-
Ever wanted an RSS feed of all your favorite gaming news sites? Go check out our new Gaming Headlines feed! Read more about it here.
Digital Foundry: Control PS5 vs Xbox Series X Ray Tracing 'Benchmark' - Unlocked FPS In Photo Mode!
- Thread starter chandoog
- Start date
OP
OP
Text Article:
www.eurogamer.net
Control Ultimate Edition's photo mode doubles up as a ray tracing benchmark
Just when we thought that our coverage of Remedy's Control for next generation consoles was complete, we received a cur…
www.eurogamer.net
A framerate unlock option make sense here. People with VRR displays can enjoy the game to it's fullest on the X.
OP
OP
This video shows the consoles have the overhead to do a 1080p/60 FPS RTS mode if Remedy puts it in with optimizations.
I don't think this is true in this case.
Going by the video, the XSX has an advantage of 3% to 11% to 20% depending on a scene.
I don't think this is true in this case.
Going by the video, the XSX has an advantage of 3% to 11% to 20% depending on a scene.
Yeah this seems bigger then most comparisons between the two consoles.I don't think this is true in this case.
Going by the video, the XSX has an advantage of 3% to 11% to 20% depending on a scene.
I feel like this is one case, as bizarre as it is, where the difference is actually quite large
Well tbh the X's GPU is better at RT iircI don't think this is true in this case.
Going by the video, the XSX has an advantage of 3% to 11% to 20% depending on a scene.
I don't think this is true in this case.
Going by the video, the XSX has an advantage of 3% to 11% to 20% depending on a scene.
while static and not having action where the cpu would also play a bigger role. cpu is miniscule difference. this explains why xsx was behind in various comparisons. faster clocks better tools make the difference when consoles are close.....
I wouldn't say 16% is a small difference. That is just the average aswell. It was nearly 25 more at points
Obviously the big thing to take away from this is nothing is happening in Photo Mode. Hard to say whether or not these numbers would hold when there's actual gameplay going on. Still, I'd welcome the option to bring the resolution down to 1080p with an unlocked framerate.
The results pretty much show exactly what was speculated before the consoles released. That the XSX would have a 10%-20% advantage over PS5.
when there is action on screen, the cpu would play a role as well, and both consoles are very close on cpu. so makes sense we saw all those close comparisons so many times.....
(I already watched the video)
I still don't quite understand how a photo mode can be a correct representative of performance in a super dynamic game like Control, like, combat situations are a pure shitshow with objects flying on the screen and other effects, it's not a ''static'' image.
Of course, this is does not matter for situations that wouldn't be much different in-game, like the HUB or any location without enemies/combat.
I still don't quite understand how a photo mode can be a correct representative of performance in a super dynamic game like Control, like, combat situations are a pure shitshow with objects flying on the screen and other effects, it's not a ''static'' image.
Of course, this is does not matter for situations that wouldn't be much different in-game, like the HUB or any location without enemies/combat.
I am surprised they didn't add the option. It doesn't cost anything and is future proof (if mid gen refresh exist or just the next console on BC mode).
VRR wouldn't fix this when the game drops to just over 30 fps in some scenes. So no wonder Remedy decided to not offer an unlocked fps mode for Xbox. While the difference is not marginal, this great analysis done by DF shows that in practice (actual gameplay) that difference is less meaningful.
It's not a test or representative of performance of the game. It's not trying to be.
It's just an interesting example of a product that has the exact same settings/resolution etc between both platforms (rare) so it's a fair test benchmarking tool when you can find areas that dip below 60fps in the same scene. That's all. It's just academic. Nobody's playing the game in photo mode, obviously.
It sounds like it wasn't a case of it being impossible, just amount of manpower on this project. Adding a 1080/60 RT mode won't do much in the way of selling more copies of the game, I can't imagine why they'd justify spending another couple months of person hours pushing it out.
Because benchmarks can only be done on exact like for like situations. And the photo mode is still rendering in real time.
Dynamic gameplay throws up way more variables.
While the image seems static, it's still being rendered in real time so the console is feeding a string of frames.
Our tools can measure per pixel differences - games that use TAA have minute differences in each new frame even if a camera is still and the image is perceptually very similar to the last.
Interesting results there.
Even if there is overhead there I still expect the biggest differences to be in timed exclusives potentially and as the gen goes on the differences should also be more frequent than now I think.
Even if there is overhead there I still expect the biggest differences to be in timed exclusives potentially and as the gen goes on the differences should also be more frequent than now I think.
Yeah you are right, I had the wrong interpretation that the overall idea of the test was that if Remedy removed the frame rate cap results would be the same.
Seems about right going by raw hardware specs between the two consoles.
I would imagine these are the types of gains we continue to see moving forward (which frankly still isn't a ton. Smaller than the PS4 Pro / XB1X gap.)
