Release Message
The Nineties! Break down the wall, surf the Internet with 56k, watch the Fresh Prince, ... Future we are back again :3

Kudos
Again, thanks everyone for your contributions!

Download links:

• Windows installer
• Windows portable
• Source (tar.lzma)
• Source (tar.gz)
• MacOSX

Major changes:

• fix Linux sync errors
• fix QTPFS deadlocks
• + all of 90.0

Full changelog here.

I love Summer.

Thx for fast reaction to all contributors and I hope this will go live ASAP. Its quite a long time since all games were running one stable version :).

Really fast! Great work!

Yeah, its been amazing to see bugs reported fixed the next day! Insanely awesome work guys <3

Thanks for fixing so much stuff and also making new release with fixes

buntu packages ready: https://launchpad.net/~spring/+archive/ ... /+packages

will zk use 91?

Ask in the zk subforum

Probably not, 91.0 bombers don't work properly...

They don't return to reload sometimes and instead jump on ground and they dont bomb sometimes.

Just to say that roughly the same issuse with bombers occur in BA, even with the relevant bits of lua disabled (although maybe I haven't absolutely everything that affects it). Will mantis later today.

Edit: Think this was the main issue: http://springrts.com/mantis/view.php?id=3264

When i was working on NOE AI in May, there was some engine bug crashing the game, which i located in situation:

Spring.GiveOrderToUnit + Attack + "attacking unit is bomber" + target unit was destroyed + (the order queue was empty OR the "the attack order on dead unit was second in queue") ... im not sure with the last one...

I was not able to reproduce it everytime (maybe in 10%), so i was not reporting it (yep and log is lost, too).

All autohosts of ZK running on 91.0, is it solved then?

No.

Where is the apparent disconnect between bug-testers and engine-deveopers?

I hope all mods now will use same engine version...

It's due to the lack of multi engine support in other lobbies. We can't test this multiplayer or rare issue stuff.

Trust me, if there was some organisation going on for testing, there would be much more of it.

I compiled the almost-latest git and did some testing and reported back to a thread that asked for testing. If someone had steered me to a test gathering i would have gladly joined, but it seems there is a wide split between the developers and the end users.

Why not have some sort of reward.. devs only read and reply to your feature requests if you pay with the points of test!..

Congratulations to zerver for his efforts in mt-spring, since this release i have been using it and had zero regressions or crashes compared to regular spring!

edit: forget that, i get semi-regular desyncs with mt-spring