VLS - Is it possible?

[treblesum81]

Originally Posted by polyfiller

Interesting find on the primary node alignment. In terms of implementation - are you having, in effect, two complete copies of an entire sub in the .dat, one visible, one not, and one rotated vertically ?

If so ... not sure if this will cause CTD. If you've remapped all ID's, then it should. One thing I would recommend is removing nodes with duplicate names for;

rudder, torpedo doors, propellers, bow plans. These are refereced by name and not just ID in the subs .sim file - for example where a torpedo door name in the .dat does not match the names in the .sim, you get CTD. What I think you may have is one of the above items appearing twice causing a lookup failure / issue in the .sim file.

Whenever I work on torp doors, I always have the .sim file open at the same time and step through each tube in turn, if that makes sense.

Good luck.

Actually, I created a whole new node which was rotated 90* with only a label and two sub nodes to hold the two inner doors. All three had new unique labels and I did go into the sim file and replace one of the original four torpedo tube door set references with the new ones. When this gave me a ctd, I started trying turn the second node into a copy of the first one, unfortunatly, I ran into two distinct problems... first, Silent 3ditor won't copy the parent child structure for some reason and won't allow you to simply tell something to become the child of another chunk, so I had to manually copy all of the values into each chunk... and second, the program refused to allow me to even place a 3D model chunk into the second node, let alone copy the 3D model of the sub into it.

I actually think that this isn't the way to go about it though, because I can foresee quite a few issues arising from the vertical model sticking out of the top and bottom of the boat, visible or not... the game my try to calculate collisions with the ground using the piece that's hanging down 100-200 feet, or enemies might be able to see it despite the fact that the sub is submerged because the bow is an extra 50 feet in the air. Instead, I think that the best way to really work this idea into a sub model would be to build the vertical node so that it only contains the inner and outer doors, since they are the minimum required to make the missile / mine exit the sub, or to really build the sub in seperate sections. The second option here would probably have the main node contain the parts of the boat needed to be aligned correctly (prop, rudder, planes, sail, and most of the body), and then the second node containing the sections of the boat which need to be aligned vertically. The only reason I can see to really go through this much trouble is that there might be some instability or ctd caused by the two primary nodes overlapping each other as they are in the first model.

Either way, I've still not gotten the setup to even load in game yet, so I'm not even sure that the thing will work. One part of the problem that I've found, is that the game wants the boat to settle with its primary node parallel to the horizon. I figured this out when I was testing the torpedo initial alignment theory because I had set the original primary node at a 90* angle, and when I loaded the game, the sub was pointing straight up, but then immediately began settling to horizontal. What this means is that any secondary node will need to either a) have no mass so that it will not try to settle, or b) be shaped in such a way as to have to real reference to "up" except for where the weapons launch from.

Originally Posted by keltos01

this really interests me even though my IJN subs don't need it

when the torp/missile is fired vertically, how do you intend to make it turn back to the horizontal and fly like a real missile towards the target ?

keltos

This is actually done for you in the same way that they would turn to horizontal if you were to launch a torpedo while diving or ascending. The torpedo / missile comes out aligned with the horizontal of the sub and then turns up or down or side to side as necessary. The same would be true of the vertical launch, the only difference being that it would have to go from vertical to horizontal rather than some small angle to horizontal. That being said, you would definitely have to be at a deep enough depth the launch this way, because if you were too shallow, the missile would shoot up into the air, dive straight back in as it porpoised and blow off your sail, creating a missile launch depth just like in real life.

Anyway, I'm still playing around with the dat file trying to get it to work, but I really feel like I'm just stumbling around in the dark. So without some additional pointers and direction (thank you already polyfiller), I'm not sure how long this will take, if I can get it to work at all, but that being said, at least its a start.

Greg

[treblesum81]

Also, polyfiller, were you saying that by remapping everything I've caused it to start CTDing, or that because I did that, it should be working?

