Ladle 116

A place for threads related to tournaments and the like, and things related too.

Moderator: Light

Post Reply
User avatar
Renegade
Core Dumper
Posts: 162
Joined: Wed Feb 06, 2013 7:44 pm

Ladle 116

Post by Renegade »

Sign-ups are open here here.
Image
User avatar
Titanoboa
Reverse Outside Corner Grinder
Posts: 1795
Joined: Sun Feb 22, 2009 8:07 pm

Re: Ladle 116

Post by Titanoboa »

ggs all! Thanks everyone who set up the wiki, hosted servers, randomized brackets, recorded, and most of all: played! Was nice having quarter finals again. :cat:
syllabear
Shutout Match Winner
Posts: 1030
Joined: Fri Oct 13, 2006 1:37 pm
Location: UK/HK

Re: Ladle 116

Post by syllabear »

Congrats twixted arrows, thanks to the players who played for open team!
The Halley's comet of Armagetron.
ps I'm not tokoyami
User avatar
Z-Man
God & Project Admin
Posts: 11585
Joined: Sun Jan 23, 2005 6:01 pm
Location: Cologne
Contact:

Re: Ladle 116

Post by Z-Man »

User avatar
delinquent
Match Winner
Posts: 760
Joined: Sat Jul 07, 2012 3:07 am

Re: Ladle 116

Post by delinquent »

I can't seem to watch these recordings. They start normally, but then drop to a screen stating that the connection is lost. It seems quite random, but always within the first few minutes. I also noticed timer hiccups, and this is on both 0.2.8.3.4 and 0.2.9.0.
User avatar
Z-Man
God & Project Admin
Posts: 11585
Joined: Sun Jan 23, 2005 6:01 pm
Location: Cologne
Contact:

Re: Ladle 116

Post by Z-Man »

Ah. Maybe that it did work reasonably well with 0.2.8.3.X was just a happy accident. We're building differently now, which mostly changed the Linux version I use for recording. I'll probably have to go back 0.2.8.3.5 for that, then.
User avatar
delinquent
Match Winner
Posts: 760
Joined: Sat Jul 07, 2012 3:07 am

Re: Ladle 116

Post by delinquent »

Is it concievable that GCC might play a part? I'll try with the non-gcc version later.


Edit: Yep. Non-GCC version was completely stable. Which is odd, because the opposite is true when I'm actually playing.
User avatar
Z-Man
God & Project Admin
Posts: 11585
Joined: Sun Jan 23, 2005 6:01 pm
Location: Cologne
Contact:

Re: Ladle 116

Post by Z-Man »

What non-gcc version are you speaking of? I don't think we have a recent one. The windows builds of 0.2.8.3 didn't have it in their filename, sure, but they were still compiled with it; mingw, to be precise.
User avatar
delinquent
Match Winner
Posts: 760
Joined: Sat Jul 07, 2012 3:07 am

Re: Ladle 116

Post by delinquent »

https://svwh.dl.sourceforge.net/project ... .win32.exe is the version I have installed.
https://download.armagetronad.org/stagi ... .win32.exe is the version of 0.2.9 I tried after 0.2.8 failed.

The build I used successfully was this: https://download.armagetronad.org/stagi ... .win32.zip

I'm not sure if there's any difference between the prefabbed zip and the installer, I just presumed that the lack of GCC in the filename denoted some other build.

At any rate, the only version that I could watch the entire match through was the prefabbed zip. I haven't tried 0.2.9 of any nature on Fedora yet.
User avatar
Z-Man
God & Project Admin
Posts: 11585
Joined: Sun Jan 23, 2005 6:01 pm
Location: Cologne
Contact:

Re: Ladle 116

Post by Z-Man »

That's curious, the gcc-less zip comes from the exact same build as the gcc-installer, it's just packaged differenly. I'll test it myself later. CPU differences and who knows what else also can play a part, so we'll see.
User avatar
Z-Man
God & Project Admin
Posts: 11585
Joined: Sun Jan 23, 2005 6:01 pm
Location: Cologne
Contact:

Re: Ladle 116

Post by Z-Man »

0.2.9.0 on Windows plays the recordings back fine for me.
What does not, however, is anything from 0.2.9.1. That would be the alpha branches at the time you were testing and the 'staging' branch on Steam. For me, somehow, a simple doubleclick on the recordings would initially prefer the Steam version now. Could that have happened to you? Were you on alpha or staging on Steam?
(Current Steam default is affected, too, but it hadn't been at the time of your report.)

I tracked the cause of the playback failure to a security fix hardening. The location of the vulnerability had already been fixed, but the playback breaking change would have stopped the issue from appearing in the first place and makes a function behave correctly, that is, do what its name says. It needs to stay in. The issue would have been the same had the recordings been made with 0.2.8.5.X.
User avatar
delinquent
Match Winner
Posts: 760
Joined: Sat Jul 07, 2012 3:07 am

Re: Ladle 116

Post by delinquent »

I won't support Steam, so nope. The issue persisted on 0.2.8.3.4, though, most definitely.

I'm aware of an odd issue with my machine, mind you. My processor will happily tell you it supports AVX2, depending on which library you use to ask it, when in actual fact it does not. I don't see how this would necessarily be relevant to Armagetron, but it might be indicative of another issue somewhere.
User avatar
Z-Man
God & Project Admin
Posts: 11585
Joined: Sun Jan 23, 2005 6:01 pm
Location: Cologne
Contact:

Re: Ladle 116

Post by Z-Man »

We don't use AVX2 (on purpose; compilers can auto-opimize for extended command sets sometimes), but other floating point differences may be the trigger.

Try this build. Primarily intended for more reliable debugging of recording sent by players, in case of certain problems, it switches off strict playback of the network layer. It leaves gameplay and processing of network messages from the server. Right now, it won't have the correct ping on playback, so an actively controlled cycle would be a bit off at all times, but purely spectating recordings look all right. Better than having them not work at all.
Post Reply