• 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.

Digital Foundry: Deathloop PC vs PS5, Optimised Settings, Performance Testing + More

Snake29

RSI Employee of the Year
Watch the video in the thread people are showing. It's the main ue5 architect saying it runs fine on PC and it was running fine and did all what we saw on the ps5 demo. Those things just already happened when they started with the video. So either the main architect is a liar and everything epic said about ue5 can't be believed or they were telling the truth and it runs fine on PC.

Let's stop it here please and just educate yourself. It was not running as we saw on the PS5 demo. Some of you are fucking blind or something? yes it runs "just fine" in editor mode is COMPLETELY different then equal gameplay. Sure editor mode can be taxin if you drop a lot of object or geometry in the scene, that's the purpose of a editor. It's not the purpose of a editor to run as a slideshow, otherwise how would you even want to create something?


PC editor mode demo breakdown: No physics enabled, no vfx enabled in some of the scenes, no animation, character control or physics:

Y50DKFa.png


PS5:
tTYccnE.png


PC:
KYtDQNL.png


PS5:
Q1UK2wz.png


PC:
Q5Z7GrA.png


PS5:
cIv27x8.png


This comparison is not to compare graphics, it's to compare the difference between editor mode and equal compiled playable demo that is optimized for a different platform. They didn't showed the game at the end on the PC like we've seen on the PS5, so it's really hard to tell how it will perform on the pc, especially in the last section of the demo. The most complex part was the end of the scene for the storage and streaming tech.

That's why UE5 is scalable, and doesn't mean one demo can run on something else without any work on it.
 
Let's stop it here please and just educate yourself. It was not running as we saw on the PS5 demo. Some of you are fucking blind or something? yes it runs "just fine" in editor mode is COMPLETELY different then equal gameplay. Sure editor mode can be taxin if you drop a lot of object or geometry in the scene, that's the purpose of a editor. It's not the purpose of a editor to run as a slideshow, otherwise how would you even want to create something?


PC editor mode demo breakdown: No physics enabled, no vfx enabled in some of the scenes, no animation, character control or physics:

Y50DKFa.png


PS5:
tTYccnE.png


PC:
KYtDQNL.png


PS5:
Q1UK2wz.png


PC:
Q5Z7GrA.png


PS5:
cIv27x8.png


This comparison is not to compare graphics, it's to compare the difference between editor mode and equal compiled playable demo that is optimized for a different platform. They didn't showed the game at the end on the PC like we've seen on the PS5, so it's really hard to tell how it will perform on the pc, especially in the last section of the demo. The most complex part was the end of the scene for the storage and streaming tech.

That's why UE5 is scalable, and doesn't mean one demo can run on something else without any work on it.
Uh no,
He quite literally says all that stuff ran when he ran thru it the first time hours ago and since it's all script based it happens when you get there and then it's done. Now if you want to call him a liar when he says it runs fine on PC that's your choice. If that's the case we can't believe anything epic said about ue5 and yes all bullshit including the ps5 demo. Your choice.
 

Guilty_AI

Member
Let's stop it here please and just educate yourself. It was not running as we saw on the PS5 demo. Some of you are fucking blind or something? yes it runs "just fine" in editor mode is COMPLETELY different then equal gameplay. Sure editor mode can be taxin if you drop a lot of object or geometry in the scene, that's the purpose of a editor. It's not the purpose of a editor to run as a slideshow, otherwise how would you even want to create something?


PC editor mode demo breakdown: No physics enabled, no vfx enabled in some of the scenes, no animation, character control or physics:

Y50DKFa.png


PS5:
tTYccnE.png


PC:
KYtDQNL.png


PS5:
Q1UK2wz.png


PC:
Q5Z7GrA.png


PS5:
cIv27x8.png


This comparison is not to compare graphics, it's to compare the difference between editor mode and equal compiled playable demo that is optimized for a different platform. They didn't showed the game at the end on the PC like we've seen on the PS5, so it's really hard to tell how it will perform on the pc, especially in the last section of the demo. The most complex part was the end of the scene for the storage and streaming tech.

That's why UE5 is scalable, and doesn't mean one demo can run on something else without any work on it.
wooow, are those Deathloop screenshots? It looks eerily familiar to that ue5 demo for some reason
 
Last edited:

Loxus

Member
Watch the video in the thread people are showing. It's the main ue5 architect saying it runs fine on PC and it was running fine and did all what we saw on the ps5 demo. Those things just already happened when they started with the video. So either the main architect is a liar and everything epic said about ue5 can't be believed or they were telling the truth and it runs fine on PC.
Would be nice if your source was Epic and not some random thread were anyone can say anything and make people believe it's true.
 

Snake29

RSI Employee of the Year
Uh no,
He quite literally says all that stuff ran when he ran thru it the first time hours ago and since it's all script based it happens when you get there and then it's done. Now if you want to call him a liar when he says it runs fine on PC that's your choice. If that's the case we can't believe anything epic said about ue5 and yes all bullshit including the ps5 demo. Your choice.

