Broken AI respawn in popular games since 1.68

[kremator 1064]

FIXED !  Dedi server is working again !!  Well done BIS .... and thanks for the fix.

[The Man Without Qualities 110]

On 10.4.2017 at 10:45 PM, kremator said:

FIXED !  Dedi server is working again !!  Well done BIS .... and thanks for the fix.

** only for crCTI, NOT for BECTI. Mixed test results.

Details: https://feedback.bistudio.com/T124098

 

 

[kremator 1064]

Nope disagree.  The latest PERFORMANCE binary works perfectly for Proman AND for BECTI.  Have tested both in a dedi environment.  No issues for respawning now.

[The Man Without Qualities 110]

22 hours ago, The Man Without Qualities said:

Mixed test results.

of course you disagree. That is why I wrote as quoted. Since we do not know the EXACT server/client setup it is by all means of serious testing absolutely wrong to render this issue as solved. It is solved for YOUR environment with YOUR server setup but NOT solved for my server setup with most probably DIFFERENT settings. And since the bugs apperance was highly depending on the server side DIFFICULTY settings it might have more dependencies.

Also "latest" performance binary is not accurate, I tested with v12 - 64bit - performance binary for both server and client. Other options were 32bit or profiling set.

If you say "latest" and todays date, you talking about v15? That was indeed not the version I was testing. But the version rendered as 1st cure for this bug was v12.

 

I tried to reach you via Mücke but it seems you are hard to catch. My intention is to compare setup and method of testing.

It is obvious that when I have a working setup that was working fine for many other missions and even fine for crCTI BUT not for BECTI, that it might happen to other A3 user and hence one cannot consider this bug as fully solved.

At least that is my understanding on regression testing with focus on high quality.

Please ask M for my contact details, I tried to peer on steam - no success.

I even made up a new BECTI with latest BECTI - repo content for testing so that we can share the very same mission - I failed to find a common transfer path to get it to your dev-server.

I am ready to help and conduct joint testing.

It could be a wrong setup or method on my side, but that needs to be analysed.

 

Edit: I tested with v15 - also negative for BECTI taken from git repo but positive for standard BECTI 0.97

[kremator 1064]

Your test therefore makes me think that something has gone wrong with the git repo rather than something BIS has done.  From my testing, like you, I've found no problems with Proman CTI or with Becti 0.97

[The Man Without Qualities 110]

I found the issue why I had false-negative results with a BECTI 0.97 before. I selected a mission with 0.97 in name, but it was not the very early one submitted by Benny, it was in fact a altered version.

When I selected the BECTI 0.97 from Armaholic storage the respawn works fine with v15 and also with v12 after all.

 

I am in contact with Benny already to find out if mission is bugged, my .pbo packing messed up the mission (which would be the 1st time, but never be sure) or another aspect of the spawn function is bugged.

[The Man Without Qualities 110]

@BIS: When we can expect the fix into stable?

I am asking because the game is in fact ruined for my purpose, for months now.

I moved to dev-branch but the 1.71 dev is unplayable, totally broken.

So I have no game right now.

[The Man Without Qualities 110]

@dwarden I just read the 1.70 Notes, a few respawn related fixes mentioned, but does not sound like this exact issue, can you confirm that the fix for the respawn issue is part of 1.70?