• 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.

The Callisto Protocol - The DF Tech Review + PS5 vs Xbox Series X/ S + PC #StutterStruggle Analysis

DeepEnigma

Gold Member
I kinda hope this game will be a huge commercial failure tbh.
e26.jpg
 

I Master l

Banned
We have a game here that is implementing 3 different types of RT applications. It is interesting from a tech perspective since it is heavily using a feature that is distinctly current gen. Consider how many ill informed (including DF) believed that peak teraflop and memory bandwidth spec was/are the deciding factor for RT performance, then you can understand why this is an interesting discussion to be had.

Game on PC have like 50% GPU utilisation when RT is on, Its obvious there is
something off about RT implementation here
 
Last edited:

Thief1987

Member
Game on PC have like 50% GPU utilisation when RT is on, Its obvious there is
something off about RT implementation here
Should be CPU bottleneck in case of PC. RT could be heavy on CPU, and this game also quite bad at multithreading on top of it
 
Last edited:

ChiefDada

Gold Member
Diffuse reflections, as mentioned by DF, are present on "stone, rough metals"

And you can see the example during the video that the refflection is not defined, so it being high/low resolution has no impact to the visuals




There are tons of reflective materials in the game, but Jacob's reflection on them is just an orange blob. So what's the point in discussing which console has the most defined orange blob?

But I guess I'm trying to find reason in children console warring


Because we are in a tech and platform comparison thread. Again, this game serves as a great education for those who thought theoretical GPU power determines actual performance. Those who are really interested in learning will take this info and move forward with this understanding. I'm not sure if you're suggesting that I'm console warring with your quote. If so, please point out where specifically.
 

adamsapple

Or is it just one of Phil's balls in my throat?
Because we are in a tech and platform comparison thread. Again, this game serves as a great education for those who thought theoretical GPU power determines actual performance. Those who are really interested in learning will take this info and move forward with this understanding. I'm not sure if you're suggesting that I'm console warring with your quote. If so, please point out where specifically.

A game that came in hot and needed multiple patches within 1 week of release should not be considered a great education or example for anything other than developers time and optimization.

🤷‍♂️
 
Last edited:

DenchDeckard

Moderated wildly
Because we are in a tech and platform comparison thread. Again, this game serves as a great education for those who thought theoretical GPU power determines actual performance. Those who are really interested in learning will take this info and move forward with this understanding. I'm not sure if you're suggesting that I'm console warring with your quote. If so, please point out where specifically.

Dude, you can not use this game as a measuring pole for anything...that much is obvious.

This game was made by a new studio that was obviously spread thin developing across 5 platforms and had to prioritise development based on that. The game launched in a complete mess where even the PC version has terrible reviews on steam. Its a mess of a release. Far from what you are Saying.

You want to see a competent fully manned studio that can handle a real multiplatform development environment. Check the fortnite unreal engine 5.1 DF video.
 
Last edited:

rofif

Can’t Git Gud
it's not about the game, it's about how it was launched and developed.

it's very clear that basically all the versions except the PS5 version were severely undercooked and thrown out the door in unfinished states.

and that shouldn't be supported 🤷
It kinda should be about a game. It really is quite bad :p
 

ChiefDada

Gold Member
Game on PC have like 50% GPU utilisation when RT is on, Its obvious there is
something off about RT implementation here

Memory bottleneck perhaps? If not VRAM also consider memory bandwidth contention. DF laughed at the potential of PS5 cache coherency architecture. I now laugh at their recurring befuddlement of how/why ps5 continuously performs above their expectations.
 

ChiefDada

Gold Member
Dude, you can not use this game as a measuring pole for anything...that much is obvious.

I use all games that incorporate next gen features as a benchmark for comparisons between current gen console and high end PC. Again, if you can't see why this game in particular with it's multiple RT applications is an above average candidate for benchmarking, don't know what to tell you.
 

Hugare

Member
Because we are in a tech and platform comparison thread. Again, this game serves as a great education for those who thought theoretical GPU power determines actual performance. Those who are really interested in learning will take this info and move forward with this understanding. I'm not sure if you're suggesting that I'm console warring with your quote. If so, please point out where specifically.
Wasnt talking about you, dude. Either on my original post that you replied then or just now (I think. Or do you fit the description? Havent followed your posts specifically)

Anyway, sure, you can talk about tech. But I think it isnt that interesting in this case because:

1 - People are making a lot of assumptions about the devs prioritizing PS5 (going as far as assuming that devs made the Xbox version worse) due to Striking Distance using Sony's mocap studio. Stuff that we will never know for sure.

2 - Xbox versions were obviously not optimized as much as the PS5 version. So making assumptions about the difference in hardware/performance between consoles on an unoptimized game is idiotic, imo.