Nobody is calling him a liar, but you guys just don't understand the word "running" in this context. He still confuses people with this word since people believed UE5 was only running on PS5 and nothing else. Yet nobody ever said it about the Engine, but the demo was only optimized for the PS5. Now a editor breakdown is not the same as a optimized version for specific hardware. What will happen if he is not in editor mode?

But since you guys are fucking tone deaf and don't understand the word "running" in a different mode...fuck it then.

But yeah let's keep this in the UE5 thread from now one.
 
Last edited:

Loxus

Member
My source is this video

Watch it, I'm not making anything up. Enjoy it's a good watch.

And where is the gameplay?
This isn't much different to Archicad or Revit model space.

There is not kind of gameplay physics going on in that video to counter anything is said.

If fact, what does anything in that video have to do with PS5 fast and narrow design?
My original point was that fast and narrow design helps with huge triangle count and large texture size.

But you PC guys alway come along and derail the topic.
 
And where is the gameplay?
This isn't much different to Archicad or Revit model space.

There is not kind of gameplay physics going on in that video to counter anything is said.

If fact, what does anything in that video have to do with PS5 fast and narrow design?
My original point was that fast and narrow design helps with huge triangle count and large texture size.

But you PC guys alway come along and derail the topic.
You are just too much.
Peace
 

01011001

Banned
And where is the gameplay?
This isn't much different to Archicad or Revit model space.

There is not kind of gameplay physics going on in that video to counter anything is said.

If fact, what does anything in that video have to do with PS5 fast and narrow design?
My original point was that fast and narrow design helps with huge triangle count and large texture size.

But you PC guys alway come along and derail the topic.

there was never a scenario demonstrated on PC where a fast and narrow GPU outperforms a wide and slow GPU that has the same TFLOP/s performance and uses the same architecture. it was never demonstrated. so until someone does this whole spiel of Cerny is just PR speak with nothing to back it up... and is apparently based on how GCN graphics cards worked because they had parts that simply sucked at actually utilizing the shaders fully at slow clocks, but that has been resolved with RDNA
 

Loxus

Member
there was never a scenario demonstrated on PC where a fast and narrow GPU outperforms a wide and slow GPU that has the same TFLOP/s performance and uses the same architecture. it was never demonstrated. so until someone does this whole spiel of Cerny is just PR speak with nothing to back it up... and is apparently based on how GCN graphics cards worked because they had parts that simply sucked at actually utilizing the shaders fully at slow clocks, but that has been resolved with RDNA
Jeez oh my god.

Mark Cerny was talking about 2 PS5's (36CUs@1GHz vs. 48CUs@0.75GHz) not a PC GPU.

When talking about CU count Mark said, "it's easier to fully use 36CUs in parallel than it is to fully use 48CUs when triangles are small it's much harder to fill although CUs with useful work."

Deathloop doesn't have triangles the size of a pixel or 16 billion triangles in a scene.

When talking about high clocks he said, "Teraflops is defined as the computational capability of the vector ALU. That's just one part of the GPU, there are a lot of other units and those other units all run faster when the GPU frequency is higher. At 33% higher frequency rasterization goes 33% faster, processing the command buffer goes that much faster and other caches have that much higher bandwidth and so on."

Command buffer is a temporary storage area where commands are kept.
Rasterization is the task of taking an image described in a vector graphics format (shapes) and converting it into a raster image.

Cerny's SSD dream, "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."

f:id:keepitreal:20200329135313j:plain


"If you figure that it takes half a second to turn that's 4GB of compressed data you can load that sounds about right for next gen."

The speed of rasterization, the command buffer and caches are probably to complement the speed of things coming off the SSD.

There was a talk were the PS5 GPU may become a bottleneck if it can't render stuff coming of the SSD fast enough.

Does Deathloop stress the SSD? No.
Does Deathloop have billions of triangles? No.

This is why you can't compare the PS5 GPU to a 5700XT based on Deathloop performance become the 5700XT doesn't have the hardware like Cache Scrubbers to assist with data coming off the SSD with the use of dedicated hardware decompression.

So to say Mark Cerny is lying about fast and narrow without a game that takes advantage of the hardware is just plain ignorant.
 
Last edited:

Buggy Loop

Member
What the fuck happened to this thread? Who even brought up the fucking UE5 demo for the thousandth time again?

Stop It Michael Jordan GIF
 
Last edited:

Guilty_AI

Member
Jeez oh my god.

Mark Cerny was talking about 2 PS5's (36CUs@1GHz vs. 48CUs@0.75GHz) not a PC GPU.

When talking about CU count Mark said, "it's easier to fully use 36CUs in parallel than it is to fully use 48CUs when triangles are small it's much harder to fill although CUs with useful work."

Deathloop doesn't have triangles the size of a pixel or 16 billion triangles in a scene.

