Support NeoGAF

[Dave_at_Home]

It's not a commercial gpu so technically you are right, but they definitely still have a max clock rate coming out of manufacturing and they could have decided to push those limits.

It's not really overclocked in the traditional sense. In The Road to PS5, Cerny talked about finding the perfect thermal equilibrium between the CPU and GPU. The PS5 has no problem hitting both clock speeds at the same time. Remember it's always fluctuating per-frame. He also mentioned the GPU is actually able to be clocked much higher but would cause logic issues (transistors flipping too fast causing 0 to be 1 or 1 to be 0).

It's still strange the narrative of Sony boosting clocks last minute still lurking around. It took Sony two years developing the liquid metal for PS5. 36 (active) compute units and hitting the highest clock speeds was always on the table. From that point you focus on reducing GPU stalls and any other latency across the whole system. PS5 clearly has lots of custom cache systems at work. Smartshift, GPU cache scrubbers, the entire I/O block, probably some tweaks on CPU cache (unified?), incredibly fast data fed to RAM, is why it punches above its weight class. Maybe Series X has the edge (a very slight one), but the PS5 seems, pound for pound, more efficient.

 

[Truespeed]

The gap will not closed simply because Microsoft decided to gimp their 12TF console with a 4TF console.

Right now you probably have developers putting one team to develop for both XSX and XSS and another just for the PS5.

Pretty obvious you would get unoptimized games on the XSX. Not an issue of tools but developer resources.

Microsoft need to ditch the XSS otherwise 9TF of RDNA 1.5 will continue to embarrass them. Developers are not going to allocate more resources for that piece of crap.

Not to mention ruining every single Digital Foundry next gen game comparison by interleaving the video with excruciating comparisons to the gimp box.

 

[sircaw]

I keep telling people launch games are not a good metric for performance. What is even more laughable is if the patches and update from ms erases the minor performance differences between platforms. There would be a lot of crow served.

Why would there crow be served? I really don't understand your reasoning at all, sony fans did not call there console the worlds most powerful console, or the beast or the 12 tflop monster.

Are you now saying that a patch erases the minor difference between the platforms and brings parity to the consoles its somehow a win for Microsoft.

What is going on here, stop it dude.

 

[Neo_game]

The state of Halo and their others games not ready probably means something is wrong with devkits, API. That does not necessarily mean they are going to a lot of more performance than normal. It could be that it would just take less time. Right now may be devs have to put more effort on Xbox where as PS5 is just more easier to work with.

 

[iHaunter]

So Xbox actually admits the PS5 is outperforming it? "Power Your Dreams - Later?"

 

[bentanchorbolt]

The only thing I’m certain of after reading this thread is people think a crossgen launch title on new hardware is indicative of performance.

 

[yurinka]

So Xbox actually admits the PS5 is outperforming it? "Power Your Dreams - Later?"

Wait for the power of the cloud, or Milo, or scanning your skate with Kinect, or the tools, or....

The state of Halo and their others games not ready probably means something is wrong with devkits, API. That does not necessarily mean they are going to a lot of more performance than normal. It could be that it would just take less time. Right now may be devs have to put more effort on Xbox where as PS5 is just more easier to work with.

There is nothing wrong with the devkits, API and so on because other companies are releasing games on their console and it's performing as expected. Teraflops doesn't mean console horsepower, it's the theorical peak for some GPU tasks. PS5 has more optimized memory management, a faster GPU that makes all the tasks that doesn't need to use all these extra CUs, feeds faster the GPU stuff with streaming from the SSD and has other optimizations that very likely is letting it reach a performance way closer to that potential horsepower compared to XSX.

XSX is more powerful for some things, and PS5 is more powerful in other things plus let's say it's more optimized to reach its potential. There is no devkit/API/tools that will change that. And as soon as they start using Oodle texture and non-crossgen engines that will take full advantage of the next gen streaming -as seen in the UE5 demo- to get extra detail, the benefits of PS5 may even increase.

 

[Explosive Zombie]

The only thing I’m certain of after reading this thread is people think a crossgen launch title on new hardware is indicative of performance.

What's your idea? That the console that handles cross-gen games better won't handle next gen games better?

 

[DeepEnigma]

It's not a commercial gpu so technically you are right, but they definitely still have a max clock rate coming out of manufacturing and they could have decided to push those limits.

