Operation Seelowe released.

[YazzyT]

@YazzyT The scenario is set at Emperor, so use that. Check the Rules file for the fighter flag. This is what you should see (for the He-111 as an example):
He-111, nil, 1, 36.,6, 13a,4d, 3h,2f, 4,0, 0, Ato, 000000000000001 . The "can attack air units" flag is the second from the left in the last field in the row. It should be set to '0'. If it's a '1' then there's a mistake and you should just change it. Otherwise, I have no idea what the problem might be.

I'll be reworking Sealion this fall. I rushed its release, and a number of errors have been pointed out, so I'm going to revisit it. Right now I'm finishing up Burma Campaign, which has been the most buggy, frustrating scenario I've ever done. I hope to have it back to playtesters by early September, if nothing else goes sideways. Then I'll give Sealion the same treatment. In the meantime, I hope you can fix that bug and carry on.

Rules.txt looks good. I should clarify that the unwanted shoot-downs seem to happen when the AI puts fighters with a range of 1 out past their fuel limit somehow, because the AI cheats. If it's their first turn to move then they cannot be shot down.

 

[techumseh]

Rules.txt looks good. I should clarify that the unwanted shoot-downs seem to happen when the AI puts fighters with a range of 1 out past their fuel limit somehow, because the AI cheats. If it's their first turn to move then they cannot be shot down.

But shot down by bombers? Weird. I must say that since I started using ToTPP for scenarios, all kinds of strange bugs have occurred. I don't think it's as stable as the Nameless One assumed.

 

[YazzyT]

But shot down by bombers? Weird. I must say that since I started using ToTPP for scenarios, all kinds of strange bugs have occurred. I don't think it's as stable as the Nameless One assumed.

I mean, I've seen similar stuff happen in MGE. May be the weirdness of how AI handles air units. I don't know if making RAF fighters have a range of zero would help? Even ground and sea units can attack fighters after they end their turn outside a city. I have seen this sort of thing in other scenarios. I also feel like the RAF is slightly too strong and that the range of RAF fighters may be too big. I think also the range of German bombers should be increased dramatically, for realism and to allow more strategic bombing operations.

 

[CurtSibling]

But shot down by bombers? Weird. I must say that since I started using ToTPP for scenarios, all kinds of strange bugs have occurred. I don't think it's as stable as the Nameless One assumed.

Perhaps it is time for a fresh install of your CIV2?

I have to admit, I have not seen much odd happenings with ToTPP at all. I've never seen a non-fighter aircraft take out a fighter even in my MGE days.
I stick with version 13 of ToTPP, which I consider the most stable. After that, I was aware some bugs had appeared...

 

[tootall_2012]

I worked with ToTPP 15.1 for Napoleon and didn't encounter any unexpected issues or problem with it. As far as I recall there are no ToTPP features related specifically to air roles so I don't see how it could be related to what you are experiencing.

 

[YazzyT]

Bombers who stop their turn outside cities cannot get shot down, only fighters. Maybe it has to do with the role assigned to the units? Are the fighters air superiority or attack?

 

[techumseh]

I can't say that I've had YazzyT's experience with air units. However, my experience with ToTPP is very unsatisfactory re: events. (please spare me any lectures about the advantages of Lua. I will get there eventually, but right now I just want to finish this damn Burma scenario.)

My first attempt to make the Burma Campaign scenario was 3 years ago. I got quite far along, even sending a beta version out to playtesters. Just about that time, something went haywire with the events. Sometimes the events file would load normally, and sometimes different events would run instead. These were an older version of events that had somehow become imbedded in the Scn. file itself. They would run even though the original events file had been changed, even if there was no events file in the scenario folder at all. In ToT, unlike some earlier FW versions, the events are not supposed to be saved in the Scn. file, but load each time the scenario is launched. After weeks of frustrating struggle to find a solution, I abandoned the project.

My next project was Operation Sealion. While I didn't have the problem of phantom events, I did have unusual problems with the events. While there were a number of errors in the events file, which Prof. Garfield has found (thank you PG), some perfectly valid events just wouldn't run. These seemed to be mainly flag/ mask events using the 'continuous' modifier. I was forced to simplify the events and used technologies as triggers instead of flags in a number of places. Now another bug has popped up with the scenario, and I'm inclined to believe that ToTPP is the culprit, rather than something on YazzyT's computer.

