Spring 89.0!
Re: Spring 89.0!
Please can we fix http://springrts.com/mantis/view.php?id=3184 before release. I believe this issue is going to come up a lot but is currently masked by affected users already disabling eaxreverb due to #2368 and #2928
I'm not sure exactly what causes the issue but I do know it can be avoided by not using the EAXREVERB filter on OpenAL Soft (See https://github.com/SpliFF/spring/commit ... 3aa035a1bd)
I'm not sure exactly what causes the issue but I do know it can be avoided by not using the EAXREVERB filter on OpenAL Soft (See https://github.com/SpliFF/spring/commit ... 3aa035a1bd)
Re: Spring 89.0!
89.0 Debian packages are in experimental
Re: Spring 89.0!
No, plan is to go live with 89.0 as is. 90.0 may follow in 1-2weeks.zerver wrote:You are welcome.
AFAIK 90.0 is the release that will go live.
Re: Spring 89.0!
Good job! Thanks to all spring community!
-
- Moderator
- Posts: 2464
- Joined: 12 Oct 2007, 09:24
Re: Spring 89.0!
I don't think 89.0 is usable because there are a few bugs that would really annoy players.
The selection size bug looks bad because the tooltip flickers and it is hard to select and give orders to units when zoomed out.
People always notice and complain about bugs with targeting.
These bugs are already fixed so why not integrate them in to a new release?
The selection size bug looks bad because the tooltip flickers and it is hard to select and give orders to units when zoomed out.
People always notice and complain about bugs with targeting.
These bugs are already fixed so why not integrate them in to a new release?
Re: Spring 89.0!
(imo) because the fixes already could have broken different things. we still have no regression testing.Google_Frog wrote:These bugs are already fixed so why not integrate them in to a new release?
-
- Moderator
- Posts: 2464
- Joined: 12 Oct 2007, 09:24
Re: Spring 89.0!
Well we just have to hope that non-important things are broken. Also that automated regression testing sounds like a waste of time.
Re: Spring 89.0!
sorry google but... thats both pretty stupid.
Re: Spring 89.0!
Can anyone provide a patch to make it compile with boost 1.49/1.50?
Re: Spring 89.0!
http://anonscm.debian.org/gitweb/?p=pkg ... perimentalburp wrote:Can anyone provide a patch to make it compile with boost 1.49/1.50?
Re: Spring 89.0!
I have mixed feelings on the attack ground thing. On one hand I like that the units target the ground and keep shooting like they are attacking a unit. On the other, I wish it there was some toggle beyond stop command
Re: Spring 89.0!
Just remember to back up your uikeys.txt before you update. That's right, installer WILL OVERWRITE YOUR FILE.
Then after install, first thing you should do is delete the new uikeys.txt and restore the old from backup, otherwise you will have to program it all from scratch.
Then after install, first thing you should do is delete the new uikeys.txt and restore the old from backup, otherwise you will have to program it all from scratch.
Re: Spring 89.0!
thanks for reporting, will try to fix that before next release: http://springrts.com/mantis/view.php?id=3194Jools wrote:Just remember to back up your uikeys.txt before you update. That's right, installer WILL OVERWRITE YOUR FILE.
Re: Spring 89.0!
Installer overwrites my local springsettings.cfg too
-
- Moderator
- Posts: 2464
- Joined: 12 Oct 2007, 09:24
Re: Spring 89.0!
There are always broken things in an engine release. Sometimes they are only annoying to game devs and I much prefer this to things which are annoying to players. As for regression testing I should probably post in that other thread but in short I agree with what knorke said.
I have mixed feeling about this "unit's keep their target" feature. At one point I wanted this feature. But then Spring.SetUnitTarget was added and I implemented it myself in lua so I don't think the engine needs an implementation. My lua implementation draws a command-like line for the target and has it's own command and stop command which have no effect on the rest of a unit's orders. I'll have to see if this engine implementation conflicts poorly with my implementation.