They are well within the limits of what that silicon can do. Even Cerny mentioned they could’ve went even higher, and well, the new cards coming out show this.

Next-Gen PS5 & XSX |OT| Console tEch threaD

Oh another episode of Phil feeling the need to play the police man, how cute If he want to give lessons on those "muh console war is bad" he should've start with his friends: Matt made a good point, but don't think those Phil saying are innocent, he wouldn't gave a flying f*ck about all...

www.neogaf.com

 

[Codeblew]

Right. What is it?

I don't think you know what overclocking is with this response.

 

[bentanchorbolt]

What's your idea? That the console that handles cross-gen games better won't handle next gen games better?

Imagine for a second that you met someone on these forums that doesn’t give a damn about the console war. I don’t care which one is performing better, but I’m for sure convinced that anyone who puts a ton of stock into how each of these systems run these games as evidence of one being more or less powerful, is just lowering the threshold of what constitutes solid evidence. They are looking for an outcome that agrees with their narrative.

We are in the launch window of brand new consoles. Is this all your guys first console launch or am I missing something?

 

[Explosive Zombie]

Imagine for a second that you met someone on these forums that doesn’t give a damn about the console war. I don’t care which one is performing better, but I’m for sure convinced that anyone who puts a ton of stock into how each of these systems run these games as evidence of one being more or less powerful, is just lowering the threshold of what constitutes solid evidence. They are looking for an outcome that agrees with their narrative.

We are in the launch window of brand new consoles. Is this all your guys first console launch or am I missing something?

You're just yammering dude, not interested, either stick by your arguments or don't quote me.

 

[Aladin]

It's not really overclocked in the traditional sense. In The Road to PS5, Cerny talked about finding the perfect thermal equilibrium between the CPU and GPU. The PS5 has no problem hitting both clock speeds at the same time. Remember it's always fluctuating per-frame. He also mentioned the GPU is actually able to be clocked much higher but would cause logic issues (transistors flipping too fast causing 0 to be 1 or 1 to be 0).

It's still strange the narrative of Sony boosting clocks last minute still lurking around. It took Sony two years developing the liquid metal for PS5. 36 (active) compute units and hitting the highest clock speeds was always on the table. From that point you focus on reducing GPU stalls and any other latency across the whole system. PS5 clearly has lots of custom cache systems at work. Smartshift, GPU cache scrubbers, the entire I/O block, probably some tweaks on CPU cache (unified?), incredibly fast data fed to RAM, is why it punches above its weight class. Maybe Series X has the edge (a very slight one), but the PS5 seems, pound for pound, more efficient.

There are perfect equilibrium(s) or multiple sweet spots. Xbox invested 50 bucks or so in a bigger better gpu, while ps5 invested in a better ssd and adaptive triggers. Products will always be differentiated. Both companies have done a good job. Adaptive triggers wont away people to playstation those who are happy with future of gamepass. Similarly a slighter better gpu wont sway those who love playstation experience.

 

[yurinka]

Even though PS5 is not outperforming xsex in all games either.

Cod seems to run better in xsex except the 120fps mode. Same as DMC5, better 60fps mode on xsex, better 120fps for ps5.

Another game was better on xsex in gameplay but fail im cinemaics or killcams or something else

Other games do perform better on PS5.

Although it seems that people just love to say that ps5 is outperforming on multiplats. It is not a clear advantage or a widespread one.


We will need a bit more time to see the real pattern.

Meanwhile I will try getting the best version of the multiplats I care about.

In CoD PS5 had some fps drops in a few cinematics or specific spots, but they released recently a patch to address that.
DMCVSE got a patch too, to add the 60fps selector that it missed, and VRR is also coming via an OS patch.
They also just released an ACV patch to address XSX tearing issue.
So well see how the comparisions remain after all the patches that they are releasing. I assume that they will close the lead each console had to basically zero or almost.

Xbox invested 50 bucks or so in a bigger better gpu

It's better in some areas, worse in others. Results doesn't show any important performance difference between consoles, and in many cases performs (slightly) worse than the PS5 one

• On paper, Series X really is the more powerful console - this isn't an invented number, it's maths.

No, it isn't. 3% more CPU doesn't do a shit and the teraflops difference/having more CUs only affect some GPU tasks while many others GPU tasks will perform better because the GPU is way faster and its memory management is more optimized. On paper there isn't a clear winner regarding which console is more powerful, and real world results show that -other than a bug/bad optimization here or there, VRR still not implemented in PS5 and so on- the results are pretty much the same in both consoles.

