MightySquirrel
Banned
Spoiler: I tried 1080p resolution on 3080 and that low texture streaming bug? is still there.
-
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.
[NxGamer] Spider-Man Remastered PC vs PS5 vs Steam Deck vs PS4 vs 750Ti - Technical Review & Comparison
- Thread starter Tqaulity
- Start date
Spoiler: I tried 1080p resolution on 3080 and that low texture streaming bug? is still there.
Md Ray
Member
Is gaming or synthetic benchmarks?
ChiefDada
Member
The bolded is just about the only thing I don't understand/agree with your argument; The gpu and vram are intertwined and you are commenting as if it's not. I really wish console had dedicated RT cores, but that doesn't mean I'm against comparing 30 series with console for RT heavy workloads. All hardware design choices have sacrifices.
01011001
Banned
well targeting 1440p makes a lot of sense. 1080p and 1440p are by far the most used resolutions on PC after all. and for good reason, at the seize of the typical PC monitor (25" to 27") 1440p is more than enough and will also help you keep settings high without introducing unwanted res scaling.
the bigger issue might be more vram demanding games of the future. but maybe Nvidia was putting its cards on direct storage helping their cards out in the long run
edit: Steam HW Survey August 2022
Last edited:
playsaves3
Banned
That’s strange Nx gamer mentioned it was high on the fidelity mode specifically and the ps5 never drops to 35fps i literally don’t know if you saw something different but the absolute lowest it can hit is right at 40
playsaves3
Banned
I legit never used neogaf before a month or so ago I only lurked here since last year. I also doubt I talk like any person herePoor yamaci17 getting bombarded by some guys alt
playsaves3
Banned
Im talking about the equal performance thing
yamaci17
Member
okay bro, it never drops to 35 and drops to 40, okay? nx gamer is free to prove that ps5 uses high weather particles by making cross comparisons with PC
Last edited:
SlimySnake
Flashless at the Golden Globes
Cinebench. They tried to do ingame benchmarks but the chip only supports PCIE 2.0 bandwidth which destroyed the performance of their 3090 systems so any gaming tests were inconclusive. And since AMD disabled the iGPU cores, they werent able to test any games on the actual chip itself. AMD does not recommend this CPU to be paired up with gaming anyway. Its for desktop applications use only.
As you can see, the 2700x is roughly 4% faster than the PS5 CPU and the 2700 is roughly 5% slower than 2700x, so its a very good CPU to pair up with the 32 CU 10.6 tflops 6600xt or the 40 CU 13.1 tflops 6700xt which can be downclocked to get to 10.23 tflops. The 6700xt has a higher bandwidth which should bring it in line with the PS5 bandwidth so that might be just a bit more accurate, but i havent seen any 6600xt vs 6700xt benchmarks be hampered by its lower ram bandwidth.
Last edited:
s-bojan
Banned
I have 8700k, 3000 ram, 3070 which I run in x8 slot (as I’ve damaged x16 one).
In most games I get better performance than on my ps5. Sometimes it’s close, sometimes the difference is really large.
Far Cry 6 was the first game ps5 offered something that my pc could not - high textures.
I was not able to run spiderman with RT in locked 1080p/60, even with textures set to high. This was the first game where I’ve felt that playing on ps5 would be preferable.
My pc has many potential bottlenecks. Honestly I am not even sure why performance is like that. But I really hope that this is not a sign of things to come.
In most games I get better performance than on my ps5. Sometimes it’s close, sometimes the difference is really large.
Far Cry 6 was the first game ps5 offered something that my pc could not - high textures.
I was not able to run spiderman with RT in locked 1080p/60, even with textures set to high. This was the first game where I’ve felt that playing on ps5 would be preferable.
My pc has many potential bottlenecks. Honestly I am not even sure why performance is like that. But I really hope that this is not a sign of things to come.
Last edited:
playsaves3
Banned
Dude im not personally offended and I’ve appreciated having this discussion with you even with our disagreements. More people have the oppurtunity to do the tests on pc vs ps5 vrr modes but no one but Nx has bothered so he is left as our sole source
playsaves3
Banned
6650xt is the closer match to ps5 I’d say 6700xt is more series x levelCinebench. They tried to do ingame benchmarks but the chip only supports PCIE 2.0 bandwidth which destroyed the performance of their 3090 systems so any gaming tests were inconclusive. And since AMD disabled the iGPU cores, they werent able to test any games on the actual chip itself. AMD does not recommend this CPU to be paired up with gaming anyway. Its for desktop applications use only.
