• 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.

DF: Control PS5 Vs Xbox Series X Raytracing Benchmark

directly from the mouth ofLance Mcdonald about the DF test

''This is fascinating. It's important to remember that when photo mode is engaged, CPU load is absolutely miniscule compared to gameplay (something I learned while debugging the game myself), so this is basically just a fantastic measurement of the differential in pure graphics hardware at play here. Really cool test.''

For your information he is the guy that made 60 fps Bloodborne possible


But I'm sure we should trust more an absolute nobody who writes on a forum in the throes of a fanboy crisis because he does not accept that his favorite box dosnt performs as he would like instead of Lance and all of Digital Foundry
You don't say.

LOL. Lance's post was referred to Alex statement about possible CPU bottleneck ( Lance was mentioned in the beginnig of the thread). There is no CPU bottleneck in photomode. It is a specific scenario for GPU. As DF pointed out - Ray Tracing. You are damn delusional
 
Last edited:

MonarchJT

Banned
Isolating the gpu from the cpu in a smartshift setup makes this an invalid analysis. The system isn't a traditional one where it receives constant power. Although it can, its designed to pass off power to the chip that needs it, if one is being stressed. Normal usage is happening in this comparison which wouldn't trigger the need to draw power from the CPU. This could be a case of the GPU working at a stock/non boosted clock. Also note how consistent the frame graph is on the ps5 vs SX.

60A4zmE.jpg
60A4zmE.jpg

Na. when cpu doesn't require power gpu get all it need
 

pixelbox

Member
Na. when cpu doesn't require power gpu get all it need
That's not what's happening though. It's not like the cpu is doing nothing at all plus that's not how Smartshift works anyway. This was a bad comparison as both these machines have different philosophies.
 
Last edited:

MonarchJT

Banned
Like i've said, Lance's commet was referred to CPU bottleneck. Cheers, Xbox fan. You're embarassing yourself
Do you have the ability to read what he wrote?
It is clear to everyone that the cpu is not at stake in this test carried out by DF but lance also says:

"so this is basically just a fantastic measurement of the differential in pure graphics hardware at play here. Really cool test."
It literally say what it say and it's different from what you saying
But clearly You can bend and spin it to make it better for you warrior
 
Last edited:

DJ12

Member
So why even capped Series X has framerate drops even without the UI bug they talked?

Imagine when the tools can only be used in Photomode...

The reality is something else is a bottleneck in Series that doesn’t allow the use GPU of the GPU like dozen or games already evidencies that and this “benchmark” finally showed what we speculated since launch of the consoles.
The IO 'bug' that DF have since admitted is shader compilation on PC and absolutely nothing to do with whats going on with Series X.

He threw PC has the same problem in there to valid what MS requires of them. Series X just cannot keep up with PS5 when it comes to IO.
 
That's one interpretation of their choices, but it's not the only one and not what the people who designed the machine say.
They said the "split memory" which isn't actually split in the traditional sense of the word is down to developers asking for faster optimized Vram, since say Valhalla on PC only uses about 6.5gb and that's the most I've heard of so far then I seriously doubt it's a problem, maybe it takes more work but that doesn't make it a bottleneck. I've got a PC with 8gb of Vram, I've never seen it get anywhere near that.

As for why Xbox hasn't shown an advantage in every game well traditionally that hasn't happened in previous gen, there were occasions PS2 outperformed Xbox, Xbox One on PS4 etc, it's not unheard of, that didn't make the PS4 or original Xbox badly designed.

My take as I've explained before is that the games so far have all started as last gen games, to get ready for launch and to mitigate the problems of working from home Devs have taken Pro and X1X versions and upgraded them for next gen, just because a game released after the new consoles doesn't make it next gen.
Xbox has had a bigger change in developer environment and if you look at the specs of going from Pro to PS5 in compute units etc it would be a lot easier to port and unlock extra performance where on Xbox you have a different setup of memory and a wider GPU.

Once we had a game that's wasn't rushed for launch in Hitman 3 then Xbox had a big advantage at 44% resolution advantage and higher shadow settings, a bigger metric than the paper metric .

This is why riky is a hilarious walking contradiction .
You make excuses for Xbox behind behind ps5 by blaming the tools or games starting off as last gen games even though they have specific next gen versions.

