It's a PS4 game at its heart which means it should scale with all types of builds and maybe some legacy(like PS4) PC parts.
With "fake resolution and FPS"
I recommend putting him on ignore as he's a rampant troll.
Plus it's got industry's fastest data streaming I/O.
that has nothing to do with ps5 getting close to 3070 or 3080. it has to do with 3070 and 3080 getting downgraded performance due to heavy VRAM bottlenecks in nixxes ports.
PS5 upscales to 4K though. This has a cost.
I don't believe it scales up to 4k, it is 1800p checkerboard. Which does have some cost, but it's nothing on the level of temporal upscalers.
Ff 16 seems more cpu than gpu limitedConsole version is usually 1440p 30fps (can drop to 1080p at times and below 30 even), so you need at least 2x the GPU for 4K (4070S) and another 2x for 60FPS.
So ~4x the power of PS5, 4090 won't be able to do it without upscaling:
It's not, really (the first game that will allow for the comparison). First of all, we're talking about different iterations of Decima. But beyond that, when it comes to reconstruction, Death Stranding on PS5 may use some form of it, but it's not CB at least according to John from DF.
www.eurogamer.net
He's mocking DLSS upscaling and frame generation, yet if/when similar technology come to the PS5 pro, watch him change his tune on them completely.
If a technique is done to simulate a resolution or add fps it's not true/native and is fake.
That's almost every game on PS5? FW and ZD both use checkerboard rendering. FF XVI uses FSR and the list goes on.
I didn't put it down or downplay it.
............
I don't think it's just vram pressure though it certainly plays a big part always appreciate the informative posts you could easily take a negative tone but your always respectful unlike others on this site thank youthat has nothing to do with ps5 getting close to 3070 or 3080. it has to do with 3070 and 3080 getting downgraded performance due to heavy VRAM bottlenecks in nixxes ports.
it is like claiming 4060ti magically performs like 4070
what you say, on paper, would be true, but technically untrue. as you can see here due to extreme vram pressure, 4070 and 4060ti looks like performing the same.
yet when you remove the vram limitation out of the equation, you get performance like expected
what part of this is so hard to understand that I had to repeat this for at least 10 times now? how much can I be clear? I don't understand. I even give examples between GPUs and showing actual outlier with actual example in the actual game you've been talking about.
PS5 is not getting magical sauce better GPU performance than a 3070 or 3080. if you believe that, then you would have to believe 4060ti gets that over 4070 too, which is nonsensical. the way their engine handles thing lower your GPU bound performance greatly if you're VRAM bound, which is where 3070 and 3080 are at (7 and 9 GB DXGI budget respectively)
if 16 gb 4060ti performs %30 faster than ps5 in these games, your logic falls on its face. because it is generally accepted that 4060ti=3070 in %99 of the cases. and by the virtue of their actual GPU chip, it is true.
as I said, you're confusing things.
It's not an exact comparison. The DRS on PS5 is more aggressive than on PC. Nixxes told Alex that on PS5, the DRS has certain levels it immediately drops to when it cannot maintain the target frame rate. For instance, if the target frame rate is 60fps and it's at 1440p but drops to 57fps, it'll immediately lower to the next level which might be 1080p. On PC, it's more granular and will always attempt to maintain the highest resolution even if the target frame rate isn't met.can you give me any sensible explanation why 4060ti 16 gb performs 1.5x faster than the 8 gb version?
NVIDIA GeForce RTX 4060 Ti 8 GB Specs
NVIDIA AD106, 2535 MHz, 4352 Cores, 136 TMUs, 48 ROPs, 8192 MB GDDR6, 2250 MHz, 128 bitwww.techpowerup.com
NVIDIA GeForce RTX 4060 Ti 16 GB Specs
NVIDIA AD106, 2535 MHz, 4352 Cores, 136 TMUs, 48 ROPs, 16384 MB GDDR6, 2250 MHz, 128 bitwww.techpowerup.com
exact same GPU chip. same GPU core count, same frequency, same bus, same bandwidth, same cache, same TFLOPS, same architecture, literally the same GPU
if you are not going to make any efforts to understand how VRAM bottleneck plays into the performance in ratchet and spiderman: just say so and I will not bother you anymore. because this is getting real tiring.
