Spring RTS Engine

In the past I've done my best to optimize things like the pathfinding calculations that you have to do the first time you load any map with a mod for the first time. I do this by reducing the number of pathfinding types by condensing things which have very similar footprints or slope tolerances into a single pathfinding type.

Now I'm thinking though, why bother? It only has to be done once and it could compromise gameplay a lot. There are certain cases where I've got two units using the same pathing calculations even though they should really be seperate. So a certain unit or units is amphibious when that wasn't in the original plan.

So what's more important, significantly shorter loading time or slightly improved gameplay?

This is another one of those threads where you phrase the question in such a way that the result is obvious, and you can feel legitimized doing something you would have done anyway. Hah, thread broken!

That being said, I'd gladly wait a few seconds longer for each map if it helps gameplay. Don't go overboard though, I don't want to click every buildable unit and press f2 every time i get a new map just to see what can go where

BTW, it doesn't care about the footprint of the movetype.

I've been getting away with just using 4 movement classes...

Sometimes less is more.

I was living off of an Intel P2, SDRAM, 8mb Vid card, and ect for most of my life, so Ive come to just not care about load times.

Do you really _need_ 1203810238 different movetypes or can you get away with 8-12?

AA's currently got 20, I think, and they're all vital... that's with compromises like units being amphibious when I didn't really want them to be, etc.

20 sounds like a lot. I mean, what are those? Tank, Kbot, Hover, Amphib and the special unit stuff like huge unit (for crushing stuff), ATV (like Spider) and... uh... what else?

KDR_11k wrote:BTW, it doesn't care about the footprint of the movetype.

In case you still don't get it, that means units of different sizes can share the same movementclass.

There's only the crushing, the slope tolerance and the min and max water that needs movementclass. So if you have 20 movementclass, you're clearly doing it wrong, and that poll is just a lame excuse to avoid the work of refactoring your movementclass.

A more honest poll would have been:

• 15 mins work for Caydr, strictly identical gameplay, and 15 mins less pathing per day for everybody
• Make every AA player lose 15 mins a day to save Caydr 15 mins work once.

zwzsg wrote:

KDR_11k wrote:BTW, it doesn't care about the footprint of the movetype.

In case you still don't get it, that means units of different sizes can share the same movementclass.

There's only the crushing, the slope tolerance and the min and max water that needs movementclass. So if you have 20 movementclass, you're clearly doing it wrong, and that poll is just a lame excuse to avoid the work of refactoring your movementclass.

A more honest poll would have been:

• 15 mins work for Caydr, strictly identical gameplay, and 15 mins less pathing per day for everybody
• Make every AA player lose 15 mins a day to save Caydr 15 mins work once.

QFT

zwzsg wrote: A more honest poll would have been:

• 15 mins work for Caydr, strictly identical gameplay, and 15 mins less pathing per day for everybody
• Make every AA player lose 15 mins a day to save Caydr 15 mins work once.

While I somewhat agree with the sentiment, this poll is in wo way more honest than the first one.

KDR_11k wrote:BTW, it doesn't care about the footprint of the movetype.

In my experience, it does. Example: I have a factory with enough space in its yardmap to produce 2x2 units. When those units have 2x2 in movetype and in FBI, all is good. However, if they have 4x4 in movetype (yet still 2x2 in FBI), they cannot leave the factory. Those units were ships, the factory was a shipyard, maybe it's different for land units.

don't make flashes/instigators incapable of driving in their current sea depth, or you'll fuck up a lot of maps with shallow water bridges (and more importantly, two of mine. )

Just start a thread and oh, I don't know AA in the moding section and do polls there. It is a known fact that most of the player base does not really browse the forum.

I am going to say no, it doesn't really matter because you are doing what you want anyway. That is ulimately your call as a content developer. Just stop making all of these polls. Project discussion goes in the moding subforum. If you afraid that you may muck up that forum then do all this stuff in your aa thread.

smoth wrote:Just stop making all of these polls. Project discussion goes in the moding subforum. If you afraid that you may muck up that forum then do all this stuff in your aa thread.

But the constant creation of polls in General Discussion helps giving more exposure to AA, as well as ensuring that everybody know that Caydr is back in town. Creating polls about your plans is much more effective than just doing them without telling anyone. Only loser "niche" mod use the mods subforum anyway, AA on the other hand obviously pervade every subforums, because Spring is AA, after all!

heh, funny how I shove gundam down people's throats...

I agree that this is just as annoying as the "THERE ARE OTHER MODS THAN BA AND SPEEDMETAL" orgies. Caydr does have the bonus that most people here spent most of their spring careers actually playing his mod and thus probably care about it somewhat.

I'm not quite sure how many pointless poll threads that this is good for, though. My estimate is somewhere around, let's say, twelve.

Ten to go, Caydr. I suggest the next one to be "Good Graphics vs. playability on Pentium II", use it to legitimize 24-sided cylinders and huge effects.

So what you are saying boirunner is the fact that some people are huge fanboys of caydr's makes this aceptable?

Why is it that you can pick up when I'm somewhat subtly making fun of you, but you can't when I'm even less subtly making fun of someone else?