As you can see, the 2700x is roughly 4% faster than the PS5 CPU and the 2700 is roughly 5% slower than 2700x, so its a very good CPU to pair up with the 32 CU 10.6 tflops 6600xt or the 40 CU 13.1 tflops 6700xt which can be downclocked to get to 10.23 tflops. The 6700xt has a higher bandwidth which should bring it in line with the PS5 bandwidth so that might be just a bit more accurate, but i havent seen any 6600xt vs 6700xt benchmarks be hampered by its lower ram bandwidth.
yamaci17
Member
he's free to compare rx 6800/ryzen 3600 to ps5 in the exact same intro scene. somehow he eluded from doing that. do you think it is convenient?
if rx 6800 ends up being much better than ps5 in matched scenes, then it clearly proves my point that PS5 is not overachieveing, but it is the 3070 that is underperforming. only way for PS5 to overachieve itself is to get equivalent RT performance to that of 2080ti's, but it is not happening either
Last edited:
yamaci17
Member
you will probably hate me for saying it but at this point i dont give any respect towards 6600xt / 6650xt either. initially i also thought they'd match ps5, but not anymore. at 1080p/1300p, they can be, but at 4K, 6600xt/6650xt shits the bed with their 128 bit interface and weak memory bandwidth one being 280 gb/s (6650xt) and other being 256 gb/s (6600xt). 6600xt and 3060ti literally ties together at 1080p yet at 4K, 3060ti roflstomps it. there is something very wrong with those GPUs at higher resolutions that bogs down their compute capability
6600xt literally gets destroyed in spiderman at 4k, due to how badly its memory subsystem bottlenecks it
at this point only proper GPUs we can compare to PS5 is to a point, is 6700xt (you can calculate how better 6700xt is to PS5 and how much better it should theoritically be, and make your overachieving calculations from there). on NVIDIA, we have 3060 that undershoots, and 3080 that overshoots
Last edited:
playsaves3
Banned
He mentioned in the video he wasnt testing every last configuration and then doing comparisons cause that would make this a well over an hour video you can look at rhe results though he mentioned it’s 10-20% faster on average and it holds up in the framerate hovering in the 50-55 región mostly
yamaci17
Member
btw i confirmed that ps5 fidelity mode uses high particles. I retract what I said about that
I now started to wonder if it is really native 4k or not with that VRR mode. it is interesting. supposedly fidelity mod has a dynamic range between 1584p and 2160p, with a cap of 30, to actually hold the 30 frames. Is there someone who can share a couple of lossless screnshots with VRR/Fidelity mode unlocked FPS mode from spiderman remastered? I can do the pixel counting if someone can provide shots. I tried to get some estimations from the elanalista and nxg videos but videos are too lossy to do it.
Last edited:
SlimySnake
Flashless at the Golden Globes
yeah, thats why i suggested they downclock the 6700xt GPU to get to 10.2 tflops.
I think you should always use the best cpu available so as not to hamstring the gpu. You are not really testing the gpu if it gets limited by the cpu.
Md Ray
Member
Cinebench is not representative of real-world gaming performance, don't judge gaming perf out of CB as it is pretty much confined to be dependant on core/thread count and to be very sensitive towards the core clocks. Things like system memory bandwidth, amount of cache and its latency, DRAM latency do not matter at all, whereas for gaming, it absolutely does! Gaming workload is often not like CB. Spider-Man for e.g. heavily thrashes system memory BW. Had the PS5 SoC with its GDDR6 NOT been limited by PCIe 2.0, it would have performed significantly faster than 2700X with DDR4 in this game due to vastly more system mem BW offered up by GDDR6 for the CPU. So I don't think you can say PS5 CPU performs roughly equivalent to 2700 just from looking at their Cinebench perf.Cinebench. They tried to do ingame benchmarks but the chip only supports PCIE 2.0 bandwidth which destroyed the performance of their 3090 systems so any gaming tests were inconclusive. And since AMD disabled the iGPU cores, they werent able to test any games on the actual chip itself. AMD does not recommend this CPU to be paired up with gaming anyway. Its for desktop applications use only.
