• 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] Dragon's Dogma 2 - Patch 2.03 Is this really a PS5 Pro focused update?

ChiefDada

Gold Member


With a new console incoming next month in the PlayStation 5 Pro, Capcom have ensured they have something to show off on it. Earlier this week they dropped s new patch, complete with new menu names for Dragons Dogma 2 graphics modes on Consoles. I was keen to test it out across PS5 and PC and compare the visual quality and performance to the PS5 Pro version. What does this patch change over the names? how well does it run on PC? and is the PS5 Pro already delivering above expectations.


As a full time Software/Hardware Engineer for the past 35+ years I run this channel as a side project to my full time job. As such I have and remain fully independent and always sim to deliver concise, accurate, factual review, news and data to inform you as much as I can. Aside my supporters on Patreon and the decreasing revenue from YouTube I also fund everything myself.


As such, if you do enjoy the content on the channel and site, then please consider Subscribing, Liking & sharing it where possible as all these things help increase my view, reach and play into the YouTube Algorithm, thank you.

TLDR: PS5 Pro has better IQ and 33% performance edge than "similarly spec PC" (6800/5600x) in CPU bound areas
 
Last edited:

Bojji

Member
Really a what? Don't keep me in suspense.

raven-anxious.gif
 

Dorfdad

Gold Member
I like the 60% better than base ps5 line myself. I am excited but all you skeptics who think you know but don't keep hating on real results.

You look at specs alone which doesn’t tell you the full story anymore. You have to take the entire pipeline and development api into consideration.

I believe the ps5 Pro will eventually maintain and even larger set of improvements as people really get to understand how to work better.
 
So he says it has better IQ while providing 33% better performance....in CPU bound areas...

...so take the graphics off interlaced mode and improve the IQ.... The GPU is sitting around waiting :rolleyes:


And all he's going on is Sony's b-roll which for all we know may already be on a different build from this patch that just released.

What a useless video... and not to mention like I said... PS5 Pro with PSSR definitely has lower settings. He's not getting anything to match up.
 
Last edited:
I like the 60% better than base ps5 line myself. I am excited but all you skeptics who think you know but don't keep hating on real results.

You look at specs alone which doesn’t tell you the full story anymore. You have to take the entire pipeline and development api into consideration.

I believe the ps5 Pro will eventually maintain and even larger set of improvements as people really get to understand how to work better.

That was exactly my point...

Some experts here have been saying that the CPU is too weak in CPU bound scenarios....

LOL

It's going to be debunked badly
 

Dorfdad

Gold Member
That was exactly my point...

Some experts here have been saying that the CPU is too weak in CPU bound scenarios....

LOL

It's going to be debunked badly
yeah wasn't on you sorry if it came off like that, just that we get posters saying this and that comparing stuff. It is not apples to apples anymore, you can get an IDEA sure but proof is in the developers hands. Some developers will do a shit job and we will say PSSR sucks while others will perform magic with it.
 

Bojji

Member
That was exactly my point...

Some experts here have been saying that the CPU is too weak in CPU bound scenarios....

LOL

It's going to be debunked badly

CPU limited games will still be CPU limited on Pro, just 10% less. This is how our physical world works, both consoles have the same CPU.

Unless they are coded differently, for example PS5 version may be calculating BVH on CPU while Pro is doing that on GPU, it's possible but we don't have any proof to something like that so far.
 
Last edited:

SweetTooth

Gold Member
That was exactly my point...

Some experts here have been saying that the CPU is too weak in CPU bound scenarios....

LOL

It's going to be debunked badly

Well, some of the experts here never left their mom's basement! its truly on you taking their word vs somone who actually knows what he is doing (Cerny).

Always bet on the Cerny
 
Last edited:




TLDR: PS5 Pro has better IQ and 33% performance edge than "similarly spec PC" (6800/5600x) in CPU bound areas

It's only the beginning. Just be ready for the PS5 Pro to regularly compete against 4070 and 4070 super. And this is what happen when you dont use a 4090 combined with a $1000 CPU to create a CPU narrative.

Good reminder that DF carefully didn't compare performance with their Ratchet and Clank DLSS vs PSSR analysis and "They need to see more PSSR titles" as in "we need to compare performance using the worst cases, this game runs too good on PS5 Pro".
 
