RNG Research - Page 2

Red_Phoenix · Sep 17th, 2009, 12:43:10 PM

Ok, I want to help too. I've been testing this new RNG on the eevee that Bill gives you.
And I can confirm that this eevee is abusable (I got a shiny one) and is generated as a method 1 pokemon with a starting frame of 3 (not confirmed).
Well, if someone could double check the starting frame it would be great since I am not 100% sure, I just did the save state as soon as I entered the game.
As ClikC (thank you) said the lower right radio station advance the frame by 1. (that's how I advanced the frame)
Also, there is a girl in Bill's house who can advance the frame, but I took the eevee before she moved.

I'll post if I find something new, good luck to everyone who is testing. ;)

LightningFusion · Sep 17th, 2009, 12:46:40 PM

Now that you mention a pokemon you are given, I did some tests on the shuckle you receive. The shuckle IVs are preset. I am not sure if it is the same in all the games, or if it is generated when you defeat the gym leader or at some other point in time.

ClikC · Sep 17th, 2009, 1:06:33 PM

Quote:

Originally Posted by Fat LightningFusion

Now that you mention a pokemon you are given, I did some tests on the shuckle you receive. The shuckle IVs are preset. I am not sure if it is the same in all the games, or if it is generated when you defeat the gym leader or at some other point in time.

I haven't checked, but according to certain sites. You get given that pokemon to look after from an in game trainer. It will most likely have predetermined nature and stats, as-well as that trainers OT and IDno. Probably preset in the ROM data of the game.

Red_Phoenix; can you do a quick check of your map for me? I want to know how many roaming pokemon are on your map screen (Both Kanto and Jhoto), particularly Jhoto, where you get that Eevee given to you.

Red_Phoenix · Sep 17th, 2009, 1:10:40 PM

Quote:

Originally Posted by Fat ClikC

I haven't checked, but according to certain sites. You get given that pokemon to look after from an in game trainer. It will most likely have predetermined nature and stats, as-well as that trainers OT and IDno. Probably preset in the ROM data of the game.

Red_Phoenix; can you do a quick check of your map for me? I want to know how many roaming pokemon are on your map screen (Both Kanto and Jhoto), particularly Jhoto, where you get that Eevee given to you.

I still haven't unlocked Kanto since I just have four badges. As of now I have two roamers (Raikou and Entei) on my map.

Dance Soldier · Sep 17th, 2009, 1:12:34 PM

Well normally I would say that the Shuckle's IV's are set because it gets
the IV's of the Traded Pokemon. That was the case always before.
So try trading a Flawless Pokemon to recieve a Flawless Shuckle.
Maybe it's different now but from what I know that type of In game Trades
never changed so I would say:
Traded Pokemons IV's = Recieved Pokemons IV's

and btw Thx for the explanation with the method K. I resetted the Mewtwo normally and it has pretty decent IV's so I more than fine with it.

ClikC · Sep 17th, 2009, 1:14:38 PM

Quote:

Originally Posted by Fat Red_Phoenix

I still haven't unlocked Kanto since I just have four badges. As of now I have two roamers (Raikou and Entei) on my map.

I think it's safe to say that the two roamers have advanced your frame for Eevee, and in-fact it's starting frame is 1.

If you want to confirm that and have back-up saves to work from, defeat/ capture Entei and Raikou, then collect Eevee again and check.

Quote:

Originally Posted by Fat Dance Soldier

Well normally I would say that the Shuckle's IV's are set because it gets
the IV's of the Traded Pokemon. That was the case always before.
So try trading a Flawless Pokemon to recieve a Flawless Shuckle.
Maybe it's different now but from what I know that type of In game Trades
never changed so I would say:
Traded Pokemons IV's = Recieved Pokemons IV's

and btw Thx for the explanation with the method K. I resetted the Mewtwo normally and it has pretty decent IV's so I more than fine with it.

You don't trade for shuckle, you are given it to look after.

Wichu · Sep 17th, 2009, 1:16:29 PM

Quote:

Originally Posted by Fat ClikC

Using the PokéGear's radio function, I alternated between clicking the bottom right hand button, and using the stylus to de-tune the radio (moving the circle to the center) and then clicking the lower right hand button again. This advanced the frame 1 at a time once I was close enough to my target frame using save states. However, as I previously showed, different stations may advance the RNG at different rates, so we need to test and see if a reliable way to advance frames can be found. It should be noted, when de-tuning the radio that accidently touching another station will advance the RNG, so be very careful.

Could you alternate between two stations using the 4 buttons instead of detuning it? Also, if you repeatedly tap a radio station button, does that advance the RNG?

Red_Phoenix · Sep 17th, 2009, 1:21:10 PM

Quote:

Originally Posted by Fat ClikC

I think it's safe to say that the two roamers have advanced your frame for Eevee, and in-fact it's starting frame is 1.

If you want to confirm that and have back-up saves to work from, defeat/ capture Entei and Raikou, then collect Eevee again and check.

My bad, I completely forgot about the roamers that advance your starting frame.
Yeah, eevee's starting frame should be 1, thank you for correcting me.

ClikC · Sep 17th, 2009, 1:22:29 PM

Quote:

Originally Posted by Fat Wichu

Could you alternate between two stations using the 4 buttons instead of detuning it? Also, if you repeatedly tap a radio station button, does that advance the RNG?