Basically they achieved the same horsepower results with different strategies.

 

[bentanchorbolt]

You're just yammering dude, not interested, either stick by your arguments or don't quote me.

You quoted me. I responded, restating what I originally said. I’m new here so I won’t flame you, but holy crap you can’t be for real lol.

 

[Explosive Zombie]

You quoted me. I responded, restating what I originally said. I’m new here so I won’t flame you, but holy crap you can’t be for real lol.

No you said a bunch of bogus crap. Your original comment was attempt to deride the thread for thinking how a cross-gen game performs is indicative of a console's performance, as if the console lagging behind on cross-gen titles will suddenly play next-gen only titles better, which doesn't follow.

 

[Hendrick's]

I don't think you know what overclocking is with this response.

I know what it is. I said it isn’t really technically overclocking in this instance, but they could still pushing the chips limits. And the chip does have a base clock, that’s what I was asking.

 

[bentanchorbolt]

No you said a bunch of bogus crap. Your original comment was attempt to deride the thread for thinking how a cross-gen game performs is indicative of a console's performance, as if the console lagging behind on cross-gen titles will suddenly play next-gen only titles better, which doesn't follow.

Damn, I said all that? So you told me not to quote you (you quoted me first), you said I wasn’t being consistent (even though expanded upon what I initially said), and now I said all that(I didn’t btw)?

I don’t know dude, I feel like engaging with you further would be a huge waste of my time if these early interactions are an indicator of anything.

Last thing I’ll say and then never acknowledge you again: I don’t think these launch titles are fair representations of either consoles power. I don’t think we necessarily learned anything about which console is more powerful. That is far too much that we don’t know for anyone to make any one certain claim, especially given the trend to devalue any source that offers potential insight. We will see which console is stronger in a year, and it might be PS5 and it might be XSX. The only thing we will know is that nobody is going to be referencing performance numbers of launch titles as an indicator of power lol.

 

[Tschumi]

I can only speculate.. since the deep dive, cerny made it clear that they wanted the thing to be easier to make games for, and wayyy more efficient/fast, then ever before. The XSX appears to be a much more standard, albeit supercharged, system. Saying XSX is getting outperformed is probably incorrect, it's probably just the PS5 performing at a super high level.. like, the buffest gym beast in the world might lose an arm wrestle to any old joe schmoe if that other guy knows the technique inside out.. Olympians only beat other Olympians in very evenly matched contests because they're more precise, more naturally gifted, more professional.. I'm not poooing on XSX, I'm just talking up the strengths of the PS5.

I think we have a long way to go in terms of upside for PS5 games.. the stuff cerny said about rendering worlds in the time it takes to turn around simply hasn't been communicated yet. I think we're probably going to first see it on a game like Journey, nothing too ambitious.. a game where wild things will happen simultaneously all around the player, just maybe not with psychotic textures or anything~

 

[Codeblew]

I know what it is. I said it isn’t really technically overclocking in this instance, but they could still pushing the chips limits. And the chip does have a base clock, that’s what I was asking.

So how does a chip have a base clock? Is there some mathematical formula or something? Or do the people that build the chip do testing on it and decide what clockrate they want to use based on test results? The base clock by definition is the clock rate that they are using on the PS5. If someone hacks the PS5 and raises the clock rate, then you can call it overclocked. SMH.

 

[bentanchorbolt]

In CoD PS5 had some fps drops in a few cinematics or specific spots, but they released recently a patch to address that.
DMCVSE got a patch too, to add the 60fps selector that it missed, and VRR is also coming via an OS patch.
They also just released an ACV patch to address XSX tearing issue.
So well see how the comparisions remain after all the patches that they are releasing. I assume that they will close the lead each console had to basically zero or almost.


It's better in some areas, worse in others. Results doesn't show any important performance difference between consoles, and in many cases performs (slightly) worse than the PS5 one


No, it isn't. 3% more CPU doesn't do a shit and the teraflops difference/having more CUs only affect some GPU tasks while many others GPU tasks will perform better because the GPU is way faster and its memory management is more optimized. On paper there isn't a clear winner regarding which console is more powerful, and real world results show that -other than a bug/bad optimization here or there, VRR still not implemented in PS5 and so on- the results are pretty much the same in both consoles.

