:O thanks Radnortuws. that will help alot
-
Check out the relaunch of our general collection, with classic designs and new ones by our very own Pissog!
-
Welcome to Smogon! Take a moment to read the Introduction to Smogon for a run-down on everything Smogon, and make sure you take some time to read the global rules.
Most Generation PRNG Help / Information
- Thread starter mingot
- Start date
Currently in my calibration I'm getting consistent seconds and offset. Just not delays. Seconds have always been 13, and offset has always been 5. Delay has always been even, and from 600-610.
You're supposed to go through the title screen and hit "Continue" as quickly as possible. The quicker you do it, the closer to 600 you'll be.
Well, without searching though all of them there are just not that many results to be honest. Finding something good would likely require jiggling those values for long periods of time.
Also, minutes and hours are not logged with a hatched Pokemon so I can't really think of a reason to limit it. Date is somewhat questionable and if someone had a really good reason for opening it I will hear them out.
Looking forward to it! If we can get this down to one variable it will be very reasonable.
Maybe at least a range for dates?
I'm not sure how having a range for dates would help...honestly more often I want more results instead of less, helps you find elusive spreads. By using my method 12 searches will exhaust all possibilities for a single spread/offset/second combination possible, letting you know all your options. If I want anything, it's the ability to set seconds to all since I can reliably control that.
Yes, but why? Are you just wanting to hit an exact target date?
Also, you mentioned you were looking for work earlier. I do have some :)
Looking for 15-20 Uxie/Azelfs captures with a syncher and following information logged:
Nature of Syncher (use one please)
Nature of Pokemon Caught
IVs of Pokemon Caught
Initial Seed
This sort of ties in with the above, but basically for you, if you want the maximum amount of spreads, choose November. At that point opening the seconds does not matter because you will actually get all spreads that would even if you could do ranges of seconds.
Remember when initial seed is calculated the date/month/minutes/seconds all all combined to form a number from 0-255. I do believe November hits them all.
=o
thanks
so, i've pressed a lot of times (like 500)
just cuz i took too much time to press?
xD
hello..?
Hey mingot, great program, thanks for all your effort. Just wanted to suggest maybe altering the program so the initial seed generator can accept a range of IVs rather than absolute values, and then brute force them until a seed with an acceptable delay is found (perhaps include a delay range field as well). Generally there should only be about 32 possible IV spreads if you catch something around level 50, and most spreads will probably be invalid. Or you could just output every possible seed, and leave it up to the user to figure out which one looks correct (which is basically the same as what already happens, except with more false positives- but it should still be pretty obvious which one is the correct one). This is what I've been doing manually, as I don't want to waste rare candies, and IV battles take too long to do (I'm also on Pearl as I haven't gotten a shiny in Platinum yet so my SID is unknown there, so the highest level wild pokémon is only 60ish). It would also speed up the initial seed generation process as solving every IV takes time, even with rare candies.
A suggestion for the DPPt Time Finder...
Can it be set to look for a specific pattern (like A\0\B\B\11\23), with the ability to limit the date\hour\minute to reduce the search space? On top of that, be able to specify a range for the offset too? It would make finding the egg offset a lot easier.
EDIT: So, I just discovered "Find Adjacent Spreads". I guess that takes care of the offset range, but still - specific patterns would be helpful.
Can it be set to look for a specific pattern (like A\0\B\B\11\23), with the ability to limit the date\hour\minute to reduce the search space? On top of that, be able to specify a range for the offset too? It would make finding the egg offset a lot easier.
EDIT: So, I just discovered "Find Adjacent Spreads". I guess that takes care of the offset range, but still - specific patterns would be helpful.
Any wild pokemon battle should work (not the SR/roaming ones though).
So yeah, you can save in front of, or on the water and fish up a high level Magikarp.
Anyways, I also have a question:
This was my first egg result for finding your delay and offset:
Initial Seed: 6711025A (from the caught pokemon)
Egg Frame/Offset: 5
Delay: 593
Seconds: 13
So, now I need to do this a few more times, and then use the results I got the most?
Just to confirm, do I have all needed information for the egg above?
Has anyone found a perfect spread as in 31/31/31/31/31/31 with an offset of 3?
EDIT: nevermind i found it
EDIT: nevermind i found it
You don't need to switch to it, but you MUST switch away from the happiness checker.
Soon? :D
I'm not 100% sure how to make it work right. I wrote some sort of a guide. It's pretty much the same as Johnnys though.
I'm working on Infernape right now(not shiny) and I'm trying to get the game to give me a consistent offset of 5. If I get my spread then I'll post my guide.
Thanks.
Right now the plan for to make this easier actually involves putting a simple IV checker (single level, characteristic text, and hidden power if known, + IVs and nature) onto a new tab on the seed finder dialog. At that point I would do exactly as you describe and brute force through all of the combinations (perhaps with a warning if it's more than some reasonable number of combinations).
One thing I have already done to get ready for this is to clean up the seed finding code quite a bit. It will no longer search as deep for an initial seed. Delays are being capped at 3,600 which will give you 60 seconds from reset until entering the game world. Currently it's at 36,000 and this leads to a load of false positives that only serve to confuse. Offsets go back by 10,000 which, again, is completely over the top. This is getting trimmed to 1,000 which is still quite high. I've been testing with these new constraints and my results are usualy ONE initial seed, which will help out with confusion.
Next to go is the line item for the monster all together. Having to know whether to select a "monster" or "initial" seed just leads to questions and I can't really see any value in this.
I guess my long winded comments here were just to say "yes, I agree and I am working on it" but I wanted to at least put out there what the plan was to make sure no one has any huge objection to the changes.
I take it that initially you were trying to figure out what you actually DID hit when you collected and egg and wanted to narrow down the search range and get tighter control over how the searching worked?
Glad you found the Adjacent function, then heh. I do agree that it would be nice to be able to specify a hard IV and NOT have it include those that are inherited. This is more important now that we're resetting for the shiny nature and can't really stop and breed intermediate parents with the IVs that we need and are forced to use what we have.
I will need to think of a way to handle this easily. Perhaps a few more compare functions:
A
B
A or B
and then some way to split the current "=" into a hard equal (matches non-inherited only) and a soft equal (current functionality).
Also, one thing I would consider doing to help you searching in the adjacent is to export the file to CSV and open it in notepad. A simple text search is sometimes very helpful.
You do have the correct information, and yes you should do it a few more times. If your delays are always in the 590's that's where you are you are going to want to search. You are also "practicing" to get a consistent value for seconds here.
Opening the Pokedex does not consume any seeds.
I tested this by standing next to the egg man while he has an egg and saving, then resetting and immediately bringing up the menu, then opening and closing the pokedex ten times. I then received the egg and flew to the resort area to catch a Magikarp to calculate my seed, then hatched the egg to determine my offset. My offset ended up being 7, so obviously opening and closing the dex did not advance RNG by one each time. More as I get it, I'll do an arbitrarily large number of pokedex open/closes just to make sure there is no effect.
I tested this by standing next to the egg man while he has an egg and saving, then resetting and immediately bringing up the menu, then opening and closing the pokedex ten times. I then received the egg and flew to the resort area to catch a Magikarp to calculate my seed, then hatched the egg to determine my offset. My offset ended up being 7, so obviously opening and closing the dex did not advance RNG by one each time. More as I get it, I'll do an arbitrarily large number of pokedex open/closes just to make sure there is no effect.
Also test things in the pokedex and see if that advances it. Like searching, clicking a pokemon and looking at its area.
Did anyone managed to get BOTH odd and even delays on the same game? I need someone to confirm.
My game only produces odd delays(617,619 etc). Some people said only get even delays. Is that clearly dependent on each ones game?
That means that I can only work with spreads with odd delays on my game.
Also, it's not hard to predict it. I can get 617, 619 or 621 easily and those are my most common delays.
With that said, you should only target the spreads that have a delay you can hit consistently. For example, I would never go for a spread with a 624 delay because I can never get one.
I'm still trying to figure how to get a consistent offset without going through the "reset-catch a pokemon" phase though.
My game only produces odd delays(617,619 etc). Some people said only get even delays. Is that clearly dependent on each ones game?
That means that I can only work with spreads with odd delays on my game.
Also, it's not hard to predict it. I can get 617, 619 or 621 easily and those are my most common delays.
With that said, you should only target the spreads that have a delay you can hit consistently. For example, I would never go for a spread with a 624 delay because I can never get one.
I'm still trying to figure how to get a consistent offset without going through the "reset-catch a pokemon" phase though.
Read my mind, already in the process of doing that. Update in a few minutes...
And Darklord, I have had both 597 and 594, so yes you can have both odd and even in the same game.