As you can see, the 2700x is roughly 4% faster than the PS5 CPU and the 2700 is roughly 5% slower than 2700x, so its a very good CPU to pair up with the 32 CU 10.6 tflops 6600xt or the 40 CU 13.1 tflops 6700xt which can be downclocked to get to 10.23 tflops. The 6700xt has a higher bandwidth which should bring it in line with the PS5 bandwidth so that might be just a bit more accurate, but i havent seen any 6600xt vs 6700xt benchmarks be hampered by its lower ram bandwidth.
yamaci17
Member
sure as hell ps5 outperforms 2700x in this game
my 2700x tops out at 55-65 FPS cpu bound with RT enabled, whereas PS5 can hit framerates upwards of 75+ easily with its unlocked VRR mode at dynamic 1440p. maybe decompression thing hammers the CPU, that could be a reason too, but in the end it is what it is, I would say a 3700x is a better match in the case of Spiderman ray tracing. (im specifically talking about swinging in open world. in door scenes or sticking up to a wall is another story where u get better CPU bound performance)
Last edited:
lukilladog
Member
Do you mean the HD textures in Far Cry 6?... some area in specific?. On my Pc it seems to jump between 7000 and 7500mb, it must be transferring textures all the time and your pcie port doesn´t cut it.
Mr Moose
Member
playsaves3
Banned
Is this actually true I thought amd gpus had the better bus
yeah cause pc is afforded that luxury but not console… totally equivalent though
Correct but that’s not the pcs fault. Just like we don’t handicap a console for having custom hardware.
lukilladog
Member
playsaves3
Banned
It’s not about handicap we are trying ro figure out hardware grunt in a same scenario situation
SlimySnake
Flashless at the Golden Globes
When you are doing pc gpu to gpu comparisons yes, but this DF fool doesnt realize that the zen 2 cpus in the consoles are heavily cutdown versions of the pc versions that also have to share previous memory bandwidth with the GPUs.
Using a 12 core 24 thread cpu running at 5.2 ghz in a comparison to console GPUs is fucking retarded. He should instead of handicapping the PC GPUs with shitty CPUs like the 2700 to see just where the console GPUs stand relative to pc GPUs. The guy really badly wants to push the narrative that the console GPUs are equivalent to 2060 and any time they outperform it he goes into panic mode.
Last edited:
Bulletbrain
Member
I think we all sometimes take all DF videos as "benchmarking" PS5 power Vs PC, DFs goal was different. It was about what kind of GPU and CPU would one need to get a "PS5 like" experience. That's a very different goal to pure benchmarking, because Spiderman isn't suitable for pure benchmarking.
A 2060s (not 2060) can as usual provide ballpark PS5 level experience on the GPU side in this game but with a caveat: the CPU has to be stronger than usual since Spiderman suffers from PC specific CPU side optimisation issues (related to RT, decompression etc). They illustrated this by pairing the 2060s with an overpowered CPU (thus bruteforcing through the CPU optimisation issues) and showing that unconstrained, the 2060s is enough to deliver a PS5 level experience. (Again, this was NOT a benchmarking exercise, just a analysis of the type of GPU needed to provide a PS5 like experience). Once that's done they show how a moderate CPU like the 3600 underperformes in the game. I think this is a rational way to test and illustrate their goal.
playsaves3
Banned
Eh they didn’t show that you can get a ps5 like experience on a visual front but not a performance front
I kind of buy the console can achieve more because you can code a little closer to a single console SKU vs dozens of PC configurations. Which is kind of the opposite of what you’re saying.
Bulletbrain
Member
That's cause the settings don't match. If nixxes included an "original" setting like what's available in Death Stranding or HZD, it would have been far easier. So they gave a ball park PS5 setting. In some cases those settings give better quality on PC ( better AF and better looking RT) but on others they are worse (crowd density etc). But it's close enough. Ballpark. And a 2060S CAN get those ballpark PS5 level quality, when paired with a stronger CPU to overcome this port's CPU optimisation issues. It's that simple.
Dampf
Member
Keep in mind PS5 uses lower settings than PC at max, which yoiu are probably using.