I love academic finds like this. This is great hobbyist stuff, even if it has little bearing on the normal play experience.
When I messed around in photo mode last night on my 60hz display (in the RT/30 fps mode), I was surprised how smooth it felt, and wondered if it was hitting an actual 60 fps in the area I was in. This seems to confirm that in some places, it can.
When I messed around in photo mode last night on my 60hz display (in the RT/30 fps mode), I was surprised how smooth it felt, and wondered if it was hitting an actual 60 fps in the area I was in. This seems to confirm that in some places, it can.
How will VRR work in the scenes that drop around 30 fps?
Not to downplay the work DF did in this video but I think it gives some false expectations, in reality the test sample size used in the video is too small to determine how the whole game would run in a practical environment, Control is a game of many variables regarding combat, where the character can reach (since she can basically fly) and very dynamic fx. It's a great video to compare but platforms but not to determine what the devs could or couldn't do to achieve better performance basically.
This makes sense, thank you! Anyway I suppose it would be easier to measure when just panning the camera around for some seconds, right? And that would also be more akin to "gameplay" (at least, almost like walking around with nothing else happening).
I loved the Ghost of Tsushima photo mode btw because of the option to have animated things (wind, leaves, etc), it just paused the action but you saw the world still being alive!
Just furthers the case that early games were not representative. This should be the expectation. Series X with a 10-20% performance gain over the PS5, but in most cases looking nearly identical. Just play where you want, there's not going to be a silver bullet where games look significantly better on one or the other.
I think this comparison was moreso to compare the XSX and PS5 in raw performance in like-for-like scenes without an FPS cap.
I don't think anyone should expect a 60fps RT patch to come to either console in Control.
The most interesting data we should gather here is the 16% average performance increase on Series X.
So XSX could have definitely hit 60 fps at 1080P and PS5 maybe could have with a slightly lower res than 1080. This game could have benefited a lot from a dynamic resolution.
Yep. It's like the PS4 Pro / XB1X all over again, but even smaller difference this time around.
The real fun will be when AAA exclusives arrive on each console, uniquely taking advantage of each of their particular hardware specs.
Exactly. That's my take away from it too. It sounds like 16% difference should result in greater things. But the reality is developers are often aiming for one of 2 performance modes. 30 or 60. And 16% isn't enough of an advantage to push a like for like scenario to 60 from 30. They could unlock and you'd get better results from the X in certain situations compared to a locked 30 PS5 version. But it would also have a wildly fluctuating framerate resulting in an inconsistent experience and I'm.not sure anyone wants that.
Sadly I don't see many developers going that extra mile like insomniac either really. But that could be the only way we're realistically going to see the series X TF advantage come into play at all. If games start having a 60/RT mode that renders at whatever Res the system is capable of pushing to maintain that framerate. Might be something like 1080p on series X and 900p on PS5 or something.
I hope AMD can get a DLSS like feature patched into these machines sooner rather than later if it means we could get 60FPS ray traced games.
Because as others have said, it's not like anything is going on game logic wise during photo mode...so it's not like we're close.
Because as others have said, it's not like anything is going on game logic wise during photo mode...so it's not like we're close.
I mean, 16% is the difference between 51-52 FPS and 60FPS.
It's not some crazy advantage but it's enough that it could come into play down the line, especially with VRR displays. More realistically, it's going to come into play by way of the PS5 having a minutely lower rendering resolution than XSX. Just like last-gen w/ PS4 Pro and XB1X but even less of a gap.
Overall though, I would agree. The TF advantage on XSX will likely mostly be taken advantage of in 1st party exclusives. Less so in multiplatform games. Just like the SSD I/O advantage on PS5 will be mostly taken advantage of in 1st party exclusives, less so in multiplatform games.
Last edited:
Just to put it out there, the one scene that is in the 30s is the only scene that was in the 30s on XSX. The Corridor of Doom is not representative of the majority load that the game experience. So VRR would not fix the Corridor of Doom on either console (which is so stupidly expensive on every platform), but on XSX, VRR would actually offer a meaningful improvement over the 30 fps lock I feel. XSX spends a lot of scenes in 50s.
Which should never be the priority of any developer imo..VRR is a crutch for inconsistent performance. If a Dev is doing their job well it should not need VRR at all.
the game isn't actually static, nothing is moving
given these differences, I'm quite excited to see what MS studios do with the hardware
By the time VRR is something devs care and regular consumers start to adopt, we'll probably have a mid gen refresh already. Today VRR support is only for dare I say less than 3% of the consumers on consoles.
Also have to think if Remedy used VRS Tier2 like Gears 5 giving another 10-15% more performance might be enough to get that 60fps too. Early gen games never be as good we will see next year and years later