Basically they achieved the same horsepower results with different strategies.

I feel like what launch games have come out so far probably aren’t optimized for each system all that well, and in 6 months to a year, we are probably going to see parity in nearly all cases when it comes to multiplat games. Sometimes X will perform a little better, and other times PS5 will perform a little better.

 

[DinoD]

I keep telling people launch games are not a good metric for performance. What is even more laughable is if the patches and update from ms erases the minor performance differences between platforms. There would be a lot of crow served.

Launch games were pretty spot on metrics for at least 2 previous generations of gaming consoles.
XBOX360 had lead over PS3 at multiplatforms at the start, and finished in the lead (Bayonetta etc)
PS4 had lead over XBOX one at ........................................................ and finished in the lead.

So there's that. You expecting anything different than almost a parity this time around, is not very sound logic.

 

[GloveSlap]

The proof is in the pudding. Claims of PS5 closing the gap with "special sauce" before launch was just words, but they have proven themselves as of now. The XBOX devkit issues could be true, but until they restore a performance gap....those are just words as well.

 

[ZehDon]

The PS3 never really caught up the 360 in multiplat performance, except for very rare cases like BF3 because of the particular way that engine worked.

PS3 never gained parity. It had some good showings towards the end, but by and large the multiplat performance favoured Xbox for the entire gen. PS3s GPU and RAM bottlenecks never overcame the potential on paper promise. First party games gained some great results out of the system but third parties rarely get there...

I'll repeat what I said, as you both seemed to have missed it: "First party titles in particular really looked spectacular as the generation wore on. Not every third party managed to achieve parity, but don't pretend that this didn't happen." The issue was the difficulty of the PS3 hardware. As I also said in my post, which you also both seemed to have missed: "Don't confuse "difficult to develop for" with "low power"." We're discussing the concept of "wait for developers to learn the hardware" as a basis for expecting better results as the generation goes on. As the PS3 shows, this is absolutely valid - as long as there is "on paper" power to back it up. Nothing you've said addresses the fact that the PS3 proves that this can be true. This means we can't put down a blanket dismissal for the Xbox Series X; it has the on-paper specs. Now, it's a matter of seeing if Developers can actually use it.

DX12U is additive to what is already on Xbox. You vastly overstate the amount of relearning. As stated by developers:

Q: What is it like developing on Xbox Series X?

A: Transitioning development to a new console platform, like Xbox Series X, is usually very painful. You have to deal with new tools, new workflows, new ways of thinking.

This time around the team at Xbox brought me a new toolset called the Game Development Kit, which they already had up and running on Xbox One.

Inside Xbox Series X Optimized: Dirt 5 - Xbox Wire

When it launches this holiday, Xbox Series X will be the most powerful console the world has ever seen. One of the biggest benefits of all that power is giving developers the ability to make games that are Xbox Series X Optimized. This means that they’ve taken full advantage of the unique...

news.xbox.com

On paper means little, when Cerny has stated their approach to PS5'S design is recognising that 1) you cannot compare AMD teraflops with eachother (especially when customised), that will not give you the fill story and 2) developers are hitting major barriers in system bottlenecks and freeing up those barriers improves performance

Suffice to say, game performance so far shows that Cerny hit the nail on the head and the Xbox narrative is now for us to "wait" just like PS3 and Xbox One before it

You either didn't read the article, or you've misunderstood it. Your quote is referring to the tool set - GDK - not the DX12u API. They aren't the same thing. The respondent goes on to say that they've never had tools this mature at a console launch before - but neither the author not the respondent mentions DX12u in the entire article.
Why is this an important distinction? DX12u packs in support for all of the new hardware features of the Xbox Series X. Literally none of these physically exist on the Xbox One for the existing DX12 API to interface with. DX12u brings with it VRS, Mesh Shaders, SFS, and layers to access the completely different CPU and GPU architecture - so, there's a lot that'll need to change. Check out Microsoft's Direct X YouTube Channel for a look into what DX12u changes. It's a lot.
Now, using an API means your calls can be, effectively, translated for you - the primary benefit of an API layer, after all - but DX12 offered much lower level access than DX11. DX12 was Microsoft's attempt to push back against AMD's Vulkan and Sony's PlayStation API. Dev's that have made use of that lower level access on the Xbone will need to alter their surrounding proprietary code and logic within their game engines, to correctly use the Xbox Series new hardware capabilities. This will require a lot of experimentation and research, as it does with every generation. As I said above, it's now a matter of seeing if Developers can actually make use of it. Time will tell.

