Alright then. 0.30.0. The usual posting; slightly late.
Not much to say about 0.30.0. We have pretty much done all the major tasks. What is left now is the less glorious task of filling in small and medium holes and fixing bugs. The later is particularly important since we have accumulated a large number of bugs.
No point in me highlighting specific tasks. At this point everything is about equal important, so take your pick.
OpenCS is also making progress, but still at a slower rate. We are still several releases away from being feature complete.
We have more or less abandoned the idea of monthly releases, but one release every two months isn't bad either. We should try to stick to that.
OpenMW 0.30.0
Re: OpenMW 0.30.0
Zini, can you talk a little more about what is missing in terms of save/load? Do you have any estimation when our save file format might be finalized?
Also, I know I am the biggest proponent of waiting to optimize until as late as possible, but I'm curious if we have any ideas about that now, about when it will be good to start or where the big bottlenecks are?
I ask because at some point we should start doing playthroughs to find bugs, but playthroughs will be rough without a complete save/load system, and with the low fps we have right now. I am probably jumping the gun here, but just wanted to get an idea of what our timeline looks like.
Also, I know I am the biggest proponent of waiting to optimize until as late as possible, but I'm curious if we have any ideas about that now, about when it will be good to start or where the big bottlenecks are?
I ask because at some point we should start doing playthroughs to find bugs, but playthroughs will be rough without a complete save/load system, and with the low fps we have right now. I am probably jumping the gun here, but just wanted to get an idea of what our timeline looks like.
Re: OpenMW 0.30.0
Most of the missing saved game file features are on the tracker. I don't plan to work on this feature in the near future (OpenCS takes all of my attention), so it depends on if other coders take up the tasks. At least some of them are a lot more beginner friendly than the rest of the remaining tasks, so that is actually kinda useful (we already had one new coder taking up one of the tasks).
Its still slightly early for optimisation. But we could start with it now, if we find someone who is willing to do the profiling.
On the graphics side of things we definitely will need more batching. Currently we are only batching statics, which is insufficient. I don't expect a huge speedup for Morrowind.esm, but content that is more interactive (and therefore has to utilise other types more) should benefit greatly.
Ogre 2.0 should bring a significant fps improvement, but we may not be able to wait for that. There are other options, but I leave that to scrawl.
On the CPU side we have vague reports about bottlenecks in the case folding code and in the NIF loader. Nothing we can act upon, which is why we need a profiling session (unfortunately a time consuming endeavor).
Its still slightly early for optimisation. But we could start with it now, if we find someone who is willing to do the profiling.
On the graphics side of things we definitely will need more batching. Currently we are only batching statics, which is insufficient. I don't expect a huge speedup for Morrowind.esm, but content that is more interactive (and therefore has to utilise other types more) should benefit greatly.
Ogre 2.0 should bring a significant fps improvement, but we may not be able to wait for that. There are other options, but I leave that to scrawl.
On the CPU side we have vague reports about bottlenecks in the case folding code and in the NIF loader. Nothing we can act upon, which is why we need a profiling session (unfortunately a time consuming endeavor).
Re: OpenMW 0.30.0
I've never done it but happy enough to have a go. I did kick off a profile run but MSVC won't play friendly. Apparently it won't support x64 without some tricks so I'm trying to get a x86 one built, but then ffmpeg causes trouble (something about SAFESEH). Once I get something useful I'll report back.
Re: OpenMW 0.30.0
Thanks for the update Zini, and thanks for looking into profiling cc9cii! I'm eagerly awaiting doing a playthrough.
Re: OpenMW 0.30.0
cc9cii, did you manage to get the profiling tool in MSVC express?
Re: OpenMW 0.30.0
I have the full version. Don't know if the express one comes with a profiler (prob. not).
- psi29a
- Posts: 5362
- Joined: 29 Sep 2011, 10:13
- Location: Belgium
- Gitlab profile: https://gitlab.com/psi29a/
- Contact:
Re: OpenMW 0.30.0
How are doing in terms of play-ability? Can we play through the main quest now?
When should we think about moving OpenMW away from experimental and into unstable? Wait until 1.0 ?
When should we think about moving OpenMW away from experimental and into unstable? Wait until 1.0 ?
Re: OpenMW 0.30.0
Feel free to try. IIRC I fixed one blocker at the vivec informants quest not long ago. I haven't tried going further yet.How are doing in terms of play-ability? Can we play through the main quest now?
Re: OpenMW 0.30.0
Either 1.0 or one of the last "polishing-only" releases before that.BrotherBrick wrote:How are doing in terms of play-ability? Can we play through the main quest now?
When should we think about moving OpenMW away from experimental and into unstable? Wait until 1.0 ?