3 - The final results arent that big (after patches). Not a Bayonetta situation where we had 30 FPS on PS3 and 60 on X360. Right now it runs almost as good as PS5 on the Series X, with worse looking RT reflections that 98% of the players will barely notice after the mirror present on the tutorial area.

So, thats my point of view. But if you guys want to keep discussing, keep at it.
 

adamsapple

Or is it just one of Phil's balls in my throat?
So, thats my point of view. But if you guys want to keep discussing, keep at it.

Like I said, a game that needed multiple patches within days of release is not a good indicator of a game that's properly utilizing all the next gen features regardless of platform.

The Fortnite UE5 update is a much better thing to look at as DenchDeckard DenchDeckard mentioned, as it not only utilizes all the next-gen UE5 features, but runs extremely well on each console, so theoretically it's doing a better job maximizing each consoles potential.
 

Hugare

Member
Like I said, a game that needed multiple patches within days of release is not a good indicator of a game that's properly utilizing all the next gen features regardless of platform.

The Fortnite UE5 update is a much better thing to look at as DenchDeckard DenchDeckard mentioned, as it not only utilizes all the next-gen UE5 features, but runs extremely well on each console, so theoretically it's doing a better job maximizing each consoles potential.
Exactly my point

I think its worth discussing the tech in Callisto, that's awesome.

But performance? Not so much
 

solidus12

Member
It doesn't. As Control, Metro Exodus, RE 2, RE 3, RE7, RE8, Doom Eternal, DMC V and Fortnite all prove. All of which either run better on the Series X or have higher resolution.

I love how people pick the exception to the rule and think it is actually the rule.
Xbox has yet to have an exclusive with a ray-tracing implementation as good as Ratchet and Clank or Spider-Man.

Btw what happened to the Minecraft raytracing update?
 

ChiefDada

Gold Member
Like I said, a game that needed multiple patches within days of release is not a good indicator of a game that's properly utilizing all the next gen features regardless of platform.

Your logic here is mindboggling. RT is a current gen feature, period.

You were expecting their "fix" to make Series X RT at least on par with PS5 and since that didn't pan out, you now claim that this game can't be used as a basis for tech comparisons between platforms. Your bias couldn't be more apparent.
 

DenchDeckard

Moderated wildly
I use all games that incorporate next gen features as a benchmark for comparisons between current gen console and high end PC. Again, if you can't see why this game in particular with it's multiple RT applications is an above average candidate for benchmarking, don't know what to tell you.

Really, lmao.

You're officially high if you think this game is any kind of benchmark. Infact the only benchmark is how not to release a game.

The only thing you need to tell me is you smoke crack.
 
Last edited:

adamsapple

Or is it just one of Phil's balls in my throat?
Your logic here is mindboggling. RT is a current gen feature, period.

You were expecting their "fix" to make Series X RT at least on par with PS5 and since that didn't pan out, you now claim that this game can't be used as a basis for tech comparisons between platforms. Your bias couldn't be more apparent.

There doesn't need to be any bias to see a clear mismatch in development when one of the console versions didn't even have RT at launch lol.

You are free to use this one example where your preferred plastic box is doing better RT while ignoring the half a dozen others before that have different results, you're also free to use a game that came in hot as a benchmark of either consoles max capabilities.

It's short-sighted, but you are free to do what you want.
 
Last edited:

S0ULZB0URNE

Member
A game that came in hot and needed multiple patches within 1 week of release should not be considered a great education or example for anything other than developers time and optimization.

🤷‍♂️
Or their is a issue with the hardware's abilities or the tools(that I was told came)
With the below comparisons being after patches.
tNCLDcD.jpg

sR6IPOZ.jpg
 

adamsapple

Or is it just one of Phil's balls in my throat?
Or their is a issue with the hardware's abilities or the tools(that I was told came)
With the below comparisons being after patches.
tNCLDcD.jpg

sR6IPOZ.jpg


Aye man you're very welcome to use a game that released with missing features and is getting one patch every 2nd day on average since release as an example of "hardware ability" all you want :messenger_tears_of_joy:

It'd be dumb but it's your prerogative.

Or you can go check out the Fortnite comparison, a UE5 release with all the next gen features that launched feature and parity complete on all the new gen consoles.
 
Last edited:

S0ULZB0URNE

Member
Aye man you're very welcome to use a game that released with missing features and is getting one patch every 2nd day on average since release as an example of "hardware ability" all you want :messenger_tears_of_joy:

It'd be dumb but it's your prerogative.

Or you can go check out the Fortnite comparison, a UE5 release with all the next gen features that launched feature and parity complete on all the new gen consoles.
I absolutely will use it as will anyone without bias and who isn't in denial.

Patches and still coming up short sounds like trying to get blood from a stone.

I don't need to talk about other games in this games tech thread.
 

adamsapple

Or is it just one of Phil's balls in my throat?
I absolutely will use it as will anyone without bias and who isn't in denial.

