magnitude of changes in ba:test (Re: BA sea balance)

magnitude of changes in ba:test (Re: BA sea balance)

Classic game design, maintained to please you...

Moderator: Content Developer

Locked
User avatar
Silentwings
Posts: 3720
Joined: 25 Oct 2008, 00:23

magnitude of changes in ba:test (Re: BA sea balance)

Post by Silentwings »

Split from viewtopic.php?f=44&t=36514&start=20#p584109 since I've no wish to discourage discussion of sea balance.
Ares wrote:changes ... 10,000+
[Fx]Doo wrote: in fact, 8k commits total ... 2.1k since 9.46 ... Talking about how far new ba will be from 9.46, based on the amount of commits is biased

rumors of BA being completely different ... please, at least provide ... relevant data.
What would you regard as a relevant data? Commit count has its flaws but is commonly used as a very rough guide to the magnitude of changes. I'll offer my best attempt: in this case my preferred choice of data would be the changelog itself - https://github.com/Balanced-Annihilatio ... ngelog.txt - which records individual changes that devs felt should be communicated to players. Eyeballing the changelog, the magnitudes/types of change recorded per line seems very consistent between releases, so this seems to be a fair comparison. I've no wish to judge the quality of the changes.

A quick bit of command line magic produces: BAs 9.46->dev changelog currently lists 144 (lines of) changes. The average number per previous stable release, based on the most recent 20 releases, came out as 6.15, so my script concludes that a stable release based on the current ba:test will be approximately 2,250% more different than 'usual'.

Of course, "completely" different is not something I can define and test for so easily... but seems clear the current situation is well outside of normal. It is without question the largest set of changelog entries ever made to BA-dev versions without a stable release, and by a comfortable margin. To find comparable events we need to go back to 2010/11: 7.50->7.60 has 90 changelog entries, covering approx 3 months of 2011, and 7.04->7.1, with 84 changelog entries, covering some unrecorded period of 2010. These are the 2nd and 3rd largest changelogs ever. Even if, for some generous reason, you ignore the lines referring to sea from the current changelog (there are only 10), it remains by far the largest ever.
I would appreciate that you would stop judging a work in progress based on your expectations of an "Official Release"
I appreciate your point, but In the absence of any new release for well over a year, which I believe is also an all time record, I can see why even people with the best of intentions will be trying.

Worse, by not producing a release that is Spring 104 compatible (and which the community wants to play) BA hosts are left vulnerable to a serious DOS exploit in Spring 103 - see e.g. viewtopic.php?f=1&t=36547#p584103.
User avatar
Forboding Angel
Evolution RTS Developer
Posts: 14673
Joined: 17 Nov 2005, 02:43

Re: BA sea balance

Post by Forboding Angel »

Sometimes changes are made and committed, and then changed back. This results in 2 commits for a null change.

That sort of relevant data.

Counting commits is kind of stupid really, considering the context of a spring game.

Sorting the unitdefs into proper folders resulted in massive commits, yet there was no actual change.

Anything done behind the scenes might result in a ton of commits that don't actually effect anything except stability of the game or it's elements while yielding no noticable changes to the end user.

What puzzles me is that you know better than to assert that counting commits is some sort of reliable data of change. Commits show activity, not how much actual change has occurred.
User avatar
very_bad_soldier
Posts: 1397
Joined: 20 Feb 2007, 01:10

Re: BA sea balance

Post by very_bad_soldier »

Forb, read his post again. He did not count commits but the number of changed lines between 9.46 and current dev branch.
User avatar
Forboding Angel
Evolution RTS Developer
Posts: 14673
Joined: 17 Nov 2005, 02:43

Re: magnitude of changes in ba:test (Re: BA sea balance)

Post by Forboding Angel »

That's even MORE irrelevant.
User avatar
triton
Lobby Moderator
Posts: 330
Joined: 18 Nov 2009, 14:27

Re: magnitude of changes in ba:test (Re: BA sea balance)

Post by triton »

CHANGELOG is big they mean, and yes, big and ambitious. So what?
User avatar
Silentwings
Posts: 3720
Joined: 25 Oct 2008, 00:23

Re: magnitude of changes in ba:test (Re: BA sea balance)

Post by Silentwings »

So what?
I have no strong opinion on the "so what" part.

@vbs/Forb: it look as though you are both mistaken, as said above, my script counted the number of changes recorded in changelog.txt, for past and present stable releases.
User avatar
Silentwings
Posts: 3720
Joined: 25 Oct 2008, 00:23

Re: magnitude of changes in ba:test (Re: BA sea balance)

Post by Silentwings »

Fyi, my reason for writing the OP is viewtopic.php?f=44&t=36553

It looks blindingly obvious to me that BA should release a stable update for 104 without further delay. I don't care whether it is done through major reverts, migration of testing versions/commits into branches, or actual progress.
User avatar
Jools
XTA Developer
Posts: 2816
Joined: 23 Feb 2009, 16:29

exploit in 103 (split)

Post by Jools »

Is the exploit in 103 relevant to other games as well? Just asking because nobody wants to have a game with an exploit.
User avatar
Silentwings
Posts: 3720
Joined: 25 Oct 2008, 00:23

Re: magnitude of changes in ba:test (Re: BA sea balance)

Post by Silentwings »

Yes, but there have not been any reports of it being used elsewhere.
User avatar
Forboding Angel
Evolution RTS Developer
Posts: 14673
Joined: 17 Nov 2005, 02:43

Re: magnitude of changes in ba:test (Re: BA sea balance)

Post by Forboding Angel »

Or, ya know, you could do your damn job and permaban them from the spring server altogether, which should be done regardless of what engine version BA is on.

Instead, you want to put it on the BA team as if the ball is solely in their court? Please. Do your job or notify someone else who will.

User was warned for this post, felonies 4,6. (Silentwings)

User received mandatory ban for multiple warnings in a short time period.
User avatar
Silentwings
Posts: 3720
Joined: 25 Oct 2008, 00:23

Re: magnitude of changes in ba:test (Re: BA sea balance)

Post by Silentwings »

As said, we issue bans to people caught using the exploit, when it is reported to us.

Unfortunately, due to VPNs and proxies, as well as the fact that the exploit can be triggered without any traffic passing through our own servers - this is not effective against determined offenders.

It is not reasonable to expect moderators to become a catch-all system for dealing with security holes long after fixes have been issued for them. (And I doubt anyone seriously expects us to be.)

Consequently, it is strongly recommended that all affected games update to Spring 104. The public post viewtopic.php?f=44&t=36553 is the result of them not doing so.

Locking my own thread.
Locked

Return to “Balanced Annihilation”