No, it isn't. 3% more CPU doesn't do a shit and the teraflops difference/having more CUs only affect some GPU tasks while many others GPU tasks will perform better because the GPU is way faster and its memory management is more optimized. On paper there isn't a clear winner regarding which console is more powerful, and real world results show that -other than a bug/bad optimization here or there, VRR still not implemented in PS5 and so on- the results are pretty much the same in both consoles.

Basically they achieved the same horsepower results with different strategies.

The Xbox One S demonstrates what a difference 3% CPU clock differences can make in a closed console environment. Your attempt to hand wave away differences you don't understand doesn't make those difference go away, and just demonstrates that you're not interested in discussing specifics, you just want to console war. Given that you seem to think that the Xbox Series X's 40% additional GPU parallelization is "the same horse power" shows me you really don't know what you're talking about. Educate, yourself friend.

 

[wachie]

I don't think it's complicated: developer talent goes a long way. Sony has always boasted some genuine lookers despite each of their consoles having on-paper hardware disadvantages.

The point is that the PS5 is outperforming the Xbox so far in 3rd party titles. When Sony's big hitters come, the graphical difference would be much bigger. Just look at TLOU2 last gen when the peak TF gap was ~45%.

 

[THE DUCK]

The point is that the PS5 is outperforming the Xbox so far in 3rd party titles. When Sony's big hitters come, the graphical difference would be much bigger. Just look at TLOU2 last gen when the peak TF gap was ~45%.

Again, feel free to show with examples of every multiplatform game out now that the ps5 "outperforms" the series x by any sort of margin. So far they are pretty much neck and neck overall, if you can't see that you are blind, or just flat out biased. So much bs in this thread.

 

[Explosive Zombie]

Damn, I said all that? So you told me not to quote you (you quoted me first), you said I wasn’t being consistent (even though expanded upon what I initially said), and now I said all that(I didn’t btw)?

I don’t know dude, I feel like engaging with you further would be a huge waste of my time if these early interactions are an indicator of anything.

Last thing I’ll say and then never acknowledge you again: I don’t think these launch titles are fair representations of either consoles power. I don’t think we necessarily learned anything about which console is more powerful. That is far too much that we don’t know for anyone to make any one certain claim, especially given the trend to devalue any source that offers potential insight. We will see which console is stronger in a year, and it might be PS5 and it might be XSX. The only thing we will know is that nobody is going to be referencing performance numbers of launch titles as an indicator of power lol.

A whole bunch of nonsense just to say the same thing again. Why should it be any different a year from now? Can you name other launches where the console that had an edge at launch didn't later on? Did the PS3 ever catch up to 360 on multiplats? Did the X1 ever catch up to PS4? Did the Pro catch up to the 1X? Glad to have another new member who spouts bs.

 

[MatrixReloaded101]

The tell tale sign of open world games this generation is the lack of actual life in these open worlds. This is all down to the weak CPU and basic storage those systems. I hope the Cyberpunk 2077 update for PS5/Series X adds more NPC and traffic. Basically I want open worlds on the new systems to look like actual cities pre-covid.

You are not going to get that until much later into the PS5/Series X life cycle when games will be exclusively developed for these consoles.

 

[Md Ray]

Please post a list of how the ps5 is outperforming the series x "in nearly every game"

Note:
Be sure to be specific, since its not true the task is impossible.

I'm starting to think we should start moderating the forum like Twitter has been moderating ....

" this post may contain potentially misleading information about xbox series x"

Let's see. So far we've had comparisons of:

DmC5
AC: Valhalla
Dirt 5
COD:CW

DmC5: One of the modes in DmC5 has a minor advantage on SX, PS5 outperforms it in the other mode. But I'd say they're basically the same.

AC:V: It's a clear win for PS5, as of now. Plenty of complaints about screen-tearing on SX which is due to constant frame-rate dips, so many regretted their decision for going with XSX due to this. PS5 is solid, holds 60fps more often, has less screen-tearing. The superior version on console.