When talking about high clocks he said, "Teraflops is defined as the computational capability of the vector ALU. That's just one part of the GPU, there are a lot of other units and those other units all run faster when the GPU frequency is higher. At 33% higher frequency rasterization goes 33% faster, processing the command buffer goes that much faster and other caches have that much higher bandwidth and so on."

Command buffer is a temporary storage area where commands are kept.
Rasterization is the task of taking an image described in a vector graphics format (shapes) and converting it into a raster image.

Cerny's SSD dream, "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."

f:id:keepitreal:20200329135313j:plain


"If you figure that it takes half a second to turn that's 4GB of compressed data you can load that sounds about right for next gen."

The speed of rasterization, the command buffer and caches are probably to complement the speed of things coming off the SSD.

There was a talk were the PS5 GPU may become a bottleneck if it can't render stuff coming of the SSD fast enough.

Does Deathloop stress the SSD? No.
Does Deathloop have billions of triangles? No.

This is why you can't compare the PS5 GPU to a 5700XT based on Deathloop performance become the 5700XT doesn't have the hardware like Cache Scrubbers to assist with data coming off the SSD with the use of dedicated hardware decompression.

So to say Mark Cerny is lying about fast and narrow without a game that takes advantage of the hardware is just plain ignorant.
Its ok man, we've all been misled by these cool tech talks at some point in out lifes. Time to put down the gun and move on

68747470733a2f2f73332e616d617a6f6e6177732e636f6d2f776174747061642d6d656469612d736572766963652f53746f7279496d6167652f5877696b5a3441496b6f476379413d3d2d3333323330333036302e3134383561643366626433643631386632313036393432393633352e676966
 

Kenpachii

Member
Let's stop it here please and just educate yourself. It was not running as we saw on the PS5 demo. Some of you are fucking blind or something? yes it runs "just fine" in editor mode is COMPLETELY different then equal gameplay. Sure editor mode can be taxin if you drop a lot of object or geometry in the scene, that's the purpose of a editor. It's not the purpose of a editor to run as a slideshow, otherwise how would you even want to create something?


PC editor mode demo breakdown: No physics enabled, no vfx enabled in some of the scenes, no animation, character control or physics:

Y50DKFa.png


PS5:
tTYccnE.png


PC:
KYtDQNL.png


PS5:
Q1UK2wz.png


PC:
Q5Z7GrA.png


PS5:
cIv27x8.png


This comparison is not to compare graphics, it's to compare the difference between editor mode and equal compiled playable demo that is optimized for a different platform. They didn't showed the game at the end on the PC like we've seen on the PS5, so it's really hard to tell how it will perform on the pc, especially in the last section of the demo. The most complex part was the end of the scene for the storage and streaming tech.

That's why UE5 is scalable, and doesn't mean one demo can run on something else without any work on it.

Believes the demo ran on a PS5 when developer says it does even while there is zero proof or ways for you to play it or see it on your PS5.
Doesn't believe the same developer when he says pc runs more complex enviroments even while it demonstrates it in multiple video's and actually ships a playable demo fully playable on PC that is extremely unoptimized as he states on top of it that runs on potato pc's.

Not going to spoil for you the ram usage the v-ram usage and the SSD usage. It was exactly what u would expect, not impressive even remotely. Guess why they never told you that information in the presentation and after it? it took all the way until the PC demo came out and people could see it.

Also whoever says the PS5 GPU is a 5700xt is a idiot.

5700xt can't RT for shit.
 
Last edited:

Zuzu

Member
NX Gamer has recently uploaded a video which includes a comparison between the PS5 and his overclocked RTX 2070.

At approximately the 7:35 minute mark he begins a statement where he says that the PS5 performs approximately 30% better than the 2070 at equivalent quality settings when the character is standing still and between 28-42% better when the character is moving. At approximately the 10:53 minute mark he begins a comparison between the 2070 running the equivalent ray traced settings to the PS5 and concludes that the PS5 is outperforming the 2070 at up to 20% better in its ray tracing mode.

This is obviously a different conclusion to Alex's video. I know there's been debate on this forum regarding the possibility of NX Gamer's cpu being a bottleneck to his RTX 2070. But would this really be the case for the RTX mode where both machines are dropping below 30fps? That seems to be a GPU bottleneck. This seems to provide some strong evidence that the PS5 is more powerful than an overclocked RTX 2070.

NX Gamer's video is here:

 
Last edited:

Snake29

RSI Employee of the Year
Believes the demo ran on a PS5 when developer says it does even while there is zero proof or ways for you to play it or see it on your PS5.
Doesn't believe the same developer when he says pc runs more complex enviroments even while it demonstrates it in multiple video's and actually ships a playable demo fully playable on PC that is extremely unoptimized as he states on top of it that runs on potato pc's.

Not going to spoil for you the ram usage the v-ram usage and the SSD usage. It was exactly what u would expect, not impressive even remotely. Guess why they never told you that information in the presentation and after it? it took all the way until the PC demo came out and people could see it.

