descentchampions.org

Yeah, there are too many reactor cubes. The level won't load. I never did figure out exactly what the limit was, but it seemed to be somewhere around 50.

I don't know how technically feasible it would be to raise that limit in future versions of Retro, because I don't know why it happens in the first place. It probably won't be necessary, though; the next version should have full blown obs mode, and these obs levels are just a stopgap until then.

In the mean time, I think you can probably use the reactor cubes you do have more efficiently.

- I see you have the cubes stacked in a 3-layer deck. If you could make that 1-layer everywhere but vertical transitions, that would save you a lot of cubes that need to be covered! If you look at my obs levels, there are a lot of areas where the cubes (from a side view) do what I've pictured here. You really want to keep things a single layer as much as possible, to save on both reactor cubes, and the effort the system takes rendering the level for the observers.

- I see you mostly put reactor cubes above walls, so they don't interfere with spawning in the lower level -- good! Sometimes you can't help it, though, especially with as many open spaces as Black Rose has. If you have to sometimes put a reactor cube adjacent to the lower level to cover a bunch of the upper level . . . well, it's better than not having an obs level! I had to make this compromise a few times in Take2.

- If you can't cover the entire upper space with spawn interdiction, that's okay, too. If you leave the uncovered spaces as far away from the entrance as possible (ideally 15-20 cubes), they are less likely to be hit.

- A few missiles will probably spawn up there no matter what you do. If the algorithm can't find a good place to spawn a missile, it panics and picks a RANDOM cube. What I've done is include a "missile return" somewhere in the upper space: a long tunnel you can shoot missiles down (so they're guaranteed not to lock onto players below), somewhere near the entrance to the area (to increase the odds of the missile spawning below, rather than above.) Observers have to return missiles sometimes, though it's best to keep this to a minimum!

On the whole, it's awesome that you did black rose! That's been needed! I bet you can get it functional with another revision. Impressive work so far, though! Making obs levels is very technically demanding, as level-building endeavors go!

Ok, this has gone so far off-topic (my fault ), we should take this somewhere else. Maybe on Mumble?

Hmmph, I don't feel good about further littering this thread with off-topic-stuff, but PMs do not allow zip-files, so here goes:

Ok, I've tampered with the code and it now allows to host a game in either the "Classic Descent" Object Spawn Mode or with a (non-default) "Smooth Distribution". One could think about adding helpful explanations to some of the options in the menu, btw.

Attached are the changed files and a small level with >100 missiles in one spot for testing. There is also access to an observatory in the same room, directly under the "Danger"-Sign. From there the difference is rather striking.

I've also made sure, that random drops happen only in accessible areas, there is a flag THIS_IS_NOT_DESCENT_1 so that the behaviour changes and can very easily be adapted to D2 (I haven't tried it yet, but it should work with just changing the flag to true).

All changes are marked by TAMPERING by nube begins ... ends here, one should find them easily.

What I have not done is change the _sock_thingy for the netgame-tracker-information, this option will not show.

changed_source_files_in_d1_main.zip

(78.43 KiB) Downloaded 139 times

nike.zip

(12.7 KiB) Downloaded 139 times

Looks pretty functional. Nice work!

This sort of thing is generally outside the scope of what I want to do with Retro, though. I want to keep Retro focused on solving problems the community has. One of my highest priorites is stability, and I don't just mean that in the sense of "the code doesn't crash". I mean it in the sense of "the game doesn't change without a darn good reason." I've been asked to provide a lot of fun, experimental features in Retro, and I've generally said no. Retro is about representing the original Descent experience as much as possible, and only changing it if it is practically universally agreed that the original experience sucks.

I mean, I'm not above doing things just because they're fun if there's a wide, widespread desire for it, and it's pretty easy to talk me into changes that enhance the competitive game. But as a rule, anything that will actually change the gameplay has to be something a LOT of people want. Not even just a majority of the community. Like, a huge supermajority. Everyone but maybe a dissenter or two. And top pilots get extra votes.

I don't get the sense that this change actually solves an existing pain point. It's just change and options for the sake of change and options, something there's already -- IMO -- a little too much of.

-------------

I could be wrong . . . If this is something pilots as a group want, speak up. I can put it in.

But at least speaking as a pilot myself, I feel like we already have more ways to modify levels than I'm completely comfortable with, and I don't want the spawning algorithm to be a variable I have to think about when hosting and negotiating a game. The spawning algorithm is fine. I like it. I'm used to it. Let sleeping dogs lie. I don't want "Logic with uniform spawns" to be another version of Logic I have to worry about getting good at, alongside Logic Mx4 and Logic No Vulcan and Logic Spawn Previews. I don't see it as potentially interesting enough to be worth that.

I would be more comfortable with giving level designers control over it. Then the spawning just becomes part of the level, and something I learn to expect when I play it. Sort of like weapons mods in D2.