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

Status
Not open for further replies.
Interesting thought (well I thought so lol) for those dissing all these comments regarding the next gen consoles 'being close'...

How many 'insiders' have said otherwise? There's literally no-one saying 'PS5 struggling' or 'one console has clear edge'...well no-one I can think of, but happy to be proven wrong :)
No one was saying that last gen either
 
Well, that was a whole lot of nothing.
Except for the part where he shoots down the massive 30% performance difference that the Phil Spencer statements and the Github Gospel suggest?

So either:
1 - Github Gospel doesn't tell the whole story of PS5's final specs (more ALUs per CU/WGP? more than 18 WGP enabled in the final version? higher clocks?)
2 - Phil Spencer didn't actually mean the SeriesX has 12 TFLOPs RDNA, and it's much closer to 9.2 TFLOPs than it is to 12.
3 - Combination of the above.
 
Somebody leaked me the TFs today and I have very strong reason to believe this individual.
PS5 target specs: 13 TF in Navi architecture (most recent devkit is reaching 12.6 TF but Sony is aiming to pump it even higher)
Xbox Series X target specs: 12 TF also Navi (the latest devkit is hitting 12.1 TF and MS still has room to wiggle further)
The person gave me a cookie afterwards and told me everything is gonna be just fine.
Source
IllegalImmenseGallowaycow-size_restricted.gif
 
You can try arguing from the POV of value all you like, but that doesn't change the engineering costs. If Lockhart is going to be $200, that needs to be explained in terms of cost to build, losses taken, and overall business model.

Then it's not a streaming box but a console.

I just did explain above with it being attached to a subscription to the Game Pass / Project X Cloud.

The rumour was always that it was a hybrid. If you think that does not make sense, welcome to the club :LOL: This whole thread was started because of me trying to make sense of Lockhart in the first place.
 
Last edited:
Somebody leaked me the TFs today and I have very strong reason to believe this individual.
PS5 target specs: 13 TF in Navi architecture (most recent devkit is reaching 12.6 TF but Sony is aiming to pump it even higher)
Xbox Series X target specs: 12 TF also Navi (the latest devkit is hitting 12.1 TF and MS still has room to wiggle further)
The person gave me a cookie afterwards and told me everything is gonna be just fine.
Source
IllegalImmenseGallowaycow-size_restricted.gif
Daily "PS5 is slight more powerful than XSX leak", version 21.
 
Except for the part where he shoots down the massive 30% performance difference that the Phil Spencer statements and the Github Gospel suggest?

So either:
1 - Github Gospel doesn't tell the whole story of PS5's final specs (more ALUs per CU/WGP? more than 18 WGP enabled in the final version? higher clocks?)
2 - Phil Spencer didn't actually mean the SeriesX has 12 TFLOPs RDNA, and it's much closer to 9.2 TFLOPs than it is to 12.
3 - Combination of the above.

I don't think he ever shoots 30% in raw TFlops down. He talks about balance, so it might have those extra, but memory and other stuff does not help it make use of it.
 
I don't think he ever shoots 30% in raw TFlops down. He talks about balance, so it might have those extra, but memory and other stuff does not help it make use of it.

I think that's what he does when he says one console runs the game at 60FPS and the other also runs it at the same framerate.

It's the exact same narrative we've been hearing from all the trusted insiders, from forum users that have been vetted by mod teams to renowned journalists, and it's one that is completely incompatible with SeriesX = PS5×1.3.


There's nothing else to be extra, at least not if the SeriesX is 320bit and the PS5 is 256bit. Unless all of a sudden those games are all fillrate limited which I find very hard to believe.
 
1 - Github Gospel doesn't tell the whole story of PS5's final specs (more ALUs per CU/WGP? more than 18 WGP enabled in the final version? higher clocks?)
2 - Phil Spencer didn't actually mean the SeriesX has 12 TFLOPs RDNA, and it's much closer to 9.2 TFLOPs than it is to 12.
3 - Combination of the above.

There's a conflict in the data between GitHub and the Sony presentations - Github suggests that there is no ray tracing, surely that's enough to give everyone pause?

