I wonder if 60 fps is possible in a performance mode.

@aenea you are mostly wrong. It seems you took offense to the fact I said upscaling, which some say only applies to scaling where pixels are guessed based on surrounding pixels. That said, frankly, claiming anything other than the originating pixels as native, just isn't true and honestly, claiming it is not upscaling is basically just a marketing ploy.

The thing is:
1. Xbox one also has a hardware based scaling (or whatever term you prefer) that is very similar to checkerboarding, where data is pulled from previous frames rather than interpolation.
2. Checkerboarding itself can reuse frames up to 10 frames old, so hardly the same as the originally pixels, especially in motion. If you stand completely still, you are correct, it is 100% native 4k, but in motion checkerboarding is definitely not the same as 4k, and frankly is hardly better than standard upscaling IMO (at least in first person shooters) due to increased and literal artifacting.
3. The lower the framerate, paired with checkerboarding (seems like 30fps in this case) the more noticeable the transferred frames are.
4. Obviously checkerboarding is not the end-all, because many games, specifically where there is high motion, like an first-person shooter have opted to standard scaling on PS4 pro. (ie. COD WW2)

In other words 1440p upscaled using Xbox's hardware scaling solution (similar to checkerboarding) so if we are claiming the form of producing missing pixels is the reason it is better, you argument falls kind of flat considering they both use methods that use pixels from previous frames.

Look, the PS4 pro is fine, no need to pretend that the Pro and One X are equals though, that is just a lie, and pretty stupid considering all the trolling about resolution between XB1 and PS4 at launch, now suddenly scaling solutions are better than higher native resolutions -_- you guys...

Basically, in sorts, the originating resolution for PS4 is 1080p, and 1440p for One X, the checkerboarding is solution is basically the same on each, that is a bit off, for 2 reasons
1. the pixels in all are 4k when rebuilt, though each originating image is 1080p/1440p
2. The PS4 checkerboarding is superior to One X version (at least at launch before patches) according to most image interpretation experts. (ie. Digital Foundry)

Again, TBH, they are going to be VERY similar, though, like I said above, that is likely impacted largely on the fact that this game is probably bottlenecked on the CPU. But the One X is doing the more intensive process regardless of any other argument relative to quality of its rendering process.

Also, checkerboarding does not have to be 2x2, it is frequently 4x4 with 4 rendering phases on PS4 Pro, so the argument that it is rendering half of 2160p is not true. Is is rendering 1/4th of 2160p in this case, and that is equivalent to 1080p.

But also, Sony's checkerboard rendering isn't quite 2x2 or 4x4, it is slightly different (and superior) though that part is beyond me, there are articles explaining it.

That said, I am assuming it is 4x4 rendering, you are assuming 2x2, frankly, I guess we will have to wait to see who is right, but I am completely confident it is 4x4, considering 2x2 would be a higher base render than One X. Maybe I am making that assumption wrongly though.

@ Aenea

I guess that means that Quantum Break looked as good as native 1080p according to your theory too?

@Aena... checkerboarding from 1080p up to 2160 is NOT way better than native 1440p..

checkerboarding from 1440p - 1800p may be way better... but not from 1080p.. it may look slightly better... but not way better

lol How many people arguing about the resolution in the topic are actually interested in the game itself?

@GottaBjimmyb stop making stuff up. In fact you don't know, you are right. WW2 uses half width interpolation with full 2160 lines. Yes, it does indeed use CB on a 60fps first person shooter.

Also, wth is 1080p CB? This has never been used on a Pro. The lowest CB framebuffer is 1800p (GT Sport @ 60fps, Mass Effect 1800pCB@30fps), most use a full 2160 lines buffer with the X resolution anything from 900 - 3840; Not sure if there is even a title at 1600pCB (BF2?).

I doubt however that the X is 1440p and the Pro is indeed 2160CB. Yes the Pro has HW extensions where it does not require shader resources for CB (to that extend), but I doubt it can offset that bandwidth deficit in full. If it does, it would be a first. But, who actually cares. It'll look fine on both on a 4K screen.

BTW: QB's resolution was ridiculously low to get the game running on a XB1. Sure, it looked great but it was <720p which means it uses quite large pixels to interpolate. It also was awfully blurry. You can't compare a 4x (actually more than that) pixel density with what QB does. The lower the res, the more pronounced are the artifacts of the CB renderer.

