Multi-platform Game Comparisons - Console Wars Redux

Yeah, it’s really impressive how much detail there is, and like I said, it’s all tagged and searchable super easily. It is slightly buried under a non-default tab when you look at the game info, but that’s one of those ‘discoverability’ non-issues.

To me this was the most interesting Control comparison yet.

@stusser notice after comparing the Series X and PS5 in identical scenarios with identical quality settings, they try to do the same with a PC. Except PC has 2x the ray tracing resolution and that can’t be changed. But it’s interesting that the AMD 6800XT can’t match the performance of even the PS5, even though it’s supposed to be a better GPU. They speculate that it’s the 6800XT drivers, and I can’t think of any other explanation myself, so that must be it.

I watched the digital foundry CONTROL video, then cracked into the game on m XSX. I swear I put raytracing on in the options before I started,. Played a bit, it didn’t look much different than when I started last month. I got to the first control point, then decided to go look again, and it turns out I had it set to Performance. So I switched to Graphics, and it was like night and day (you can switch on the fly). Been playing that way since, and I’ve reaffirmed that I really don’t care much about framerate. Solid 30 vs. 60? I wouldn’t say I can’t tell, but it just doesn’t matter at all, and is not affecting my playing in the least.

Considering the early benchmarks favored the PS5 it’s interesting to see the Xbox Series X having an average advantage of 16% better performance in Control Ultimate.

This would be pretty universal if so right?

It’s clear there are two bottlenecks in this generation; the Xbox Series S and the PS5. It’s likely that developers will just ignore the Series S and get it to work “as well as they can”, but might have to do things they didn’t want to do w/re to memory allocation on the X and PS5 to fit the Series S.

OTOH, they’re likely to target the PS5 as the performance standard, which means the X’s hardware performance advantage won’t show up. Basically it’s going to be the PS5’s generation and the Xbox is just going to be living in it. The problem for MS here is that they’re going to need 1st party games or special deals with devs to show off the X’s hardware… which, for 15%-20% increase, isn’t going to be all that special anyway since devs seem to want to lock performance to 30 or 60.

Yes, I wonder it seems like that is a software related issue though. In other words, if Microsoft feels that the size of the games is an issue couldn’t they develop a better compression algorithm to fix the issue?

It’s only during photo mode when none of the gameplay simulation is actually running. There are also some screen comparisons that were posted on reddit showing the XSX version missing certain details, both in the RT reflections, and other areas, so it’s not an exact like for like test.

I thought Remedy indicated that the settings on both were identical. I would assume those issues are small bugs, and likely don’t impact performance significantly for the purposes of the test.

They’re supposed to be the same, but the missing details could definitely impact performance.

The details were minimal. I mean, these are two different platforms - hardware differences, software differences - but with the dev saying settings are identical, this is more or less a great benchmark for what they were testing, which is rendering minus the computation of the remaining game elements.

In a PC setting, setting things identically across different hardware you still may see differences solely because of the different rendering pipelines. I don’t think what you’re highlighting is that meaningful. I think this test - for this title, testing in this way - can be taken largely at face value.

Remedy said they are the same and DF concluded they were like for like. If they had spotted noticeable differences, which is what they do for a living, they would have said so.

Fact is, PS5 is going to be the slightly lower target for devs. I disagree it’s really going to constrain the Series X. We’ll likely get slightly higher res and a few better settings on Series X. Nothing game changing though, as the machines are still fairly close.

As for the PS5, it seems like it will deliver the best load times and awesome first party titles.

And while we wait for the Xbox exclusives (a bit of a familiar tune), we have more games than we can play on GamePass.

All in all, seems like both platforms are fine and neither is going to feel constrained by or in the shadow of the other.

DF aren’t infallible. And it doesn’t really matter if the devs say the settings are the same if the visual difference is easy to see. We’re not talking about minor gamma or color differences.


Whoa. I think Brad has convinced me. I think i finally realize how those old folks complaining about how the number of rivets on the plane models felt in a flight sim. Now that it’s been pointed out, I can’t unsee it. CONTROL on the Xbox is clearly inferior, I have no idea why I’m still playing it.

My apologies for introducing facts and evidence to the discussion. I know how much you guys hate that.

Just so we’re clear, you’re contending that the Xbox gets some kind of performance boost because Remedy has changed the assets and/or ray tracing rendering on the Xbox version of the game, which they have denied. You think this because of some screenshots folks took of a video on YouTube.

What do you say when I present a screenshot I just took, in that area, that shows that reflection being there?

Obviously that you are a Microsoft shill with photoshop skillz.

Try the other side of the room. I read last night the bug might only affect one side of the pillar there.

And just so we’re clear, I’m not contending Remedy “changed” anything. I’m questioning the validity of a performance comparison when the benchmark video itself displays divergent visual results.

For right now, I think which console sells more each month is going to come down to availability as much as anything else.