Digital Foundry Microsoft Xbox Scorpio Reveal [2017: 04-06, 04-11, 04-15, 04-16]

But that was kinda my point earlier. If nobody but MS first party titles (such as Forza Tech Demo that we're getting our performance ideas from) use DX12 features, aren't we then using an example that is a best case scenario but unlikely to occur in reality?
if nobody but MS first party titles are using dx12 features on essentially a dx12 only console that would be an issue. That would really suggest that they didn't design dx12 very well. Lol. There are feature flags and other things. Considering both PS4 and XBO are effectively the same GPU this seems unlikely. The XPA initiative helps to combat the issue of dx12 on PC; MS tied two dx12 ecosystems together.
 
Last edited:
Something interesting that I haven't seen anyone mention:

Ps4 -> Ps4 Pro CPU frequency: 1.6Ghz -> 2.13Ghz = 33% Increase
Xbox One -> Scorpio CPU frequency: 1.75Ghz -> 2.3Ghz = 31% Increase

So the Scorpio holds pretty much a similar CPU advantage over the Pro as the X1 did over the Ps4.
 
Something interesting that I haven't seen anyone mention:

Ps4 -> Ps4 Pro CPU frequency: 1.6Ghz -> 2.13Ghz = 33% Increase
Xbox One -> Scorpio CPU frequency: 1.75Ghz -> 2.3Ghz = 31% Increase

So the Scorpio holds pretty much a similar CPU advantage over the Pro as the X1 did over the Ps4.

Yeah but both scaled relatively the same with the GPU.

I'm still confused how they hit 2.3GHz...Isn't that beyond what Jaguar could go? Or is it the max?
 
After digesting the info we have from Forza running on Scorpio hardware some more I feel a bit more hesitant to draw too many conclusions. I predicted myself that if there was a game that was going to hit native 4K it was likely Forza (non Horizon series). This engine seems incredibly optimized to being with and is probably atypical of what will be seen. Who knows if other games will scale this well.

I'm going back to my old prediction of 4K dynamic res/1800p titles for a significant number of Scorpio titles to maintain consistent framerates.
 
Yeah but both scaled relatively the same with the GPU.

I'm still confused how they hit 2.3GHz...Isn't that beyond what Jaguar could go? Or is it the max?

There's no real max. The max comes from the slowest timing path in the design and the process node. With process scaling the same design should be able to hit higher clocks, with re-balancing the timing paths you can further increase clock. Sometimes that's not practical if it were to be a significant change, or the schedule doesn't allow it even if it's easily doable. This is probably one of the things that AMD improved in the year between the Pro and Scorpio.
 
Are there any examples of X360/X1 titles that show what a 16xAF will look like compared to the original image?
 
Are there any examples of X360/X1 titles that show what a 16xAF will look like compared to the original image?

http://international.download.nvidi...shared/images/embed/aa-af-guide/filtering.jpg

http://www.vgamuseum.info/images/tech/anisofiltercompare.jpg

What do you guys think about anadtechs take on the spec reveal?

http://www.anandtech.com/show/11250/microsofts-project-scorpio-more-hardware-details-revealed

They seem to think that devs will not be able to fully utilize the bandwith and will be limited to about 218GB/sec.

They just wonder if they'll be able to, but it's too soon to draw conclusions as not enough info is available.
 
But that was kinda my point earlier. If nobody but MS first party titles (such as Forza Tech Demo that we're getting our performance ideas from) use DX12 features, aren't we then using an example that is a best case scenario but unlikely to occur in reality?

I think you're probably onto something, and that many games - at least initially - won't use the customisations to greatest effect.

In the case of Xbox features that save CPU time and that (probably) aren't equalled on PS4/Pro, you have to wonder whether it'll be a priority for third party developers to spend resources using these when X1 and Scorpio already have faster CPUs than their Playstation counterparts. Probably better to focus on changes that would help all systems at once - particularly the system with the slowest CPU.
 
From the 7970 review:
Ryan Smith said:
Clearly the crossbar design has improved ROP performance through the amount of memory bandwidth they can access, but did it impact anything else

Ok!

From the Scorpio article:
Ian Cutress said:
what we saw from the 7970 is that the extra bandwidth + crossbar setup did not offer much of a benefit over a straight-connected, lower bandwidth configuration.
.......huh?

Is Ian referring to Tonga's 256-bit bus? (while Tonga had v1 DCC)
 
Examples please.

Errors: 360 GFlops; 360 edram BW (they mistakenly use inter die BW); X1 esram BW, X1 TFlops.

Shaky suppositions: suggesting crossbar may bring no extra BW; saying 16 ROPs were a "big part of the reason" that X1 commonly uses 900p ("it has been a pretty universal suspicion").

It's a messy rushed article that's adding noise to the debate, and not really up to Anand's and Ian Cutress' normal high standards.
 
Errors: 360 GFlops; 360 edram BW (they mistakenly use inter die BW); X1 esram BW, X1 TFlops.

Shaky suppositions: suggesting crossbar may bring no extra BW; saying 16 ROPs were a "big part of the reason" that X1 commonly uses 900p ("it has been a pretty universal suspicion").

It's a messy rushed article that's adding noise to the debate, and not really up to Anand's and Ian Cutress' normal high standards.

Thank You
 
Back
Top