I have mixed feeling about this "unit's keep their target" feature. At one point I wanted this feature. But then Spring.SetUnitTarget was added and I implemented it myself in lua so I don't think the engine needs an implementation. My lua implementation draws a command-like line for the target and has it's own command and stop command which have no effect on the rest of a unit's orders. I'll have to see if this engine implementation conflicts poorly with my implementation.
- very_bad_soldier
- Posts: 1397
- Joined: 20 Feb 2007, 01:10
Re: Spring 89.0!
I love SSMF + shadows so much in LOS view! Thanks!
Re: Spring 89.0!
yeah, that's a very nice improvement. Not to mention the new soundwet sounds, that's great.very_bad_soldier wrote:I love SSMF + shadows so much in LOS view! Thanks!
Re: Spring 89.0!
This targetting 'feature' lets my com keep on dgunning-repeat
-
- Posts: 451
- Joined: 03 Jul 2011, 11:54
Re: Spring 89.0!
In last release (88) all zk hosts upgraded half a week before the server upgraded, hence linux users could not play online if they did not compile from source. Thankfully there are some very helpful peeps at #zk who helped me through my first compile ordeal back then.
Now when 89 was released, zk again bumped their version half a week before the server upgraded which was fine, i had compiled before, i could do it again. But then they decide to revert to 88 once the server (and spring ppa) had gone online with 89.
Since Spring does not support 'make uninstall' it is a real pita to clean up after a compile from source so i did not try bother fetching an old version from git and compile that, and just waited it out to get the zk servers back to 89, which finally happened yesterday. (edit: seems to have once again have been reverted except for a single server).
My question is: how hard is it to coordinate the releases with spring devs / zk/autohost maintainers so these things would not happen? And how hard would it be to make spring support 'make uninstall'? Someone said it would be trivial, then why is it not done?
Some helpful dude in #zk (sorry i forgot your nick) made me some script that i did not understand a bit of, but it did do the same as make uninstall.
My last bit of concern is the ubuntu ppa maintainer. If he would have kept 88.0 in there, and just uploaded a new 89.0 i could have specified the old version to be installed and uninstalled via package management. However 88 was replaced with 89, but for some reason there still was the possibility to get 85? Is this because 85 is from ubuntu ppa and 89 is from spring ppa? Or can there be multiple versions of the same package available in a single ppa?
These are minor annoyances to some hardcore experts, but something that in my opinion could improve the user experience dramatically with minor effort if fixed.
TL/DR; 89 is a great release but the logistics in deploying it seem failed to me.
Now when 89 was released, zk again bumped their version half a week before the server upgraded which was fine, i had compiled before, i could do it again. But then they decide to revert to 88 once the server (and spring ppa) had gone online with 89.
Since Spring does not support 'make uninstall' it is a real pita to clean up after a compile from source so i did not try bother fetching an old version from git and compile that, and just waited it out to get the zk servers back to 89, which finally happened yesterday. (edit: seems to have once again have been reverted except for a single server).
My question is: how hard is it to coordinate the releases with spring devs / zk/autohost maintainers so these things would not happen? And how hard would it be to make spring support 'make uninstall'? Someone said it would be trivial, then why is it not done?
Some helpful dude in #zk (sorry i forgot your nick) made me some script that i did not understand a bit of, but it did do the same as make uninstall.
My last bit of concern is the ubuntu ppa maintainer. If he would have kept 88.0 in there, and just uploaded a new 89.0 i could have specified the old version to be installed and uninstalled via package management. However 88 was replaced with 89, but for some reason there still was the possibility to get 85? Is this because 85 is from ubuntu ppa and 89 is from spring ppa? Or can there be multiple versions of the same package available in a single ppa?
These are minor annoyances to some hardcore experts, but something that in my opinion could improve the user experience dramatically with minor effort if fixed.
TL/DR; 89 is a great release but the logistics in deploying it seem failed to me.
Re: Spring 89.0!
Since the Shard issue is not resolved, can it be removed from the v89 installer before further damage is done to its reputation? The mingw32 build in the installer is just plain broken and does nothing but annoy/confuse, it would be best for all if it were removed from the main releases until a solution is found