[DEPRECIATED] Campaign 4 (Old thread)

[Rman Virgil]

Bad news, folks. Looks like Cam 4 on 3.1+ is going to require a total reset of the data files (meaning it needs STOCK files) that it uses in order to work. I'm not getting any useful debug messages from Master versions of Warzone in order to fix whatever is wrong.

The departure from 2.3.9 to 3.2 is too vast of a chasm to bridge.

What I intend to do is reset using 3.1 data, and then build back up from there. I fear the next release will be a long, long way off -- but it's completely necessary from where I sit in order to progress.

Play Release-13 on 2.3.9 as much as you like, everyone. Guess it turned out to be unlucky after all.

What leads you to point the finger at your data set as the sole culprit?

Because the original campaign works and the only diff between it and yours is the data set?

Is that the only difference, I wonder, that could result in breakage?

Thinking aloud.

With v.3.1+ a moving target, what specific ongoing changes from v.2.3.9 could make this a recuring pattern going forward?

You say there is no useful debug info to work from so you will introduce new data piecemeal and see if you can isolate any culprits that way?

Perhaps there is not enough debug support even if you proceeded to introduce new data piecemeal? (There are already mods introducing new data with scripting to v.3.1+ that are running - so there must be more at work than that alone, I would hazard to speculate.)

Again, just thinking aloud. (And wishing there was something along the lines of a Rational PurifyPlus debuging tool to work with. Seems too much now like trying to hit a target in pitch black without night goggles or laser scope.)

.

[Goth Zagog-Thou]

I've been thinking about the problem a bit more, and I have a couple of different routes I can follow at this point.

I'll update when I have more info.

[Goth Zagog-Thou]

My original hypothesis (and bug report) was correct after all. The problem is definitely features. The three compiles of Cam 4 on 3.1+ *would* have worked, if not for this features issue.

What I did:

1) I stripped the map bare, and replaced cam1a with it. I retooled the cam1a and cam1a-ai scripts to account for this (aka pretty much stripping them bare also).

2) Placed some units on the map. No problems.

3) Placed some Scavenger stuff on. No problem there.

4) Placed some of the 3.0_b4 features on the map (the new ones). Still good.

5) Copied then pasted all of Cam 4's map stuff onto the map. Removed all T3 units and structures, leaving T1 and Features. Crash.

6) Removed all features, leaving a few T1 structures and units. Works.

7) Added a single oil, then another with an oil derrick. Works fine.

So something is amiss concerning features. And since this is using a working base.wz (stock except for the map and script changes) I can only assume that it's an engine issue concerning either original features themselves (since the new ones work), or my application of those features. I'll test and see for myself, and I'll have a better idea what's going on.

If this turns out to be a hard limit in the number of features that is allowed on a map, I can certainly accommodate that. If it's something a bit deeper than that, hopefully I've provided enough information to give the Dev Team a direction to follow.

[Rman Virgil]

My original hypothesis (and bug report) was correct after all. The problem is definitely features. The three compiles of Cam 4 on 3.1+ *would* have worked, if not for this features issue.

What I did:

1) I stripped the map bare, and replaced cam1a with it. I retooled the cam1a and cam1a-ai scripts to account for this (aka pretty much stripping them bare also).

2) Placed some units on the map. No problems.

3) Placed some Scavenger stuff on. No problem there.

4) Placed some of the 3.0_b4 features on the map (the new ones). Still good.

5) Copied then pasted all of Cam 4's map stuff onto the map. Removed all T3 units and structures, leaving T1 and Features. Crash.

6) Removed all features, leaving a few T1 structures and units. Works.

7) Added a single oil, then another with an oil derrick. Works fine.

So something is amiss concerning features. And since this is using a working base.wz (stock except for the map and script changes) I can only assume that it's an engine issue concerning either original features themselves (since the new ones work), or my application of those features. I'll test and see for myself, and I'll have a better idea what's going on.

If this turns out to be a hard limit in the number of features that is allowed on a map, I can certainly accommodate that. If it's something a bit deeper than that, hopefully I've provided enough information to give the Dev Team a direction to follow.