Patches and still coming up short sounds like trying to get blood from a stone.

I don't need to talk about other games in this games tech thread.


Callisto Protocol the game works better on PS5 compared to Series X right now, anyone can see that. Just like anyone can see how when one version was released with missing features and is still being patched to add those features in is not indicative of that consoles potential.

I totally understand why you're latching onto this result as the end-all be-all, but please don't call yourself as someone without a bias, that's hilarious :messenger_tears_of_joy:
 
Last edited:

S0ULZB0URNE

Member
Callisto Protocol the game works better on PS5 compared to Series X right now, anyone can see that. Just like anyone can see how when one version was released with missing features and is still being patched to add those features in is not indicative of that consoles potential.

I totally understand why you're latching onto this result as the end-all be-all, but please don't call yourself as someone without a bias, that's hilarious :messenger_tears_of_joy:
It's been patched and still looks vastly worse.
Sounds like trying to get blood from a stone to me.

What's hilarious is the denial and excuses.
 

adamsapple

Or is it just one of Phil's balls in my throat?
It's been patched and still looks vastly worse.

"VASTLY" worse .. oh man :messenger_tears_of_joy:

"Been patched". The official account for the game is saying they're working daily on fixes and optimization.



-

I agree with you on the denial and excuses part for thinking that this game in its current state is some kind of final benchmark for either consoles capabilities :messenger_tears_of_joy:
 
Last edited:

S0ULZB0URNE

Member
"VASTLY" worse .. oh man :messenger_tears_of_joy:

"Been patched". The official account for the game is saying they're working daily on fixes and optimization.



-

I agree with you on the denial and excuses part for thinking that this game in its current state is some kind of final benchmark for either consoles capabilities :messenger_tears_of_joy:

Yes vastly.
You didn't actually look at the pictures I posted did you.

Constantly tweaking to try and get things closer because of the backlash from a certain group.

I will guess if they get it closer than the vast difference that currently exists that it will be at a cost.
My guess for the cost will be a reduction in resolution or performance with some added screen tearing in the mix.

I wonder if DF will remember to mention these trade offs when/if it gets improved and they do a follow up video....
 
A game that came in hot and needed multiple patches within 1 week of release should not be considered a great education or example for anything other than developers time and optimization.

🤷‍♂️

Yea, not sure what he was trying to get across there. Proper optimization done the game runs better on the more powerful console. This is one of the main reasons I can't wait for Sampler Feedback Streaming to become the norm in updated game engines. There will no more instances of the Series X GPU's running into situations where it ends up using the slower 3.5GB of RAM that runs @ 336GB/s. With Sampler Feedback the entire game will be able to fit exclusively inside 10GB with RAM to spare.

It's the primary reason for why we ever see strange things like this.
 

adamsapple

Or is it just one of Phil's balls in my throat?
Yea, not sure what he was trying to get across there. Proper optimization done the game runs better on the more powerful console. This is one of the main reasons I can't wait for Sampler Feedback Streaming to become the norm in updated game engines. There will no more instances of the Series X GPU's running into situations where it ends up using the slower 3.5GB of RAM that runs @ 336GB/s. With Sampler Feedback the entire game will be able to fit exclusively inside 10GB with RAM to spare.

It's the primary reason for why we ever see strange things like this.

It's not even sampler feedback, the game is very clearly unpolished on 2 of the 3 platforms it's on (not even counting last gen), and being patched almost daily, outside of comparing in a vacuum only an idiot would use this game as a definitive look into what either console can ideally do.
 
Last edited:
It's not even sampler feedback, the game is very clearly unpolished on 2 of the 3 platforms it's on (not even counting last gen), and being patched almost daily, outside of comparing in a vacuum only an idiot would use this game as a definitive look into what either console can ideally do.

Oh no, I know it's unpolished and not properly optimized, but I will admit that the PS5's straightforward unified 16GB of RAM with 448GB/s of memory bandwidth is easier to build games for.

The 10GB @ 560GB/s 6GB @ 336GB/s split for the Series X will introduce issues for games where the data isn't properly organized a certain way in memory. It adds an extra layer of complexity to Series X development. It can be worked around, sure, but if Sampler Feedback Streaming were being used it wouldn't have to be worked around. The faster 10GB of RAM could literally behave like an effective 20GB+ of RAM due to all the memory that would be saved from textures thanks to to Sampler Feedback Streaming.

That's what Sampler Feedback is capable of, but the engines need to be designed around it. I don't believe it's some experimental thing either. I legit believe it does exactly as advertised.
 

I Master l

Banned
Yea, not sure what he was trying to get across there. Proper optimization done the game runs better on the more powerful console. This is one of the main reasons I can't wait for Sampler Feedback Streaming to become the norm in updated game engines. There will no more instances of the Series X GPU's running into situations where it ends up using the slower 3.5GB of RAM that runs @ 336GB/s. With Sampler Feedback the entire game will be able to fit exclusively inside 10GB with RAM to spare.

