Support NeoGAF

[Gaiff]

So, some cool stuff in there. Rich has save files on PC meant to be benchmark runs. They're all scripted in-game and real time sequences to stress the PC. He used the cross-save feature to export those saves to Series consoles and PS5. The scripting remained intact and they serve as perfect 1-to-1 benchmarks.

The interesting bits.

• The benchmarks for Quality/RT and Performance mode are locked to 30 and 60fps respectively
• Both consoles output at 1800p using FSR2 with a maximum internal resolution of 1440p
• Series X in performance mode has a lower bound of 1152p vs 1008 for the PS5
• In the RT mode, the performance difference is irrelevant. They both lock to 30
• First benchmark and least demanding one: Series X drops 36/3600 frames (1%), PS5 25/3600 (0.7%)
• Second benchmark: Series X drops 74/3100 frames (2.4%), PS5 34/3100 (1.1%)
• Third benchmark: Series X drops 1264/15500 frames (8.2%), PS5 32/15500 (0.2%)
• Performance can be up to 33% better on the PS5 in matched frames with Series X having lows of 45fps
• Rich also used an AMD 4800S desktop kit which has a CPU identical to the Series X paired with a 6700 downclocked to 2.23GHz to match the PS5's clocks
• They tweaked the PC version to be a match for the PS5's settings and DRS range
• They ran the third benchmark on all three: Series X 1264/15500 (8.2%) dropped frames with a 55.1fps average, PC 218/15500 (1.4%) for a 59.2fps average, PS5 leading with 32/15500 (0.2%) with a 59.9fps average, and finally series S 432/15500 (2.8%) with a 58.3fps average
• Pairing the 4800S with a 7900 XTX resulted in no dropped frames, confirming that the results on consoles are GPU-limited
• Also tweaked the 4800S/6700 to match the slightly higher DRS of the Series X. PC was still ahead
• Series X/S seem to be using VRS

Edit: DRS windows

 

[Mr.Phoenix]

Nice summary, just watched this and was about to make a thread lol.

It would have been great if the consoles had an unlocked mode.

 

[Gaiff]

Nice summary, just watched this and was about to make a thread lol.

It would have been great if the consoles had an unlocked mode.

I was actually waiting for adamsapple 's summary but decided to get off my ass and do it.

 

[Horseganador]

Well, the performance difference is too big, between 10-15 fps in favor of PS5 in this test.

 

[jm89]

• Performance can be up to 33% better on the PS5 in matched frames with Series X having lows of 45fps

 

[Gaiff]

Well, the performance difference is too big, between 10-15 fps in favor of PS5 in this test.

Yep. PS5 is rock solid in that sequence. A couple of frame drops here and there but by and large, sticks very close to 60fps. Series X is all over the place with a lot of screen tearing and tons of drops into the 40s. This is even more puzzling because a PC with DX12 and what should be significantly more overhead actually outperforms the Series X at matched settings and DRS.

A console should at the very least match an identically-specced PC. It loses a lot of its appeal if it just performs worse.

 

[Gamer79]

 

[tusharngf]

Well, the performance difference is too big, between 10-15 fps in favor of PS5 in this test.

resolution is also different for both machines

 

[analog_future]

resolution is also different for both machines

That doesn't account for the performance difference though.

 

[Gaiff]

resolution is also different for both machines

Sure, but it's really not a massive difference.

So, either the PS5 is much more aggressive with its DRS and consistently runs it at a much lower resolution to maintain 60fps, or the Series X just massively underperforms. I think it's the latter since Olivier did a comparison and besides the lower resolution floor, the PS5 doesn't have a consistently much lower resolution.

Hell, the VRS in stills is more noticeable.

Xbox loses that one pretty badly. That benchmark run would probably look worse for the Series X if the PS5 wasn't capped to 60fps.

 

[Insane Metal]

Another one and it isn't close in this case.

 

[rodrigolfp]

A console should at the very least match an identically-specced PC

If DSR is working identically.