Having released Sealion (prematurely it now seems) I started Burma over from scratch. This time, I was careful to save the original pre-events sav. file so I could remake the scenario file if something went wrong. And it did. At first, everything went swimmingly, but then it happened again. This time, I was more methodical (and calmer). I relaunched the scenario 20 times and found that the phantom events loaded about 50% of the time. I tried to narrow down the commonalities between the processes of creating the two versions to the best of my recollection . Two things struck me. In both cases, the problems began after I had added lettering using the 'landmarks text' feature of ToTPP. Secondly, I had saved the game as a scenario, then launched that scenario and saved a game from that, then used that save game to create another scenario file. The advantage of this is that it got rid of a bunch of "We love the Admiral" texts that the "Clear all Disorder and We Love the King" key didn't, and saved the scenario parameters. So, I remade the scenario from the original save file, following the process above. I made sure that there was no events file in the folder during the process, adding it only after I had created my scenario file. And I've refrained from adding Landmarks Text. The phantom events problem has not recurred - yet. But...

 

[techumseh]

There are serious problems with the events. First, flags that are supposed to be 'continuous' don't stay on. I tested this with little text messages. Some stay on for a few turns, others just for one. And some simple events don't work every time. For example, on turn 4, the Chinese are supposed to get 3 T-26 tanks. Most games, they get only 2, and sometimes none. I can live with random tanks for the Chinese, but the continuous flag failure will require yet another re-working of the events and the loss of several important strategic features, eg. the ability of the British to cut Japanese supply lines.

I think the phantom events bug may be quite different from the other type of event failure. But I place the blame for both squarely on the Patch Project. I have never had these problems before patching my ToT exe. I suspect that the flag/mask fail may be caused by the expandable events memory of ToTPP. In both Sealion and Burma I utilized well over 200kb of events memory, over double the standard amount. And I used a number of 'delay' events, which require additional memory. It seems to me that the game just isn't reading some of the events, or may not be able to store all of them.

Btw, I've used two different computers to work on and play these scenarios. My old desktop still uses XL, and my laptop runs Windows 7. Same issues on both. I used ToTPP v.14 on both scenarios. Ciaran, I'm interested in why you think v.13 was more stable. Can you elaborate? Thanks.

 

[JPetroski]

There are issues with ToTPP for sure. The question is if we can trouble shoot what they are and find a solution. Prof. Garfield was able to figure out one issue that was causing us to have headaches with certain tiles not "reading" certain events (in lua however) and I wonder if this has anything to do with some of your issues, such as the tanks. Did they always appear on a particular tile or was there a chance that they could appear on one of several random tiles? If random, have you noticed which tiles they will show up with and which they won't?

I've also experienced weird things with Red Front though I suspect I could solve it by adding a Cosmic 2 field (units have infinite movement so they're probably defaulting to that since Cosmic 2 is not in the rules text normally). I just haven't been bothered to figure it out yet.

I know you don't want a lua lecture but I will say that after having tested things pretty thoroughly for about 1.5 years now through 2 scenarios of my own as well as Napoleon, I think they are fairly stable, work as intended, and can be troubleshot. Macro.txt might not work very well with new ToTPP versions. Then again if Curt has found earlier, non-lua versions to be more stable, there's no reason one can't design their scenario with that if they aren't ready to make the jump to lua just yet. This is an easy game to copy and paste so it's possible to have a few folders.

 

[Knighttime]

I used ToTPP v.14 on both scenarios. Ciaran, I'm interested in why you think v.13 was more stable. Can you elaborate? Thanks.

@techumseh This thread mentions that other people have also found TOTPP 0.14 to be "buggy". (That also happens to be the first version that permitted Lua scripting.) So my recommendation would be to get off of 0.14 as soon as possible. Personally I'm using the latest release, 0.15.1, and that would probably be my first recommendation -- and for anyone who wants to create or play a scenario that uses Lua events, this is definitely the best version to use. If you don't need Lua support, then you could also consider version 0.13.2, which was the last non-Lua release.

 

[techumseh]

Thanks Knighttime, I'll give it a try. Will I be able to continue to work on and play scenarios made with 0.14 if I uninstall it and replace it with 0.13.2?

 

[JPetroski]

Why not just have a folder for each? Do a fresh install and then copy the whole civ2 folder before applying any patch.

