• Please make sure you are familiar with the forum rules. You can find them here: https://forums.tripwireinteractive.com/index.php?threads/forum-rules.2334636/

Regarding performance

The main thing I'm worried about is how effective the optimization is going to be.

On every setting at the lowest, I get an average 25 fps. Max 30, min 15.

My specs:
AMD Phenom II X6 1090T 3.2Ghz
GeForce GTX 460
8GB DDR3

So if I understand correctly, I just have hardware that can't handle the game? Either upgrade my computer or wait for performance optimization? Will that optimization be decent, or am I looking at a 5-10 fps increase at the lowest settings?
 
Upvote 0
I can't actuallt stand with Ramm statement........are you saying that (for example) Crysis 1, Crysis 2 or Arma 2 are utilizing less polygons or drawing less things?:eek::eek::eek::eek: and they work better than RO2:eek::eek::eek:

Cammon guys, a gtx570 and I5 2500k should eat this game, EAT it!!!!!!

I don't want to complain, I love you guys and game is amazing, but I can't understand........
 
Upvote 0
I can't actuallt stand with Ramm statement........are you saying that (for example) Crysis 1, Crysis 2 or Arma 2 are utilizing less polygons or drawing less things?:eek::eek::eek::eek: and they work better than RO2:eek::eek::eek:

Cammon guys, a gtx570 and I5 2500k should eat this game, EAT it!!!!!!

I don't want to complain, I love you guys and game is amazing, but I can't understand........


yeah, my FPS doesn't falter in arma 2 unless i'm playing campaign, and i think it's the AI optimizations that kill the FPS in that one.
 
Upvote 0
EVERYONE relax..

..and PLEASE stop talking about BC2 and BF3. Those games are more fun to run around in then to actually shoot in. The hit registration is a deal breaker on both of those games and it will never be fixed.

If you want to talk about both games in the same sentence, wait til you encounter the bug in RO2 were no bullets register at point blank range. Including melee attacks. :D

Can't really say much about the bug other than whats been said on the forums though, again, haven't been able to play the game yet.
 
Upvote 0
you. you make sense and i understand that this game is rendering individual bricks and all that, but i am sticking to my guns on this one and regardless of cited examples and their shortcomings, this game should have, at the very least, a more stable fps than what i'm getting on the game's default settings. sorry if i sound like a dick or anything in previous [psts, but the game isn't up to snuff (performance-wise) in my opinion, and a lot of people are having a similar experience.

it's not like i signed up on the forums today and this is my first post. i've been reporting my findings almost daily since final phase of the beta.

I don't think anyone is disagreeing that the game should have better performance. If they are then they are the lucky ones who have great performance.
 
Upvote 0
I think people who claim 15 fps with those juiced systems are quite frankly lieing and telling their worst case when playing on a loaded server while holding the sniper scope while looking at pavlov's house from a distance.

Thats why they want your perflog because getting a steady 15 fps with a jacked system defies logic and physics.

Dont confuse HITCHING with fps drops they arent remotely the same thing

Firstly I said I had 15fps DROPS, not that I had 15fps all the time.
In fact if you only consider the numbers, my FPS are ok (well 30fps with my rig is not ok, but it should be enough to play).
What is not ok is the constant stutters I have : every 5 seconds or so I go from 30fps to 20fps (or less) which is making the game REALLY unpleasant to play (when not just impossible, I died a lot of time because I was stuttering)
Secondly, as you are calling me a liar I filmed my screen with a camera (as capping with FRAPS would only make things even worse) and as you would admit this is really not pleasant AT ALL and far away from the kind of performance I should get with my computer.

Watch in 720p or you won't see anything and no the lagging is not due tothe poor quality of my camera.
Red Orchestra 2 poor performance. - YouTube
 
Last edited:
Upvote 0
Simply turning OFF framerate stabilizer, vsync and the one frame thingy improved my fps by 10, and removed most of the unplayable constant stutter. It still stutters the first 3 minutes of a map, but then its silky smooth with all at ultra except character detail and detail....... detail, for the sake of a little bit extra fps.

Core 2 Duo E2600, 560TI, 6gb DDR2, Win7
 
Last edited:
Upvote 0
So how can you help? By getting us perflog dumps which will help us get the information we need to improve the performance. Please check out this link here for information on how to do this:

[url]http://forums.tripwireinteractive.com/showthread.php?t=61789[/URL]

The more information we have, the easier it will be to track down these issues. So thank you for your patience while we get these issues sorted. And rest assured we're hard at work to make that happen.

I'll get right on it!
 
Upvote 0
I think people who claim 15 fps with those juiced systems are quite frankly lieing and telling their worst case when playing on a loaded server while holding the sniper scope while looking at pavlov's house from a distance.

Thats why they want your perflog because getting a steady 15 fps with a jacked system defies logic and physics.

