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

[Rangers]

And changing any clock means you stray away from your calculated cost/performance analysis. Changing the clock would be based instead on better than expected process results.

I really expect 16GB to 20GB for both, but memory amount is the one thing that has been changed twice in the past based on leaks, if one have 32GB and the other have 16GB, developers would eventually hint with subtlety and decorum.... "dude, if you have 16GB you're done!"... And they would want to be close enough not to have a clearly visible texture/asset detail gap on third parties. The nvme speed is putting a big question mark on real memory needs though.

I imagine Cerny might be thinking about this, if the PS5 lags so much in TF it might be his only chance to pull a rabbit out of a hat, kind of. But I dont know if it's even technically feasible. If it is, I cant imagine it happening because of cost. 32Gb in ANYTHING is insane, only some very expensive desktops and laptop have this. Also, Cerny is not in ultimate charge, the Japanese higher ups are likely not worried about power, they likely think (perhaps correctly) they will win regardless.

Honestly I often think about the move fron 4GB to 8GB in PS4. I think it was a great one, and like the 360 moving from 256 to 512, probably decided that generation. That said, I sometimes wonder if things would have been materially different with a 4GB PS4 with a superior GPU vs an 8GB One with a inferior GPU. I can see a reality where it played out about the same.

 

[Proelite]

Story time:

Let's say MS went to devs in late 2017 and early 2018 and shared these target specs.

CPU: 4x over XB1X
RAM: 1.5 over XB1X
GPU: 2x over XB1X
SSD: 40x over XB1X

I don't think we should doubt this, as it's the same multipliers that MS is still using to this day.

Now if you're a developer or insider and saw these targets prior to 2019, what's your only interpretation? These are clearly perf multipliers. The GPU is 12TF GCN. No one outside of people at AMD, or a few at MS/Sony might be privy to the massive leap in architecture that Navi was. Even AdoredTv, an AMD insider with a great track record, was down on Navi.

Story cont.

One day at office, you come to work and everyone is gathered around a shiny silver PC tower. Confused, you ask what the commotion is. All your colleagues turn to you and collectively chant. "PS5 devkit with a ~13TF Vega GPU".

Excited, you message all your friends, "13>12".

At a certain time before or after you first laid your eyes on the shiny Cerny tower, you acquire an updated / different set of targets from MS. 10+TF Navi.

"Ah you think to yourself, 12GCN translates to ~10TF Navi. Good to see that Navi is fine. AdoredTV must be a fraud!"
"Wait a minute." you realize. "This means that PS5 will be around ~10.8TF. That's better than Stadia!"

You message your retired journalist friend. "Scarlett and PS5 >= 10TF. PS5 > Scarlett by 5-10%."

Around similar time, your office receives the updated PS5 devkit. "Wow that things looks substantial, all these vents must meant it's very stackable! Wait only 9.2TF? It performs better than the Vega 64 kits though. The RDNA jump is bigger than expected! This means the final Xsx is likely under 9tf, or more likely there will be an upgraded PS5 dev kit down the line over 10TF."

Fast forward to December:

Articles come out with 12TF Navi rumor for Xsx. You read them and chuckle. "These fools thinks 2x 1x is 12TF RDNA." Later after the Xsx reveal, Phil corroborate your secret knowledge by again using the same multipliers you knew for 2 years!

12/25th: All hell breaks loose. How could AMD screw the pooch like this! You saw that Xsx's CU count is 56 Cus and you begin sweating. Maybe you thought, "you just screwed over your ex-journalist friend who is currently in a bunker somewhere in a remote part of America".

 

[Rangers]

Are we also assuming the iphone 12 has already started production in December 2019 for a release 11 months later in October 2020? With a N5 SoC that has barely any test chips available yet?

I bet the chip is locked in, yes.

For annual iterative phones, they must be working even now on technology designated for late 2023, 2024, 2025 etc. They know long in advance what they are planning.