Also whoever says the PS5 GPU is a 5700xt is a idiot.

5700xt can't RT for shit.

I ran the demo myself, you don’t have to explain me that. The demo scene The Acient wasnt more impressive then the PS5 demo, and my fps was higher in editor mode. You could mess with the environment that were the ram usage will go sky rocket.

When i ran the playable demo, the fps was a bit lower since ghe scenes geometry and texture detail changed, vfx was added on top of it and you had a playable character.
 

SlimySnake

Flashless at the Golden Globes
NX Gamer has recently uploaded a video which includes a comparison between the PS5 and his overclocked RTX 2070.

At approximately the 7:35 minute mark he begins a statement where he says that the PS5 performs approximately 30% better than the 2070 at equivalent quality settings when the character is standing still and between 28-42% better when the character is moving. At approximately the 10:53 minute mark he begins a comparison between the 2070 running the equivalent ray traced settings to the PS5 and concludes that the PS5 is outperforming the 2070 at up to 20% better in its ray tracing mode. This is obviously a different conclusion to Alex's video.

NX Gamer's video is here:


Nx gamer uses a pretty shit cpu. Don’t take his comparison too seriously.

Ps5 is probably closer to a 2070 super in standard rasterization but it is by no means 20% faster than a 2070 in rt. That should’ve raised a red flag right there
 
Last edited:
NX Gamer has recently uploaded a video which includes a comparison between the PS5 and his overclocked RTX 2070.

At approximately the 7:35 minute mark he begins a statement where he says that the PS5 performs approximately 30% better than the 2070 at equivalent quality settings when the character is standing still and between 28-42% better when the character is moving. At approximately the 10:53 minute mark he begins a comparison between the 2070 running the equivalent ray traced settings to the PS5 and concludes that the PS5 is outperforming the 2070 at up to 20% better in its ray tracing mode.

This is obviously a different conclusion to Alex's video. I know there's been debate on this forum regarding the possibility of NX Gamer's cpu being a bottleneck to his RTX 2070. But would this really be the case for the RTX mode where both machines are dropping below 30fps? That seems to be a GPU bottleneck. This seems to provide some strong evidence that the PS5 is more powerful than an overclocked RTX 2070.

NX Gamer's video is here:


NX Gamer needs to upgrade his CPU or his comparison videos are a waste of bandwidth at least if he is comparing a specific GPU to the PS5.

Go look at the Horizon Zero Dawn thread for how wrong his conclusions were. I was getting almost double his frame rates with a 2070s and the same settings.
 

Armorian

Banned
NX Gamer has recently uploaded a video which includes a comparison between the PS5 and his overclocked RTX 2070.

At approximately the 7:35 minute mark he begins a statement where he says that the PS5 performs approximately 30% better than the 2070 at equivalent quality settings when the character is standing still and between 28-42% better when the character is moving. At approximately the 10:53 minute mark he begins a comparison between the 2070 running the equivalent ray traced settings to the PS5 and concludes that the PS5 is outperforming the 2070 at up to 20% better in its ray tracing mode.

This is obviously a different conclusion to Alex's video. I know there's been debate on this forum regarding the possibility of NX Gamer's cpu being a bottleneck to his RTX 2070. But would this really be the case for the RTX mode where both machines are dropping below 30fps? That seems to be a GPU bottleneck. This seems to provide some strong evidence that the PS5 is more powerful than an overclocked RTX 2070.

NX Gamer's video is here:



RT adds a lot of load on the CPU. He really needs to change his CPU to do this comparisons.

Minimum framerate is almost doubled on better CPUs and this is without RT:

SRorMI1.png
 
Last edited:

SlimySnake

Flashless at the Golden Globes
NX Gamer needs to upgrade his CPU or his comparison videos are a waste of bandwidth at least if he is comparing a specific GPU to the PS5.

Go look at the Horizon Zero Dawn thread for how wrong his conclusions were. I was getting almost double his frame rates with a 2070s and the same settings.
Yep. His comparison to the 5600xt should’ve been the first clue that something was off.

VHPIQfK.png
 
Last edited:

NEbeast

Member
I would say it's mostly irrelevant and ~10TF RDNA 1/2 card should be comparable in raster as long as it's not memory BW limited or something.



This is also what alex says:

INsSPXK.png


But I don't say that PS5 is worse than 2060S in raster. You can see where PS5 lands based on pure specs alone and that in most games I think it shows results similar to DL.



RDNA1 and RDNA2 cards can be totally compared to each other and PS5, it's the same fucking arch.

The ps5 has kept up with the SX in 90% of tests, performed better on alot of those, I hope you aren't team green and somehow think all your posts are "another one"
 
NX Gamer has recently uploaded a video which includes a comparison between the PS5 and his overclocked RTX 2070.