 

[JohnnyFootball]

This is a SHOCKING result. I recently completed a Cyberpunk 2077 playthrough on my PS5 and I was amazed at how well it performed and how good it looked. I played it on my XSX a year and a half ago.

I didn't expect their to be an advantage for the PS5.

 

[adamsapple]

Yep. PS5 is rock solid in that sequence. A couple of frame drops here and there but by and large, sticks very close to 60fps. Series X is all over the place with a lot of screen tearing and tons of drops into the 40s. This is even more puzzling because a PC with DX12 and what should be significantly more overhead actually outperforms the Series X at matched settings and DRS.

A console should at the very least match an identically-specced PC. It loses a lot of its appeal if it just performs worse.

The game has always had higher average fps on the PS5 in all of its previous tests, results wouldn't change because they imported PC benchmark specs.

But CDPR have switched out Red Engine for UE5 for their future projects so this is probably their last game with this kind of difference.

 

[JohnnyFootball]

Yep. PS5 is rock solid in that sequence. A couple of frame drops here and there but by and large, sticks very close to 60fps. Series X is all over the place with a lot of screen tearing and tons of drops into the 40s. This is even more puzzling because a PC with DX12 and what should be significantly more overhead actually outperforms the Series X at matched settings and DRS.

A console should at the very least match an identically-specced PC. It loses a lot of its appeal if it just performs worse.

A console should wipe the floor with an identically-specced PC.

 

[YOU PC BRO?!]

A very large difference. It certainly appears like Series X requires further optimisation. It’s becoming clearer by the day that PS5 is the lead console platform.

 

[JohnnyFootball]

The game has always had higher average fps on the PS5 in all of its previous tests, results wouldn't change because they imported PC benchmark specs.

But CDPR have switched out Red Engine for UE5 for their future projects so this is probably their last game with this kind of difference.

Moving to UE5 is a not a move that I am in favor of. I have not been a fan of that engine as so many games have been a stutter fest. Epic needs to prioritize fixing that issue.

 

[MisterXDTV]

A very large difference. It certainly appears like Series X requires further optimisation. It’s becoming clearer by the day that PS5 is the lead console platform.

But I thought Series X was Microsoft's mid-gen console?

How is it possibile that it performs worse than the base PS5???

 

[Gaiff]

A console should wipe the floor with an identically-specced PC.

No. This isn't 2001.

If DSR is working identically.

They should make it more aggressive on Series X in that case.

 

[Mr Moose]

A very large difference. It certainly appears like Series X requires further optimisation. It’s becoming clearer by the day that PS5 is the lead console platform.

Didn't Xbox have marketing with this game? And I thought PC games were easier to port to Xbox?

 

[anothertech]

Series S?

 

[Mr Moose]

Series S?

Better than Series X, worse than PS5, but it's at a lower res and stuff.

 

[Xyphie]

Well, the performance difference is too big, between 10-15 fps in favor of PS5 in this test.

Benchmark is also clearly non-deterministic given that there are cars in the XSX view and not in the PS5 one. So you have to find shots that are roughly equivalent if you want to do a fair comparison.

 

[Gaiff]

Series S?

Runs better than Series X. I updated the summary.

432/15500 dropped frames with a 58.3fps average vs 1264/15500 dropped frames with a 55.1fps average in favor of the Series S.

 

[MikeM]

Blows my mind that this game does not have a vrr/uncapped option. Wtf devs…

 

[winjer]

Benchmark is also clearly non-deterministic given that there are cars in the XSX view and not in the PS5 one. So you have to find shots that are roughly equivalent if you want to do a fair comparison.

But after a few minutes, the average should even out.
One moment the PS5 might have more cars, other times, the Series X has more.

 

[Gaiff]

Benchmark is also clearly non-deterministic given that there are cars in the XSX view and not in the PS5 one. So you have to find shots that are roughly equivalent if you want to do a fair comparison.

There are run-to-run variances but come on, they won't explain the massive advantage of the PS5.

 

[Xyphie]

There are run-to-run variances but come on, they won't explain the massive advantage of the PS5.