Dirt 5: Again, the PS5 clearly outperforms SX and has a higher, more stable frame-rate in 60Hz mode. Both XSX and PS5, I believe have identical graphics quality and resolution in this mode, IIRC. SX has the advantage in the 120Hz mode but that's clearly due to downgraded graphics, and resolution (or a bug) on XSX so performance is smoother on SX whereas on PS5 it holds the target 120fps at least 90% of the time. A patch is on its way though.

COD: 120Hz mode - PS5 outperforms SX at identical graphics, and resolution. 60Hz mode is identical. SX doesn't outperform PS5 here as there's some sort of a bug that causes frame-rate to dip below 60fps randomly on PS5 in scenes where it previously did not. See below for stats and comparisons:

The frame-rate bug from two different sources (Digital Foundry & VG Tech):

So yeah, hence the phrase: "PS5 is outperforming XSX in nearly every game".

 

[Explosive Zombie]

You forgot the Observer game runs better on PS5.

 

[DinoD]

You forgot the Observer game runs better on PS5.

Wait until that "Dolphin" game gets released on XBOX XS

 

[Hendrick's]

So how does a chip have a base clock? Is there some mathematical formula or something? Or do the people that build the chip do testing on it and decide what clockrate they want to use based on test results? The base clock by definition is the clock rate that they are using on the PS5. If someone hacks the PS5 and raises the clock rate, then you can call it overclocked. SMH.

No. That the max clock rate. The base clock rate is what it is when it's not boosted by Smartshift.

 

[Sanpei]

The gap will not closed simply because Microsoft decided to gimp their 12TF console with a 4TF console.

Sony didnt?

 

[DinoD]

No. That the max clock rate. The base clock rate is what it is when it's not boosted by Smartshift.

You are right. This is kind of luster a real 12 TF machine without smoke and mirrors can give to you

 

[Hendrick's]

You are right. This is kind of luster a real 12 TF machine without smoke and mirrors can give to you

So much fear. Just go enjoy your favorite box.

 

[Md Ray]

On paper, Series X really is the more powerful console - this isn't an invented number, it's maths.

Elsewhere in the GPU the PS5 has a 22% advantage in pixel fillrate, rasterization. Perhaps one of the reasons why it's outperforming SX. TF isn't the absolute indicator of GPU perf.

 

[Md Ray]

I'll repeat what I said, as you both seemed to have missed it: "First party titles in particular really looked spectacular as the generation wore on. Not every third party managed to achieve parity, but don't pretend that this didn't happen." The issue was the difficulty of the PS3 hardware. As I also said in my post, which you also both seemed to have missed: "Don't confuse "difficult to develop for" with "low power"." We're discussing the concept of "wait for developers to learn the hardware" as a basis for expecting better results as the generation goes on. As the PS3 shows, this is absolutely valid - as long as there is "on paper" power to back it up. Nothing you've said addresses the fact that the PS3 proves that this can be true. This means we can't put down a blanket dismissal for the Xbox Series X; it has the on-paper specs. Now, it's a matter of seeing if Developers can actually use it.


You either didn't read the article, or you've misunderstood it. Your quote is referring to the tool set - GDK - not the DX12u API. They aren't the same thing. The respondent goes on to say that they've never had tools this mature at a console launch before - but neither the author not the respondent mentions DX12u in the entire article.
Why is this an important distinction? DX12u packs in support for all of the new hardware features of the Xbox Series X. Literally none of these physically exist on the Xbox One for the existing DX12 API to interface with. DX12u brings with it VRS, Mesh Shaders, SFS, and layers to access the completely different CPU and GPU architecture - so, there's a lot that'll need to change. Check out Microsoft's Direct X YouTube Channel for a look into what DX12u changes. It's a lot.
Now, using an API means your calls can be, effectively, translated for you - the primary benefit of an API layer, after all - but DX12 offered much lower level access than DX11. DX12 was Microsoft's attempt to push back against AMD's Vulkan and Sony's PlayStation API. Dev's that have made use of that lower level access on the Xbone will need to alter their surrounding proprietary code and logic within their game engines, to correctly use the Xbox Series new hardware capabilities. This will require a lot of experimentation and research, as it does with every generation. As I said above, it's now a matter of seeing if Developers can actually make use of it. Time will tell.


The Xbox One S demonstrates what a difference 3% CPU clock differences can make in a closed console environment. Your attempt to hand wave away differences you don't understand doesn't make those difference go away, and just demonstrates that you're not interested in discussing specifics, you just want to console war. Given that you seem to think that the Xbox Series X's 40% additional GPU parallelization is "the same horse power" shows me you really don't know what you're talking about. Educate, yourself friend.