Excellent trouble-shooting Goth.

If recall from my work on Aqua Coop / ManGodAi4X there is a hard coded limit on each of the original features - think it was 255 but i'm not sure. I exceeded that number by one of my features but I was running in v.2.3.x and just got an exception warning as such while the engine still allowed the level to be loaded fine..... and of course, any of the features above that number were not placed in game.

Like you said.... you may have isolated the culprit enough for the devs to address the issue.

.

[Goth Zagog-Thou]

Yep, which leads me to believe that it's something in the engine causing the above issues.

Attempts to re-base using the "crashy" version that I submitted with my bug report did not work, even after removing all features. It was worth a try, I guess.

So, I plan to re-base it (again) on Cam 1's stuff, and proceed along that route. At least I know it works for sure. It may be a while for the next release, so I plan to take my time on this. I've run myself ragged and I need some "fun mode" again.

At least I can put the *new* features on the map now.

[Rman Virgil]

Yep, which leads me to believe that it's something in the engine causing the above issues.

Attempts to re-base using the "crashy" version that I submitted with my bug report did not work, even after removing all features. It was worth a try, I guess.

So, I plan to re-base it (again) on Cam 1's stuff, and proceed along that route. At least I know it works for sure. It may be a while for the next release, so I plan to take my time on this. I've run myself ragged and I need some "fun mode" again.

At least I can put the *new* features on the map now.

The fun side of it all has to come first. Even at this remove I could feel these issues would be draining the fun of your creative experience and that's the vital fuel you can't do without.
.

[vexed]

Seems it is pretty self-explanatory once you see output that isn't covered up by using --debug=all.

Code: Select all

info    |11:29:10: [_imd_load_polys:96] Invalid polygon size (4)
info    |11:29:10: [_imd_load_polys:96] Assert in Warzone: imdload.cpp:96 (npnts == 3), last script event: '<none>'
error   |11:29:10: [_imd_load_level:589] (_load_level) unexpected directive 6a00 4
never   |11:29:10: [res_parse:118] file: IMD fxgrdexs.pie

Fix your .pie files.
Game tries to load them, it can't, as pointed out above, and then continues to try and load the other files... so you have lots of .pie files that need converting, as was mentioned in the ticket.

[Goth Zagog-Thou]

I thought I was using 3.1+'s .pies, in my crashy version that I uploaded for the devs to check out.

In fact, I specifically recall replacing them all with 3.1+ versions.

Weird.

[Goth Zagog-Thou]

Anyway, after a little breather I'm back and working on Release-14. We're still on 2.3.9 (and I'll keep it civil............). Changes include:

- More informative and instructional dialogue on M3 and M4.
- Better attack code during M7.
- More map & mission fixes for gameplay flow (esp. in M3).
- Added new Achievements for doing some things earlier than expected.
- Added Mission Failure condition if the player loses the NexusLink gained during M3.
- Script Markers enhanced on M1 and M8 maps. Everything should be there that's supposed to be there now.

I'll release it when it's ready for prime time.

[aubergine]

Is there a name for Campaign 4? Also, are there names for the first 3 Campaigns?

[Goth Zagog-Thou]

Nope, no name for Cam 4. Nor do I know of any proper names for the original campaigns other than Alpha, Beta and Gamma, if we go by the Team Names. In this naming convention, Cam 4 would be (of course) Delta.

[aubergine]

Would "Vendetta" be a suitable name? Or is there a new faction in this cam - if so it might be worth naming each campaign after the main new faction in it?

Ponderance on cam names and numbering here: viewtopic.php?f=30&t=9500

[Goth Zagog-Thou]

Feel free to name it something appropriate.

[aubergine]

I'm currently naming the other cams based on the main "new" faction in them -- is there a major new faction in this cam?

[Goth Zagog-Thou]

Now that i think about it, a good (working) title for this cam could be "Reckoning".

And no, no major new faction unless we count the Coalition and the Red Alphas, comprised of remnants of the Military and Politicians from elsewhere (much like The Project) who never really got past Tech 0 (Scavenger Tech).