OpenMW Optimization?

General discussion regarding the OpenMW project.
For technical support, please use the Support subforum.
Esoptros
Posts: 3
Joined: 30 May 2014, 04:04

OpenMW Optimization?

Post by Esoptros »

I am particularly curious about potential optimization updates and or enhancements. Perhaps I'm being blind, but I couldn't find really any info with what's going on with increasing optimization so I do apologize beforehand if this is an old question.


I have play tested the newest update for OpenMW [mainly recently due to my current involvement with modding Skyrim] I just returned to Morrowind. While I am not a coder/scripter I have to be impressed by the work done here. But to me Optimization should be a very important aspect of it. And forgive me if this is harsh, but the optimization is horrid. My computer can handle a heavily modded Skyrim graphics wise with no lag, but when playing OpenMW with just basic graphics and reflective water causes lags, that's a big issue.

My point is, what is going on with the Optimization side of things for OpenMW, any updates or news?
ezze
Posts: 513
Joined: 21 Nov 2013, 13:20

Re: OpenMW Optimization?

Post by ezze »

Speed concerns are mainly pushed after ver. 1.0; overall considering it is possible to play with reasonable hardware.
Esoptros
Posts: 3
Joined: 30 May 2014, 04:04

Re: OpenMW Optimization?

Post by Esoptros »

ezze wrote:Speed concerns are mainly pushed after ver. 1.0; overall considering it is possible to play with reasonable hardware.

Either I am daft or that was a vague answer. I didn't quite catch what you meant.
User avatar
sjek
Posts: 442
Joined: 22 Nov 2014, 10:51

Re: OpenMW Optimization?

Post by sjek »

before graphical optimisation it's better to have whole system in place before starting to breaking it to spieces would say .p
and it largely depends how the ogre2 will come along this year.
and bullet3 also is coming more with backward compability.
basically waiting for the engine before can make more precise tuning to car

http://www.ogre3d.org/2015/01/04/review ... -into-2015

with skyrim and such there's usually unity4 or unreal engine 4 in the background which are provision or license based and runs on big money. it's very hard to compete with those but with better script compiler, features and editor the content can be much much wider than graphical mods.

bethesda has had biggest limitations in these parts as the core game mechanics has been kept hardcoded for commercial purposes.

at least i think that's the reason. otherwise some game in which you can easily mod kinda any sort of game into would have been already done. in which part comes that this will be kinda much game engine more rather than just a game.

but as said, before that the groundwork has to be properly done with bugtracking that it doesn't crash heavily modded.
and of course the mod choosing is for the player : )

the optimisation as of long lasting saying is to be made before 1.0 is released. it's gonna cause some bugs witching to ogre 2.0 later on so it depends.
Esoptros
Posts: 3
Joined: 30 May 2014, 04:04

Re: OpenMW Optimization?

Post by Esoptros »

sjek wrote:before graphical optimisation it's better to have whole system in place before starting to breaking it to spieces would say .p
and it largely depends how the ogre2 will come along this year.
and bullet3 also is coming more with backward compability.
basically waiting for the engine before can make more precise tuning to car

http://www.ogre3d.org/2015/01/04/review ... -into-2015

with skyrim and such there's usually unity4 or unreal engine 4 in the background which are provision or license based and runs on big money. it's very hard to compete with those but with better script compiler, features and editor the content can be much much wider than graphical mods.

bethesda has had biggest limitations in these parts as the core game mechanics has been kept hardcoded for commercial purposes.

at least i think that's the reason. otherwise some game in which you can easily mod kinda any sort of game into would have been already done. in which part comes that this will be kinda much game engine more rather than just a game.

but as said, before that the groundwork has to be properly done with bugtracking that it doesn't crash heavily modded.
and of course the mod choosing is for the player : )

the optimisation as of long lasting saying is to be made before 1.0 is released. it's gonna cause some bugs witching to ogre 2.0 later on so it depends.

So in other words, optimization is gonna be the most difficult part of this project and will need to be finished after the base of everything else is complete, to avoid major bugs? lol Least that's what I got from all that.
User avatar
sjek
Posts: 442
Joined: 22 Nov 2014, 10:51

Re: OpenMW Optimization?

Post by sjek »

heh 8-) : D
optimization is gonna be the most difficult part of this project and will need to be finished after the base of everything else is complete
that's right but changing the places of optimisation and everything is a bit of taste question xP
https://bugs.openmw.org/projects/openmw/activity

and btw are you using MGSO on your install. it has several bad leftovers in it's models which are causing fps hits
Tinker
Posts: 231
Joined: 06 Mar 2013, 17:35

Re: OpenMW Optimization?

Post by Tinker »

Optimisation is always the last thing in software development, not much much point in optimising something that may not survive a few rebuilds.

As I understand it a lot of performance will come from Ogre updates to start with and another problem is most active members are working on Linux and performance under Linux is always better than windows so they need more feedback from windows users. Specifically where problems are and under what conditions.
User avatar
psi29a
Posts: 5356
Joined: 29 Sep 2011, 10:13
Location: Belgium
Gitlab profile: https://gitlab.com/psi29a/
Contact:

Re: OpenMW Optimization?

Post by psi29a »

Tinker wrote:Optimisation is always the last thing in software development, not much much point in optimising something that may not survive a few rebuilds.

As I understand it a lot of performance will come from Ogre updates to start with and another problem is most active members are working on Linux and performance under Linux is always better than windows so they need more feedback from windows users. Specifically where problems are and under what conditions.
Normally, I would agree with this but there is a bit of nuance.

Windows is the largest target, so we do take the time to figure out problems that happen there.

We do acknowledge that performance is underwhelming, and that is what Tinker said... we are focusing more on stability, good code and functionality. Once we we get close to 100% feature complete, then we have to take a good long look at optimization and fix any bottlenecks. That is the moment when code re-writes happen and we follow the rabbit hole down development hell... that last 2% of a project that usually lasts the longest.

We rather have something playable than something shiny, but of course, we want both. As with most things, just give us time please. :)
dteviot
Posts: 144
Joined: 17 Dec 2014, 20:29

Re: OpenMW Optimization?

Post by dteviot »

psi29a wrote:Windows is the largest target, so we do take the time to figure out problems that happen there.
One nice thing about MSVC 2013 community edition, it looks like it contains a profiler.
Menu -> Analyze -> Profile.

There's also a static code analyzer (which does the same sort of thing as Coverty).
It takes ages to run, but someone might like to try it and send the output. It might find a few potential issues.
robcbwilson
Posts: 39
Joined: 04 Dec 2014, 14:15

Re: OpenMW Optimization?

Post by robcbwilson »

I am running OpenMW on a Wintel PC, so I will be more than happy to help with any Optimisation work that is needed once we are at 1.0.

I agree totally that we need a stable and solid game before we optimise anything. Morrowind on my PC is far from stable, so am so excited to see this project reach 1.0.
Post Reply