Xbox One S CPU has a 10% advantage, not 3%.

 

[BeardGawd]

No you said a bunch of bogus crap. Your original comment was attempt to deride the thread for thinking how a cross-gen game performs is indicative of a console's performance, as if the console lagging behind on cross-gen titles will suddenly play next-gen only titles better, which doesn't follow.

It actually does follow. These games are using last gen effects and engines. Next-gen effects (think Ray tracing) and engines could presumably benefit from more CUs vs higher clocks.

 

[Explosive Zombie]

It actually does follow. These games are using last gen effects and engines. Next-gen effects (think Ray tracing) and engines could presumably benefit from more CUs vs higher clocks.

Some of these games do have ray-tracing.

 

[Md Ray]

could presumably benefit from more CUs vs higher clocks.

By that logic the Series S is at an even more disadvantage than previously thought since it has significantly less CUs than SX.

 

[Bogroll]

Let's see. So far we've had comparisons of:

DmC5
AC: Valhalla
Dirt 5
COD:CW

DmC5: One of the modes in DmC5 has a minor advantage on SX, PS5 outperforms it in the other mode. But I'd say they're basically the same.

AC:V: It's a clear win for PS5, as of now. Plenty of complaints about screen-tearing on SX which is due to constant frame-rate dips, so many regretted their decision for going with XSX due to this. PS5 is solid, holds 60fps more often, has less screen-tearing. The superior version on console.

Dirt 5: Again, the PS5 clearly outperforms SX and has a higher, more stable frame-rate in 60Hz mode. Both XSX and PS5, I believe have identical graphics quality and resolution in this mode, IIRC. SX has the advantage in the 120Hz mode but that's clearly due to downgraded graphics, and resolution (or a bug) on XSX so performance is smoother on SX whereas on PS5 it holds the target 120fps at least 90% of the time. A patch is on its way though.

COD: 120Hz mode - PS5 outperforms SX at identical graphics, and resolution. 60Hz mode is identical. SX doesn't outperform PS5 here as there's some sort of a bug that causes frame-rate to dip below 60fps randomly on PS5 in scenes where it previously did not. See below for stats and comparisons:

The frame-rate bug from two different sources (Digital Foundry & VG Tech):

So yeah, hence the phrase: "PS5 is outperforming XSX in nearly every game".

Funny how up to 8% advantage (i think from memory) in one of the modes in DMC has turned into minor this gen. Last gen if Pro was a frame or 2 ahead for 2 seconds despite being up to 50% lower resolution it was a win for the Pro v X. "Mmmmmmmmm!" As I scratch my chinny chin chin.

 

[nemiroff]

Slight offtopic :

it is absolutly amazing to me that all of those xbox Boys have such punchable faces. Those bellow look like the pure Betasquad

Why is that?

This is a good example of how it is to try to wade through certain threads on gaf to try to find a decent on-topic discussion. I don't need a sterile environment, but man.. the endless sea of bad jokes and animosity is messing up the readability of the forum.

 

[TheAmazingSpiderMan]

Touched up last gen games which is all we've had so far probably like higher clocks. When next gen engines come around it'll be a different story, with the extra power and the full suite of RDNA2 hardware features the Series X will be the performance king easily.

Lol what imaginary engines are you talking about now?

Also when does that Crystal ball of yours say that series S will start pushing PS5

 

[Md Ray]

Funny how up to 8% advantage (i think from memory) in one of the modes in DMC has turned into minor this gen. Last gen if Pro was a frame or 2 ahead for 2 seconds despite being up to 50% lower resolution it was a win for the Pro v X. "Mmmmmmmmm!" As I scratch my chinny chin chin.

Before these comparisons, the claim from green team was that SX would be 20-30% faster on avg. sometimes even larger numbers were quoted by the likes of longdi IIRC.

In comparison, that single digit adv. is indeed minor.

 

[Explosive Zombie]

You shouldn't even give X any edge on DMC V, they just patched in the option to run it at 60FPS on PS5, this option wasn't available when these comparisons were done.

 

[Md Ray]

You forgot the Observer game runs better on PS5.

Yep, I hear WD:Legion also has better perf on PS5. But I purposefully didn't include it. Because waiting for VG Tech, NX Gamer, and DF's take on it for confirmation.