From DF. Try use them.
That is the case with nearly all games there is BTW. People compare PC performance at max settings to PS5 and wonder why the PS5 is so much faster, when the PS5 is actually running a combination between low, medium and high settings.
Last edited:
DenchDeckard
Moderated wildly
Would be really funny if nixxes do some extra work and manage to patch performance. I wonder if nxgamer would then revisit....
Last edited:
Dampf
Member
Yes I agree with the others here that this is likely a VRAM constraint.
Setting the 2070 to 4K resolution will completely overflow it's VRAM, even with DRS on. Remember, setting it to 4K also means the textures are using the correct mip level bias for 4K res, regardless if DRS enabled or not.
The PS5 should have around 7-9 GB available as video memory, and the game is apparently set to use 6.7 GB at maximum on 8 GB PC GPUs, so the reason why the PS5 performs much better in NXGamer's video is quite obvious.
NXGamer
Were you using the PS5 equivalent settings I mentioned above? Could you try using them again, but this time use low texture quality and report back your findings? Thank you!
Setting the 2070 to 4K resolution will completely overflow it's VRAM, even with DRS on. Remember, setting it to 4K also means the textures are using the correct mip level bias for 4K res, regardless if DRS enabled or not.
The PS5 should have around 7-9 GB available as video memory, and the game is apparently set to use 6.7 GB at maximum on 8 GB PC GPUs, so the reason why the PS5 performs much better in NXGamer's video is quite obvious.
NXGamer
Were you using the PS5 equivalent settings I mentioned above? Could you try using them again, but this time use low texture quality and report back your findings? Thank you!
Last edited:
tommib
Banned
Loxus
Member
I decided to re-watch Digital Foundry's Control Performance reviews and realized something.
In this video, we see the PS5's RT uncapped performance.
Most of their benchmarks show the PS5 sitting in the mid 40s, 45 fps on average.
In this article by Digital Foundry, we can see the PS5 console settings applied to the 6600XT. Which gives an average of 32 fps.
AMD Radeon RX 6600 XT review: ray-tracing performance
Control Benchmarks
If the Console settings were to be applied across the board. Will this give us a proper example of the PS5's GPU RT performance compared to other GPU's?
Console settings seems to give an 11fps improvement.
6600XT: 32fps
6700XT: 36fps
6800: 44fps
6800XT: 49fps
6900XT: 52fps
What's even more bizarre, this is the area used for the benchmarking by Digital Foundry.
And this is the same area on PS5.
There is one problem though.
The RT on PS5 is checkerboarded, while on PC it's full resolution and cannot be lowered.
So to be fair, I added 10fps in an attempt to match PS5's checkerboarded RT.
Console settings with checkerboarded RT?
6600XT: 42fps
6700XT: 46fps
PS5: 49fps
3060: 50fps
6800: 54fps
6800XT: 59fps
6900XT: 62fps
All this above is just a thought experience.
In my opinion with good optimisation.
PS5 RT off = 2070 Super - 2080 Super range.
PS5 RT on = 2060 - 2070 range.
One thing to note, the PS5 performance is equal to the sum of all it's part working together to punch above it's weight. This applies to the Xbox Series Consoles as well.
Last edited:
yamaci17
Member
no, i'm exactly using ps5 equivalent settings xdKeep in mind PS5 uses lower settings than PC at max, which yoiu are probably using.
From DF. Try use them.
That is the case with nearly all games there is BTW. People compare PC performance at max settings to PS5 and wonder why the PS5 is so much faster, when the PS5 is actually running a combination between low, medium and high settings.
yamaci17
Member
this has always been the case, and still is the same case. ps5 performs just a tad bit above 3060 in spiderman with RT enabled. since spiderman's RT implementation is rather lighter than most other games, PS5's rasterization can shine through. you can see this with AMD GPUs versus NVIDIA GPUs in light RT games such as Far Cry 6. Even in this game, rx 6700xt and above is highly competitive in terms of ray tracing against NV cards
what happens here that people misuses this event to their own arguments however is that 2070 being 2 times slower than PS5 in an abnormal case of VRAM bottleneck and people taking that fact as claiming PS5 is 1.5-2 times faster than 2070 even with RT.
individually, it is correct. at native 4k with very high textures, PS5 does indeed beat 2070. but it still does not change the fact that PS5's performance level is still near a 3060, which was practically a 2070 if not for VRAM bottlenecks
let's imagine we have 3 persons:
Person A can prepare 10 dishes per hour (PS5)
Person B can prepare 10 dishes per hour (2070)
Person C can prepare 10 dishes per hour (3060)
Person D can prepare 15 dishes per hour (3070)
Person E can prepare 15 dishes per hour (2080ti)
Person B (2070) and Person C (3060) are twins, but one of them buckles under stress, the Person C.
