1080p+RT+60fps seems like a possibility with some optimization.
That would be my preferred way to play it.
That would be my preferred way to play it.
-
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
It's been stated numerous times that they used photo mode as a benchmark for testing. It's academic, and made for some interesting, if expected, results. That's it.
Photomode has the same settings as the game in action but has the 30fps cap removed. This comparison shows how the game would run on each console if they took that cap away
DF probably would've tested the gameplay, if there was any way to unlock the framerate while RT is active. But unfortunately there is no way to do this on console. The test is interesting, because it's the first time the ports have the exact same resolution and settings, while offering a unlock framerate. It's a benchmark, nothing else.
You should watch the video or read the article before commenting in the future
Probably, anyway, based on Alex's CPU testing on PC and the games running on last gen consoles, but I think it's fair to point out that in photo mode certain game logic that would run on the CPU may well not be running.
That's already the case with first-party stuff. Games like Miles, Demon's Souls, Sackboy, and Astro's Playroom typically have 2-5 seconds load times, which is definitely faster than Series X first-party games (certainly in part because they're not made ONLY for the Series X).
Although, for whatever reason, Series X load times in BC titles are almost always faster than PS5 BC titles.
Yeah in the last Control DF thread a poster was warned for cherry-picking a screenshot that basically derailed the thread. I really don't understand why people do this and others quote them with serious replies.
I think the truth of the matter thus far is that the difference in third party games, whether it's load times, resolution, framerates, effects quality, texture quality, shadow quality, ray tracing, etc.. etc.. has been virtually indistinguishable between the PS5 and XSX in the vast majority of circumstances.
That should really be our main take away from the first few months of this generation. And that's a beautiful thing.
I was a browsing reddit when I found this (please focus on the screenshots):
www.reddit.com
It seems like a nice benchmark but in the end I think it's pointless as a measure of RT power, it needs more parity, I guess. Any opinions?
Reddit - Dive into anything
www.reddit.com
It seems like a nice benchmark but in the end I think it's pointless as a measure of RT power, it needs more parity, I guess. Any opinions?
I was a browsing reddit when I found this (please focus on the screenshots):
Reddit - Dive into anything
It seems like a nice benchmark but in the end I think it's pointless as a measure of RT power, it needs more parity, I guess. Any opinions?
Looks like temporal artifacts. Dictator?
Were you able to confirm that Xsx uses the exact same settings as the PS5?
Edit: Apparently yes.
Last edited:
Interesting.I was a browsing reddit when I found this (please focus on the screenshots):
Reddit - Dive into anything
It seems like a nice benchmark but in the end I think it's pointless as a measure of RT power, it needs more parity, I guess. Any opinions?
I was a browsing reddit when I found this (please focus on the screenshots):
Reddit - Dive into anything
It seems like a nice benchmark but in the end I think it's pointless as a measure of RT power, it needs more parity, I guess. Any opinions?
interesting, will look
The point of this thread is literally a perfect case of screenshot cherry-picking as that is what the photomode is about. This lets us benchmark the consoles GPU.....
I'm surprised no one mentioned so far that PS5 variable frequency, could be in play here. I'm not saying that it's happening but it potentialy could.
I've watched DF video but unfortunately, they also didn't bring it up.
I've watched DF video but unfortunately, they also didn't bring it up.
What are you talking about? The PS5 GPU is working as intended and these are the numbers being presented. That's the whole story.
Just like we all figured in the PS5 hardware thread, 16-17% average would be the gpu related difference.
interesting, will look
I pointed this out on the PS5 side in the last thread
Digital Foundry: Control Ultimate Edition on PlayStation 5: The Next Generation Tech Review
Whats happening here? The reflection is totally wrong, no?
www.resetera.com
interesting, will look
The Ps5 looks more like wrong reflection to be honest
EDIT: I see the poster two posts above me pointed out this
I don't know about this. Gsync has been great for me on PC. If I had to lock every game to 60fps I wouldn't be able to play at the settings I'm playing at. If you're locking at 60 then you need to sacrifice a lot in order to have the overhead for demanding moments in the game. I don't understand this absolutist view. VRR is great and should be utilized by those who have it. They should still have a "locked" mode for those who don't of course.
I suppose it's a crutch if that's the only mode you have, but the option should be there for those who want it.
I would say that if the game was using some form of resolution scaling and target 1440p60 with RT then XSX would likely run at a higher average resolution than PS5. Not by much but noticeable sometimes.
Which is pretty much what was expected from their specs really.
Which is pretty much what was expected from their specs really.
It is actually almost an exact difference in their peak flops.
Are we looking at the same pic? The PS5 has "wrong" reflections?
I don't know why you quoted me there. That's pretty much what I've been saying.
And I'm also saying, that taking the minimum and maximum achieved fps isn't important when comparing two systems together. It's more important comparing the relative/percentage difference between them. As that is what translates to real world gains or losses.
Because we can and will always find outlier situations.
Bulby's post shows a wrong reflection, but I don't think the one posted here is wrong.
The thickness of the plot has increased.I was a browsing reddit when I found this (please focus on the screenshots):
Reddit - Dive into anything
It seems like a nice benchmark but in the end I think it's pointless as a measure of RT power, it needs more parity, I guess. Any opinions?
Maybe the raytracing is just buggy and sometimes it doesnt catch some geometry. We did get remedy saying they had the same settings iirc.
Not sure if serious. Only one, maybe two comparisons are actually valid and hopefully Dictator posts about this and shares his thought. I think, yes there are some incredible minor diffences, however overall the video was great.
