Request: 0.76b2

[Acidd_UK]

I don't know how much effort is involved in producing a release, but with the seeming fix found for the ground decal bug (All hail Imbaczek, Kloot and Tvo) and some other little bigfixes, maybe we could get a 0.76b2? Pretty please?! Things just arent the same with ground decals off!

[el_matarife]

I don't know how much effort is involved in producing a release, but with the seeming fix found for the ground decal bug (All hail Imbaczek, Kloot and Tvo) and some other little bigfixes, maybe we could get a 0.76b2? Pretty please?! Things just arent the same with ground decals off!

Yeah, a 76B2 with a ground decals fix, the Settings++ fixes, and any of the other big bug fixes would be great.

I'm going to go through the revision list and pick out a few other big bug fixes that are critical, but 5244 and the other network fixes are the ones I'm really thinking about.

[imbaczek]

http://gitorious.org/projects/spring-sv ... /logs/next

please use this as a reference, perhaps maybe create your own tree on gitorious.

EDIT: a new build: http://spring.jobjol.nl/show_file.php?id=997

[el_matarife]

I'm going to go through the revision list and pick out a few other big bug fixes that are critical, but 5244 and the other network fixes are the ones I'm really thinking about.

I took a look at the WebSVN changelog and there's soooo many fixes. Honestly, just go for the code fixes listed as resolved in Mantis, which is almost 30 bugs. Add in the network code fixes from the first few weeks after release, and this dangling pointer fix and I think we have a new .76B2 release with a lot better stability. Still, there may be enough changes by now to prompt a .77B1 release, though I'm not sure. There aren't any specific goals for the next release right?

[KDR_11k]

Careful, there's still some crashbugs left in SVN I think.

[ZellSF]

Careful, there's still some crashbugs left in SVN I think.

I'm assuming they, like me, don't want a release right now, just for the release process to actually start.

[smoth]

it isn't that simple zell.

[Pxtl]

Careful, there's still some crashbugs left in SVN I think.

I'm assuming they, like me, don't want a release right now, just for the release process to actually start.

Here's the problem: while there is effort to work on old crash bugs, there is also effort to add new features. Those new features can introduce new crash bugs, and they may be completely half-done right now.

Now, to work around this, most shops will have a "development" branch and a "stable" branch, to which "development" bugfixes are backported. However, that backporting is work. Work nobody wants to do. It means maintaining code repositories. So, do you want somebody who might be excited to work hard on new and interesting Spring features to instead spend their time on uninteresting and thankless drudgework like that? No. Not only does it consume their (limited) time, it also destroys their interest, which effectively reduces their overall contribution to the engine.

In short: wait until all the devs agree to go into feature-freeze/bugfix mode.

[Acidd_UK]

Yes, but that is what I am asking for. Since turning off grounddecals, I have had no crashes out of maybe 10-15 games. Before this I was crashing out of maybe 30% or more games. So to me, this is a fairly significant bugfix that warrants a new beta release. Doubly so when (afaik) ground decals are turned on by default and so new players may experience the crash. On top of this, if I understand it correctly, the chance of experiencing the bug depends upon whether you computer attempts to render a particular frame of animation. As hardware gets faster, framerates increase, increasing the number of people that experience the bug. Finally, since the bug will generally crash out several people from a single game, it can effectively ruin games for everyone, even those ppl that have the workaround (grounddecals off). As for suggesting that backporting the dev bugfix patches is masses of thankless work, I have to disagree with this. I (and everyone else that experiences this bug) will thank the ppl that make this happen.

[smoth]

for this one bug.

but also consider it is dull and uninteresting. Most of the devs are doing code that interests them. Repo work is not interesting.

[Acidd_UK]

Well maybe we need to look at why people work on Spring? Is it for their own amusement, or is it to produce a stable and fun game for other people to enjoy?

I think it has to be a balance between the two, but in the same way as we all like new features, we also all like fewer crashes. I think suggesting that something should not be done because it is 'dull and uninteresting' is a bad response. Much coding is dull and uninteresting, I'm sure tvo's many refactoring commits are the result of long dull hours, with nothing 'pretty' to show the users, but he knows it's a necessary thing to be done to allow future development. In the same way, bugfixing results in more users (or rather, not bugfixing results in fewer users).

[smoth]

I am a coder and it is not dull or uninteresting to me. fiddling around with a repository is one of the less interesting things to do. Well that and complying to coding standards like bracket arrangement oh and fucking java docs.

[Satirik]

I am a coder and it is not dull or uninteresting to me. fiddling around with a repository is one of the less interesting things to do. Well that and complying to coding standards like bracket arrangement oh and fucking java docs.

where is tower of defense btw ?

[smoth]

where is tower of defense btw ?

lol

TD will receive more attention after gundam RTS 1.2 gets released. There is a looooong thread where I explained my computer exploding(not literally) etc. which at up a bunch of free time that I allocated to TD. Right now TD is chilling until I or one of the team gets back onto it.

[ZellSF]

it isn't that simple zell.

Uh, where did I say it was simple?

[YokoZar]

If a second beta is too much of a hassle, why not push up the timetable for 0.77 drastically?

[lurker]

All these fixes are already backported. Read imbaczek's post. The main issue is that making a release is a big process afaik.

[smoth]

this isn't a matter of time tables, people are doing this work in their spare time. They cannot just say, hey let's make a push for a release because X person wants it. Believe me, I just found a fucking fail new bug just last night, I would love to see a fix with roll out but it is not that simple. With all fixes and improvements can come new bugs, some of which may make spring less stable.

FURTHER MORE if the developers push themselves to hurry up and make a release it will take an emotional toll on them. NOT ONLY are they giving up their spare time but you want them to hurry it up? Shit dude, that is a good way to loose volunteers. Rushing a timetable only ever means putting more man hours in over a shorter period of time.

In the terms of a regular work week you are asking for over time. and you are asking for it from volunteers.

[imbaczek]

i think telling yoko that's it's about free time is... unnecessary he's a MOTU, after all.

anyways a 76b2 will happen, hopefully sooner than later.

[smoth]

the post was more fore zell.