At approximately the 7:35 minute mark he begins a statement where he says that the PS5 performs approximately 30% better than the 2070 at equivalent quality settings when the character is standing still and between 28-42% better when the character is moving. At approximately the 10:53 minute mark he begins a comparison between the 2070 running the equivalent ray traced settings to the PS5 and concludes that the PS5 is outperforming the 2070 at up to 20% better in its ray tracing mode.

This is obviously a different conclusion to Alex's video. I know there's been debate on this forum regarding the possibility of NX Gamer's cpu being a bottleneck to his RTX 2070. But would this really be the case for the RTX mode where both machines are dropping below 30fps? That seems to be a GPU bottleneck. This seems to provide some strong evidence that the PS5 is more powerful than an overclocked RTX 2070.

NX Gamer's video is here:


Man, he really needs to stop doing PS5 vs PC comparisons. Imagine believing that the RX 5600XT actually outperforms an RTX 2070. Fix your PC NXGamer NXGamer .
 

hlm666

Member
There's a reason people were still primarily going with intel until the AMD 3000 series, you would think he would have known he should have got an intel cpu when he bought that 2700x. Look at that 9700k, and thats basically a rebadged 8700k he could have got when he got his fine wine 2700x.
 

Darius87

Member
Nx gamer uses a pretty shit cpu. Don’t take his comparison too seriously.

Ps5 is probably closer to a 2070 super in standard rasterization but it is by no means 20% faster than a 2070 in rt. That should’ve raised a red flag right there
NX Gamer needs to upgrade his CPU or his comparison videos are a waste of bandwidth at least if he is comparing a specific GPU to the PS5.

Go look at the Horizon Zero Dawn thread for how wrong his conclusions were. I was getting almost double his frame rates with a 2070s and the same settings.
it's very similiar CPU's to PS5 except cores differences but everything else is better then PS5 CPU cache sizes, boost clocks etc, and anyway it doesn't matter because Deathloop isn't CPU bound.
DF actually uses better CPU's and GPU's then PS5 and compares it to if that makes you feel better. People who thinks 3700x is PS5 equalent CPU are delusional it's way better then PS5 CPU.
 
NX Gamer has recently uploaded a video which includes a comparison between the PS5 and his overclocked RTX 2070.

At approximately the 7:35 minute mark he begins a statement where he says that the PS5 performs approximately 30% better than the 2070 at equivalent quality settings when the character is standing still and between 28-42% better when the character is moving. At approximately the 10:53 minute mark he begins a comparison between the 2070 running the equivalent ray traced settings to the PS5 and concludes that the PS5 is outperforming the 2070 at up to 20% better in its ray tracing mode.

This is obviously a different conclusion to Alex's video. I know there's been debate on this forum regarding the possibility of NX Gamer's cpu being a bottleneck to his RTX 2070. But would this really be the case for the RTX mode where both machines are dropping below 30fps? That seems to be a GPU bottleneck. This seems to provide some strong evidence that the PS5 is more powerful than an overclocked RTX 2070.

NX Gamer's video is here:


This is exactly why I said we better wait NXGamer video as his comparisons are fair while DF comparisons usually are often designed to make the PC GPU look better, somehow, mostly by cherry picking one particular scene (like the way they did actually in their video).

RT adds a lot of load on the CPU. He really needs to change his CPU to do this comparisons.

Minimum framerate is almost doubled on better CPUs and this is without RT:

SRorMI1.png
But if you want to compare GPUs you need to use similar CPUs. If you use a scene limited by the CPU you'll benchmark the CPUs using that particular scene, not the GPU.
 

Zuzu

Member
Yep. His comparison to the 5600xt should’ve been the first clue that something was off.

VHPIQfK.png
Ok wow, this does bring the conclusions of his video into serious doubt. He probably should change his CPU to something that's newer.
 
Last edited:

Md Ray

Member
Nx gamer uses a pretty shit cpu.
Man, this whole "NXG uses a pretty shit CPU" is really blown out of proportion like he's using a Jaguar CPU or something in his PC. It's a Ryzen 7 2700 CPU, while yes it's slower than today's CPUs, but it's still plenty capable of delivering at least 60fps in a game like Deathloop that leans more towards GPU than CPU. In GPU-bound scenarios, both Ryzen 2700 and PS5's CPU should deliver more or less similar perf, at the end of the day.

Here's how the Zen+ 2700 non-X compares against Zen 3 5800X under 4K GPU-bound conditions:

QXlMDVE.png


On avg. the Zen+ CPU is able to deliver 99% of the 5800X's perf, is this not enough? CPU isn't really an issue at 2160p which is what most devs are shooting for on consoles via upscaling techniques, dynamic res scalers. In the case of Deathloop - Arkane is using both dynamic res and FSR, so it's predominantly GPU-bound.
 
Last edited:

Zathalus

Member
NX Gamer has recently uploaded a video which includes a comparison between the PS5 and his overclocked RTX 2070.

