Past Gen RNG Research

Kaphotics

Remodeling Kitchens
is a Top Researcher Alumnusis a Top Contributor Alumnus
Oh my goodness! I forgot about my waiting Victini!

Anyways, Omega, I am kind of wondering if the breeding side of things is just out for now because when I enter my seeds into the seed to time finder I get "this seed doesn't work" message even though I quintuple check my numbers/letters for the seed and MAC address.
Breeding isn't in yet (inheritance). Still hasn't been figured out.

Seed to Time only works for C-Gear seeds.

mattj, if only I knew of that earlier LOL, I manually searched through seeds by changing the delay +1... :(
 
Breeding isn't in yet (inheritance). Still hasn't been figured out.

Seed to Time only works for C-Gear seeds.

mattj, if only I knew of that earlier LOL, I manually searched through seeds by changing the delay +1... :(
Actually, the Seed to Time works for reg seeds too, I just put in zero for delays and it allows you to run a regular set of frames, as if the C-Gear is off.
 

ΩDonut

don't glaze me bro
is a Programmer Alumnusis a Forum Moderator Alumnusis a Top Researcher Alumnusis a Top Contributor Alumnus
Actually, the Seed to Time works for reg seeds too, I just put in zero for delays and it allows you to run a regular set of frames, as if the C-Gear is off.
No, it doesn't. Non-C-Gear seeds are generated very differently than C-Gear seeds; there's some SHA-1 encryption involved, the exact details of which we don't know. Seed to Time might list some "adjacent" seeds, but they're all wrong and will definitely not give you the correct results.

If we knew what was being encrypted, we wouldn't need to use the C-Gear at all. We could just start our game at a particular time, advance the RNG a little, and that'd be the end of it.
 
No, it doesn't. Non-C-Gear seeds are generated very differently than C-Gear seeds; there's some SHA-1 encryption involved, the exact details of which we don't know. Seed to Time might list some "adjacent" seeds, but they're all wrong and will definitely not give you the correct results.

If we knew what was being encrypted, we wouldn't need to use the C-Gear at all. We could just start our game at a particular time, advance the RNG a little, and that'd be the end of it.
Yeah, lol. EE and I figured that out. Too bad though.
 

Kaphotics

Remodeling Kitchens
is a Top Researcher Alumnusis a Top Contributor Alumnus
Hell yeah! and how are the roamer's IVs are screwy? I didn't get the 2,5,6 or whatever meant.
Instead of IV frames 1,2,3,4,5,6 when you have a frame count of 0270, they will be IV frames 2,3,4,7,5,6. Effectively shifted +1 and rearranged. (HP, Atk, Def, SpA, SpD, Spe)

==

If a sync works, it works for all natures. PID is unaffected by sync, but certain PIDs cannot be synced for different natures.

This is how I was able to catch all those Dittos with the same spread, same PID, but different nature. Took me 14 encounters until I got a sync accepting frame with an encounter slot holding Ditto, then I just repeated with different sync natures.

==

MAC Address (123456, DeSmuME) IV Spreadsheet [Lite 1%]

There's a larger 100% seeds searched on the way. Does not include lowest Atk IV for lowest confusion recoil, or 0 Speed spreads.
 

mattj

blatant Nintendo fanboy
Frames

  1. 13
  2. 30
  3. 19
  4. 31
  5. 15
  6. 04
  7. 09
  8. 12
  9. 17
  10. 00
  11. 08
  12. 16
  13. 07
etc, etc...


For all captured mons AND for Fossil Mons, the IVs will be frames 1, 2, 3, 4, 5, 6 in that order, so they would be 13 / 30 / 19 / 31 / 15 / 04(HP / Atk / Def / SAtk / SDef / Spd).

For your roamer, the IVs will be frames 2, 3, 4, 7, 5, 6 in that order, so they would be 30 / 19 / 31 / 09 / 15 / 04 (HP / Atk / Def / SAtk / SDef / Spd).

For eggs, the "base egg IVs" will be frames 8, 9, 10, 11, 12, 13 in that order, so they would be 12 / 17 / 00 / 08 / 16 / 07 (HP / Atk / Def / SAtk / SDef / Spd).

Things that we know can change that are:

  1. walking 128 steps, which advances your starting IV frame forward by 1 for every mon in your party
  2. dragging and dropping a mon from your PC into your party (can be repeated as many times as you want), which advances your starting IV frame forward by 7 each time you do it
 

Kaphotics

Remodeling Kitchens
is a Top Researcher Alumnusis a Top Contributor Alumnus
Eggs are actually just like the Withdrawing advancement except IVs are generated (second).

Instead of
Code:
Regular [1-6]
Starting  0 (One advance to get egg IVs)
Regular ->1,  2,  3,  4,  5,  6

Eggs [8-13]
Starting  0
Placing  +7 (happens before base IVs grabbed, (One advance to get egg IVs))
Ending  ->8,  9, 10, 11, 12, 13 (base IVs grabbed)
I guess the reason it does this is because of inheritance playing a role in the resulting IVs. This is why Gift mons aren't like eggs.