Up until that stress point, people who support Person A are constantly being reminded how they can only prepare as much as dishes as Person B and Person C.
Suddenly, Person B gets buckled under stress and starts to prepare 5 dishes per hour. It is astounding, everyone knew him as the person who did 10 dishes per hour.
Person D also drops from 15 dishes to 10 dishes, which puts him in a comical situation.
But in the same stressful situation, Person B and Person E just do their usual dishes per hour. Person A also keeps doing their exact same 10 dishes per hour. Person A's and Person C and E's relations stay exactly the same.
In the end: Person A is not "overperforming", neither Person C and Person E.
If Person A was preparing 15 dishes per hour like the Person E all of a sudden, you could say Person A did more work to get a better result. However that is not the case.
The entire case is Person B and Person D being heavly buckled under stress (huge vram bottleneck)
As I said countless other times; there are people in this very thread who keeps asking reviewers to use 2700x to do a proper match, yet now all of a sudden "but but 3070 is a package!!1 it is not sony's fault!1" Whenever I told tham that a high end rig is also a package that comes with a faster Zen 3 CPU, they retaliate saying it is unfair. Now when you ask them to fairly compare PS5 with a GPU that has enough VRAM budget, they refuse to acknowledge it.
yamaci17
Member
P
playsaves3
my final post to you, to make you understand real good this time. do note that i have a 2700x, the so called ps5 capable CPU
lets see. in this scene we supposedly see ps5 being %38 faster than 2070. therefore, NX gamer claims ps5 is almost performing like a 3070 here.
let's see. with low textures but matched ps5 settings (high particles, high RT, 10 rt object range and so on)
3070 renders a whopping 56 frames. this is the raw brute power 3070 has. it is %30 faster than ps5, which is what it should be.
if i use very high textures;
i lose %36 frames.
practically the entire advantage 3070 has over PS5.
now, what I'm trying to convey you is that if 3070 had 12 GB memory this would not happen. because, this does not happen to 2080ti, and it always performs %30 over PS5, in almost every case.
in that shot with 2070 and PS5, 2070 is handicapped by %30. if it did not have the huge VRAM bottleneck, it would perform almost similar to PS5, maybe only tailing %5 behind it, instead of tailing %38 behind it.
as i said, this is a VRAM memory bottleneck issue. it does not have to do with architectural differences. VRAM bound bottlenecks are known to cause severe framerate drops since 2005s. this is a wide knowledge that this is what happens when you run out of VRAM
this is clear as I can get.
both my 2700x and my 3070 is capable of pushing 56 framerate average there. if i had a 2080ti instead of a 3070, I would get 56-57 framerate average with VERY HIGH textures. but I cannot. I'm hugely constrained by VRAM. a very simple concept misused by NX gamer to make irrevelant foundations his comical deductions
When we compare PS5 to equivalent GPUs, we compare their raw strenght and power. but when that power is strangled by an other factor, ie VRAM, things change. 3070 is not performing like the usual 3070 in this case. if it had more VRAM, it would.