Why not? XSX code runs at a ~26.4% higher lower-end resolution which is pretty massive knowing how GPUs scale on PCs. Knowing that pixel delta you can find an equivalent scene which frame drops below 60FPS on both systems and you can roughly calculate a performance delta between the two.

 

[sendit]

Well, the performance difference is too big, between 10-15 fps in favor of PS5 in this test.

Was this update pushed with the latest Microsoft Xbox GDK commit? Without this info, I think we can safely say this comparison is flawed.

OCTOBER_2023_GDK_UPDATE_1 · microsoft/GDK@279af1f

Microsoft Public GDK. Contribute to microsoft/GDK development by creating an account on GitHub.

github.com

/s

 

[Denton]

Nice that PC performs identically to consoles at same specs and resolution. Very good DX12 optimization by CDP.

I hope CDP programmers will improve Unreal engine significantly for their game.

 

[Gaiff]

Why not? XSX code runs at a ~26.4% higher lower-end resolution which is pretty massive knowing how GPUs scale on PCs. Knowing that pixel delta you can find an equivalent scene which frame drops below 60FPS on both systems and you can roughly calculate a performance delta between the two.

Decreasing the resolution by 14% per axis won't shoot your performance up by over 20% on a 6700-tier GPU. When the PS5 drops to 57fps, the Series X tanks to the mid-40s. And that's not even getting into VRS.

 

[Tripolygon]

Why not? XSX code runs at a ~26.4% higher lower-end resolution which is pretty massive knowing how GPUs scale on PCs. Knowing that pixel delta you can find an equivalent scene which frame drops below 60FPS on both systems and you can roughly calculate a performance delta between the two.

VRS is used on XSX which explains the higher lower-end DRS. Instead of dropping the output resolution, certain parts of the image are shaded at quarter resolution while PS5 remain at native resolution. Still does not explain the performance discrepancy.

Simple answer, some engines and games favor certain hardware setup. You win some, you lose some.

 

[Mr Moose]

Why not? XSX code runs at a ~26.4% higher lower-end resolution which is pretty massive knowing how GPUs scale on PCs. Knowing that pixel delta you can find an equivalent scene which frame drops below 60FPS on both systems and you can roughly calculate a performance delta between the two.

45 v 57. So yeah dropping the res would likely fix it which I think is what Richard said in the video.
Edit: I forgot about VRS.

 

[acm2000]

i swear MS stipulates that the lower end of drs must be higher than ps5 or something since this happens a lot, just drop the drs to the same as ps5 lowest res and be done with ti.

 

[YOU PC BRO?!]

Didn't Xbox have marketing with this game? And I thought PC games were easier to port to Xbox?

Did someone say PC games weren’t easier to port to Xbox? Point em out. I’ll laugh at them with you.

Turns out that games require console specific optimisations. Hasn’t that been true since… forever? In programming, making code ‘generic’ can often come at the cost of performance. Unfortunately, none of us here can determine the cause of the low frame rates we are seeing on the Series X. 30% - 40% performance differentials to the PS5 are clearly not typical. That much is clear.

 

[SkyHighTrees]

I'm just waiting for those walls of text, ya feel me?

Best believe the performance delta is about to be explained

 

[YOU PC BRO?!]

But I thought Series X was Microsoft's mid-gen console?

How is it possibile that it performs worse than the base PS5???

Wait… is this passive aggressive fanboying? Is this the new meta?

Are you suggesting that the poor performance on display in this single game is indicative of Series X performance across all cross-platform games?

 

[Topher]

Wait… is this passive aggressive fanboying? Is this the new meta?

Are you suggesting that the poor performance on display in this single game is indicative of Series X performance across all cross-platform games?

It's the kind of logic thrown out every single time one console comes out on top of the other. Last time was Alan Wake 2. Rinse....repeat.

 

[Darsxx82]

The game has always had higher average fps on the PS5 in all of its previous tests, results wouldn't change because they imported PC benchmark specs.