 

[Rangers]

He is the former editor for a gaming magazine IIRC

He is a former editor at Gamefan, which according to Wiki ran from 1992 to 2000. So it's been 20 years. That magazine was mostly about 16 bit in it's heyday. I read it voraciously. Hard to believe that in my little circle he is still relevant in 2020. Anyway, he is a lone beacon for sure, will be very interesting how he spins it if he is wrong. Although he is so vague and has already given himself vast amounts of cover..

 

[BRiT]

You both have MS information right?

 

[PSman1700]

32Gb in ANYTHING is insane, only some very expensive desktops and laptop have this.

It is the least im going to have, theres already a game wanting that as recommended.

Story cont.

So, what do you mean exactly, that XSX is about 10TF RDNA?

 

[Proelite]

So, what do you mean exactly, that XSX is about 10TF RDNA?

According to Phil, "If you do the math, we're at 2x over Xb1x (gpu)".

 

[JoeJ]

"currently in a bunker somewhere in a remote part of America".

This happens to people fucking up the true meaning of "Teraflops"

 

[Deleted member 13524]

You both have MS information right?

I have exclusive information from both a gitbin and a pastehub posts from January 2019 that details all the consoles' ultra final specs down to memory KHz. It was there for months for everyone to see and many saw it but now it was taken down so it must be real and although I did a backup I'm obviously not going to share it with anyone else because reasons.

 

[AbsoluteBeginner]

I have exclusive information from both a gitbin and a pastehub posts from January 2019 that details all the consoles' ultra final specs down to memory KHz. It was there for months for everyone to see and many saw it but now it was taken down so it must be real and although I did a backup I'm obviously not going to share it with anyone else because reasons.

A bit too arrogant considering people have been talking about this for a week. I love it that some people just pop out like you just did thinking everyone before you was so dumb to buy anything that is posted on the internet without thinking "Perhaps leak of this scale is fake?".

https://www.eurogamer.net/articles/...laystation-5-xbox-series-x-spec-leak-analysed

The scale and scope of this latest leak is remarkable and the origin of the new information seems even more far-fetched than the Gonzalo story, leading many to believe that the entire thing may be a work of fiction. However, having looked into the situation and independently verified the source, the overwhelming evidence is that the data does indeed originate from AMD - and hasn't been doctored. We're lacking crucial context for sure but the reasons to doubt the veracity of the leak are somewhat thin on the ground.

From what I can gather, someone at AMD's ASIC validation department used GitHub to store fragments of internal testing data from a range of work-in-progress Team Red projects. The leaks include testing of next-gen desktop and mobile Ryzen APUs along with some deep-dive testing on the PS5 chip, now codenamed Oberon. While the data is not public, it's clear that the GitHub test data has travelled far and wide: further details from the leak mentioned in this article are being discussed at length on ResetEra, for example. The genie is out of the bottle.

I already said it reminded me of this gem from April 2004

https://forum.beyond3d.com/threads/xenon-system-block-diagram.10197/

Full 360 system diagram leaked 18 months before system was released.

 

[iroboto]

You both have MS information right?

Information is always changing. I prefer to just stick with the shop talk and see the discourse from that. Teasing people or trying to direct a conversation isn’t particularly helpful. There’s no agenda here. Hell even Phil Spencer will make an announcement and later change it. If he Can change his mind on things and he is _the source_ then anything else is subject to change.

so it’s much better to stick with talking shop. Trying to figure out if an insider has real information is a waste of time.

let’s assume I had information to share. I would want to share it with places that have journalistic integrity like DF; because they will do the work to Confirm it with other sources they have. Then that information is useful to everyone. Richard and team have had regular dialog with both Cerny and Xbox teams; it’s not going to be a witch hunt if they ask questions. Putting it out on an open forum is much less effective and much more dangerous.

 

[PSman1700]

Ok then, 9TF for PS5 (or high 8 range), but that is RDNA i think. XSX is obviously going to have more raw power then that.

 

[Deleted member 13524]

A bit too arrogant considering people have been talking about this for a week.

Dude chill it was a joke. Have you read the thread title or the posts in some of the first pages of this thread?

As for the leak itself, what everyone seems to be missing from the article is:

We're lacking crucial context for sure

Hell even Phil Spencer will make an announcement and later change it. If he Can change his mind on things and he is _the source_ then anything else is subject to change.