GPU has the grunt, it just cannot show it because the GPU stalls and waits for VRAM to do its job. its not a normal thing to happen, you never want this to happen. you either lower texture resolution or game resolution.
yes, in the end, 8 GB 2070 model cannot match PS5 experience at 4K. no one denies this. But it does not change the fact that PS5 is not overperforming. It is 8 GB GPUs underperforming instead, compared to PS5.
my final post to you, to make you understand real good this time. do note that i have a 2700x, the so called ps5 capable CPU
lets see. in this scene we supposedly see ps5 being %38 faster than 2070. therefore, NX gamer claims ps5 is almost performing like a 3070 here.
let's see. with low textures but matched ps5 settings (high particles, high RT, 10 rt object range and so on)
3070 renders a whopping 56 frames. this is the raw brute power 3070 has. it is %30 faster than ps5, which is what it should be.
if i use very high textures;
i lose %36 frames.
practically the entire advantage 3070 has over PS5.
now, what I'm trying to convey you is that if 3070 had 12 GB memory this would not happen. because, this does not happen to 2080ti, and it always performs %30 over PS5, in almost every case.
in that shot with 2070 and PS5, 2070 is handicapped by %30. if it did not have the huge VRAM bottleneck, it would perform almost similar to PS5, maybe only tailing %5 behind it, instead of tailing %38 behind it.
as i said, this is a VRAM memory bottleneck issue. it does not have to do with architectural differences. VRAM bound bottlenecks are known to cause severe framerate drops since 2005s. this is a wide knowledge that this is what happens when you run out of VRAM
this is clear as I can get.
both my 2700x and my 3070 is capable of pushing 56 framerate average there. if i had a 2080ti instead of a 3070, I would get 56-57 framerate average with VERY HIGH textures. but I cannot. I'm hugely constrained by VRAM. a very simple concept misused by NX gamer to make irrevelant foundations his comical deductions
When we compare PS5 to equivalent GPUs, we compare their raw strenght and power. but when that power is strangled by an other factor, ie VRAM, things change. 3070 is not performing like the usual 3070 in this case. if it had more VRAM, it would.
GPU has the grunt, it just cannot show it because the GPU stalls and waits for VRAM to do its job. its not a normal thing to happen, you never want this to happen. you either lower texture resolution or game resolution.
yes, in the end, 8 GB 2070 model cannot match PS5 experience at 4K. no one denies this. But it does not change the fact that PS5 is not overperforming. It is 8 GB GPUs underperforming instead, compared to PS5.
avin
Member
I'm playing on 12700k - 3080 - 32ddr4 and have to play at 1080p 50hz with their own upsampling method enabled (looks best to me) to eliminate framerate drops at max settings.
It doesn't matter what upsampling method i enable, the game still drops below 60 frequently at 60hz.
Still, i can't complain, the game looks and plays great at 50hz for a Bfi Oled addict like me
It doesn't matter what upsampling method i enable, the game still drops below 60 frequently at 60hz.
Still, i can't complain, the game looks and plays great at 50hz for a Bfi Oled addict like me
Last edited:
MightySquirrel
Banned
I decided to re-watch Digital Foundry's Control Performance reviews and realized something.
In this video, we see the PS5's RT uncapped performance.
Most of their benchmarks show the PS5 sitting in the mid 40s, 45 fps on average.
In this article by Digital Foundry, we can see the PS5 console settings applied to the 6600XT. Which gives an average of 32 fps.
AMD Radeon RX 6600 XT review: ray-tracing performance
Control Benchmarks
If the Console settings were to be applied across the board. Will this give us a proper example of the PS5's GPU RT performance compared to other GPU's?
Console settings seems to give an 11fps improvement.
6600XT: 32fps
6700XT: 36fps
6800: 44fps
6800XT: 49fps
6900XT: 52fps
What's even more bizarre, this is the area used for the benchmarking by Digital Foundry.
And this is the same area on PS5.
There is one problem though.
The RT on PS5 is checkerboarded, while on PC it's full resolution and cannot be lowered.
So to be fair, I added 10fps in an attempt to match PS5's checkerboarded RT.
Console settings with checkerboarded RT?
6600XT: 42fps
6700XT: 46fps
PS5: 49fps
3060: 50fps
6800: 54fps
6800XT: 59fps
6900XT: 62fps
All this above is just a thought experience.
In my opinion with good optimisation.
PS5 RT off = 2070 Super - 2080 Super range.
PS5 RT on = 2060 - 2070 range.
One thing to note, the PS5 performance is equal to the sum of all it's part working together to punch above it's weight. This applies to the Xbox Series Consoles as well.