==

Here's the MTRNG position with using a C-Gear. It looks like the C-Gear seed is hit, and then advanced twice from the initial (0270), and thus transformed. Every 0270 it transforms. Trash value.

Code:
031B4645 upon hitting --> 7b29148E
Timer Sync Value 00000B62 (play timer)

624*	Value
-1	031B4645 	Initial Seed
0	7b29148e	Transformed, +2 C-Gear advance
1	b0190a6b	Battle Process
2	2b623239	...
3	7f889586	...
4	01beb8a5	...
5	6b75a8a9	...
6	5b590699	...
7	32d50ced	...
8	75fbf79b	...
9	a8fe98e1	...
10	92a2be02	...
11	7a70da8e	...
12	5bfa8bdb	Flee late in the 0270 counter
13	6FA2811e	Exiting battle passes 270 and  ^ transforms
 
I have noticed something interesting while fiddeling with the C-gear, your seed also changes when returning to C-Gear from the Wi-Fi settings screen. I don't know if this could have any pratical use... I also haven't verified if it uses the same formula to determine the seed when this happens or whether this is a frame advancement or not.

EDIT: Time to Seed accepts the resulting seed as a valid seed, so I guess it's a C-gear seed?

EDIT2: Seed to time is not quite reliable yet, right? It's accepting some seeds while rejecting others...
 

ΩDonut

don't glaze me bro
is a Programmer Alumnusis a Forum Moderator Alumnusis a Top Researcher Alumnusis a Top Contributor Alumnus
Okay, let me make this perfectly clear once and for all: The MTRNG check code does NOT show you the initial seed.

What it shows you the first value in the table of 624 values that the MTRNG uses. This value just *happens* to be equal to the initial seed, but only if you start the game without turning on the C-Gear.

It will not show you the seed if you use the C-Gear, because using the C-Gear modifies all the values in the table so you can't guess the initial seed.

It will not show you the seed if the RNG advances beyond frame 624, because it recalculates the entire table.
 
Okay, let me make this perfectly clear once and for all: The MTRNG check code does NOT show you the initial seed.

What it shows you the first value in the table of 624 values that the MTRNG uses. This value just *happens* to be equal to the initial seed, but only if you start the game without turning on the C-Gear.

It will not show you the seed if you use the C-Gear, because using the C-Gear modifies all the values in the table so you can't guess the initial seed.

It will not show you the seed if the RNG advances beyond frame 624, because it recalculates the entire table.
Ah, I see. So what would the change of the first value of that table indicate then?

EDIT: It seems to rewrite the entire table when exiting the Wi-Fi menu, would this mean it could basically have the same effect as turning on the C-Gear? I there a way I could verify this?
 

Kaphotics

Remodeling Kitchens
is a Top Researcher Alumnusis a Top Contributor Alumnus
The 0270 frame counter advancing at least once.

It starts at 0270 (equivalent to 0000), +7 it changes to 0007. Since it passed 0270, it is recalculated.

For when you hit the C-Gear seed, I think it is just advancing twice after you hit the seed. The same deal of advancing past 0270 causes it to change to the recalculated value that doesn't predict anything.

I just don't see why the C-Gear would seed to 0002 instead of 0270 anyways.

You can use that first C-Gear value to verify that you hit your C-Gear seed, so long as you verify it first by encountering a Pokemon for the IVs.

The effect of it changing has no change on the IVs, the initial seed recalculated (no c-gear) will predict all the IVs for the playthrough. For C-Gear seeds, even leaving the C-Gear on after hitting, the IVs will be predicted by that initial C-Gear seed no matter the frame (transformations)
 
The 0270 frame counter advancing at least once.

It starts at 0270 (equivalent to 0000), +7 it changes to 0007. Since it passed 0270, it is recalculated.

For when you hit the C-Gear seed, I think it is just advancing twice after you hit the seed. The same deal of advancing past 0270 causes it to change to the recalculated value that doesn't predict anything.

I just don't see why the C-Gear would seed to 0002 instead of 0270 anyways.

You can use that first C-Gear value to verify that you hit your C-Gear seed, so long as you verify it first by encountering a Pokemon for the IVs.

The effect of it changing has no change on the IVs, the initial seed recalculated (no c-gear) will predict all the IVs for the playthrough. For C-Gear seeds, even leaving the C-Gear on after hitting, the IVs will be predicted by that initial C-Gear seed no matter the frame (transformations)
The frame goes to two after exiting the Wi-Fi menu, so I guess it's the same thing happening like when you turn on the C-gear. So I suppose there's not much to be gained from this.
 

Kaphotics

Remodeling Kitchens
is a Top Researcher Alumnusis a Top Contributor Alumnus
I believe it is launching Wi-Fi communications that reseeds the MTRNG. But the only practical way of reseeding the game is with the C-Gear.

I saw the same thing and tested it out with the Wi-Fi rooms in the PokeCenter to see for other ways to advance the frames awhile ago :P
 
I believe it is launching Wi-Fi communications that reseeds the MTRNG. But the only practical way of reseeding the game is with the C-Gear.

