PID%6 and Natures are the only way you can check.
Hm, I wonder if immediately capturing a Pokemon right after the event and then doing PID%6/Nature frame deduction would help. The rain would have to stop of course, but after a few attempts you should get a better idea where you hit.
That's actually a pretty great idea. I was already capturing something on Route 7 right after Thundurus flew off and using its IVs, Hidden Power, and later just its characteristic (once I knew what adjacent seeds gave me) to check the seed. If I can use that to also narrow down a PID frame, it should help a bunch in finding the starting frame for the roamer. Having the encounter slot, ability, gender, nature, etc. to work with will definitely help with that, too. Thanks for the idea!
Duplicates aren't weird for the roamers. It's probably exactly the same frame again. While you can hit the same frame consistently the frames don't quite advance consistently. For example I would do 100 frame advancements and only end up 96 frames higher. I figured it had something to do with the NPCs on that route 'cause there's NPC advancement when you exit the house and for a split second after you talk to the roamer though they're frozen for the rest of it.
Ah, I see. So, probably what happened with that second seed where I got Rash 3 times in a row would have been that with 0 advancements the NPCs advanced 2 times before the PID was generated, with 1 advancement the NPCs advanced 1 time before, and with 2 advancements the NPCs didn't advance the PID frame at all. Something like that, anyway... Either that, or I had two Rash frames close together. I figured I was getting duplicate frames (ie., the same frame) but couldn't figure out why.
Though, the fact that some frames possibly cannot be hit due to the NPCs is a little worrying when trying for a shiny. Hopefully I never run into that problem.
And if you can't figure it out with the PID%6 method then obviously you're doing something wrong. Your nature appears about 3-4 times for every 100 frames. So lets say you're checking a range of 400 frames. That's about 16 frames to check to see if they have a matching characteristic. It really shouldn't take you a long time to do that if you have a calculator in front of you and out of those 16 you'll find only 1-3 that are a match. Then you just do +5 frame advancements catch the roamer again and check via PID%6 again. Obviously you'd expect the second matching frame to be very close to the first matching frame and that's how you check to know if you've got the correct starting frame.
Before I make myself sound like more of an idiot, let me just make sure I have the PID%6 method correct...
After getting the remainder of the PID, I check that with the IVs that are tied. In this case, I have speed, special attack, and special defense. The characteristic matches with special attack if the remainder is 4, special defense if the remainder is 5, and speed if the remainder is 0, 1, 2, or 3. Right?
Assuming I got that right, the problem I was running into with the PID%6 method is that I got a ton of 0-3 results and not very many 4 or 5 results. For example, of the first 16 Rash results (ranging from frame 50 to 500), only five of those results had a 4 or 5 as a remainder. That doesn't help much to narrow it down, with the other 11 results being valid. I suppose it helps more to try to narrow down one of the PID%6 results of 4 or 5 instead, now that I think about it.
I was also advancing by 1 each time instead of 5 or so like you suggest, and I was looking for a sequence (though not an exact sequence, but something close to it), which is likely what threw me off. Next time I try this I'll do 5 advances instead and see if that helps.
Anyway, thanks for the help, you two. I hope next time is a little easier.
ok i just have one question, is a frame of 692 a good frame to try and hit when going after tornadus? because i believe i saw somewhere the ivs and pid are seeded after 465 frame and then i could just advance from 465 to 692 obvsly after finding my initial frame after 465
That sounds like it could work. After looking into it some more, I think my starting frame was 459 or 462, so 465 shouldn't be a bad estimate of a starting frame. Keep in mind that I was doing this when there was no rain on Route 7 prior to the roamer showing up, so it might be different if there is rain or snow on the route beforehand.