Hmmm.. why are you posting PC Ultra RT benchmark vs PS5 RT knowing well that you can get well over 100 fps on PC at such low settings.
playsaves3
Banned
Bit off topic but it’s weird that Nx gamers framerate average is lower than elanalistabits for the ps5 vrr ID assume Nx gamers is more accurate thoughP playsaves3
my final post to you, to make you understand real good this time. do note that i have a 2700x, the so called ps5 capable CPU
lets see. in this scene we supposedly see ps5 being %38 faster than 2070. therefore, NX gamer claims ps5 is almost performing like a 3070 here.
let's see. with low textures but matched ps5 settings (high particles, high RT, 10 rt object range and so on)
3070 renders a whopping 56 frames. this is the raw brute power 3070 has. it is %30 faster than ps5, which is what it should be.
if i use very high textures;
i lose %36 frames.
practically the entire advantage 3070 has over PS5.
now, what I'm trying to convey you is that if 3070 had 12 GB memory this would not happen. because, this does not happen to 2080ti, and it always performs %30 over PS5, in almost every case.
in that shot with 2070 and PS5, 2070 is handicapped by %30. if it did not have the huge VRAM bottleneck, it would perform almost similar to PS5, maybe only tailing %5 behind it, instead of tailing %38 behind it.
as i said, this is a VRAM memory bottleneck issue. it does not have to do with architectural differences. VRAM bound bottlenecks are known to cause severe framerate drops since 2005s. this is a wide knowledge that this is what happens when you run out of VRAM
this is clear as I can get.
both my 2700x and my 3070 is capable of pushing 56 framerate average there. if i had a 2080ti instead of a 3070, I would get 56-57 framerate average with VERY HIGH textures. but I cannot. I'm hugely constrained by VRAM. a very simple concept misused by NX gamer to make irrevelant foundations his comical deductions
When we compare PS5 to equivalent GPUs, we compare their raw strenght and power. but when that power is strangled by an other factor, ie VRAM, things change. 3070 is not performing like the usual 3070 in this case. if it had more VRAM, it would.
GPU has the grunt, it just cannot show it because the GPU stalls and waits for VRAM to do its job. its not a normal thing to happen, you never want this to happen. you either lower texture resolution or game resolution.
yes, in the end, 8 GB 2070 model cannot match PS5 experience at 4K. no one denies this. But it does not change the fact that PS5 is not overperforming. It is 8 GB GPUs underperforming instead, compared to PS5.
playsaves3
Banned
Amd cards dont actually run the game that well either so the dudes statement holds up
Hmmm.. why are you posting PC Ultra RT benchmark vs PS5 RT knowing well that you can get well over 100 fps on PC at such low settings.
playsaves3
Banned
Nx gamer actually tested a lighter setting in his benchmark than what the ps5 uses since it’s a custom in between to drive his point even further (for example one if the settings in perf rt is between med and high so he used the medium setting on the pc so there are no excuses for it) and he usually does this in his settings benchmarks to show a bare minimum difference between pc and console. So it’s showing you didn’t watch the video and take whatever DF says at face value
Bulletbrain
Member
Nxg testing methodologies are extremely suspect as many have pointed out already (with examples). I'm not going to repeat them. But hey, you are free to believe what you want to believe.
playsaves3
Banned
No your original claim was that it was unfair to pc cause he was using a higher setting there than on console when it’s actually the opposite he alleviates any extra load for pc when there is discrepancies. In actuality if there was an exact setting for what the ps5 uses in that mode on pc the performance would be EVEN WORSE than what he showed on screen
Rubim
Member
The unfair side is more of it is relevant to his Graphics Card being severed limited by his CPU.
But that Mipmap issue happens evens on a 3080TI 24GB.
So is it a memory issue?
Is it a PCI Budget issue?
Whats actually going on?
playsaves3
Banned
I wasnt talking about mitmaps in my post I was sorely talking about graphical settings and a custom preset on console. About cpu limits that gpu is being limited by it just as much as the ps5 is by it’s cpu so it’s still equal in the end
Loxus
Member
Post like these are how you know the person didn't even bother to read what you posted.
Hmmm.. why are you posting PC Ultra RT benchmark vs PS5 RT knowing well that you can get well over 100 fps on PC at such low settings.
If you did bother to read everything, you would have understood I used PS5 console settings from Digital Foundry with has the 6600XT @ 32fps and the PS5 @ 49fps.
It can be that hard, all of us went to school correct?
Last edited: