• 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] Deathloop - The Technical Review

Blizzje

Member
- performance mode target 4K/60fps
- medium resolution 1296-1440p (1080p possible)
- quality mode 1836p target 60fps
- RT mode target 4K dyn/30fps
- lowest resolution 1836p
- performance mode is locked at 60fps 99% of the time
- quality mode is more like 40-50fps
- RT mode is locked at 30fps but have frame pacing issues
- PS5 performance in non-RT modes is similar to RX 5700 XT and RTX 2070 Super
At what point in the video do they talk about the PS5 performance compared to RX 5700 xt anf 2070super? I missed it.
 

Evilms

Banned
At what point in the video do they talk about the PS5 performance compared to RX 5700 xt anf 2070super? I missed it.
"PS5 performance in non-RT modes seems to scale in line with a Radeon RX 5700 XT or GeForce RTX 2070 Super, to give you some idea of the overall GPU burden."

 

DenchDeckard

Moderated wildly
It obviously needed the delays. Fingers crossed some patches can improve performance. Another call for VRR in the quality mode.
 

iHaunter

Member
Im sort of surprised we are still seeing low resolutions for high frame rates. Obviously 4k 60 is expensive, but i would have assumed it could do better than what deathloop seems to be putting out. If this had the visual fidelity of say, demons souls i would get the trade off
Demon's Souls had 8k Textures and 4K (Dynamic) + 60 FPS... Not sure what Bethesda's excuse is? Textures = RAM
 
Well CPU is probably hardly a limiting factor here...
How do you know? For the 30fps mode I'd say yes, but the 60fps mode? Try the game with PS5 settings using a 3.5ghz AMD 2700 with non-unified L3 cache (8MB total instead of 32MB) and let's see if the performance is identical in all scenes to the same settings with a 5ghz high end CPU (so higher IPC, and much more unified L3 cache).

As usual his comparison is flawed when it's about PS5 vs PC. Only NXGamer is doing fair comparisons.
 
Last edited:

pratyush

Member
So digital foundry just do pixel counting and frame rate testing. That video was way too long. Should be at best 5 min.

Game art direction is a turn off for me. I am sure its a good game to play but will wait it for gamepass to try. Cant risk wasting money on this. Same as i did for Returnal. Waiting for price drop on that as well.
 

Black_Stride

do not tempt fate do not contrain Wonder Woman's thighs do not do not
How do you know? For the 30fps mode I'd say yes, but the 60fps mode? Try the game with PS5 settings using a 3.5ghz AMD 2700 with non-unified L3 cache (8MB total instead of 32MB) and let's see if the performance is identical in all scenes to the same settings with a 5ghz high end CPU (so higher IPC, and much more unified L3 cache).

As usual his comparison is flawed when it's about PS5 vs PC. Only NXGamer is doing fair comparisons.
Are we still calling the PS5 CPU a Zen+ chip?
It might not be the full Zen2 found in desktops but it should easily walk Zen+
 
Is anyone talking about the pop in/missing textures/blinking textures on PS5’s performance mode? I noticed it as soon as I launched the game. When he first wakes up on that beach if you move around while looking at the water it’s like whole parts of the area that pops in really bad.

I double dipped, been one of my most anticipated since the reveal and my PC has a 3090 and 5950..I
Can pretty much beast the game to run but it still has really bad microstutters when arriving at a new area, picking up certain items(usually trinkets), and one time my FPS dropped to like 15fps when I got invaded and it was stuck there. Other times I got invaded you could feel the game stutter/tank FPS, ruining the “surprise” of the invasion.

i love the game, been one of the main people cheering for it on here in the past year, but I don’t know how in the hell reviewers gave it a perfect score with the performance issues and AI.
 
Last edited:

JimboJones

Member
And smaller team. I never had high Hope's for the PS5 Version. Smaller teams usually dont use console hardware efficiently.. the ger it running and then care for other stuff. Never a good comparision against the respective PC version..
However, laughed out when by the end of the vid I realized that Alex won't talk about load times..
The one Area where PS5 has an natural advantage over ANY PC out there is not talked about.. hehe who would have thought...🤣

The small PC loving Twat will play this Game and marginalize PS5s I/O miracle right until Direct Storage is out and has its first couple of proper supported games with RTX I/O and what not . He THEN will make big comparison sections in his review videos where he shows how PC (LIKE 5% OF ALL PCs but will be portrayed as all PCs)is easily faster in loading than even a PS5 ...
The PC loving litte Turd haha😅
What is the load times like on PS5?
 