[treblesum81]

One last question for the moment: does SHIV have a debug mode, or at least is there a way to find out the cause of the crashes? I'm using Windows 7 (could be part of the problem, but SH4 runs clean in it so I doubt it) which, unfortunately, doesn't readily provide information about what caused the ctd. If I could find out where the conflict is, I could much more easily determine the solution...

Any thoughts?
Greg

[polyfiller]

Sorry - by remapping ID's I meant that by doing so you should avoid crashes i.e. if you don't remap Id's (more than own ID the same in the .dat or .sim file can cause CTD's).

Sadly there is no debug mode I've been able to find.

Given what you say about the verticle node settling to the horizontal, I reckon you're flogging a dead horse here. There is a controller for torpedo launching (torpedo_eject.dat) but this doesn't look to have anything about launch angle and given the reference in it to @torpedo_eject, I reckon that the reference is hard coded i.e. if you do change this, it will efect all torpedoes.

If you want to persist with the duplicate aligned "ghost" node, start with a basic node, import the 3d model, with NO OTHER nodes attached to it, then load the game - this should work, if it doesn't then what you're trying doesn't fly. If it does load, begin to add some nodes and see what happens.

As I said before good luck ... I have a vested interest in this being figured out - I want destroyers in the TSWSM to launch their torps sideways and not for/aft.

[treblesum81]

Originally Posted by polyfiller

Given what you say about the verticle node settling to the horizontal, I reckon you're flogging a dead horse here. There is a controller for torpedo launching (torpedo_eject.dat) but this doesn't look to have anything about launch angle and given the reference in it to @torpedo_eject, I reckon that the reference is hard coded i.e. if you do change this, it will efect all torpedoes.

If you want to persist with the duplicate aligned "ghost" node, start with a basic node, import the 3d model, with NO OTHER nodes attached to it, then load the game - this should work, if it doesn't then what you're trying doesn't fly. If it does load, begin to add some nodes and see what happens.

After what I've learned in my investigations, I can tell you that the torpedo_eject.dat is only the effect of the bubbles and sound that plays when the torpedo is launched out of the tube and has nothing to do with the actual firing of the torpedo.

Also, I keep running to this problem with silent 3ditor where if I create a second primary node and then try to place a 3D Model child chunk into it, it refuses to make the parent child association and simply places it on the same level. Can you tell me if this is something I'm doing wrong, or something inside of S3D that is not working correctly, or even if its doing something right and you just can't do that?

Some more messing under my belt since the last bit of writing here... I've tried just adding a new node into the existing structure, which leads also to a CTD. As a result, I firmly believe that I'm not doing something correct in the process leading to dangling ends in the data that are causing the CTDs. Aside from adding a node and naming it, what else is required to make that function without issue? I've added the entries into the sim file for the new torpedo tube and doors, but what else needs to change? Do I also need to go into the various upc and cfg files to make alterations? Are there other parts of the dat/sim/zon/etc. structure that I need to change as well?

Finally, is there some documentation about how to do this somewhere, or even perhaps a tutorial? While you might be right that this really isn't possible, I won't be willing to accept that until I know that it isn't some stupid mistake I'm making every time I try to make an alteration.

Greg

[ETR3(SS)]

Like polyfiller said I too have a vested interest in this project. I'd like to include this in the 688i mod as well. Version 4 for stock SH4 should be up by the end of the week. Not sure what boat(s) you're testing this with but figured I'd throw that info out there if you wanted to use that as a test bed.

[treblesum81]

Well, I've actually been using ver 3 at the moment, only because this way, if I get the ctd figured out, then I can work on fitting the launch points to the right place. all of this being said, I'm still not able to get the ctd issue solved, and the more I work at it, the more I think its something I'm doing wrong and not something that can't be done. What I've arrived at now is that I can manipulate any value in any existing chunk without a problem (however this doesn't produce the correct effect), but if I try to add anything at all, label, node, 3D model, etc., then all of the sudden I start getting the ctd, and always at the same point... the loading bar fills completely and then several seconds later, before the scene loads, the game crashes. So, I need to figure out still what I'm not adding, editing, or changing when that I need to be when I add in a new chunk.