 

[jroc74]

I remember when devs were saying PS5 or Series X was easy to develop for. PS5 being easy was stated first. Then some Xbox fan had to find an article saying Series X was easy to develop for...

I remember this article:

www.wccftech.com/microsoft-we-could-have-used-variable-clocks-for-xsx-but-were-not-interested-in-tflops-numbers/amp/

We focus on optimizing the developer experience to deliver the best possible experience for players, rather than trying to 'hunt' down certain record numbers. We've always talked about consistent and sustained performance.

We could have used forced clocks, we could have used variable clock rates: the reality is that it makes it harder for developers to optimize their games even though it would have allowed us to boast higher TFLOPS than we already had, for example. But you know, that's not the important thing. The important thing is the gaming experiences that developers can build

Just one that has aged like milk. Jason Ronald, folks.

When there was an article about PS5 being easy, Xbox fans would say "so, Series X is too!!" And vice versa.

....so far one of those thing is true. So, which is it: that narrative was completely false, and we wait on the tools, dev kits were late? Devs were just wrong?

Jason Ronald was flat out wrong?

On a related note....some of us said before the face offs...its gonna be a rough gen for some fans.

Once again, like a broken record, "closer than 2 consoles have ever been". Early results are showing this.

Might be time for some ppl to just accept it. Because with how well PS5 is running 3rd party games, Series X getting better is not gonna equal PS5 running worse.

It will be Series X finally living up to all the hype.

That's not the win some were looking for. And that's ok, it'll be alright. ❤

 

[longdi]

Before these comparisons, the claim from green team was that SX would be 20-30% faster on avg. sometimes even larger numbers were quoted by the likes of longdi IIRC.

In comparison, that single digit adv. is indeed minor.

I'm innocent. I did think 30% is the best case.
No larger

 

[wachie]

Again, feel free to show with examples of every multiplatform game out now that the ps5 "outperforms" the series x by any sort of margin. So far they are pretty much neck and neck overall, if you can't see that you are blind, or just flat out biased. So much bs in this thread.

What are these arbitrary margins? The Xbox had a peak theoretical TF advantage which so far has failed to materialize. We've had more 3rd party showdowns that have favored the PS5 over the Xbox. Either PS5 is punching above it's peak TF advantage or Xbox is significantly punching below it's peak TF advantage. Again that is not the only spec that matters but for Microsoft to boast that spec since March, the trend so far is really not a good look for them.

You can claim bs all you want and try to spin all you want, the fact that there are articles like this in the first place means something hasnt gone as per plan for Xbox.

 

[THE DUCK]

Let's see. So far we've had comparisons of:

DmC5
AC: Valhalla
Dirt 5
COD:CW

DmC5: One of the modes in DmC5 has a minor advantage on SX, PS5 outperforms it in the other mode. But I'd say they're basically the same.

AC:V: It's a clear win for PS5, as of now. Plenty of complaints about screen-tearing on SX which is due to constant frame-rate dips, so many regretted their decision for going with XSX due to this. PS5 is solid, holds 60fps more often, has less screen-tearing. The superior version on console.

Dirt 5: Again, the PS5 clearly outperforms SX and has a higher, more stable frame-rate in 60Hz mode. Both XSX and PS5, I believe have identical graphics quality and resolution in this mode, IIRC. SX has the advantage in the 120Hz mode but that's clearly due to downgraded graphics, and resolution (or a bug) on XSX so performance is smoother on SX whereas on PS5 it holds the target 120fps at least 90% of the time. A patch is on its way though.

COD: 120Hz mode - PS5 outperforms SX at identical graphics, and resolution. 60Hz mode is identical. SX doesn't outperform PS5 here as there's some sort of a bug that causes frame-rate to dip below 60fps randomly on PS5 in scenes where it previously did not. See below for stats and comparisons:

The frame-rate bug from two different sources (Digital Foundry & VG Tech):

So yeah, hence the phrase: "PS5 is outperforming XSX in nearly every game".

So the odd scene where one console happens to drop frames vs the other scene where the other drops frames? Image quality and frane rate is near identical, in all these games, to the average player who isn't scanning every frame they largely look and play the same, plus or minus 5% either way depending on the scene.

The word outperform means to do better than - in most of these cases if you put gamers in from of all of these games without watching the YouTube videos first, I bet they couldn't tell you which is which.