret insider there, so I wouldn't past them to do similar again. Amount of rumors that pop up daily there is pretty staggering not to be suspicious, and these insiders are anything but specific.
 
I wonder if GAF moderators and/or Evilore are actually participating in next gen hyping, especially for PS5, because amount of traffic that site has got duo to Osiris/Tommy Fischer/Heisenberg/BG/O'dium/etc. is pretty big. Certain gaming sites have even written articles about these insiders originating from GAF, therefore they actively spread good PR for Evilore and co.

I know last gen mods/Evilore created CBOAT character and role played secret insider there, so I wouldn't past them to do similar again. Amount of rumors that pop up daily there is pretty staggering not to be suspicious, and these insiders are anything but specific.
If i'm Sony's PR team i would be facepalm now because these "insiders" just set the hype for PS5 become unrealistic high just like what Klobrille said:
 
It is up to each own to believe these rumors that claim close to xsx etc. You cant blame sony’s pr team later on, they havent done anything yet, and i doubt their behind all this regarding insiders.
 
Status
Not open for further replies.
Back
Top