#cough#windows-mixed-reality-on-scorpio#cough#

 

[AbsoluteBeginner]

Dude chill it was a joke. Have you read the thread title or the posts in some of the first pages of this thread?

As for the leak itself, what everyone seems to be missing from the article is:

Yes, of course we lack context. When something like this leaks out, there really is a context missing. Not like someone leaked Durango/Orbis dev documentation, just straight out repo from AMD ASIC team doing regression test on upcoming PC and console chips.

WGP = 18 (36CUs)
Theoretical BW 448GB/s (14Gbps chips) - >500GB/s for Oberon.

Its a huge data dump. More info on chip performance then anyone of us would actually need.

https://www.resetera.com/threads/ps...unch-thread-ot9-for-flops-sake.162235/page-21

 

[PSman1700]

Its a huge data dump. More info on chip performance then anyone of us would actually need.

He needs more flops

 

[Proelite]

Yes, of course we lack context. When something like this leaks out, there really is a context missing. Not like someone leaked Durango/Orbis dev documentation, just straight out repo from AMD ASIC team doing regression test on upcoming PC and console chips.

WGP = 18 (36CUs)
Theoretical BW 448GB/s (14Gbps chips) - >500GB/s for Oberon.

Its a huge data dump. More info on chip performance then anyone of us would actually need.

https://www.resetera.com/threads/ps...unch-thread-ot9-for-flops-sake.162235/page-21

Why are you posting fake made up data? You know that MS owns Github right?

 

[Betanumerical]

Yes, of course we lack context. When something like this leaks out, there really is a context missing. Not like someone leaked Durango/Orbis dev documentation, just straight out repo from AMD ASIC team doing regression test on upcoming PC and console chips.

WGP = 18 (36CUs)
Theoretical BW 448GB/s (14Gbps chips) - >500GB/s for Oberon.

Its a huge data dump. More info on chip performance then anyone of us would actually need.

https://www.resetera.com/threads/ps...unch-thread-ot9-for-flops-sake.162235/page-21

We actually don't know if the CUs and WGPs changed between Ariel and Oberon just like how the bandwidth did. There was a rumour previously that Ariel was an earlier chip that got chucked out in favour of something faster (Oberon).

 

[Proelite]

We actually don't know if the CUs and WGPs changed between Ariel and Oberon just like how the bandwidth did. There was a rumour previously that Ariel was an earlier chip that got chucked out in favour of something faster (Oberon).

If something did change, it's not evident in these results, despite the fact that these are Oberon regression tests, and we see the stepping letters, and also the improved bandwidth numbers.

Extremely comprehensive GPU tests with results, but coincidently omits 2+ WGPs for some reason!

 

[Betanumerical]

If something did change, it's not evident in these results, despite the fact that these are Oberon regression tests, and we see the stepping letters, and also the improved bandwidth numbers.

Extremely comprehensive GPU tests with results, but coincidently omits 2+ WGPs for some reason!

The CU's and WGPs for the PS5 chip are not even mentioned except in the theoretical performance for Ariel column, we would not know if they did or not did not change. There is no test that shows the number of SP's for the PS5 that only exists for Arden and Sparkman in the leaks for some reason.

 

[Proelite]

The CU's and WGPs for the PS5 chip are not even mentioned except in the theoretical performance for Ariel column, we would not know if they did or not did not change. There is no test that shows the number of SP's for the PS5 that only exists for Arden and Sparkman in the leaks for some reason.

They're in the igpu and igpu_295 sheet in the Oberon regression test files. First column.

Cache Bandwidth
I$ (per SQ)
K$ (per SQ)
GL0 (per V$)
GL1 (WGP/SA = 1)
GL1 (WGP/SA = 2)
GL1 (WGP/SA >= 4)
GL2 (WGP=1, SA=1) (Read)
GL2 (WGP=9, SA=2) (Read)
GL2 (WGP=18, SA=4) (Read)
GL2 (WGP=1, SA=1) (Write)
GL2 (WGP=9, SA=2) (Write)
GL2 (WGP=18, SA=4) (Write)