• Hey, guest user. Hope you're enjoying NeoGAF! Have you considered registering for an account? Come join us and add your take to the daily discourse.

[NX Gamer] AC: Valhalla - XsX vs PS5 - The first REAL head to head battle.

Im just wondering if the improvements will be noticeable since it's already so close and the PS5 will get better with time.
Don't know to be quite honest. I do expect resolution and fps to be better on xbox series x, but load times / pop ins to be better on PS5. Also, just to set the record straight, I'm getting neither console. I am getting xbox series s just because I'm cheap and want to play all the xbox game pass games. So I have no stake in the XSX/PS5 fight despite my username lol.
 
Last edited:
Don't know to be quite honest. I do expect resolution and fps to be better on xbox series x, but load times / pop ins to be better on PS5.

I've seen people make claims of a 50% difference but I don't think it's going to be anything near that especially with such a small difference in the beginning.
 

rnlval

Member
Nope, variable pervasive fine gated clocks are part of big Navi and RDNA2, point 1 below and per CU based (could be DCU)

A slde that poeple missed or did not get from AMD about big NAVI

Qcbe7Rt.png
Note that PS5's Shader Engine per CU count ratio (two shader engines, 18 active CU per SE) is the same as RX 5700 (two shader engines, 18 active CU per SE) and RX 6800 XT (four shader engines, 18 active CU per SE).
 
Hell no, people are on crack if they think that on either side imo.

It was a theory combining the extra flops with a full set of RDNA2 features. I've seen it mentioned plenty of times in the speculation thread. But then again I've seen people claim that the PS5 was going to have broken multiplats and I haven't seen that yet.
 

rnlval

Member
You dont know ps5 L2 cache, so incorrect.

You dont know ps5 L1 cache structure either. You dont know how ps5 cache coherency and scrubbers work, the specfific architecture and effects on effective bandwidth, so again you dont know.

All we know is a 36 CU part, which is clocked higher,l runs this game better than a 20 % bigger die XSX with 52 CU.

And its too late to make hardware claims and how it might affect performance, we can all see the results.

Ccjbn15.png
 

MrFunSocks

Banned
It was a theory combining the extra flops with a full set of RDNA2 features. I've seen it mentioned plenty of times in the speculation thread. But then again I've seen people claim that the PS5 was going to have broken multiplats and I haven't seen that yet.
Anyone claiming that is an idiot, frankly. It’s annoying because all of Team Blue use those figures to try and then bash the Xbox, when those figures were never even remotely possible.
 
It was a theory combining the extra flops with a full set of RDNA2 features. I've seen it mentioned plenty of times in the speculation thread. But then again I've seen people claim that the PS5 was going to have broken multiplats and I haven't seen that yet.
Nah, don't see that happening. The only main difference for me at the end of the day is one company provides xbox game pass and a cheaper console. Call me a casual, but I don't care. Still love my nintendo switch too besides being under power.
 

ethomaz

Banned
Nah, don't see that happening. The only main difference for me at the end of the day is one company provides xbox game pass and a cheaper console. Call me a casual, but I don't care. Still love my nintendo switch too besides being under power.
Perfect.

$399 for a PS5 is really a steal... it is the same price of the console in 2013... Sony maintained the same price.

Performance, graphics, speed, etc all too notch at $399... no limitation in hardware.
 
Last edited:
Perfect.

$399 for a PS5 is really a steal... it is the same price of the console in 2013... Sony maintained the same price.

Performance, graphics, speed, etc all too notch at $399... no limitation in hardware.

Lol, I see you're about the performance/graphics, but it still missing xbox game pass which will save me hundreds if not thousands of dollars on games. Not to mention xbox one series s is 299.
 
Last edited:
Remember back in 2013 the PS4 had the obvious edge on the likes of COD, BF4 & AC Black Flag at launch over the Xbox One. The reality of multiplatform parity hitting some folk hard this launch considering how GAF has been these past few months

tenor.gif
 

ethomaz

Banned
If anything MS invested a lot more on helping Ubisoft pushing Valhala out than Sony did. Marketing rights meaning it’s going to be the single biggest game running a marketing campaign around the Xbox brand this season.

Looking at how slow the loadings are on PS5, it’s clear they didn’t bother much there.
The marketing deal just prevented the PS5 version to be showed and/or sent to journos before launch.
 
Last edited:

JonkyDonk

Member
tJvlc4f.png


The red indicators are also associated with a significant frame rate drop on PS5
I'm pretty sure it's the mix of the explosion effect, the crowd and the blur that causes the drop. In this instance the particular effect only happened on PS5. In the previous image, the same effect happened on both but slightly different times, and one held up better than the other.
 

Md Ray

Member
Come on now. The PS5 drops just as low just after that scene. But there is actually a scene that I find puzzling, where PS5 has a definitive advantage: in the city with plenty of NPCs. PS5 never drops here and it's not about GPU. It's about CPU and both CPUs are virtually identical.