Yet your more than willing to brag about Xbox outperforming ps5 in games using backwards compatibility which clearly isn’t a real benchmark since both consoles aren’t even being properly coded for.

You can’t have both.
 
Last edited:
Do you have the ability to read what he wrote?
It is clear to everyone that the cpu is not at stake in this test carried out by DF but lance also says:

"so this is basically just a fantastic measurement of the differential in pure graphics hardware at play here. Really cool test."
It literally say what it say and it's different from what you saying
But clearly You can bend and spin it to make it better for you warrior

Pretty much just a stress test between the two GPUs and shows us what the differences can be. But like all tests this one needs to be taken with a grain of salt because the results might not translate to the actual gameplay. Something to keep in mind for future comparisons.

It would have been really cool to have a RT 60FPs mode with the game. That way we can really speculate on where performance drops for each system and why.
 

phil_t98

#SonyToo
But why didn't that translate to the gameplay?

That's my question.

Your suggesting that developers could have pushed the settings higher but they chose to go with parity. I don't believe that's the case.

Is that simple enough for you?

so you won't answer the question about this benchmark?
 

Loope

Member
Man, you fucking smashed it.
No,he didn't. Sony fanboys as always trample all over any thread that isn't giving an advantage to PS5, even though plenty of people said it was interesting although does not invalidate the findings of the first DF video. The first few pages is riddled with triggered Sony fanboys, just like the 1st few pages of the Nintendo thread about Japan sales or the thread comparing a pc to a ps5. And they're always the same group
 

FALCON_KICK

Member
exactly for this reason he call the test a pure gpu hw benchmark
For a non-dynamic scene which means the 52 CUs of Xbox Series X are being with filled with work.

This can be noticed in 3D modeling/design softwares too, where a high poly object will show maximum fps but as soon as you pan across and rotate to bring the other sides of the object into view, the fps will tank while moving eventually settle to maximum fps as soon as you stop moving.
 
Last edited:

MonarchJT

Banned
Pretty much just a stress test between the two GPUs and shows us what the differences can be. But like all tests this one needs to be taken with a grain of salt because the results might not translate to the actual gameplay. Something to keep in mind for future comparisons.

It would have been really cool to have a RT 60FPs mode with the game. That way we can really speculate on where performance drops for each system and why.
this is why ...taking into account that the game will require the same bandwidth for the cpu on both consoles and that the cpus are in both exactly the same (we can exclude that hundred hZ more for the xsx) this test lets you imagine what can be achieved by optimizing the game ... the power of the gpu is that.

ps. no I'm not on the bandwagon of those who boast non-existent hw bottlenecks. And if we want to be honest the xsx also has more bus bandwidth to start with .. so people should stop with this fud. This is just an non-optimal code case probably due to the fact that the consoles have just arrived and that control is basically a ps4 xbox game.
 
Last edited:

phil_t98

#SonyToo
So why not higher settings?

Like Hitman 3 for example. They do have access to the additional CUs. Fotomode already proves that.

My guess.

for the reason its a rushed patch on both systems, the PS5 could also do better than what it is currently. look at load times to, it aint utilising the SSD properly on PS5
 
Last edited:
this is why ...taking into account that the game will require the same bandwidth for the cpu on both consoles and that the cpus are in both exactly the same (we can exclude that hundred hZ more for the xsx) this test lets you imagine what can be achieved by optimizing the game ... the power of the gpu is that.

ps. no I'm not on the bandwagon of those who boast non-existent hw bottlenecks. And if we want to be honest the xsx also has more bus bandwidth to start with .. so stop with this fud. This is just an non-optimal code case probably due to the fact that the consoles have just arrived.

Only for 10GBs though the other 6GBs has less bandwidth than the PS5. Obviously the decision was made to reduce cost.
 
for the reason its a rushed patch on both systems, the PS5 could also do better than what it is currently. look at load times to, it aint utilising the SSD properly on PS5

But it's definitely using the PS5s compression. Otherwise it wouldn't be 40% smaller than the XSX version. So that there is a little bit confusing.
 

phil_t98

#SonyToo
But it's definitely using the PS5s compression. Otherwise it wouldn't be 40% smaller than the XSX version. So that there is a little bit confusing.

yeah it seems odd that it aint as small on ps5 and not xbox but we dont haver any explanation for it yet. until we do its all guess work
 