At approximately the 7:35 minute mark he begins a statement where he says that the PS5 performs approximately 30% better than the 2070 at equivalent quality settings when the character is standing still and between 28-42% better when the character is moving. At approximately the 10:53 minute mark he begins a comparison between the 2070 running the equivalent ray traced settings to the PS5 and concludes that the PS5 is outperforming the 2070 at up to 20% better in its ray tracing mode.

This is obviously a different conclusion to Alex's video. I know there's been debate on this forum regarding the possibility of NX Gamer's cpu being a bottleneck to his RTX 2070. But would this really be the case for the RTX mode where both machines are dropping below 30fps? That seems to be a GPU bottleneck. This seems to provide some strong evidence that the PS5 is more powerful than an overclocked RTX 2070.

NX Gamer's video is here:


This seems a bit odd. The PS5 GPU is 30-40% better then a overclocked 2070? That would put it around 2080 Super to 2080ti territory. Also 20% better in RT mode? Once again as fast as a 2080 in RT performance.

None of that lines up with what we know about either consoles GPU.
 

Snake29

RSI Employee of the Year
And where is the gameplay?
This isn't much different to Archicad or Revit model space.

There is not kind of gameplay physics going on in that video to counter anything is said.

If fact, what does anything in that video have to do with PS5 fast and narrow design?
My original point was that fast and narrow design helps with huge triangle count and large texture size.

But you PC guys alway come along and derail the topic.

Because these are the same frustrated people in the UE5 thread who keep spamming with :messenger_tears_of_joy: :messenger_tears_of_joy: :messenger_tears_of_joy: :messenger_tears_of_joy: ....but have no arguments.
 

Md Ray

Member
RT adds a lot of load on the CPU. He really needs to change his CPU to do this comparisons.

Minimum framerate is almost doubled on better CPUs and this is without RT:

SRorMI1.png
I don't really see an issue. Deathloop is targeting 60fps on PS5, here even a 4770K quad-core CPU that is almost a decade old is able to deliver 60+fps. NXG was also targeting 60fps in his tests anyway.
 
Last edited:

hlm666

Member
I don't really see an issue. Deathloop is targeting 60fps on PS5, here even a 4770K quad-core CPU that is almost a decade old is able to deliver 60+fps. NXG was also targeting 60fps in his tests anyway.
If that's the case what exactly is he doing to make the 5600xt match the 2070, the site you linked to has numbers that contradict that at the very least.
 
Nx gamer uses a pretty shit cpu. Don’t take his comparison too seriously.

Ps5 is probably closer to a 2070 super in standard rasterization but it is by no means 20% faster than a 2070 in rt. That should’ve raised a red flag right there
I was new to this whole PC gaming thing but I did my research and threw in a Ryzen 7.
 

NXGamer

Member
Man, he really needs to stop doing PS5 vs PC comparisons. Imagine believing that the RX 5600XT actually outperforms an RTX 2070. Fix your PC NXGamer NXGamer .
I really, truly wish people would stop being so "offended" by facts. A Console is more than a GPU a PC is more than a GPU, when I test PC versus consoles these are based on Specs that are more reflective AND real. As I have said before many upgrade GPU before CPU, some and many will have a Zen2700 CPU and then upgrade from a 2018 built machine with a RX580 to a 2070 or 2070Super etc. When others test JUST the GPU and then stick a 5900x or I710900k with is really creating a single test. My tests are designed to reflect real machines and 2 specs and the results show that (as I always keep saying) that reductive conversation on many forums (including this one) reduce these conversations to such a high level. PC is Faster than Console, the same as saying Bikes are faster than Cars, it is true but the real answer is it depends on the bike or the PC spec.
 

NXGamer

Member
Ok wow, this does bring the conclusions of his video into serious doubt. He probably should change his CPU to something that's newer.
It really does not, this is a graph or what scene, how long, what happened.

I can test scenes on a 5600XT with an unlocked frame-rate than shows averages above 70fps. People using these as some basis are just looking for confirmation bias, these are only good to compare GPU to GPU or CPU to CPU so long as they are all identical shots. This does not reflect anything more than the individual sites tests with no pictures we can only guess what that was.
 

Armorian

Banned
I don't really see an issue. Deathloop is targeting 60fps on PS5, here even a 4770K quad-core CPU that is almost a decade old is able to deliver 60+fps. NXG was also targeting 60fps in his tests anyway.

You can't guarantee that it won't go below 60 on 4770k. Test was done in some place for some time etc. Just to show CPU scaling and stuff.
 