4VeCnRH.jpg


DeepEnigma DeepEnigma is faster than me!
It's not CPU. This same exact scene with NPCs on-screen runs between 30-27fps on base PS4, the console with the slowest Jaguar CPU. And the Zen 2 inside SX should easily be more than 4x faster than PS4's CPU. So it's definitely not CPU.

Come on now.

Source: from 5:10
 
Last edited:

MrFunSocks

Banned
Look at the frame graph, the same effect happened a bit earlier on XSX and the frame drop there was bigger compared to PS5. All the way down to 50 in fact.
I think you’ll find the poster was showing how stupid all these cherry picked screenshots are. The game performs essentially exactly the same on both consoles.
 

JonkyDonk

Member
I think you’ll find the poster was showing how stupid all these cherry picked screenshots are. The game performs essentially exactly the same on both consoles.
I agree with that for the most part honestly. But I still think in the context of so many people shouting that the PS5 is underpowered for so long, this comparison is a massive turnaround and vindicates everyone who said the consoles are basically identical in power. And that shouldn't be downplayed or swept under the rug.

In other words, the gloating is well deserved!
 
I agree with that for the most part honestly. But I still think in the context of so many people shouting that the PS5 is underpowered for so long, this comparison is a massive turnaround and vindicates everyone who said the consoles are basically identical in power. And that shouldn't be downplayed or swept under the rug.

In other words, the gloating is well deserved!

Take into account that alot of people were afraid of some of the negative rumors turning out to be true. Especially since they were being spread so much on twitter. Many thought the PS5 would end up being an extremely underpowered broken overheating console. But thankfully that didn't turn out to be the case.

Seeing multiplats perform so well just makes me confident that I'm going to get a well designed system.

Anyways for those that believed the FUD hopefully they learn a valuable lesson from this. And that's not to believe any of those ridiculous twitter users that spread those rumors in the first place. And also that marketing is just marketing. The real world results can be very different than what's advertised.
 

JonkyDonk

Member
Take into account that alot of people were afraid of some of the negative rumors turning out to be true. Especially since they were being spread so much on twitter. Many thought the PS5 would end up being an extremely underpowered broken overheating console. But thankfully that didn't turn out to be the case.

Seeing multiplats perform so well just makes me confident that I'm going to get a well designed system.

Anyways for those that believed the FUD hopefully they learn a valuable lesson from this. And that's not to believe any of those ridiculous twitter users that spread those rumors in the first place. And also that marketing is just marketing. The real world results can be very different than what's advertised.
Very true. And to be fair I still expect that Xbox will have its wins along the way in some places, as will PS5. That's how close it is. I think there will be many instances where within the same game, some sections will perform better on one console and then it will flip in another section.
 
Last edited:

inflation

Member
It's not CPU. This same exact scene with NPCs on-screen runs between 30-27fps on base PS4, the console with the slowest Jaguar CPU. And the Zen 2 inside SX should easily be more than 4x faster than PS4's CPU. So it's definitely not CPU.

Come on now.

Source: from 5:10

Very interesting results... It seems PS4 Pro maintains 30FPS the whole time, kinda matching what DF said in WD:L comparison that Ubisoft's main optimization target is PS4 Pro.

The first platform dropped frames is Xbox One S, then shockingly Xbox One X, after that base PS4. That brings more questions than answers...🧐
 
Last edited:
Very true. And to be fair I still expect that Xbox will have its wins along the way in some places, as will PS5. That's how close it is. I think there will be many instances where within the same game, some sections will perform better on one console and then it will flip in another section.

Honestly the only questionable console here where it comes to power is the XSS. The PS5 and XSX both have enough juice to last a while. The XSS might be a different story.
 

Md Ray

Member
I've also read read that developers still have to get used to the inner workings of the PS5 so things will get better for Sony as well. But so far it seems like the PS5 is a very well designed system.
I got the same impression after we got the full specs of both machines. Especially after the HotChips conference where MS went into more detail regarding their GPU... And then when I compared it to PS5, it was clear to me PS5 was a better designed machine overall. And the games showcasing this to be true is the proof of that.

That said, there will be situations where SX will pull ahead in scenes with ALU bound and there will be scenes where PS5 will pull ahead (like in AC:Valhalla) if it's fillrate bound. And will likely be close and within margin of error if the scene is both ALU and fillrate heavy.
 

Md Ray

Member
Very interesting results... It seems PS4 Pro maintains 30FPS the whole time, kinda matching what DF said in WD:L comparison that Ubisoft's main optimization target is PS4 Pro.