MonarchJT

Banned
Only for 10GBs though the other 6GBs has less bandwidth than the PS5. Obviously the decision was made to reduce cost.
Absolutely yes. the calculate also that basically on both system + or - 3 gb would go to the OS so that speed would have been objectively useless. So we are left with 13 giga ... 10 very fast 3 less but also in this case (it is clear that they did thisas you said to keep costs down) they hope that SFS just allow it (no game has used it yet) to load much less data so they probably hope 10 is enough for the vast majority of games .
ps. please note that not all data is bw intensive. the sounds the logic etc etc does not need that speed those 10 giga are practically used to move the graphics
 
Last edited:
No,he didn't. Sony fanboys as always trample all over any thread that isn't giving an advantage to PS5, even though plenty of people said it was interesting although does not invalidate the findings of the first DF video. The first few pages is riddled with triggered Sony fanboys, just like the 1st few pages of the Nintendo thread about Japan sales or the thread comparing a pc to a ps5. And they're always the same group
I've been saying this forever. You rarely ever see it happen the other way around. They are like the spoiled kids, that always got what they wanted, and just never grew up to what reality has in store for them. No matter what kind of thread I enter, they come in swarms and attack others, bring up exclusive games card, spam emojis, etc. If you look in the PC threads I was just in, you'll see it on every one of my posts.

I just wish these same guys could just watch the video and not turn it into a attack & cope session. There is some very interesting technology that went into this game, and it's definitely ahead of it's time. It should be appreciated, and not used as console war fodder.
 
xbox memory is faster though, I can't see this game pushing either console in that respect though

But it still gets the data from the I/0 though. The bandwidth is higher but the I/O is slower if we are just looking at paper specifications.

To avoid confusion I'm not talking about load times. I'm talking about how fast the data goes from the SSD to the ram.
 
Last edited:

phil_t98

#SonyToo
But it still gets the data from the I/0 though. The bandwidth is higher but the I/O is slower if we are just looking at paper specifications.

yeah but how much data are games actually using. this is the thing, that super fast SSD is awesome but what games are streaming that much data that quick. I honestly dont think any game this gen will require that much speed from the SSD. always good to have overhead though
 
yeah but how much data are games actually using. this is the thing, that super fast SSD is awesome but what games are streaming that much data that quick. I honestly dont think any game this gen will require that much speed from the SSD. always good to have overhead though

It really sounds like something that Sonys 1st party studios will try and prove.

20200329135313.jpg


So when I talked about the dream of an SSD part of the reason for that 5 gigabytes a second target was to eliminate loads, but also part of the reason for that target was streaming as in what if the SSD is so fast that as the player is turning around. It's possible to load textures for everything behind the player in that split second.

That's kind of their reasoning behind it.
 

phil_t98

#SonyToo
It really sounds like something that Sonys 1st party studios will try and prove.

20200329135313.jpg




That's kind of their reasoning behind it.

so with a game such as Control, 25gb install and yes there is compression, how much do you think is being streamed at any one time per second on screen? curious
 
so with a game such as Control, 25gb install and yes there is compression, how much do you think is being streamed at any one time per second on screen? curious

I'm not sure about that but Alex said there's a 400% improvement in NiOhs load times. They also mentioned that games might have to be built for the PS5s SSD for it to be fully utilized.

The much smaller file size is definitely weird with control.
 

phil_t98

#SonyToo
I'm not sure about that but Alex said there's a 400% improvement in NiOhs load times. They also mentioned that games might have to be built for the PS5s SSD for it to be fully utilized.

The much smaller file size is definitely weird with control.

when you look at games like god of war which looks amazing and how slow HDD were then factor how much data was streamed, I think the 6gbs per second won’t be utilised this gen. as I said it’s always good to have over head. Now something like Gran turismo may do that depending how the game is built as that can have a lot of detail and textures around a track but again look what was produced last gen
 

AmateurPong

Member
I've been saying this forever. You rarely ever see it happen the other way around. They are like the spoiled kids, that always got what they wanted, and just never grew up to what reality has in store for them. No matter what kind of thread I enter, they come in swarms and attack others, bring up exclusive games card, spam emojis, etc. If you look in the PC threads I was just in, you'll see it on every one of my posts.