It's the primary reason for why we ever see strange things like this.

Is it possible to lock the bandwidth intensive data on the faster portion of the vram?
 

Zathalus

Member
Xbox has yet to have an exclusive with a ray-tracing implementation as good as Ratchet and Clank or Spider-Man.

Btw what happened to the Minecraft raytracing update?
Using exclusive games to compare the differences between consoles is pretty pointless is it not? Is it due to the power of the console or testament to the skill of the developer? Most Sony Studios are some of the most skilled developers out there, so of course the games will look amazing on PS5, we just don't know if they would be any better if designed for the XSX first.

That's why people use multi-platform games to compare, which XSX has almost always been superior barring this singular example in RT, which people in this thread are trying to use as some sort of definitive proof of the PS5 being superior in RT.
 

Gaiff

Gold Member
This game is fucking busted at least on NVIDIA hardware.

The 4090 gets 149fps at 4K max settings.

HcaVAaU.png


Throw in RT and the frame rate tanks by over 100fps.

CXFT8pw.png


Notice, the 7900 XTX's performance drops by 60% but the 4090 which is far stronger in RT sees its performance drop by over 67%. It crushes the 7900 XTX by 49% in rasterization but in RT where its lead should further extend, it actually shrinks to a mere 19%. The 4080 is 3% faster in rasterization than the 7900 XTX, but suddenly the 7900 XTX is 15% faster with RT.

The same story can be seen across the 4080 beats the 6900 XT by a meager 22%. The 3090 Ti basically matches the 6800 XT.

Then seeing these shit reflections on Series X tells us this game is completely broken on anything that isn't a Playstation 5. What a piss-poor job by the developer. They should be ashamed of themselves for hyping up such a game only to deliver an unfinished product but even if it was finished, it's still a mediocre as hell game.
 
Last edited:

Bojji

Member
This game is fucking busted at least on NVIDIA hardware.

The 4090 gets 149fps at 4K max settings.

HcaVAaU.png


Throw in RT and the frame rate tanks by over 100fps.

CXFT8pw.png


Notice, the 7900 XTX's performance drops by 60% but the 4090 which is far stronger in RT sees its performance drop by over 67%. It crushes the 7900 XTX by 49% in rasterization but in RT where its lead should further extend, it actually shrinks to a mere 19%. The 4080 is 3% faster in rasterization than the 7900 XTX, but suddenly the 7900 XTX is 15% faster with RT.

The same story can be seen across the 4080 beats the 6900 XT by a meager 22%. The 3090 Ti basically matches the 6800 XT.

Then seeing these shit reflections on Series X tells us this game is completely broken on anything that isn't a Playstation 5. What a piss-poor job by the developer. They should be ashamed of themselves for hyping up such a game only to deliver an unfinished product but even if it was finished, it's still a mediocre as hell game.

RT is also heavy on CPU so maybe CPU limit on 4090?
 

Gaiff

Gold Member
RT is also heavy on CPU so maybe CPU limit on 4090?
Most likely. Apparently, the 4090 sits at a very low GPU usage and a single CPU thread gets maxed out while the rest remain idle. It appears to be even worse on NVIDIA hardware than with AMD, not unlike Gotham Knights which is also fucked even on consoles with poor CPU usage.

The Steam reviews are still poor a week after launch too. Their patches haven't fixed the game. It's still broken, albeit, less than on day 1.
 
Last edited:

DenchDeckard

Moderated wildly
Using this game as any benchmark is like reviewing the ability for a cooker to cook chicken where the chef has only focused on one cooker and made sure the food has been cooked for the correct time while the other cookers have only had half the time....

Then looking at the one cooked chicken and labelling all the other cookers shit because they've served half cooked chicken.....in half the time.......

...I'm hungry and I want chicken.
 

Patrick S.

Banned
I got the sense from this video that he wasn't actually distraught over the condition of the game. It sounded more like he was complaining to Rich about his job. He hates analyzing the vast number of modern PC games because so many of them recently have increasingly bad shader compilation-based stutter.
So he was essentially crying to the boss about his job being hard.
It must have been a slap to the face when Rich then suggested that he do a big writeup about the issue.
Honestly, I think there's more drama behind the scenes of this group than anyone knows.
I agree. On their weekly shows, I’ve been noticing that John often seems annoyed whenever Alex is welcomed to the episode. John also almost never engages with Alex. There’s barely any back and forth between them. Whenever Alex talks, John just sits there with a blank stare. John doesn’t laugh at Alex’s jokes, or react to him at all, mostly. Richard is the only one talking to Alex 99% of the time. The chemistry among the group seems to be way off.
 
Top Bottom