I really, truly wish people would stop being so "offended" by facts. A Console is more than a GPU a PC is more than a GPU, when I test PC versus consoles these are based on Specs that are more reflective AND real. As I have said before many upgrade GPU before CPU, some and many will have a Zen2700 CPU and then upgrade from a 2018 built machine with a RX580 to a 2070 or 2070Super etc. When others test JUST the GPU and then stick a 5900x or I710900k with is really creating a single test. My tests are designed to reflect real machines and 2 specs and the results show that (as I always keep saying) that reductive conversation on many forums (including this one) reduce these conversations to such a high level. PC is Faster than Console, the same as saying Bikes are faster than Cars, it is true but the real answer is it depends on the bike or the PC spec.
If you were just comparing two PCs, it'd be fine since it would simply show the CPU limitations and that's an interesting thing to know. But you're comparing a PC to a PS5, and then people come to weird conclusions like "PS5 is 40% faster than a 2070" when it simply isn't true. If you want a real PS5 vs PC benchmark, stick the 2070 into your other computer which has the Zen 2 CPU.
 

Armorian

Banned
I really, truly wish people would stop being so "offended" by facts. A Console is more than a GPU a PC is more than a GPU, when I test PC versus consoles these are based on Specs that are more reflective AND real. As I have said before many upgrade GPU before CPU, some and many will have a Zen2700 CPU and then upgrade from a 2018 built machine with a RX580 to a 2070 or 2070Super etc. When others test JUST the GPU and then stick a 5900x or I710900k with is really creating a single test. My tests are designed to reflect real machines and 2 specs and the results show that (as I always keep saying) that reductive conversation on many forums (including this one) reduce these conversations to such a high level. PC is Faster than Console, the same as saying Bikes are faster than Cars, it is true but the real answer is it depends on the bike or the PC spec.

In dx12 games Nvidia driver produce more overhead than AMD driver (tests on HU) so that may explain why you see different results than other outlets testing GPUs in this game. The weaker the CPU the more affected it is.

And CPU is very important in gaming, i had 3600 that is better than 2700 for gaming in every way and there were games dripping below 60 on it. And that were pure CPU related drops. Now that consoles don't have jaguars anymore CPU is more important than ever.
 

NXGamer

Member
In dx12 games Nvidia driver produce more overhead than AMD driver (tests on HU) so that may explain why you see different results than other outlets testing GPUs in this game. The weaker the CPU the more affected it is.

And CPU is very important in gaming, i had 3600 that is better than 2700 for gaming in every way and there were games dripping below 60 on it. And that were pure CPU related drops. Now that consoles don't have jaguars anymore CPU is more important than ever.
If you were just comparing two PCs, it'd be fine since it would simply show the CPU limitations and that's an interesting thing to know. But you're comparing a PC to a PS5, and then people come to weird conclusions like "PS5 is 40% faster than a 2070" when it simply isn't true. If you want a real PS5 vs PC benchmark, stick the 2070 into your other computer which has the Zen 2 CPU.
But that is not my aim, I am trying to show the audience information and how it may run on their machine. You and other here only want to see a Win in favour of your platform as others will want for the PS5. I am not interested I am only interested in the results and the data. Hardware is much more than one part, software is more than just hardware and the final answer to all these questions is, its complicated. Just see above for the exact thing I am trying to demonstrate and talk about all the time. Doing a technical analysis is more than just picking the best parts to confirm your bias (see should never be this).
 
Last edited:
Deathloop doesn't have triangles the size of a pixel or 16 billion triangles in a scene.

Rasterization is the task of taking an
Does Deathloop stress the SSD? No.
Does Deathloop have billions of triangles? No.

What's with all this billions of triangles talk? I'm pretty confident the ps5 has never and will never have billions of triangles in a scene(frame)
 
But that is not my aim, I am trying to show the audience information and how it may run on their machine. You and other here only want to see a Win in favour of your platform as others will want for the PS5. I am not interested I am only interested in the results and the data. Hardware is much more than one part, software is more than just hardware and the final answer to all these questions is, its complicated. Just see above for the exact thing I am trying to demonstrate and talk about all the time. Doing a technical analysis is more than just picking the best parts to confirm your bias (see should never be this).
I get what you're saying, but I feel like the vast majority of your audience isn't interested in how a game runs on a specific machine. They want to know how the PS5 compares to a PC. And your videos are also framed like that. So why not do comparisons with similar hardware? Why give the PC a huge handicap when doing the comparison? Comparing similar hardware would actually be the opposite of bias.
 
Last edited:

martino

Member
But that is not my aim, I am trying to show the audience information and how it may run on their machine. You and other here only want to see a Win in favour of your platform as others will want for the PS5. I am not interested I am only interested in the results and the data. Hardware is much more than one part, software is more than just hardware and the final answer to all these questions is, its complicated. Just see above for the exact thing I am trying to demonstrate and talk about all the time. Doing a technical analysis is more than just picking the best parts to confirm your bias (see should never be this).
Why do you expect gamer will have more 2700 than 3600 cpu especially for gaming when 8 core is less than half of 6 core one on steam survey ? (30% vs 12%)
Why pair a less successful card with a more successful cpu ? (on steam survey 0.36% for 5600xt vs 1.77% for 2070)
Where is the data that lead to think your choice reflect a good reality ? point them to me ....steam survey is what it is but still the source gathering most data easily accessible to get an idea and it make your choice of reality again a strange one.
 