I saw the same thing and tested it out with the Wi-Fi rooms in the PokeCenter to see for other ways to advance the frames awhile ago :P
Well, then it relaunches them for the C-Gear after returning from the menu.
 

ΩDonut

don't glaze me bro
is a Programmer Alumnusis a Forum Moderator Alumnusis a Top Researcher Alumnusis a Top Contributor Alumnus
I just don't see why the C-Gear would seed to 0002 instead of 0270 anyways.
It's entirely possible this was done to make it more difficult for people to figure out how the MTRNG was involved in C-Gear seeds, so they wouldn't be able to, say, program RNG Reporter to be able to predict them.
 

Bond697

Dies, died, will die.
kaph isn't kidding, there's a MASSIVE spreadsheet coming. i've had RNG reporter churning on a server for a little over 2 days so far and it's 1/3 of the way done.

e: i guess since i'm posting i might as well ask.

i've had a hell of time hitting seeds for some reason. i'm using desmume and monitoring all relevant addresses, so i know i'm hitting everything properly. in almost 3 days of trying i've hit the right seed 7 times. that's it. kaph is able to hit it pretty often, and i don't expect to be that good, but if i'm using frame advance and closely monitoring, i should be hitting it. i even experimented with freezing some of the addresses to guarantee that i would hit it no matter what. it still didn't work. the couple times i did hit the seed i had different results than kaph from the cgear-transformed seeds, too. i had to save state and catch a poke to verify.

the seeds i do end up hitting are no where near the target. i'm not even hitting adjacent ones. anyone else having trouble like this?
 
kaph isn't kidding, there's a MASSIVE spreadsheet coming. i've had RNG reporter churning on a server for a little over 2 days so far and it's 1/3 of the way done.

e: i guess since i'm posting i might as well ask.

i've had a hell of time hitting seeds for some reason. i'm using desmume and monitoring all relevant addresses, so i know i'm hitting everything properly. in almost 3 days of trying i've hit the right seed 7 times. that's it. kaph is able to hit it pretty often, and i don't expect to be that good, but if i'm using frame advance and closely monitoring, i should be hitting it. i even experimented with freezing some of the addresses to guarantee that i would hit it no matter what. it still didn't work. the couple times i did hit the seed i had different results than kaph from the cgear-transformed seeds, too. i had to save state and catch a poke to verify.

the seeds i do end up hitting are no where near the target. i'm not even hitting adjacent ones. anyone else having trouble like this?
I don't know the answer to your question but may I ask you which version of desmume you're using? That info might be helpful for others who read your question.

Now, this is hilarious, I was trying to catch a shiny cobalon but the expected PID is different by one value if I try to catch it in the "proper" save file...

C5586198 is what I expected.
45586198 is what I get in the save file which would make the first one shiny.

EDIT: It reminds me of Kaphotics' post about zekrom, but I didn't use a save editor, I made it all the way to cobalon on my 44th save file ç_ç
 

Kaphotics

Remodeling Kitchens
is a Top Researcher Alumnusis a Top Contributor Alumnus
It's for DeSmuME 0.9.6, well it's tailored for the XX-XX-XX-12-34-56 MAC address range.

I'll test out Cobalon...



Looks like you got pwn'd by this.
(Wild Pokemon: XOR'd with 0x80010000 or 0x00010000, one was missing a 0)
 
does this mean that Zekrom and Reshiram are infact possible to obtain shiny?
no, there was no mistake there.
According to Serebii there seam to a way to catch Zekrom/Reshiram shiny, maybe someone should test it..

Serebii:
With thanks to the recently published guide, we now know that if, when you reach the end of the game, you have a maxed party and maxed boxes, the game will progress without you needing to capture Reshiram/Zekrom (previously, the game will not continue until it is captured, and you will be able to capture Reshiram/Zekrom in Black & White respectively in the Dragon Spiral Tower. They will once again be at Level 50 and this will allow them to be Shiny. Previously, the Reshiram/Zekrom in N's Castle was discovered to be unable to be Shiny but these can. The Legendary Pokémon page has been updated accordingly

 

Bond697

Dies, died, will die.
that sounds like it's straight out of 1998. it even has no cited source.

slowly filling up a save via pokegen. wow is this taking awhile... 24 boxes is a lot.

e: well, i filled up all 24, but the game would only show me the first 16. it was kind of odd. N still refused to fight me unless i encountered the legendary, though. after i did, i said i wanted to send it to the box(not the party). after me refusing twice, the game MADE me put it in my party. only thing was, after i chose a spot to replace, i went into the battle with N and the zekrom was nowhere to be found in my party. the poke i chose was never replaced. so i thought that maybe it was true and once i beat the game i'd find zekrom in the tower, etc, since there was no room anywhere with me. i tried this twice, and both times the game froze after i beat it. i can't see this actually working.
 

Users Who Are Viewing This Thread (Users: 1, Guests: 0)

Top