This is topic the mechanics of DT in forum Research Lab at The Azure Heights Forum.

To visit this topic, use this URL:;f=1;t=000516

Posted by laZeye (Member # 464) on 11-27-2000, 06:02 PM:
I tried looking at azure's research on it, but the only results they had were pretty inconsistent. Does anybody know how DT works, such as how much the evade is raised per use?

(o) ( .)

Captain Terrible of the ATF

Posted by cfalcon (Member # 19) on 11-27-2000, 10:53 PM:
Well, assuming there actually IS an evade byte (likely) you could probably trail it with an emulator or something... heck, a shark might be able to do it.

That might be a neat code to find: more importantly because we'd have numbers in memory.

It should be noted, however, that for practical purposes the percentages are better than any in game data: the findings are pretty damn good, I've always thought.

Perhaps someone already did it, and will post it... otherwise I'll add it to my "shit to look for" list.

Also note that it is possible that the "Evade" concept is just an index to a small list of numbers, with each number representing a different "to-hit" mask (mask because you have to roll "to-hit" normally, then against your accuracy, then against their evade, if I recall correctly).

Either way, the results will be the same: your evade mask goes up from zero with DT and friends, and their accuracy mask goes down (from 100%) with ever Sand-Attack and friends. The regular "to-hit" is determined by the move used. So the end result no matter how it is done is a pair of tables with numbers representing how often something happens... and we have those now. In fact, I bet if you multiplied those numbers by 255 and divided them by 100 you would end up very close to whatever happens in game (scaling them to per 255, instead of per 100).

But I haven't done the tests, so I could always be talking out of my ass... if someone with a bigger clue would chime in, that'd be nice.


"It's not a sentence.
Do I understand it all?
Difficult to say." -Mr. K

Posted by laZeye (Member # 464) on 11-27-2000, 11:11 PM:
Eh, the reason I was asking is because the stuff I got from azure heights is pretty inconsistent... I was just wondering if anyone had some more specific figures.

BTW, I tried my best to make something consistent and still get close to the values listed at azure heights, so here's what I did:

Of course, a big problem is that the DT/Sand attack info isn't exactly accurate, and it won't be until I can find some (duh) accurate information about them.

(o) ( .)

Captain Terrible of the ATF

Posted by Mr. K (Member # 2) on 11-27-2000, 11:27 PM:
laZeye: Eh, the reason I was asking is because the stuff I got from azure heights is pretty inconsistent...

Well, someone did point out that there's a reasonably logical patten in the percentages:

1/3, 2/4, 2.5/4.5, 3/5, 4/6, 5/7

The third one is obviously a little odd, but you can see how they might have tossed it in to make sure DT didn't get even more powerful.

I think they should have added another stutter step in there anyway...

Save big money at Menard's!

Posted by laZeye (Member # 464) on 11-28-2000, 04:40 PM:
Hey, you're right! Thanks Mr. K. The only problem with that is that it'll be hard to implement, unless I just make 6 if statements. Ah well, if statements it is.

(o) ( .)

Captain Terrible of the ATF

Posted by White Cat (Member # 42) on 12-07-2000, 01:37 AM:
Try setting up an array with the seven values (0, 1/3, 2/4, etc.) and then make a pointer that moves around between them.

"I like to do politics with people from the East. Joe Clark and Stockwell Day are from Alberta. They are a different type. I'm joking. I'm serious. Uh..."
-- Canadian Prime Minister Jean Chrétien

Karpe Diem