Dont confuse HITCHING with fps drops they arent remotely the same thing

If you want to see juiced system specs, i'll gladly send you mine. Also I've never seen over 31 FPS, and I will screenshot it if you have doubts. Usually the game sits at 23 dipping to 4-5 FPS when it wants to.
 
Upvote 0
Regarding performance

We're working diligently on performance, but improvements/fixes for performance are going to take a bit longer.[...]

I appreciate & commend your well articulated, and honest, communication. And you're not even some marketing/PR specialist, are you? :) So often I find that companies employ omission of truths and outright lies, all to make something appear as positive to themselves as possible. But TW seems nothing like that. So refreshing.

[...]This will be ongoing until we get to the point where we feel the majority of users are reasonably happy with the performance they are getting.

So how can you help? By getting us perflog dumps which will help us get the information we need to improve the performance. Please check out this link here for information on how to do this:

http://forums.tripwireinteractive.com/showthread.php?t=61789[url]http://forums.tripwireinteractive.com/showthread.php?t=61789[/URL]

The more information we have, the easier it will be to track down these issues. So thank you for your patience while we get these issues sorted. And rest assured we're hard at work to make that happen.

I have suggestions regarding how to accellerate the performance improvement process. I don't know to 100% how your situation is like and what is technically possible or not. But my perception should be somewhat near reality. :)

You lack wide variety of machine combinations required for acquiring data. You require this data to be volunteered by the players. Meet a typical (passionate) player (me). I have the great hardware with nasty performance problem. I want to help you out. But the process is perceived by me (and probably thus many others like me) as too much hassle & complexity. First reading up on the process, then manually running the command during specific scenes, then going to find the logs, attaching this and that, preferably with comments etc. Then not being sure if it was done right and uncertainty regarding much help my particular effort will have given.

I propose the creation and usage of timedemos. This mode may be triggered by players through a desktop or steam shortcut, or a "run performance test" in the main menu > options. Or (and this is far less ideal) as a console command.

The timedemo runs for one minute each. Varying amounts of AI bots repeating some simple action. Artillery. Smoke. Post processing effects. A fixed camera or moving one. Fov changes. Sudden direction changes. Does not have to be perfect or cover every possible situation. The conditions are identical for each test run. The logs should try to retrieve as much of the relevant specs of the machine automatically. Any files needed for submission should be unified as one file. Or more ideal - set up a machine to receive the logs from the RO2 client on the players machines.

Positive:

Your own developers will appreciate access to a steady enormous amount of reliable, standardized performance data. The infrastructure to keep doing it for the entire life of the product. Interpreting the data will be much easier and efficient, without need to figure out what players are really sending them. Work should proceed faster, less resources needed to fix the problems, higher results achieved, less patches required, game will be perceived to be of higher quality and this results also in higher revenue for the company.

The players will appreciate it for doing own performance comparisons, for finding the best setting for themselves, for being able to help you (and as such themselves) with the performance problems, and for being able to reliably track performance gains/losses from different game versions rather than relying on their own vague (and often incorrect) perception on the state of performance.

Negative:

Creating timedemos and the basic infrastructure to have them generate & send the data you need, will require developer resources right now which will not yield anything helpful until after it is finished and pushed out to the players. It may seem more tempting to just scramble for the 'urgent' fixes now for instant gains, rather than delay for much greater gains in the longer run.
 
  • Like
Reactions: Rumo
Upvote 0
My personal performance observations from my first playthrough last night. I'm typing this up at work while on my iPhone, so forgive me if I forget the name of some settings.

I just recently set up this new rig, AMD Phenom 944 quad core, sapphire radeon 6850, 4 gigs of RAM. 64 bit windows 7.

I started the game up and went into single player and sat at a solid 65-66 FPS during training. When I hit the actual missions I saw between 45-50. I decided to leave the settings at the default high settings and go into multiplayer on a 64-man server. I saw on average 30 FPS on high settings so I turned shadows down to low, turned off AA, motion blur, framerate smoothing, and the other option at the end (can't recall the name right now).

In wide open space I'm about 45-50 FPS, in closed quarters 30. And when the artillery goes off I've hit 15-20. Dropping graphics to medium shows no improvement.

I have also tried many of the suggestions here such as turning off the steam in game community and depth of field in the config files. No real improvements when tweaking. I'll have to try a few more things later.

Bottom line. It runs. Does it run well? Not really, but I can play it with a few snags when things get bad.

Maybe I should try a 32-man server?
 
Upvote 0
I think the people who can't start the game at all, or having menu glitches needs to be addressed first.

Performance can be addressed after that. Most of the poor performance issues seems to be stemming from some users used to console-ported Unreal Engine 3 games made on an older, less graphically-intensive and less-capable build of UE3. So they unrealistically expect similar performance on a game that not only pushes the graphics in terms of the engine being used, but the amount of players playing on a server at any given time on said engine.
 
Last edited:
Upvote 0