The interesting thing is that the performance in XSX in the old/original zone was better before the Phantom Liberty update. But even more curious is that in the new area that includes Phanton liberty (much more demanding as Rich says in the video) the performance between XSX and PS5 is equal. In fact, a comparison of that area would have been interesting as well, and even more so when Rich himself recognizes that it is the most demanding.

And if we want to complicate things in strange situations... We have the Phanton Liberty comparison before launch where the XSX version performed better than the PS5 version.




The RedEngine has been a rather strange engine in its performance on consoles. The Witcher 3 also offered very different results where XSX clearly won in more situations and yet PS5 had better performance in the city.

 

[Md Ray]

Cyberpunk 2077 'Performance Mode' console equivalent settings for those curious:

Resolution:

Graphics:
Min res: 56% is 1008p. Change it to 64% for 1152p (if you want to replicate Xbox Series X's lowest resolution).

 

[Mr.Phoenix]

The only thing I love more than these data-laden DF threads is listening to the reasons people come up with to discredit the results.

I would think that after 3 years we could all have agreed that there is very little between these two consoles. Sometimes on is better, sometimes the other is... and even when one is better, its seldom to the effect that it is glaringly obvious.

 

[shamoomoo]

That doesn't account for the performance difference though.

It does to a degree but I'm not sure how big of a gap 1008p vs 1152p is.

 

[Mr Moose]

The interesting thing is that the performance in XSX in the old/original zone was better before the Phantom Liberty update. But even more curious is that in the new area that includes Phanton liberty (much more demanding as Rich says in the video) the performance between XSX and PS5 is equal. In fact, a comparison of that area would have been interesting as well, and even more so when Rich himself recognizes that it is the most demanding.

And if we want to complicate things in strange situations... We have the Phanton Liberty comparison before launch where the XSX version performed better than the PS5 version.




The RedEngine has been a rather strange engine in its performance on consoles. The Witcher 3 also offered very different results where XSX clearly won in more situations and yet PS5 had better performance in the city.

How accurate is Toms counting? They were really 900p?

 

[Damigos]

I usually see better frame rates on PS5 in comparison to XSX. I wonder if this is because of the unified memory or of the way the SSD works. I m curious since XSX is about 2 Tflops better than PS5

 

[Gaiff]

The only thing I love more than these data-laden DF threads is listening to the reasons people come up with to discredit the results.

I would think that after 3 years we could all have agreed that there is very little between these two consoles. Sometimes on is better, sometimes the other is... and even when one is better, its seldom to the effect that it is glaringly obvious.

I'm confused with the SX results though. The PC version using the PS5 settings easily beats it, and using the Series X settings, it still wins comfortably, albeit by a slimmer margin. I thought maybe it was some DX quirk but this doesn't appear to be the case.

 

[Skifi28]

Rdna 1.5 delivering the goods.

 

[skyfinch]

PlayStation is unbelievably poweful. I might get one instead of a pc next gen.

 

[Darsxx82]

How accurate is Toms counting? They were really 900p?

I dont know how accurate he was, perhaps in that pre-release version the resolution was the same on both consoles? That said, it is also true that in the Phantom Liberty zone the performance of the XSX and PS5 is the same and that video perhaps only reflects that.
The comparison in the Phantom Liberty area would have been interesting.


Then, in the video it seems that there are visual inconsistencies in the PS5 version. I don't know if the game generates details randomly or Is popping but there are more visual details in XSX in the same scenes.

 

[GHG]

Just checking, does this qualify as a "real next-gen" game?

 

[Xyphie]

Decreasing the resolution by 14% per axis won't shoot your performance up by over 20% on a 6700-tier GPU. When the PS5 drops to 57fps, the Series X tanks to the mid-40s. And that's not even getting into VRS.

It's roughly the difference between going from say DLSS/FSR Quality (0.66x) to Balanced (0.58x), 0.66/0.58 = 1.137. Just ran it and it's solidly ~20% faster on my 3080. Not surprising, GPU-limited perf tends to scale fairly linear with resolution.