• 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/

Why did the beta run better?

I think all of us who were in the preorder beta know exactly what I mean. The performance was better (on my end), the pings were certainly better.
Is it the stats system that is bogging down the game?

The only beta patch that really screwed with the performance was the one that introduced punkbuster, and probably the first statistics functionality. So it's possible.

Though I'm not sure why you even bother with posting here, according to your signature "Tripwire ignores the problems of this game". Even though they've been doing the exact opposite :)
 
Upvote 0
without a doubt there was a period in the beta where it ran fantastically

To the point where leading targets with a bolt action on fallen heroes was no issue and you could make hit after hit regardless of the server you are on.

I dont remember which patch it was but it was though. but it was sometime in the middle of the test. i cant remember if it was before or after they fixed the CTDs

but everything , the performance of the graphics , to the netcode were FLAWLESS.

I would like to see the changelists for the beta test over time so we can narrow down what build it was
 
Last edited:
Upvote 0
I think we're talking about an illusion here. There have been numerous posts about beta performance getting worse and the actual game being worse than beta.

I think it's a map issue, at least partly. I'm not talking about all of the individual cases, but map issues cover a big bunch of cases.

The first beta had three maps: fallen fighters, apartments. Both ran well - still do. Second set of maps introduced barracks and grain elevator. Both also ran quite nicely - still do. But the last set of maps introduced both troublesome maps: station and pavlovs. This leads to an illusion that the performance is getting worse.

And I know I missed a few maps, but they aren't significant anyways.
 
Upvote 0
I think we're talking about an illusion here. There have been numerous posts about beta performance getting worse and the actual game being worse than beta.

I think it's a map issue, at least partly. I'm not talking about all of the individual cases, but map issues cover a big bunch of cases.

The first beta had three maps: fallen fighters, apartments. Both ran well - still do. Second set of maps introduced barracks and grain elevator. Both also ran quite nicely - still do. But the last set of maps introduced both troublesome maps: station and pavlovs. This leads to an illusion that the performance is getting worse.

And I know I missed a few maps, but they aren't significant anyways.

i lost about 15 fps on each of the first six maps, and any random 64-player-server in the beta was more stable than the ping-wise best 24-player-server i find today.

so, yeah, has nothing to do with the maps at all...
 
Upvote 0
I still cannot understand why so many people who think they are computer savvy never have data to back up what they say:

FRAPS BENCHMARKS:

BETA: 9/10/2011 - 11:33 PM EST
Frames: 6643
Time (ms): 120000
Minimum: 14
Maximum: 62
Avg: 54.358

GOLD EDITION: 10/09/2011 - 1:41 AM EST
Frames: 7349
Time(ms): 120000
Minimum: 48
Maximum: 65
Avg: 61.242


Your memory fails you or it's a placebo effect.

All my benchmarks are run on Spartannovka in a full 32 player match.
 
Last edited:
Upvote 0
Except for one patch which introduced an issue where the game could end up in single threaded mode (and half performance), each patch from beta to current has been tested to maintain or improve performance on every machine we have in the office.

These machines range from our minimum spec to the highest end hardware we can get.

Any major loss in performance between builds is usually dependent on something local to that users machines. The most common can be security software/hardware actively scanning the game while it is running, throttling its network packets, game files being fragmented, and corrupt files.

The two major causes of performance issues in RO 2 are when the Game thread and the Rendering thread get capped or split (between multiple threads. This worsens performance as they were not designed for it, but occasionally windows will do this to try and spread cpu load around if a core is maxed out for a significant amount of time.

We are working to improve both of these situations. Work on the rendering thread involves reducing the overall amount of draw calls being made, and work on the game thread involves finding and isolating spikes in it and finding out the cause and addressing them.
 
  • Like
Reactions: Machete234 and Apos
Upvote 0
I wonder how many of you have dedicated gaming rig. I mean that only what's essential for playing games is installed and nothing more. I have it that way, not even antivir/malware programs or firewall enabled (I'm behind hardware firewall).
I have zero problems with games.
All that crap you need for browsing, playing media etc. must have some part in so many problems.
 
  • Like
Reactions: TheGoden
Upvote 0
I just got done playing what was supposed to be the November 2nd patch, and I must say that the performance has gotten a good bit better. first off, I tried lowest settings, which it looks as if the minimums and stutters have gotten better, then I tried ultra settings, which ran at about 30-45 as compared to the previous 11-25 I was getting (all on grain elevator). The next step is to create some benchmarks of a level run-through, revert to the old patch, benchmark that one, and I'll post some results for you guys.

One last thing to Yoshiro; I live about 10 minutes from your offices in Roswell, would it help if you guys could use my rig for a more diverse testbed in order to improve performance on amd crossfire machines?
 
Upvote 0