Hard Light Productions Forums
Modding, Mission Design, and Coding => FS2 Open Coding - The Source Code Project (SCP) => Topic started by: colecampbell666 on July 21, 2007, 11:48:37 am
-
I've asked this several times in several threads with no response, so I figured that I should ask formally.
What is the current planned release date for 3.6.10?
-
as soon as you shorten your signature
-
'When it's ready' is the only answer that can be given, because it's the only answer there is.
It's important people understand that, if something comes up in one of the coders lives and they have to take a few weeks off, and the project gets held up, then that's the way things go. A bug might crop up at the last minute that leads back into a huge twisted section of code, that's just the way things go. The source code does not work to dates, it works to achievements, once they have reached a level where a release is warranted and the build appears to be stable, then they will release it.
-
k.
-
If it makes you feel better, you will get an announcement when it comes out, don't worry, you won't miss out on 3.7 when it comes out, and it will be worth the wait ;)
-
If it makes you feel better, you will get an announcement when it comes out, don't worry, you won't miss out on 3.7 when it comes out, and it will be worth the wait ;)
he might miss out on 3.610 though :p.
-
LOL To be honest, I was under the impression the next official release was 3.7, but I could be wrong, so I didn't say either way :p
-
If it makes you feel better, you will get an announcement when it comes out, don't worry, you won't miss out on 3.7 when it comes out, and it will be worth the wait ;)
he might miss out on 3.610 though :p.
There will be a 3.610.0? ;)
-
If it makes you feel better, you will get an announcement when it comes out, don't worry, you won't miss out on 3.7 when it comes out, and it will be worth the wait ;)
Poor 3.6.11, no attention whatsoever... :(
-
The only people who can have release dates are people who do this stuff as their job. They get fired if they stop working.
-
but nobody has a job in this, so therefore there is no official release date.
-
Given the number of bugs in the current codebase, any official release may be a while unless we get some real bug hunting and killing done.
-
i compiled the cvs versione of fs2open and i see in the hangar fs2 open 3.6.10 :D
-
Maybe have a Code phase completely dedicated to finding, torturing and then killing bugs.
-
i compiled the cvs versione of fs2open and i see in the hangar fs2 open 3.6.10 :D
Yeah, what's up with that? Why is the fs2_open_3_6_9 branch still being commited to? Why isn't there a fs2_open_3_6_10 branch? It's a bit confusing... :nervous:
-
Everyone just kept committing to the 3.6.9 branch after it was out. :)
-
Maybe have a Code phase completely dedicated to finding, torturing and then killing bugs.
Those are the infamous code freezes. And its probably not a bad idea...
As for getting a clean copy of 3.6.9 you can still get by date, just find the release thread and get code from before that date (or within a couple days afterwards).
-
Everyone just kept committing to the 3.6.9 branch after it was out. :)
[just wondering]
Is it gonna go the same way in future releases? At least for us users it would be clearer if a branch was dedicated for bug fixes only after the release.
[/just wondering]
-
Ideally, no, it should not be that way.
Some of it depends on the stress levels and numbers of the volunteers.
-
As a person who did an appreciable share of the support on this forum under the older system I'd prefer to keep it this way with a stable and development branch.
However we shouldn't be leaving two year gaps between official releases from HEAD after 3.7.
-
What's planned for 3.6.10? Anything major?
-
all i know is that they are gonna fix the red alert bug
-
. It will have the new FS2NetD2 code that was originally slated for 3.6.9, a bunch of OpenGL fixes, a reworked ships lab, reduced memory footprint, and better multi-format file loading (particularly for music).
the big holdup now is we need to do a LOT of bug fixing before we can consider a point release
EDIT (taylor): all of that info wasn't for public eyes. :)
-
Are they also going to fix the flickering pop-up menus in OpenGL?
It is just an OpenGL problem, and it happens anytime that a popup appears on screen. There are various fixes for this, but they have never been 100%. The code was originally written with a software renderer in mind, but dealing with single/double/triple buffered screens just isn't a clean a simple task with how it's supposed to function. I'll probably end up just rewriting the entire bit of functionality rather than bothering to fix it.
-
I would bugtest and report them if someone started posting serious pre-RC builds. I'm much more hesitant to just grab a build from CVS and start testing.
You'd probably get more community bugtesting then too =)
-
Are they also going to fix the flickering pop-up menus in OpenGL?
No. At least not for 3.6.10 anyway.
-
The only people who can have release dates are people who do this stuff as their job. They get fired if they stop working.
I disagree completely. First of all yes you have release dates for a job, but:
Oh these deadlines, don't you love the wooshing sound they make as they fly by?
On the other hand there has been a movement in most major open source projects from "It is ready, when its ready." to a fixed release schedule. (Gnome, Xorg, Ubuntu, Debian, ...)
Most of the time they did not even hold the date, but they always released earlier than without a fixed release date.
There are several reasons for that (thanks goes to Michael, former debian project leader for a very nice talk about this in Girona, during Jornade SPL this year):
One is, without a date most of the time it goes like this in open source projects:
Release Manager: Okay, now we are stable enough, lets fix the remaining bugs and then do a release.
Devel 1: But well this feature I have I really just need 1 week more and I could get it in.
Release Manager: Okay, of course we can delay for 1 week.
(one week later)
Devel 2: Please, before I was not ready, but now I have made so much progress, just a little more time.
Release Manager: Okay, but then we are really going into the bug squashing phase.
Devel 3: But ...
... and so on ...
We have calculated that the thing devel 1 wanted to have in in the extremst case, was later in, then if he had waited one whole release.
Fixed release schedules (like every year or every 6 months or every 3 months) have several advantages:
- Your users know it and are not asking "dumb" questions.
- Your users anticipate it.
- Your developers know it and will not try to get that 80% working feature in, but will patiently finish it, because the next release is just around the corner anyway.
- You get more experiences with releases and as its always the same time of the year, also with the surroundings (like vacations, summer breeze, motivation, christmas eve, ...)
You do not even have to hold it to the day, but just having a feeling for when the next release will be done is very very good, because it gives _clarity_ to everyone.
And it also seems to work with the help of the subconscious, as you somehow "know" its coming and behave accordingly. Without a date there is more a feeling of - well its still months away.
Funnily enough it also gives people motivation, because they know if they get it in, before the code freeze, it will be in. And they also know that the bugs need to be squashed before the release.
Of course, this is just my opinion, but I found it very helpful.
cu
Fabian