Gametic Unlagged Testing (Current Build: Zandronum 2.0-alpha-150118-1152)

[Dark-Assassin]

It would be a better idea just to install OpenSSL
http://www.openssl.org/related/binaries.html

Edit: Maybe not. It's specific to the binary, not the one user.
You may also need ssleay32.dll as well, but not sure if necessary.

[Kara Kurt]

To Uwe L. Bendoverson: I see you've made an instagib server. Could you please also make one with buckshot? Playing with high ping would make the testing better imo.

[Uwe L. Bendoverson]

Testing so far has been fantastic. A huge improvement over ping-based unlagged from what has been seen.

[Kara Kurt]

Sick ass testing session indeed. Lookfing forward to playing some CTF next time!

[Ruin]

Sick ass testing session indeed. Lookfing forward to playing some CTF next time!

This. I would love to see a ctf session.

EDIT:
I strongly encourage those with some time to spare to give this a go.So far, I like the way the game plays now.

[Uwe L. Bendoverson]

Added an IDL CTF server, get going.

[DevilHunter]

While we are on the subject of testing this.. I can't for the life of me.. Figure out why this will not build on Linux. Ofc, we have been using Ubuntu for the whole time. And it never failed so hard as This.

Even after hunting for the right lib files, I'd expect it to work but sadly, it don't.. If anyone can help me on this, be my guest.

And yes, I understand I put this in 2.0 folder.. That was My bad, I'm burnt out right now, and I moved it so its all good on that part.

[Medicris]

Just tested CTF over the course of about an hour playing against some people who used the Gamer's Proxy. The code seems very solid and I didn't notice a single whiff.

On another note I'd go as far to say as Striker's damage numbers should be an included toggleable feature in future builds simply for the competitive applications and things like duel spectating.

[Dusk]

I must say I never expected the community to not only be able to organize a testing session on its own but also make a build of Zandronum while at it - and a sandbox version no less. I'm throughly impressed and Torr appeared to agree with my view.

I'm sorry I missed it. I was sleeping.

[Torr Samaho]

I'm very happy to hear about your testing efforts! From the feedback so far, I'm tempted to pull the gametic unlagged code into our main repository. Can you elaborate the details of the testing session? I.e. number of players, play time, game modes, weapons used, etc. Where there any complaints by any of the testers? Are you sure that the code is not overcompensating? For instance, for the SSG there is a delay that shouldn't be removed by unlagged.

While we are on the subject of testing this.. I can't for the life of me.. Figure out why this will not build on Linux. Ofc, we have been using Ubuntu for the whole time. And it never failed so hard as This.

GCC simply does not accept this code, it's not a problem with your setup. I didn't notice because I only compiled the code with VC++. Should be fixed now. Also, the gametic_unlagged hg bookmark seems to be where it should be now.

On another note I'd go as far to say as Striker's damage numbers should be an included toggleable feature in future builds

Are you talking about striker_showdamage_v3.pk3? If it works fine as mod, why don't keep it as mod?

[Ruin]

I'm very happy to hear about your testing efforts! From the feedback so far, I'm tempted to pull the gametic unlagged code into our main repository. Can you elaborate the details of the testing session? I.e. number of players, play time, game modes, weapons used, etc. Where there any complaints by any of the testers? Are you sure that the code is not overcompensating? For instance, for the SSG there is a delay that shouldn't be removed by unlagged.

From the time I was there: TDM, instagib, and ctf were tested. Both TDM and instagib sessions for me were relatively small, about 4 or 5 players. During the CTF session, we peaked at 8 players in-game with several spectators. I honestly did not feel the code was overcompensating, personally. I did not see any complaints from most of the testers. One person (capo) complained about jittery players, but that was about it. I did not notice any player jitter on my end and nobody else aside from capo made that complaint. I saw mainly positive remarks throughout.

[DevilHunter]

Thanks again Torr. I finally got it built. if anything from a Spec's POV.. it looked better then before.. no more Ghost shots, or delayed shots rather. I tried playing myself but only a lil bit because everyone left then; they been playing for awhile. I noticed that my shots connect better then before. GJ on this.

[Uwe L. Bendoverson]

We've tested the Instagib and normal DM modes for several hours, peaking at over 10 players at one point.

It worked out extremely well, very noticeable in Instagib when we tried switching between ping and gametic unlagged. In ping unlagged, people were missing valid shots, whereas in gametic unlagged, things worked much better.

We also tested highly fluctuating connections via Gamer's Proxy, and things still remained very playable.

[TheCitrusKiwi]

Here is a little demo of mine playing at ≈ 200 ping
Demo


We played mostly DM and TDM with instagib ( if you skip the 3 first maps) "demo_skiptonextmap" just incase you don't know ;)
and buckshot as modifier most of the time

also got some screenshots for some feedbacks ingame:
Screenshots

Thanks to TheEnigma for hosting the server :)

[Kara Kurt]

I'd like to get more high ping players to play CTF next time, I was the only one playing CTF. Unlagged was decent and consistent, but felt like something was wrong. Could be the server!

Edit: I also managed to make a ZD purist love Zandronum 1.4... He did say that Zan alot smoother and "In Zandro this aiming was cool" compared to ZD.

[Torr Samaho]

That sounds all very promising!

Meanwhile, I went through the implementation details once more and found one problem that happens when the "unreliable" movement packets from the server arrive at the client in wrong order. I guess this happens rarely, so fixing this shouldn't change how things behave too much. Nevertheless, I'd like to fix this problem before the next testing session, so that we can be sure that the fix doesn't break anything. Do you already have a time in mind for the next session?

EDIT: I updated the code to fix the problem with packets in the wrong order. The gametic_unlagged bookmark is updated accordingly.

[Uwe L. Bendoverson]

I'm ready tuesdays, and any weekend.

[Torr Samaho]

There may be another small problem with packets in the wrong order (this part affects only the client -> server packets), but I don't have time to take care of this today. If you do more testing in the mean time, please use the latest version of the gametic_unlagged bookmark, which I updated today.

[Danzoa]

I'm down for testing, I'll even host the testing event if needed, I'll gather up information about when people can get on or not, and put an average time for those who are willing to play during weekends or even weekdays so we can get this implemented, since it sounds very promising so far.

If this really is much better than Unlagged, then we can finally move on to a future where we don't miss shots that shouldn't have, or just have players cry about it when they're not getting it because it's just a valid excuse.

[Kara Kurt]

There may be another small problem with packets in the wrong order (this part affects only the client -> server packets), but I don't have time to take care of this today. If you do more testing in the mean time, please use the latest version of the gametic_unlagged bookmark, which I updated today.

Aw, hope it gets fixed soon!