• 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] Death Stranding Director's Cut: PC vs PS5 Graphics Breakdown

Arioco

Member
Jesus fucking Christ.

:messenger_beaming:


cTz4fgG.jpg
 
Alex is well known for being a Sony pony. Don't even act like you didn't know.
I think Alex calls it like he see's it, sometimes that lands on the MSFT side and sometimes Sony's. He's also human so I don't think he intentionally misleads anyone, he just makes mistakes and maybe forgets to include things as I'd wager making videos of such length requiring a ton of metrics and information can make it easy to forget certain talking points. From my perspective, he tends to kind of land in the middle and just call a good looking spade a spade.

Some of you become so entrenched in the Silicon Wars you guys can't see the forest for the trees.
 
Last edited:

Zathalus

Member
did i miss the point alex was trying to make with this?

why don't introduce Ampere to the mix?
->

stranding2hkjb.png
Ampere results are difficult to ascertain as while the TFLOP figures shot up (due to the doubling of the ALU) that is still hamstrung by various other factors in the GPU that didn't get doubled.

So while a 3090 is 38 TFLOP is performs roughly the same as a 26 TFLOP 6900xt.

The best thing to use for Ampere is to use 65% of the TFLOP figures to get equivalent Turing/RDNA 2 FLOPS.
 
Ampere results are difficult to ascertain as while the TFLOP figures shot up (due to the doubling of the ALU) that is still hamstrung by various other factors in the GPU that didn't get doubled.

So while a 3090 is 38 TFLOP is performs roughly the same as a 26 TFLOP 6900xt.

The best thing to use for Ampere is to use 65% of the TFLOP figures to get equivalent Turing/RDNA 2 FLOPS.

sure, everybody with the smallest amount of interest in GPUs knows this.
the best thing to do is not to compare flops between different architectures....
 

MikeM

Member
This motherfucker really reading his shit from a piece of paper?
I love his content but he missed big time with this. He doesn’t do console at all and you can tell. I’m sure if he did this with more recent games vs PC that he would have a different result. DMC is a terrible port to benchmark from.
 

PaintTinJr

Member
cold boot into a save file in this location, 10900k loads the game in 13.72 seconds

DBWrXRj.jpg


2700x COLD BOOTS into the very same location in 14 or so seconds (compared to the 10900k's 13.72 sec cold boot time)



soft boot takes 3-5 secs



this is a cpu that is literally %60-70 slower than 10th gen intel CPUs

people really go extremes to downplay PCs lmao

That info suggests that the loading is bottleneck by the RAM copying and RAM to VRAM copying on PC cold boot - whereas on subsequent loads it is already decompressed and held in an SSD's overprovisioning cache and DMAs either direct to the VRAM, or passes thru the RAM very quickly. unlike the PS5 that has unified RAM and avoids some of the copying on cold boot.

The difference in the CPUs is then reduced to their RAM interface.
 

winjer

Gold Member
cold boot into a save file in this location, 10900k loads the game in 13.72 seconds

DBWrXRj.jpg


2700x COLD BOOTS into the very same location in 14 or so seconds (compared to the 10900k's 13.72 sec cold boot time)



soft boot takes 3-5 secs



this is a cpu that is literally %60-70 slower than 10th gen intel CPUs

people really go extremes to downplay PCs lmao


Death Stranding is DX12 game, so it's probable that it's compiling shaders when it's loading each level.
So the first time will take longer to load assets and compile shaders. The second time, shaders are already compiled, so it just has to load assets.
 

Corndog

Banned
Pc comparisons are fun and all, but it is the performance of a closed system that gets the geek juice flowing. Having that hard baseline limit for comparison amongst different games is what allows me to appreciate tech achievement. How much can be extracted, how far can things be pushed.

A pc is about as interesting as duct taping a rocket onto a wheelbarrow and proclaiming to have the fastest car in the world, slow clap but yeah. You feel me?
So you are saying your computer sucks. Sorry to hear.
 
Last edited:

tygertrip

Member
It still amazes me how stout gen 9 consoles are, the gap was MASSIVE in gen 7 and gen 8, now, not so much. Obviously, someone is going to chime in with a (Insert whatever argument about the RTX 3090 demolishing Gen9 consoles).
I have a Ryzen 5800x + 3070, and I am THOROUGHLY impressed with my PS5. Not just the graphics, but the haptics and the 3D sound (I had the Aureal A3D card back in the day… I can’t believe it took TWENTY YEARS to regain true 3D sound in a non-VR game). I honestly thought the days of new console gens being impressive were long gone. I’m VERY happy to be proven wrong! Honestly, I don’t think I’ve been this impressed with a console since the Dreamcast.
 

tygertrip

Member
Pc comparisons are fun and all, but it is the performance of a closed system that gets the geek juice flowing. Having that hard baseline limit for comparison amongst different games is what allows me to appreciate tech achievement. How much can be extracted, how far can things be pushed.

A pc is about as interesting as duct taping a rocket onto a wheelbarrow and proclaiming to have the fastest car in the world, slow clap but yeah. You feel me?
Cope. Also, that sounds like one kick-ass wheelbarrow!
 

Tchu-Espresso

likes mayo on everthing and can't dance
This is on sale on the ps store atm.

Not sure if I have energy for another open world game right now but I’m interested.
 
Top Bottom