MiguelItUp

Member
It felt like waking up on another planet. And I RARELY have issues with divergence in reviews.
That's been my thing as well. Sure, I haven't played it yet, but a lot of fairly decent complaints were repeated that revolved around PC performance and AI being bad. As a result, seeing perfect scores was pretty... surprising.
 
It felt like waking up on another planet. And I RARELY have issues with divergence in reviews.
Forreal. Like I support the game, I support it so much that I had multiple people on here call me a "Sony Shill" for saying I thought it would be good when everyone else was bashing it. I bought it on PS5 and PC...if you go in the reveal thread last year i'm sure I was one of the few saying it was my most anticipated release last fall when they said it was scheduled for then. I really like the art style, combat, and so on

But with all that said I just don't see perfect scores, thats bullshit. Its no way someone played this for 30 minutes even, saw the performance issues, and said "Yeah this a 10/10". And once again I love the game, never talked bad about it, todays posts from me are literally my first time even saying anything remotely negative.

Its common, repeated technical issues that damn near everyone on the internets reporting at this point(The amount of people I saw saying they refunded the steam version until more patches hit was wild last night)...If one of you work for an outlet and have insight on reviews...what happened here?
 

Riky

$MSFT
zflP4O5.jpg


What could cause this in performance mode? Some sort of bug?
 

M1chl

Currently Gif and Meme Champion
How do you know? For the 30fps mode I'd say yes, but the 60fps mode? Try the game with PS5 settings using a 3.5ghz AMD 2700 with non-unified L3 cache (8MB total instead of 32MB) and let's see if the performance is identical in all scenes to the same settings with a 5ghz high end CPU (so higher IPC, and much more unified L3 cache).

As usual his comparison is flawed when it's about PS5 vs PC. Only NXGamer is doing fair comparisons.
I am not saying anything, but I don't believe that the game is somewhat CPU intensive and draw calls on Playstation is non-issue. Could be obviously, but if you don't see profiler hard to say. And I don't believe we have some benchmarks already... Also the overhead, especially on PS is way lower than on PC, so keep that in mind.
 
In a weeks span we went from Deathloop being seen as overmarketed trash with Redfall looking like garbage to deathloop being a goty nominee and redfall being a highly anticipated game...gaming moves so fast.
 

M1chl

Currently Gif and Meme Champion
zflP4O5.jpg


What could cause this in performance mode? Some sort of bug?
Since THEAP99 THEAP99 posted few examples of when Z-buffer gets completely fucked, in this scene, I can see situation, when it renders stuff which is behind the stars, or that wall. It's something you don't commonly see in modern engines, but it could happened. If you have few pixel wide/tall of missing geometry, you can easily achieve, that since you can basically see through the opening, it renders shit which is invisible to the eye. And I haven't seen the video, but do. they pan the camera? That would explain it...
 
Last edited:

Karak

Member
Forreal. Like I support the game, I support it so much that I had multiple people on here call me a "Sony Shill" for saying I thought it would be good when everyone else was bashing it. I bought it on PS5 and PC...if you go in the reveal thread last year i'm sure I was one of the few saying it was my most anticipated release last fall when they said it was scheduled for then. I really like the art style, combat, and so on

But with all that said I just don't see perfect scores, thats bullshit. Its no way someone played this for 30 minutes even, saw the performance issues, and said "Yeah this a 10/10". And once again I love the game, never talked bad about it, todays posts from me are literally my first time even saying anything remotely negative.

Its common, repeated technical issues that damn near everyone on the internets reporting at this point(The amount of people I saw saying they refunded the steam version until more patches hit was wild last night)...If one of you work for an outlet and have insight on reviews...what happened here?
Well it's also things like the issues with the pause in the menus for up to 30 seconds that were happening on review code. I don't know if I saw ANYONE report it at all and it was actually known to be a genuine issue that always occurred. Or the sound issues with the mixing, especially depth issues on surround sounds, or the AI.
But I said my piece really on this one. Some parts of reviews are subjective, some are objective and I do get that because I cover sound and music and voice, and difficulty curves and AI and such that possibly I see a bit more and those CAN lower a score where someone not checking on those wouldn't be affected at all, but I feel like...well...