High mostly with very high textures and maybe draw distance
Lmao it's a cross-gen game from a couple years back mate, sort your bonce out.This is the game that will bring the 4080 and 4090 cards to their knees.
No, I quoted you correctly. You were addressing some of the videos playsave3 posted and some of them were from Nxgamer so assuming you watched those, I was just telling you that his comparisons aren't even apples to apples.
it is purely vram pressure. these are the exact same GPU chips.
Looking at the requirements it doesn't seem either a 4080 or a 4090 will be brought to their knees. Unless over 4k60 counts as that.This is the game that will bring the 4080 and 4090 cards to their knees. Time to give yall that Sony port special since want that "greatness" in ya face so bad.
Just make sure you let those shaders compile first before you lose your marbles over how good it looks. I know Alan Wake 2 look good, but once this game gets moded with path tracing.....Cyberpunk who?
I get you now. I did watch those videos way before so I know their contents and I know they're not apples to apples. but even if they were apples to apples, sadly, due to how NIXXES handles their ports, you would still get weird performance penalties under VRAM limitation (which is a topic i'm trying to convey across to playsave3, otherwise he will keep claim ps5 is approaching to 3080 and on its way to conquer 3090 or something lol). how can ps5 come close to 3080 all the while still being slower than a 16 gb 4060ti.
The reason I don't think it's just vram pressure is there are some scenes in the game where 16gb+ cards still fall below 50 though not as much It should be a mix of issues depending on sceneit is purely vram pressure. these are the exact same GPU chips.
in old times, when you run out of vran, in most cases, game would tank to single digits. nowadays you either get reduced but playable performance by a specific percentage or you get less quality textures (depending on the engine, severity of this will change)
problem with you is that you take that "reduced but playable" performance as the actual performance of that chip and compare it against PS5 and come to wrong conclusions.
now tell me, 4 gb 6500xt drops to 44 fps but the exact same chip with 8 GB vram renders 64 fps in same scene. will you claim that 6500xt chip is only capable of 44 FPS and treat it as such, when you know that when it is not under vram pressure it performs much better?
Would love for them to add super textures especially cause it can get backported to the pro
these are maxed out ultra settings on top of maxed out ray traced settings that also includes ray traced shadows and ambient occlusion that are not present on PS5.... and even then those options are also maxed out (not high, but very high is used)
Not real means?
Yeah, and this game is weird because at least a few weeks after release, AF hit VRAM really hard. Usually, AF 16x on PC is basically free but in this game, it tanked the performance of cards with less than 12GB of VRAM. Even the 2080 Ti could struggle with Very High Textures+16x AF. The recommendation was to reduce the AF to 8x or textures to High. I can't recall the last time when anyone with a card released in the past decade had to forego AF.
No one thinks DLSS is native. DLAA on the other hand...
it is not really fixed, it still heavily relies on PCI-e even to this day (source, i played it 2 weeks ago on PC). it always used upwards of 2.5-3 GB shared vram with very high textures. high textures bring huge performance improvements but they are very inconsistent in quality
Cyberpunk has pathtracing lolSo is Cyberpunk that he compared.
Oops It's even older than cross gen as it was in shown off during the 360/PS3 generation.
Agreed. Personally not a big fan of the hard-on some people have for Nixxes. I think it's nice that their games scale well above PS5 in terms of visuals (Spider-Man and Rift Apart can look quite a bit better with RT) but their performance profiles leave a lot to be desired, especially for people on older systems or lower-specced machines.
eh even then
Imagine if it wasn't with its DLC putting out the best 2023 visuals!
The boss fight alone outshines anything released in 2023.