@ju On Cod WW2

"It's early days here, but our impression is that WW2 employs dynamic resolution scaling on both systems, with the lower bounds set at 1920x2160, rising to a full-fat 3840x2160. Perhaps not surprisingly, it's Xbox One X that gets closer to the higher-end of the scale, and the artefacts present in the horizontal upscale (apparent on both systems) are more obviously observed on the Pro presentation. Whether it's a checkerboard-style side effect or a product of temporal upscaling remains to be seen, but we feel there's more work to be done here in how WW2 makes the final resolve pass for 2160p output."

But let's not let facts get in the way of your propaganda and BS.

On the Checkerboarding, I explained this already in another comment. Checkerboarding can use any number of frames to make a render. So, while most PS4 Pro games use 2 frames (one current, one old) to make a 4k render some use 4 frames to make each render. (Some games woth checkerboard use dynamic ranges or special ratios, but most don't) 4 frames will each have the same number of pixels as 1920x1080. It is easier to just say 1080p, than to say an equivalent number of pixels to 1920x1080. Again though, let's not use facts, just your propaganda and hyperbolic statements and questions...

@GottaBjimmyb Your quote you just posted directly contradicts your earlier statements. Thanks for proofing my point.

@ju you clearly cannot read. Either that or you don't know the difference between "checkerboard-style side effect" and actual checkerboard resolution.

They even state plainly it is standard upscaling right at the end, most likely alluding to some kind of temporal scaling which is similar to checkerboard, hence the wording "checkerboard-style side effect"

That would be like if a doctor said coughing is a flu related symptom, but your diagnoses came back as pneumonia, so you have the flu for sure. You just suck at reading apparently.

That's not how checkerboarding works, if it's a checkerboarded 4K image it means the real resolution is 1920x2160 or exactly 2x the number of pixels per frame as 1080p. I find it hard to imagine why they'd do checkerboard 4K on PS4 Pro and 1440p on Xbox One X but it may be due to the PS4 Pro including the ID buffer, the developer being Japanese, and the majority of the focus being on the platform where it will sell best, the PS4.

"perhaps, but I can't imagine you couldn't basically copy + paste the PS4 Pro code and then do a bit of tweaking."

The Pro has GPU additions to support checkerboard rendering that are Sony specific so the code can't be simply copy and pasted. Also the Pro has been out a year longer than the One X so they had more time with it.

Of course they could have taken the time to use the One X sparse rendering method as well, but they didn't for whatever reason...

--------
@gottabejim

Ehh no, checkerboard rendering at 2160p is NOT the same as native rendering at 1080p. How on earth did you get that idea?

Checkerboard rendering divides the framebuffer into 2x2 pixel squares, half of those squares (and thus pixels) are rendered each frame, which squares are rendered are alternated every other frame. The Pro's ID buffer and other GPU additions help to determine which pixels to use from the previous frame. When there is no movement at all between 2 frames the result is a native 2160p frame! This means it renders half of 2160p per frame (1080p is a fourth instead!). Half of 2160p is also 12.5% more pixels than native 1440p.

So it's:
2160p checkerboard rendered - PS4 Pro
native 1440p - One X

or
half of 2160p - PS4 Pro
full 1440p - One X

Half of 2160p is still 12.5% more pixels than native 1440p...

--------

@shloobmm3

The One X renders 12.5% pixels less, so you're dead wrong!

@aenea you are assuming it is 2x2 checkerboard, not 4x4 (which is almost positively true considering option 2 is 1080p 60fps)

Basically, it is rendering and compiling an image from a source equal to 1080p

Though, 2x2 and 4x4 are relative, as technically that is just a way to visualize the process. What happens is the horizontal or vertical pixels are rendered in 2 phases making 1920x1080 (1080p) be built from 2 960x1080 frames overlayed, using the missing pixels from the other. In this, you sometimes have frames fail to render and can sometimes have up to 10 frames where the same data is used/reused, though very rare, usually the data is used only in the original render and the following render.

However, you can actually use this same process, but render using 4 frames. This would look horrible if you were rendering 4 frames for a 1080p image, but if you are rendering 4 frames for a 4k image, then each frame would still be 1920x1080 (3840/2 and 2160/2) and the image quality would be good enough that even if all other 3 frames failed (VERY UNLIKELY) the image is still good.

However, no matter what the process is, a still image IS actually full 4k, uncompromised (or 1080) just not so in motion. The problem with checkerboarding is the look in motion and the more frames that are reused, the less effective it is. Then again, some people like the visual effect it causes, just like some people like "motion blur," I really do not.

Now I could be wrong, maybe the checkerboarding for this game only uses 2 frames per render, and if so, it actually has a higher pixel count per frame than the One X version, but considering it is 1080p 60fps or 4k checkerboarding, I highly doubt it.