[TEC] Differences between "random" groups and scripted?

[tater]

Hey, thought someone else might have a handle on this.

I started redoing the "eastasiacampaign" layer.

Compulsive bastard that I am, I'm sitting here with a pile of books, and about 10 tabs in a browser window while I do it. I actually know some of the specific transports that went to specific invasion beaches, that level OCD, lol. For example: 1x Takao, 1xNaka, 4xFubuki, 2x Large Old Liner, 1x Modern liner at Kota Bharu the morning of the 8th (I have all the ship names, too).

There are two ways to do this.

One, I can make groups by hand, and arrange the ships. The formations move somewhat rigidly, though I think the DDs weave some.

Two, I can make a random group that appears at the time desired 100% of the time, and I can limit the number of times the group is generated to exactly 1. DDs can be assigned as members of the force, or as escorts.

Does anyone know if the 2 different groups behave differently? Is the AI more robust in the random vs scripted groups in terms of how far the DDs go while escorting, etc, or do I have to run a bunch of experiments and watch them?

tater

[bigboywooly]

From SH3 experience there is no difference between a scripted convoy and a random one
The only difference with random convoys is you can run the same convoy over the same course with frequency set by yourself
Ships can also be random also - named class in the scr convoys - giving variety

If you are planning to add single historical convoys then it doesnt really matter which method you use

Though if the convoys follow the same route but with different units it can be easier to use an rnd group which you can copy\paste ( in mission ed ) and alter the date etc as well as unit contents without having to add in the route again

[tater]

Thanks! That helps a lot.

<S>

tater

[nematode]

I actually am building a scripted campaign from TROMs, attempting to get things as historical as possible. So far I've completed maybe 2/3rds of Dec 1941 for IJN movements: Kido Butai's return to the Inland Sea and associated sorties from Bungo, Malaya invasion and followon convoys, 1st and 2nd Wake invasions, Gilberts invasion, Guam invasion, West Borneo invasion, and am about to get started on the Phillipines. I've been able to identify many of the Marus, down to SH4 equivalents. It's been a tremendous amount of work, but very educational. If it never gets finished, or presentable as a mod for release, I really don't care, its chiefly for my own interest in the history.

Altho I'm not ready to release Dec 41 as a WIP, Tater if you're interested, PM me and I can share with you what I have so far.

You'll want to use scripted groups, not random groups, for anything that you need to be at a certain place at a certain time. Random groups don't start when you want them to. Furthermore you can't control the formation of a random group, but you can drag scripted group members exactly where you want them to appear within the formation. And when you have a 50 ship convoy, like my 2nd Malaya Convoy, which stretches 20km from end to end, you need scripted placement to make sure it fits.

My SOP is all IJN and convoys scripted, solo merchant traffic random. At some point I'll tackle the random solo merchant traffic layer.

Testing, yes. Test everything that you add.

Originally Posted by tater

Hey, thought someone else might have a handle on this.

I started redoing the "eastasiacampaign" layer.

Compulsive bastard that I am, I'm sitting here with a pile of books, and about 10 tabs in a browser window while I do it. I actually know some of the specific transports that went to specific invasion beaches, that level OCD, lol. For example: 1x Takao, 1xNaka, 4xFubuki, 2x Large Old Liner, 1x Modern liner at Kota Bharu the morning of the 8th (I have all the ship names, too).

There are two ways to do this.

One, I can make groups by hand, and arrange the ships. The formations move somewhat rigidly, though I think the DDs weave some.

Two, I can make a random group that appears at the time desired 100% of the time, and I can limit the number of times the group is generated to exactly 1. DDs can be assigned as members of the force, or as escorts.

Does anyone know if the 2 different groups behave differently? Is the AI more robust in the random vs scripted groups in terms of how far the DDs go while escorting, etc, or do I have to run a bunch of experiments and watch them?

tater

scripted one gives you more control of formations so I would probably use that one.

[tater]

Yeah, that's a big benefit, control of the arrangement. OTOH, I wonder how the station keeping is in a zig-zag compared to AI (random). It's certainly easier to build the units with random though instead of placing each unit by hand.


tater

[nematode]

Originally Posted by tater

Yeah, that's a big benefit, control of the arrangement. OTOH, I wonder how the station keeping is in a zig-zag compared to AI (random). It's certainly easier to build the units with random though instead of placing each unit by hand.


tater

Just thought of something else, if you are looking to create one and only one instance of something, I'm not at all sure that can be done with a random group .. unless there's a way to "zero out" Repeat Interval. -1 won't do it.

But for me, it never gets that far, the inability to control when something is going to spawn kills random groups for me for anything but simulating merchant shipping lanes.

I'm not sure what you're wondering about with "station keeping in a zig zag", but the AI doesn't know how to do a column turn, what happens is the leader turns and the other group members skid out at max speed to the new bearing like waterskiers towed behind a ski boat.

With respect to creating decent looking zig zagging, it doesn't matter scripted or random, either way the members wil hold formation (except DDs of course) .... but subject to the problematic AI turning behavior. It becomes a question of how much wierdness you're willing to cope with. The crappy AI formation turning behavior can be ameliorated to some degree by formating multiple smaller groups so that there isn't so much "back end" in any single formation to cause its members to skid out a great distance during a zig.