Free Allegiance Forums

Ok - IB and I have a dispute about whether or not the AME documentation is correct when it states that "The aleph location randomizer is VERY smart, it keeps the distance to center within a defined range."

So I created a map to test this effect, and whilst it's no where near as sensitive as the manual makes out, it does definitely happen.

Here's the map I made to test it.

First off the sector with Aleph set far apart.


Now Alephs close together.


In game what do these look like?

Long:


Short:


About a 1k difference. Does this always happen? I tried it again.

Long:


Short:


Again, about a 1k difference. Nothing like what's shown in AME, but real and seemingly reproducible.

So if you want closer or further alephs in your custom maps, set them insanely close or far from the centre. I think that this 1k represents the maximum difference that the randomiser will allow, you can't force maps any further apart than what I have here, or any closer.

Rav

sorry but no.

the randomizer code doesn't use the stored locations at all to compute the new locations...

KGJV wrote:QUOTE (KGJV @ Apr 26 2010, 07:53 PM) sorry but no.

the randomizer code doesn't use the stored locations at all to compute the new locations...

Suck it Rav.

New question, would it be desireable for the aleph random placement algorithm to use information from the map?

Perhaps the aleph would always be placed at the (2d x-y) radius taken from the map file +-500m. So they would be randomly placed somewhere within a 1k wide ring around the middle of the map.
Perhaps Z axis magnitude could be preserved (+- some value) - so map makers could set na aleph to be close to the plane or further away (but up or down would be random).

KGJV wrote:QUOTE (KGJV @ Apr 27 2010, 03:53 AM) sorry but no.

the randomizer code doesn't use the stored locations at all to compute the new locations...

So why did it say that it did in the readme you supplied with AME for so long? Also, why do I get consistent results when testing?

I know that you know what you're talking about Kage, I just wonder if that functionality used to be there and if it was removed.

I don't recall writing this in the readme of AME. Can you quote that part ?

As for your consistent results, that seems odd, how many tests did you do and how ?

I'm pretty sure the randomize code is the same since the day I added it, so nothing was removed / changed. I've reread the code, it doesn't use at all the existing positions of the alephs (see igc\missionigc.cpp line 3838+)

The only recent modification made was to not randomize the "fixed position" alephs (R5 stuff).

The quote in the first post is from the wiki which was taken from (I think) the AME documentation. I feel it was written by UKDude not yourself though.

I've run 3 tests so far that I've actually bothered to check, each time it's been an order of 1k further apart in the long sector.

How many tests would you say I should do to be sure? Another 3 or 4?

how do you do these tests if more important than how many of them.

fly into one sector, target one aleph from the other. Fly to the other sector and repeat.