Nobody should conclude anything based on limited data. I can only think of three possibilities:

1. GitHub did not have the complete final chip
2. RT will be performed on a separate chip
3. Part of the chip was inactivated due to backward compatibility testing
 
There's a conflict in the data between GitHub and the Sony presentations - Github suggests that there is no ray tracing, surely that's enough to give everyone pause?

It should be, but for many here it isn't.

The Github Gospel doesn't test raytracing. Sony have very officially confirmed that raytracing is in the GPU hardware. Therefore the Github Gospel did not test the whole chip.
Yet because there's one description in an excel sheet saying "18 WGP = full chip" then apparently there can't be anything else in it.


I think the Gospel is testing only BC (9 WGP 800MHz / 19 WGP 911MHz) and BC Boost (18 WGP 2GHz) modes, and it's obviously only activating the minimum amount of functional units that will be used in these BC modes, regardless of it having one excel field within thousands saying 18 WGP is full chip.
 
I would not say inactivated. To me that sounds more like being disabled. It just wasn't used.
 
Considering the RT part wasn't being used, then it's not too far-fetched to assume that other parts were also unused.

I honestly don't care too much either way, I'll buy the more powerful console.

It's just there's a huge hole in the data that many seem to be ignoring.
 
2 - Phil Spencer didn't actually mean the SeriesX has 12 TFLOPs RDNA, and it's much closer to 9.2 TFLOPs than it is to 12.

I don't think anyone is basing anything JUST on what Phil said because it is so intentionally vague.

HOWEVER... It does line up with what Windows Central has reported and Richard Leadbetter indicated in a video that they have the same info and i think Tom Warren also said it was 12TF on twitter.
All quite respected people.
 
The only credible data is from GitHub. The rest is from pretenders. It's not ignoring what they say, but giving it appropriate weight that anything without backing evidence deserves. I dont think github gives the entire picture just like I dont believe anything the pretenders say is true.
 
There's a conflict in the data between GitHub and the Sony presentations - Github suggests that there is no ray tracing, surely that's enough to give everyone pause?

Nobody should conclude anything based on limited data. I can only think of three possibilities:

1. GitHub did not have the complete final chip
2. RT will be performed on a separate chip
3. Part of the chip was inactivated due to backward compatibility testing
Whoa dude! Welcome back
 
No one was saying that last gen either
Well except for the ones telling Sony that they had better double their ram or they were toast, but yeah they were instead just quietly gimping XBOne games and hoping no one would notice. We wont even go into the previous generation where the howling became so insufferable that each week was a question of what dev was going to break out into a crying rant. Come to think of it, although PS2 would eventually become the base dev platform of that gen, it sure didn't get there quietly. Nintendo of course has been assailed with impunity thru this all, justified or not. In fact, I think they're just happy to be dev ignored now. I think what the platform holders have gotten real good at is holding devs to their NDAs, we'll see if that lasts thru GDC which is always a hard time to keep things secret (as it's purposed is to bring people together to share secrets). If Sony hasn't revealed by then, watch out as they will have lost the narrative and anything is possible especially griping.
 
I think that's what he does when he says one console runs the game at 60FPS and the other also runs it at the same framerate.
But not necessarily at the same resolution. He did not state both consoles run the game the same; he said they'd run the games such that its not obvious which is more powerful. He also said to ignore the TF numbers as that doesn't really relate to what we see/experience. What he didn't say is that both consoles are basically the same, which would be a far more straight-forward post than saying, "TFs don't matter, a machine can be more efficient with less TFs, you can't really tell the difference anyway." If they were the same. If they aren't the same, but the end results aren't fairly indistinguishable, you'd expect the statement to be something like, "Yes, one will be faster than another (but you will not notice it in practice)."

Contrast, "I can tell you the machines are very close, pretty much identical in power," with "one is less powerful in TF count but in real terms, you won't even see the difference so shouldn't worry about the numbers and choose your platform based on what games you want to play or which box is cheaper or on any other than TF number."
 
Status
Not open for further replies.
Back
Top