Consumers, watchers, whatever can choose what they want to consume and watch, but after a couple of things here in the last years its been unique to say the least to be on both sides. I rarely feel like I can't see where someone is coming from. With Cyberpunk I couldn't and with this, I can't.
 
Last edited:
Well it's also things like the issues with the pause in the menus for up to 30 seconds that were happening on review code. I don't know if I saw ANYONE report it at all and it was actually known to be a genuine issue that always occurred. Or the sound issues with the mixing, especially depth issues on surround sounds, or the AI.
But I said my piece really on this one. Some parts of reviews are subjective, some are objective and I do get that because I cover sound and music and voice, and difficulty curves and AI and such that possibly I see a bit more and those CAN lower a score where someone not checking on those wouldn't be affected at all, but I feel like...well...

Consumers, watchers, whatever can choose what they want to consume and watch, but after a couple of things here in the last years its been unique to say the least to be on both sides. I rarely feel like I can't see where someone is coming from. With Cyberpunk I couldn't and with this, I can't.
Very dope post. Thank you for the insight and that makes sense. And I think its pretty dope that you get to cover sound/music/and voice/ai and such and so on. Thats pretty damn fire lol
 
So digital foundry just do pixel counting and frame rate testing. That video was way too long. Should be at best 5 min.
It's why I mostly stopped watching their videos. I know they want that ad money and that the algorithm loves 10-15 minute videos, but the information density is so low that I'd rather read the Eurogamer article in 2 minutes.
 

M1chl

Currently Gif and Meme Champion
But Arkane is already using IdTech in their games. Dishonored 2 was made on modified IdTech 6.
Yeah, but it's pretty big jump from 6 to 7. Also I know, that's why I mentioned the ID Tech, that possibly they wouldn't have to move as a team to completely new engine, which could cause additional issues.
 

mrMUR_96

Member
Looks like the same one as their previous games.

Which is weird since they really seemed to have learned their lesson after Dishonored 2, with Prey being amazingly optimized at least on PC.
I remember there being a digital foundry article where they said they could get it to 1080p/60fps on a 2Gb GTX1050 by dropping a few graphical options



Performance and quality both "target" 60fps, but only performance mode hits it consistently 99% of the time.
Quality mode usually runs in the low 50's and drops to around 45fps during heavy combat sequences, it only reaches 60fps when looking at the floor, or some wall.
Prey was using cryengine, while deathloop and dishonored 2 are using their own void engine (idtech based I believe)
 

Armorian

Banned
But Arkane is already using IdTech in their games. Dishonored 2 was made on modified IdTech 6.

It was modified id tech 5, it was running like shit on any hardware. While it6 would run like a dream...


Discussion points to conflicting reports but performance would clearly suggest it5.
 

kuncol02

Banned
It was modified id tech 5, it was running like shit on any hardware. While it6 would run like a dream...


Discussion points to conflicting reports but performance would clearly suggest it5.
Let's say, that I will believe devs over you or any other random person on the internet.
"With this engine based on Id Tech 6 we've rewritten quite everything that's 70% of the engine. And it allows us to push the boundaries in every corner."
Who to believe? Game art director or army of some random people?
 

Fbh

Member
Prey was using cryengine, while deathloop and dishonored 2 are using their own void engine (idtech based I believe)
Oh, didn't know that. They really managed to capture the look and feel of their other games despite the engine change.

I guess that explains it though
 

pratyush

Member
It's why I mostly stopped watching their videos. I know they want that ad money and that the algorithm loves 10-15 minute videos, but the information density is so low that I'd rather read the Eurogamer article in 2 minutes.
Yeah but earlier they used to cover much more but now it's basically comparison video of framerate for fanboys which can be covered quite quickly I believe. 18 min of fillers is just way too much. I used to watch their video but now only watch it if it's an exclusive game. Not interested in PS5 vs Xbox framerate comparison videos
 
Last edited:
Top Bottom