Last edited:

FireFly

Member
Good reminder that DF carefully didn't compare performance with their Ratchet and Clank DLSS vs PSSR analysis and "They need to see more PSSR titles" as in "we need to compare performance using the worst cases, this game runs too good on PS5 Pro".
How do you "compare performance" from a mode that is locked to 60 FPS?
 

ap_puff

Member
CPU limited games will still be CPU limited on Pro, just 10% less. This is how our physical world works, both consoles have the same CPU.

Unless they are coded differently, for example PS5 version may be calculating BVH on CPU while Pro is doing that on GPU, it's possible but we don't have any proof to something like that so far.
It really depends what's causing a cpu bind. 10% higher mhz doesn't translate into 10% more performance. For example if they didn't improve memory latency then a cpu bottleneck caused by cache misses might have a 0% performance improvement, and a cpu bottleneck caused by RT instructions might have a much larger improvement than 10% if there's separate hardware handling stuff like bvh construction. We really don't know how things will turn out in every game, just like how the larger cache on X3D chips don't give a 15% improvement in every game... Some games get very little improvement and some gain like 100%
 

Bojji

Member
It really depends what's causing a cpu bind. 10% higher mhz doesn't translate into 10% more performance. For example if they didn't improve memory latency then a cpu bottleneck caused by cache misses might have a 0% performance improvement, and a cpu bottleneck caused by RT instructions might have a much larger improvement than 10% if there's separate hardware handling stuff like bvh construction. We really don't know how things will turn out in every game, just like how the larger cache on X3D chips don't give a 15% improvement in every game... Some games get very little improvement and some gain like 100%

Yeah that 10% was in my post as theoretical maximum, Zen 2 scaled very poorly with CPU clock.
 
Last edited:

Gaiff

SBI’s Resident Gaslighter
CPU limited games will still be CPU limited on Pro, just 10% less. This is how our physical world works, both consoles have the same CPU.

Unless they are coded differently, for example PS5 version may be calculating BVH on CPU while Pro is doing that on GPU, it's possible but we don't have any proof to something like that so far.
Well, there is the bandwidth issue, isn't there? If the PS5 was bandwidth-bound on the CPU, that 28% extra bandwidth on the Pro might help more than just 10%. It could be 10%+whatever extra amount the higher bandwidth provides. Could it be what we're looking at here?
 

ap_puff

Member
Well, there is the bandwidth issue, isn't there? If the PS5 was bandwidth-bound on the CPU, that 28% extra bandwidth on the Pro might help more than just 10%. It could be 10%+whatever extra amount the higher bandwidth provides. Could it be what we're looking at here?
Yeah, zen2 scaled a lot with memory bandwidth so having 28% more bandwidth could be a huge buff.
 

Bojji

Member
Well, there is the bandwidth issue, isn't there? If the PS5 was bandwidth-bound on the CPU, that 28% extra bandwidth on the Pro might help more than just 10%. It could be 10%+whatever extra amount the higher bandwidth provides. Could it be what we're looking at here?

Yeah, zen2 scaled a lot with memory bandwidth so having 28% more bandwidth could be a huge buff.

Do we know how sliced GDDR6 bandwidth is between CPU and GPU? What are the limits? Or there aren't any?

RAM overclocking on CPU was also about lowering latency and GDDR6 has much higher latency than DDR4. Z2 was also limited by infinity fabric speed to ~3600MHz DDR4 speed, how that translates to much faster GDDR6? I have no idea.
 
Last edited:

DavidGzz

Member
This patch made the PC version better too. Raytracing, 4k, 90fps, and I have 30+ mods installed including custom difficulty that is a god send for this game. I hope the console version gets more QoL changes because subsequent playthroughs are a chore without them.
 

Three

Member
Well, there is the bandwidth issue, isn't there? If the PS5 was bandwidth-bound on the CPU, that 28% extra bandwidth on the Pro might help more than just 10%. It could be 10%+whatever extra amount the higher bandwidth provides. Could it be what we're looking at here?
I think optimisation and performance profiles are a lot more involved on console than people think. People tend to think of it too much like a PC and get too hung up on single components and its bottlenecks when it is a far more delicate balance to frametime on console. Cerny did a talk about optimising to power consumption/power budgets on console on base PS5.

boost.png
 
Last edited:
Top Bottom