After trying the latest build (e403c71) made available by Ace (SWE) i have seen several things that no one else has mentioned (that i can find). Don`t know if the issues i have been experiencing is unique to me or if they just aren't important at this time in the development.
I thought id mention them here and see if others have experienced the same things, and see if they should be reported in the bug tracker.
Just want to make it clear, there is a ton of issues already that as to be dealt with, and i don`t want to add more issues than necessary. So thats why i`m quality testing them here first. If someone has more to ad, or any thought on my findings they are free to do so.
1. Very poor performance in Vivec, 5 fps at the worst. This seems to have something with the ground textures since there was no slowdown present in 0.13.0 in any area.
2. On cell change the fps can drop or increase by 50% or more for no good reason. This is related to the above issue, but are happening in areas that makes no sense since there is hardly anything there that should, or could cause slowdowns.
3. The river-walk pavement textures in Balmora is very dark.
4. I get no sound in any of the windows branches that ACE (SWE) have made available. OpenMW is just spamming in the background console about OpenAL lacking resources like "several 100 times" a second, thus no sound at all. Was no issue in 0.13.0
Testing the latest windows master branch
Re: Testing the latest windows master branch
I'm confirming this is also present on linux.Rebel-Rider wrote: 2. On cell change the fps can drop or increase by 50% or more for no good reason. This is related to the above issue, but are happening in areas that makes no sense since there is hardly anything there that should, or could cause slowdowns.
I will do a bisect to find out where it was introduced.
Edit: can't reproduce it on master, but i've seen it in my water branch i think.. weird
Edit2: ran openmw in an OpenGL debugger and after crossing such a cell border that drops the fps by half, the OpenGL calls / sec almost doubled. Apparently we are indeed rendering something twice.
Re: Testing the latest windows master branch
@Rebel-Rider: Does the batch and/or triangle count change significantly for you when crossing such a cell border?
You can activate the fps display by setting [HUD] fps = 2 in your settings.cfg.
You can activate the fps display by setting [HUD] fps = 2 in your settings.cfg.
Re: Testing the latest windows master branch
Haven't heard about this issue before, can you grab a few of the messages and post them?Rebel-Rider wrote:4. I get no sound in any of the windows branches that ACE (SWE) have made available. OpenMW is just spamming in the background console about OpenAL lacking resources like "several 100 times" a second, thus no sound at all. Was no issue in 0.13.0
- Rebel-Rider
- Posts: 34
- Joined: 29 Mar 2012, 18:27
Re: Testing the latest windows master branch
Issue #4
This happens the very second i start the latest windows master branch, it`s going trough the errors so fast i had to take a screenshot to see what it said. There was no sound at any point during the testing, neither music or activation sounds (doors, pickup and so on)
Here is a screenshot of it
-------------------------------------------------------------------------
Issue #2
I just remembered that i already tried toggling FPS in the "settings-default.cfg" file when the feature was announced. Nothing shows up no matter if i set it to 1 or 2. So i have no batch and/or triangle count data to show you.
This happens the very second i start the latest windows master branch, it`s going trough the errors so fast i had to take a screenshot to see what it said. There was no sound at any point during the testing, neither music or activation sounds (doors, pickup and so on)
Here is a screenshot of it
-------------------------------------------------------------------------
Issue #2
I just remembered that i already tried toggling FPS in the "settings-default.cfg" file when the feature was announced. Nothing shows up no matter if i set it to 1 or 2. So i have no batch and/or triangle count data to show you.
Re: Testing the latest windows master branch
You are not supposed to change settings-default.cfg. There is a user settings file created in the same location as the log files (settings.cfg) this is the one you should change.
- Rebel-Rider
- Posts: 34
- Joined: 29 Mar 2012, 18:27
Re: Testing the latest windows master branch
*Face-palm*
After some thinking, i found it where it was supposed to be.
C:\Users\Myname\Documents\My Games\openmw
Ok.
When i travel outside beshara i get this when when im right outside the door before i do anything.
FPS: 27
TRI: 63586
BATCH: 202
When i travel just west of beshara over the water and stop right before the next cell is loaded i get this.
FPS: 27
TRI: 49585
BATCH: 313
Traveling "a few inches" inside the next cell i get this
FPS: 80
TRI: 61836
BATCH: 347
Doing the same towards the Telvanni cantoon, i stop right by the mushroom where the infamus "mo-slow" cell border, i get this
FPS: 27
TRI: 56918
BATCH: 440
Traveling a few inches into the infamus cell i get this:
FPS: 12
TRI: 62647
BATCH: 473
Looking right up into the sky in the same cell, i get this:
FPS: 12
TRI: 8582
BATCH: 68
After some thinking, i found it where it was supposed to be.
C:\Users\Myname\Documents\My Games\openmw
Ok.
When i travel outside beshara i get this when when im right outside the door before i do anything.
FPS: 27
TRI: 63586
BATCH: 202
When i travel just west of beshara over the water and stop right before the next cell is loaded i get this.
FPS: 27
TRI: 49585
BATCH: 313
Traveling "a few inches" inside the next cell i get this
FPS: 80
TRI: 61836
BATCH: 347
Doing the same towards the Telvanni cantoon, i stop right by the mushroom where the infamus "mo-slow" cell border, i get this
FPS: 27
TRI: 56918
BATCH: 440
Traveling a few inches into the infamus cell i get this:
FPS: 12
TRI: 62647
BATCH: 473
Looking right up into the sky in the same cell, i get this:
FPS: 12
TRI: 8582
BATCH: 68
Re: Testing the latest windows master branch
Looks like its the CPU's fault.Traveling a few inches into the infamus cell i get this:
FPS: 12
TRI: 62647
BATCH: 473
Looking right up into the sky in the same cell, i get this:
FPS: 12
TRI: 8582
BATCH: 68
- Rebel-Rider
- Posts: 34
- Joined: 29 Mar 2012, 18:27
Re: Testing the latest windows master branch
That sounds very strange, i have a AMD Phenom II X6 1100T black edition. And games that are a lot more CPU limited than this runs very well on it. I even tried to force it to run at full speed all the time to rule out any strangeness with the same result. I could make a video of it so you can see what i`m experiencing.
This being my machines fault, i`m not willing to buy that yet.
Edit1: If i look up in the sky in an "unaffected/Normal" cell i get several hundred FPS...
This being my machines fault, i`m not willing to buy that yet.
Edit1: If i look up in the sky in an "unaffected/Normal" cell i get several hundred FPS...
Re: Testing the latest windows master branch
Not blaming your system. I'm just saying that there might be a heavy cpu bottleneck somewhere in the code. Hard to say though.