NXGamer

Member
I get what you're saying, but I feel like the vast majority of your audience isn't interested in how a game runs on a specific machine. They want to know how the PS5 compares to a PC. And your videos are also framed like that. So why not do comparisons with similar hardware? Why give the PC a huge handicap when doing the comparison? Comparing similar hardware would actually be the opposite of bias.
Again, I am not handing any 'handicap' I am testing what I have access to. I have just built a new machine that I want to add into the mix and will test also when moving forward. But my PC spec is as equally valid as any other you may see. My only aim is to inform and present facts to help the audience, if you have a better machine or worse you can scale from here.
 

yamaci17

Member
okay now i see the test hes using native 4k with rt on rtx 2070 , yet it is known that ps5 drops to 1440p 1800p in that mode. without dropping the res, it is meaningless to compare both
 
Last edited:
i dont understand why some people cant accept that ps5 can outperform a 2070? why it seems like mythological or impossible for some? its not like rtx 2070 has the advantage of running with superior lowlevel metal API with superior sauce special console optimizations?
Because secret sauce doesn't exist for 3rd party games. If a game runs similarly on AMD and Nvidia hardware, the PS5 will always perform somewhere around the 2070 level. In Nvidia games, it will perform worse. In AMD games, it will perform better. It's how it goes.
 
NXGamer NXGamer
Hey few things
First you used caveats of your cpu a few times so I think you are in the right and people should stop complaining about that.
Second you can't say it performs better on AMD based on a test that uses 2 very different CPUs. I'm not saying that's not true I'm just saying your test doesn't show that.
And third your oc 2070 isn't between a 2070s and 3060ti. 3060 ti edges out the 2080 super so that's a bogus claim.
Decent video overall though and I'm happy you mentioned your CPU a few times.
 

NXGamer

Member
Why do you expect gamer will have more 2700 than 3600 cpu especially for gaming when 8 core is less than half of 6 core one on steam survey ? (30% vs 12%)
Why pair a less successful card with a more successful cpu ? (on steam survey 0.36% for 5600xt vs 1.77% for 2070)
Where is the data that lead to think your choice reflect a good reality ? point them to me ....steam survey is what it is but still the source gathering most data easily accessible to get an idea and it make your choice of reality again a strange one.
I do not assume, see my post above. I am using 2 specs that are popular within the steam survey and that I have access to. The 5600XT was a machine that I built after the 2070, currently have another build that I will have ready to use soon. Should out a smile on some people's faces I am sure.
NXGamer NXGamer
Hey few things
First you used caveats of your cpu a few times so I think you are in the right and people should stop complaining about that.
Second you can't say it performs better on AMD based on a test that uses 2 very different CPUs. I'm not saying that's not true I'm just saying your test doesn't show that.
And third your oc 2070 isn't between a 2070s and 3060ti. 3060 ti edges out the 2080 super so that's a bogus claim.
Decent video overall though and I'm happy you mentioned your CPU a few times.
Re your second point, I do say in the video the 6% delta between the 2070 and 5600 is likely down to the 3600 and would likely reduce to a net negative also.
third, it is though, at the moment it is outperforming a 2070 and a stock 2070 in this spec I am sure. which is between the 3060Ti but that is a 18-25% margin.
 
Last edited:
I do not assume, see my post above. I am using 2 specs that are popular within the steam survey and that I have access to. The 5600XT was a machine that I built after the 2070, currently have another build that I will have ready to use soon. Should out a smile on some people's faces I am sure.

Re your second point, I do say in the video the 6% delta between the 2070 and 5600 is likely down to the 3600 and would likely reduce to a net negative also.
third, it is though, at the moment it is outperforming a 2070 and a stock 2070 in this spec I am sure. which is between the 3060Ti but that is a 18-25% margin.
Must have missed that over the sound of my bidet.
Better then decent video overall. Just double check 3060ti performance. It's quite good.
 

Shmunter

Member
NX Gamer has recently uploaded a video which includes a comparison between the PS5 and his overclocked RTX 2070.

At approximately the 7:35 minute mark he begins a statement where he says that the PS5 performs approximately 30% better than the 2070 at equivalent quality settings when the character is standing still and between 28-42% better when the character is moving. At approximately the 10:53 minute mark he begins a comparison between the 2070 running the equivalent ray traced settings to the PS5 and concludes that the PS5 is outperforming the 2070 at up to 20% better in its ray tracing mode.

This is obviously a different conclusion to Alex's video. I know there's been debate on this forum regarding the possibility of NX Gamer's cpu being a bottleneck to his RTX 2070. But would this really be the case for the RTX mode where both machines are dropping below 30fps? That seems to be a GPU bottleneck. This seems to provide some strong evidence that the PS5 is more powerful than an overclocked RTX 2070.

NX Gamer's video is here:


Steam gets review bombed due to poor pc version. Alex - no no no, is all good baby…
 
Top Bottom