I just wish these same guys could just watch the video and not turn it into a attack & cope session. There is some very interesting technology that went into this game, and it's definitely ahead of it's time. It should be appreciated, and not used as console war fodder.

I'm sure the fanbase is made up of a lot of normal, well adjusted people, but some of the fanboys around here really do everything they can to repel you from the brand with their constant angry attacks and fud. I've even seen some liken this corporate cheerleading to being a sports fan. I usually add the worst trolls to the spam filter, but every couple months i get curious about what they're saying and clear the list, only to run into a thread like this that reminds me of why I had it in the first place ...
 

phil_t98

#SonyToo
I'm sure the fanbase is made up of a lot of normal, well adjusted people, but some of the fanboys around here really do everything they can to repel you from the brand with their constant angry attacks and fud. I've even seen some liken this corporate cheerleading to being a sports fan. I usually add the worst trolls to the spam filter, but every couple months i get curious about what they're saying and clear the list, only to run into a thread like this that reminds me of why I had it in the first place ...

yeah I don’t get it , some things the PlayStation does really really well but so does Xbox. I prefer to play on Xbox but I have a PS4 to, Spider-Man is amazing.
I don’t see any of the main fanboys who shit post on Xbox constantly ever praise anything about it

the problem with this place at times is there is to much fud and shit posting by certain members

don’t get me wrong some idiots on both sides
 
Last edited:

Kingthrash

Member
You didn't mention the part where the actual developers told them they are identical, you know the people who made the game.
And digital foundry...like idiots believed them....so the dev lied and they delivered that lie to us...calling it identical......I checked found the lie to be a lie made a video to inform gamers of the lie.....and you make a comment defending this lie
 

Riky

$MSFT
And digital foundry...like idiots believed them....so the dev lied and they delivered that lie to us...calling it identical......I checked found the lie to be a lie made a video to inform gamers of the lie.....and you make a comment defending this lie

People have screens showing things "missing" in the PS5 version, but they aren't really missing at all as it's randomised. Why would the Devs lie about the settings on their own game? You seem to believe that DF have a pro MS agenda, so you saying the Devs do too?
Do you think those incredibly small differences make up a 16% performance gap on average, if so how?
 

Kingthrash

Member
Have you played the game?

You’re analysing an analysis video of a highly dynamic game on Youtube. Claiming the devs are lying after that is not very serious imho. Like I said, at least start the game and have a look around, try replicate one of the scenes DF showed and you’ll understand what we’re dealing with here.

The texture was missing, a bug on one side on the pillar, not sure about the puddle and floor rail thing, some Xbox gamer can probably check that. I didn’t see anything else in your video that I couldn’t explain just from having played the game and knowing the crazy dynamic shit this game does. Besides the physics and gfx filters and all that you have smoke and ambient occlusion and particles and all kinds of dynamic effects all over the place making things look different pretty much every second unless you’re standing on the exact position and goes into photo mode on the exact same second.

I tried matching the pillar scene on PC and getting everything the same, the light, shadows and reflections etc. It was... not easy, I gave up.

Someone explained the missing objects in the mirror on PS5, which I think you forgot to mention, by talking about a randomizer which is placing out the crap you blow around when launching something. Haven’t double checked that but it’s probably right.

All in all, just chill. It’s just a weird little GPU benchmark.
Yes I've played it.....so have they. And still they went and lied ....told the masses that they were identical when it was not......I make a video pointing it out.....you mad for some reason....lol it's like you mad that I told the truth and made a real analysis video for gamers to see the truth....proof and all!
 

Kingthrash

Member
People have screens showing things "missing" in the PS5 version, but they aren't really missing at all as it's randomised. Why would the Devs lie about the settings on their own game? You seem to believe that DF have a pro MS agenda, so you saying the Devs do too?
Do you think those incredibly small differences make up a 16% performance gap on average, if so how?
Well I used the same pics that DF used ....that should tell you alot lil guy
 

Zoro7

Banned
I do wonder how many people here actually have a next gen console. Who cares if the PS5 is weaker?? It’s a fantastic console with a shit tonne of games coming out. Enjoy it lads it ain’t worth the stress of arguing with people like Ricky.
 

ErRor88

Member
FYI, according to the original glitch finder... the CPU usage does not change even in photo mode. This benchmark is a valid assertion of the GPU-CPU rendering on both consoles.

 
Top Bottom