Let's talk about the exit/enter situation. My design goal is for the new exit/enter behavior to be completely identical to what it was. One tricky part there is how to handle buffered input that reaches TW before it shuts down and restarts. I currently hand any remaining buffered data off to TWGS, which then hands it back to the new TW session when it restarts. In my tests, this worked 100% of the time. But clearly it is failing intermittently. So I need to put more work into that. Whether or not scripts are broken, this transition breaks a fundamental rule of TW IO, which is that data that you send should never be "eaten" by either the server or the game.

So for this one problem, it is my goal for TWGS to fix the problem, not Mombot or other tools.

For the other issue T0yman points out, that change was necessary to fix a bug. I suppose we could discuss other alternative solutions to that bug, but the reality is that the formatting of that screen does not fully support the range of values that it should display.

I think the waitfor / waiton was added by most people added before you went and placed the buffer to correct the issue. I never tested it afterwards since I already had it done, has anyone else?

I'd appreciate some feedback on whether the old exit/enter script works on v2.19 where I'm doing this buffer handling. It might save me some work.

I will remove the fix and test later tonight.

The majority of issues people are seeing in 2.18+ is from timing changes imo. The helpers can't keep up with the text as it passes by, especially when considering spaces for abort's.

There may be a deeper root cause, but speaking only for myself. I don't have script issues outside the changes to relog and @ in slower ping games. Servers with low emulation tend to cause me the most issues with scripts like wsst/worldtrade/anything that aborts and moves "fast".

It appears to me the client's cant keep up with the server and are missing commands. This has been verified by the swath crashing and I have to assume it's happening with TWX also.

I don't have TWX anymore, at least on my win 7 machine.

From the posts I'm reading, am I correct in assuming that TWX is not windows 7 compatible?

Also, seeing there are two versions of TWGS still out there, would it be just easier to have two versions of Mombot as well, one for each version of TWGS? While I could write some simple scripts, it seems like there is too much of a difference between TWGS v1 and TWGS v2 for one version of Mombot to work with both TWGS versions.

I know I've been out of the loop for a while and I'm still trying to catch up on older posts about what's been going on over the last year.

twx 2.05 works just fine with windows 7. www.navhaz.com you can find it there.

Mombot works in both versions for the most part. There is one issue that needs to be addressed and I have not had time to work on it. TWX 2.05 is Win7 compatible but has some issues of it's own. Swath still works like a champ.

After talking with a few people today it seems some Sysops are running servers in the < 70ms range and that is extremely unstable. I did a lot of testing and anything under 70 ~ 75 gives very radical results on some scripts(but not all). 25ms is just crazy fast and there is a warning that warns sysops that this can cause problems, warning is there for a reason.

If were playing ping wars to get players this will only last for a short time and then they will start leaving for stability.

Just a note to talk about "Ping" I think you are talking about move timing not ping emulation, correct me if I am mistaken please.

The move/attack delay is the referenced in the Center, I just forgot to actually call it that... my mistake.
The ping wars part kind of goes with that if you giving people the ability to move so fast no helpers can keep up it "may" be causing problems with scripts. I don't run a forced latency but if I recall I never could ping TWGS with <90ms (local) now you can ping servers at true ping. I can't say if that is causing a problem since I have not tested it but should not be to hard to set up and play a game. Just seems people are getting erratic results using scripts that have worked for years. Of course a few things have changed as well.... TWGS 2.19 / TWX 2.05 / OS-Win7 .... maybe we need to take a look at all the pieces and see if we can sort it all out.

I definitely suspected that fast move times were behind this, but Darkstarbase says he's not running fast moves. He's running 250 ms. He has run 0 ping times, but I'd really be surprised if that causes any instability. That wasn't put in to slow things down, it was put in to level the playing field.

I'm not sure I believe that the problems are related to speed. The timings that are in there now have been in there since v2 first released, but some of these issues are just now showing up.

Then why does a server at 0-90ms behave and respond noticablly faster than a server 100-200+. To me it would seem that if at v1 twgs the average ping was 150-300 and now were in sub 60 range on V2 that it might affect things, Is this a unfair assumption?

Ping emulation has nothing to do with data throughput. It has to do with response time. On local testing, a steady steam of commands generates about 98 Kps output with either 150 ms ping or 0 ms ping. Ping emulation does have an impact on the speed of a response from a given command, which is why I think it's better to use ping emulation, so no single player has too great an advantage because of superior network performance. But I don't see evidence that it's causing script instabilities. From what I've seen, the issues we're seeing are intermittent disruptions in large streams of output.

My understanding of emulation is that it bridges the gap of a slower player ping with a faster player ping therefore making it as even as possible.

The issue as I see it is that as the server increases response times text speeds up on player side. At some point usually sub-100ish ms; depending on the players computer irregularities start happening. The speed of the text outpaces the helper. As evidenced by swath not updating right on lower emulation servers.

Is there difference in response between a V1 server running default and a V2 server running a 0ms ping emulation?

Or do I have the whole Ping emulation thing backwards?

One of the biggest problems with scripts and the new TWGS is the way they are written.
Case in point , an old public favorite that has worked flawlessly for years is
just now beginning to show its ugly fangs ....

_ck_buydown

mainly the includes are the issue

_ckinc_CNsettings.ts
_ckinc_getInfo.ts

they were written to work with slower connections , but now that things work nicely
these scripts show their flaws.

You do not loop with kill all triggers and send the script back up to a long list
of set triggers ..

I re-wrote my 2 includes to work exactly like the original
with the exception of looping back and killing all the triggers as it went
down the list.

Now the only other real problem is the differences in TWX , the win7
version rounds up while the twx xp version rounds down.

I am still experimenting with a thought that may have fixed that too , but
till I get confirmation that this has been fixed I will not share that yet.

So with this info I hope to put J.P.'s mind to rest .. he has been chasing his tail for so long even with my advise , trying to get him to include a 10 or 12 step questioning to the user who reports a so called broken script or bug ..

I do not want to sound like I know it all , but this bit of evidence (includes)
may show others I'm on the right track .

just my take
take it or leave it

Vid