The first platform dropped frames is Xbox One S, then shockingly Xbox One X, after that base PS4. That brings more questions than answers...🧐
Yes, if you look at the GPU of both PS4 Pro and X1X, there's 2x more ROPs on Pro resulting in 55% more pixel fillrate in Pro's favor. Not all parts of the Pro's GPU is inferior.

This is why devs say TFLOPS alone don't tell the whole story. The PS5 also has pixel fillrate advantage over XSX GPU.
 
Last edited:

MrFunSocks

Banned
I agree with that for the most part honestly. But I still think in the context of so many people shouting that the PS5 is underpowered for so long, this comparison is a massive turnaround and vindicates everyone who said the consoles are basically identical in power. And that shouldn't be downplayed or swept under the rug.

In other words, the gloating is well deserved!
So the same goes for the magic SSD secret sauce too then, right? All the dozens of topics about how incredible it will be and how all games will load instantly and you'll never see a load screen on the PS5 and how it'll be 3x+ faster on the PS5 shouldn't be downplayed or swept under the rug too, correct?

This is a cross generation multiplatform launch day game. It's not indicative of anything. Launch games have never been a good indicator of how consoles will perform. Launch games are always just cross gen up-ports, generally just with better framerate and resolution because that's the easiest thing to brute force.

Anyone gloating about this being equal is dumb.
 

DeepEnigma

Gold Member
Wait for same answers magical geometry engine, cache scrabbers an new shit "l3 shared cache", turbo nitro ssd bla bla bla bla bla

Low high speed cache with low latency seems to be working, so far.

Nitti gritty inner working customization nuance does seem to be lost on shallow surface level spec sheets though.
 

Md Ray

Member
I'm expecting them to be very similar, bit not necessarily PS5 being /better/.

Sorry NXGamer has a bias. So I'll be waiting on Digital Foundry, which won't be too long now.

If there is a difference between the consoles, I might just sub to NXGamer, but for now, the bias is too off-putting.
Someone didn't catch the VG Tech vid/stats.
 

Md Ray

Member
Am I wrong to say that? Going down the spec sheet Microsoft xbox series x is more powerful than PS5 in virtually everything. Microsoft was late and therefore developers weren't able to utilize it as much. DMC is getting a late patch because Capcom just like almost every developer received the dev kit late. Not to mention this game can still get a patch for xbox series as well as PS5.
Another guy looking at just TF numbers and calling it a day.
 

Md Ray

Member
Lol average frame rate of 59.9fps with like 0.01% tearing is “broken” now? Lol

I don’t understand why people think the Xbox version is some sort of tearing me like GT5 was, but the similarly tearing ps5 version is superior? They both tear, but the PS5 version also drops frames. On the Xbox you can mitigate the tearing. on the ps5 you can’t.

VRR is great because it allows games to run higher than 30 but below 60. Games like God of war 3 ran terribly because the frame rate sat around 45fps during fights. With VRR that’s fine, plays great. Without VRR it’s all stuttery.

VRR isn’t a crutch like you think it is, it’s literally better for games and developers. It allows them to push the hardware more as they don’t have to worry about locking the frame rate.
PS5 holds 60fps nearly 100% of the time while XSX holds it 87% of the time and has 5x more tearing than PS5. The result is from VG Tech testing further into the game.
 

Ev1L AuRoN

Member
So the same goes for the magic SSD secret sauce too then, right? All the dozens of topics about how incredible it will be and how all games will load instantly and you'll never see a load screen on the PS5 and how it'll be 3x+ faster on the PS5 shouldn't be downplayed or swept under the rug too, correct?

This is a cross generation multiplatform launch day game. It's not indicative of anything. Launch games have never been a good indicator of how consoles will perform. Launch games are always just cross gen up-ports, generally just with better framerate and resolution because that's the easiest thing to brute force.

Anyone gloating about this being equal is dumb.
Loadtime is the most boring metric for the next-gen SSD's, the true potential is the stream ability the what it can do. And is in that arena that the PS5 superior technology could really paid dividends for Sony.
 

MrFunSocks

Banned
Well it definitely is a concern for Series S owners. If I build a cheap gaming PC I can't expect it to play all games well over the next 5 years.
If your cheap gaming pc has the same cpu, and same gpu architecture and features as the latest gaming pc in 5 years time, and you play at the correct resolution then yes you can.
 

Lysandros

Member
I got the same impression after we got the full specs of both machines. Especially after the HotChips conference where MS went into more detail regarding their GPU... And then when I compared it to PS5, it was clear to me PS5 was a better designed machine overall. And the games showcasing this to be true is the proof of that.

That said, there will be situations where SX will pull ahead in scenes with ALU bound and there will be scenes where PS5 will pull ahead (like in AC:Valhalla) if it's fillrate bound. And will likely be close and within margin of error if the scene is both ALU and fillrate heavy.
What makes you think that AC is fill rate (pixel or texel?) bound, just curious.
 
Top Bottom