 

[Knighttime]

Thanks Knighttime, I'll give it a try. Will I be able to continue to work on and play scenarios made with 0.14 if I uninstall it and replace it with 0.13.2?

I think so, with a couple basic qualifications:

First, if you have a .sav or .scn file that is corrupted in some way, I doubt that going back to 0.13.2 would undo those problems. But if you have a clean, working saved game file, you certainly ought to be able to keep editing and working on it after downgrading TOTPP.

Secondly, please review TheNamelessOne's list of changes that were included in the 0.14 release, which you'd be giving up by going back a version, and make sure you aren't relying on any of them in the scenario you want to build or play.

Also, I agree with the recent posts by CurtSibling and JPetroski that raised the idea of reinstalling Civ2 and having multiple versions installed in different folders simultaneously. If you leave your current installation alone, reinstall ToT from scratch into a new folder, and then perform the upgrade process and go straight to TOTPP 0.13.2., that seems the all-around safest way to get a fresh start that won't have any remnants of 0.14. If anything goes wrong or actually seems worse, you can always switch back to your current setup.

Last thing: If you think that a large amount of memory usage by the macro events is a potential issue, you might want to experiment with the @COSMIC2 parameter named "EventHeapSize". Perhaps there is an issue with the way dynamic allocation is handled? You could try setting a much larger first parameter, for initial/static heap size, and potentially also using 0 for the second parameter to turn off dynamic allocation (which is enabled by default). I've never played around with these values myself, but thought I'd point out the possibility in case you wanted to give it a shot.

 

[techumseh]

Thanks. I think I'll start with the @COSMIC2 parameter. The first parameter was set to 219,000 and change, and the scenario events currently is at 272269, so I bumped it up to 300 thousand. I think I'll keep the dynamic allocation setting for the moment, in case my delayed events end up exceeding that amount.

I'm curious about how the events 'heap' actually works. Is it possible that if it exceeds a certain size, the program itself will be unable to 'see' or handle all of the stored events?

 

[Prof. Garfield]

There are issues with ToTPP for sure. The question is if we can trouble shoot what they are and find a solution. Prof. Garfield was able to figure out one issue that was causing us to have headaches with certain tiles not "reading" certain events (in lua however) and I wonder if this has anything to do with some of your issues, such as the tanks. Did they always appear on a particular tile or was there a chance that they could appear on one of several random tiles? If random, have you noticed which tiles they will show up with and which they won't?

That issue was a bug in Lua code, nothing to do with the TOTPP itself (except that civlua.lua is shipped with TOTPP). We can change bugs in Lua code, but without TheNamelessOne or someone else of comparable competence we can only work around TOTPP bugs.

Two things struck me. In both cases, the problems began after I had added lettering using the 'landmarks text' feature of ToTPP. Secondly, I had saved the game as a scenario, then launched that scenario and saved a game from that, then used that save game to create another scenario file. The advantage of this is that it got rid of a bunch of "We love the Admiral" texts that the "Clear all Disorder and We Love the King" key didn't, and saved the scenario parameters. So, I remade the scenario from the original save file, following the process above. I made sure that there was no events file in the folder during the process, adding it only after I had created my scenario file. And I've refrained from adding Landmarks Text. The phantom events problem has not recurred - yet. But...

I don't know how TOTPP implemented landmarks, but it would not surprise me if they are simply added on to the end of the file, and an oversight by TNO has them overwrite some events. Try adding all landmarks before saving with events. Or, use my Lua converter, which shouldn't have that particular problem (Lua and landmarks haven't been a problem). Lua events seem to be reliable (except maybe for preventing diplomacy), and if my converter isn't reliable, I can fix it.

Thanks. I think I'll start with the @COSMIC2 parameter. The first parameter was set to 219,000 and change, and the scenario events currently is at 272269, so I bumped it up to 300 thousand. I think I'll keep the dynamic allocation setting for the moment, in case my delayed events end up exceeding that amount.

I'm curious about how the events 'heap' actually works. Is it possible that if it exceeds a certain size, the program itself will be unable to 'see' or handle all of the stored events?

Yes, I think if the memory required for the events exceeds the heap size, they won't be seen and/or written. That might be a problem if you have a lot of delayed code. If you want to go the Macro event route, set the parameter to something like 500k or even 1000k, so you're sure you have enough room.