I still think this is the closest we have seen a comparison outside of cutscenes yet, because usually during gameplay the comparisons differ more due to uncontrollable AI and such. Interesting nonless, but I can't deny I juggled a bit, because nobody complains about gameplay finding, which usually differ way more (NPC count, effects, ...) and the fact that in the interest of time DF can't test literally everything and can't replicate everything. No criticism of course towards DF!
Last edited:
There are dozens of minor visual discrepancies between the versions.
It's either a bug or just some kind of natural difference between the game versions. The XSX version isn't using VRS either. Also in that "missing reflection" picture, the MC is a step closer on PS5. It could be a draw distance thing or the texture simply didn't load in yet. These little "gotchas" are missing the forest for the trees - for instance, that missing texture isn't going to increase the FPS by so much lol - and just seem to be weird conspiracy theory console warrior stuff although rightfully interesting in their own right.
It's either a bug or just some kind of natural difference between the game versions. The XSX version isn't using VRS either. Also in that "missing reflection" picture, the MC is a step closer on PS5. It could be a draw distance thing or the texture simply didn't load in yet. These little "gotchas" are missing the forest for the trees - for instance, that missing texture isn't going to increase the FPS by so much lol - and just seem to be weird conspiracy theory console warrior stuff although rightfully interesting in their own right.
Last edited:
I dont understand what the issue may be here? It was expected that the XSX would pull ahead in head to heads, all things being equal, as its the more powerful console.
I expected a +15% to +25% differential in expected performance based on target settings. After early issues with the XSX tools we are starting see the power play out.
PS5 exclusives is where we will see the console truly sing imo. Same applies XSX as well, exclusives is where the consoles power is fully utilised.
I expected a +15% to +25% differential in expected performance based on target settings. After early issues with the XSX tools we are starting see the power play out.
PS5 exclusives is where we will see the console truly sing imo. Same applies XSX as well, exclusives is where the consoles power is fully utilised.
I don't think there is an issue. Just a benchmark pointing out a performance difference in a game with like for like settings.
I can't wait what arguably the most talented first party studios in their roster respectively (Naughty Dog, The Coalition, Polyphony Digital and Playground Games) will do.
Yep, I did a longer post earlier, but you can basically use these little gotcha moments all the time, outside of cutscenes then. That's not a dig at DF, because it's impossible to get the same scenarios during gameplay.
Don't get me wrong, pointing those differences out is kinda the point of these threads. However I think this is simply down to slightly different positioning (the reflection in the first pic) and not different settings, because iirc the developers said the settings are the same.
People want their box to somehow be the most powerful, when it's not. It's what it is and people need to move on from it. Both consoles this gen are freaking amazing and honestly it's great for us.
That's what's interesting though right? The delta seen in this synthetic benchmark isn't representative of gameplay where the XS consoles stutter and drop frames moderately more significantly than PS5. At least in the examples shown in the initial comparison video.
The examples shown during gameplay were mostly those ackward stutters whenever huge font was displayed right? I wouldn't be surprised to see this fixed.
Yeah I'd imagine it will be fixed. Maybe it's XDK related since it's similar to behavior in other titles as well.
I would speculate that has nothing to do with GPU/CPU specs. It literally happens when coming out of menus or when text pops up on screen. That's a different issue altogether that seems more like a bug. Normal gameplay sees parity amongst the two consoles.
Last edited:
My view on all of this: for 3rd party games you will see almost identical performance, even if XSX has say 15% overhead, it wouldn't make sense for any studio to push visual or performance fidelity to take advantage of that: time + resource = money, unless it is as easy as a toggle switch. Most, if not all 3rd party games will target the exact same baseline specs. After 3 or 4 years, when some studios are very confident on their approaches, you might start seeing differences if the resources spent don't cost much.
You will see the most differences from first party efforts, whether its taking advantage of the faster IO, or more compute units. But the problem remains, these will be absolutely different games, so there will never be 1 to 1 comparisons, as hard as platforms like DF try to analyze that. Fanboys of respective platforms will always call their winners based on art-style and other nebulous factors. Unless there is a 3rd party benchmark demo created specifically to utilize every ounce of both platforms, most people will never agree on the measures.
You will see the most differences from first party efforts, whether its taking advantage of the faster IO, or more compute units. But the problem remains, these will be absolutely different games, so there will never be 1 to 1 comparisons, as hard as platforms like DF try to analyze that. Fanboys of respective platforms will always call their winners based on art-style and other nebulous factors. Unless there is a 3rd party benchmark demo created specifically to utilize every ounce of both platforms, most people will never agree on the measures.
All games should be using dynamic resolution, a 60FPS mode that scales between 1080p and 1440p would be ideal. Dropping the resolution all the way down to 1080p when it already hits 60 at 1440p in some scenes just seems overkill. I don't understand why games like Control, Resident Evil and Devil May Cry aren't using Dynamic Resolutions.
We can't just live with the 'Remedy said they are the same' because there are differences in IQ. What's the reason, or if that difference renders the comparison unfair is the question.
On the idea 'Comparisons will never be 1:1', of course they won't, Alex is a human positioning the camera after all. Doesn't mean we shouldn't investigate why there are differences or strive for more closely matched images.
On the idea 'Comparisons will never be 1:1', of course they won't, Alex is a human positioning the camera after all. Doesn't mean we shouldn't investigate why there are differences or strive for more closely matched images.
Running DRS don't cost anything. A faster machine will hit higher resolution more often.
Another possibility is to have PS5 running a couple of settings at a lower level like low/med with XSX running at med/high.
Also shouldn't cost much since these are LODs mostly which are being implemented for PC anyway.
Lol, I know this is a typo, but it's suitable for the direction this thread is taking.