You'll want to start calibrating an hour or two before your seed comes around, depending on how long your wait time is. It helps since it gives you a feel for when you should press A, and once your seed comes you'll be ready to go. When I was getting Groudon, I failed the first time, but luckily, I got another chance half an hour later. I don't know what else to tell you, other than good luck.

Edit: Oh no! You already missed your frame. Sorry about that.

Finally found my exact RTC value! Time to get into action. Any tips you yourself can offer?

The minutes value given by the program is the amount of times the minute is increased, not an actual value like you desrcibed.

v = (1440 * d) + (960 * th) + (60 * uh) + (16 * tm) + um;
x = floor(v / 65536);
I call x the seed cycle since since it stays constant as the seeds increase from 0000 to ffff, and then it increments by one.

edit: I'm not quite sure what I mean either. Let me think about this for a while, but I feel like v isn't really how many minutes the RTC has been running, since at 23:59 on any day, the part of the equation that makes up the minutes equals 2189, but there's only 1440 minutes in a day. Do you see a problem with that?

Not quite sure what you mean by that. Would you mind explaining further?

I'm really impressed with the progress you've made on live battery abuse. But when I was working with this, I wasn't concerned with how many minutes my RTC was running, but rather what the seed cycle (x, an 8-bit number) was. If you could write a program that could spit that out, then you would only have to worry about how far off your battery is from the real time.

Awesome! I have no idea how to program, so I did all my calculations manually :( . But did you figure out how to determine what date your cart thinks it is? Its probably not synchronized with the real time, and that will mess with hitting target seeds. Mine was a couple years behind.

Oh and btw, my notes aren't really notes, but more like memory joggers. I'd have to make time to sit down and relearn everything before I can write out something that someone else can understand.

BAM! Figured out how to target seeds, I'm pretty sure. I created a program that takes care of the seed to time thing, so now all I have to do is, well, use it.
Even with the program, though, you can't get around the one-month wait.

I've kinda given up again, but I wouldn't mind seeing some of those notes you had in the end on the dreaded alive battery abuse.

A lot of calibration is required, but I think it will end up being worth it.