Well, from mine and LF's experiences, the Top Right and Bottom Left buttons give random results of about 14~16 and 3~5 frames. I tried chaining buttons together, but to no great avail. The only consistent buttons are top left (no advances) and bottom right, which only advances you 1 frame if you keep selecting it after de-tuning.

Syberia obviously did some testing on lugia, and radio reception is none existent in the whirl islands. So further testing, and hopefully finding a newer and better method of advancing the offset is required.

**ΩDonut** · Sep 17th, 2009, 1:54:18 PM

Quote:

Originally Posted by Fat LightningFusion

The Average delay is 480 - 510 about. You would find your seed by catching a method 1 pokemon. Any pokemon that go by the new method "k", you can't do until the alogorithim is found. Don't save after, obviously.

Does Method K, like Method J, basically use the Method 1 list and then have some technique to decide which Method 1 frame to land on? If so, you can find your seed catching a Method K Pokemon as well.

ClikC · Sep 17th, 2009, 2:16:18 PM

Quote:

Originally Posted by Fat OmegaDonut

Does Method K, like Method J, basically use the Method 1 list and then have some technique to decide which Method 1 frame to land on? If so, you can find your seed catching a Method K Pokemon as well.

Yes, in short. You can defiantly check what seed you are on. Of course it's not much use for knowing how far it is you have to go, which is why save states are being used.

Dance Soldier · Sep 17th, 2009, 3:22:48 PM

As of what I noticed is that the Method K advances in an unpredictable way so far casue I had differnet frame jumps
from what I just caught and stood right in front facing this "K" and the Frame "Jumps" were : 6,8,17,19,36,82,117,171 from my testing phase.

So as long as noone tells me how to look into the Gamescript I can't help
you further with that.

Hey Guys It would be great if someone could make a List of Interactable Pokes/ Gift / Legends and note their respective Method.
So a simple Question which Pokemon in the Game are generated by which Method?

Thx in advance
greets
DS

DarkLynx · Sep 17th, 2009, 3:41:48 PM

Has someone tested that thing you get from the cientist on Unown Ruins? I don't quite remember the name, but each Unown has a 'word' assigned to it. It has two pages, so I was wondering if that would advance the RNG (I just pwned Whitney, so I can't RNG abuse Eevee or Dratini yet, and I can't play right now because my DS is at home).

***mingot***** · Sep 18th, 2009, 7:08:20 PM

Method K (With Syncher) =

seed(0):
If even, synchronize
If odd:
target nature = seed(1) mod 24

And then pairs of RNG calls to generate PIDs to test.

Very much the same thing as Method J with slightly different math.

I am about 95% sure this is what I am seeing, just need to write some code and then verify. I'll probably release an interim version of RNG reporter to a few folks on IRC tonight who have time to test.

Yay me.

EDIT: For cases with no synchronizer:

target nature = seed(0) mod 24

And then pairs of RNG calls to generate PIDs to test.

**ΩDonut** · Sep 18th, 2009, 7:17:02 PM

Quote:

Originally Posted by Fat DarkLynx

Has someone tested that thing you get from the cientist on Unown Ruins? I don't quite remember the name, but each Unown has a 'word' assigned to it. It has two pages, so I was wondering if that would advance the RNG (I just pwned Whitney, so I can't RNG abuse Eevee or Dratini yet, and I can't play right now because my DS is at home).

Just tested. Unown catalog page flipping does not do anything, sadly.

Good to hear Method K's been mostly finalized. Any idea how it works without a Syncher? (Wait, never mind, it's seed(1) mod 24, isn't it?)

***mingot***** · Sep 18th, 2009, 7:20:55 PM

I strongly suspect it is just going to be the change from the old division to the MOD math, but I need to check into it. This is what happens when I get homw a few hours early and have no family around ;) They are here now, though, so I will probably have to wait until later tonight to finish testing this all.

You on IRC would be so nice. All the other researchers are there :P

**ΩDonut** · Sep 18th, 2009, 8:49:36 PM

I'm getting some inconsistent results with the radio so far.

I'm starting on frame 3 (two runners out), and the radio is preset to start on the bottom-right station.

Click the top-left station: frame 4
Click the top-left station, then click bottom-right: frame 5
Click the top-left station, then click bottom-right, then click top-left: frame 3
Click the top-left station, then click bottom-right, then click top-left: frame 4

I'll be doing some more tests tonight. Normally I have a difficult time connecting to IRC at work, which is where I am a lot because I only have dial-up at home. So that's why I'm not on IRC. But I'll see if I can be there tonight. Which room?

***mingot***** · Sep 18th, 2009, 9:15:07 PM

I have had problems getting the radio to do much of anything, but I may not be doing it right. . . #smogonwifi, #smogon, #stark, or the indie scoop if you want to help out. I just finished adding Method K to RNG Reporter, but it needs some serious testing since I am having some issues capturing legendaries.

I will PM you the link shortly.

**ΩDonut** · Sep 19th, 2009, 2:42:05 AM

After checking my original Silver (which HOLY SHIT still works, most people's internal battery died three years ago), I rediscovered that some of the radio programs change depending on the day of the week. So, for consistency, people should also record that when testing the radio. I think it's the bottom-right station that does.

Some of the programs choose Pokemon and phrases at random, which is probably why the RNG advances. It also might mean that the RNG will keep advancing if you leave a station playing.

EDIT: Yeah, I just turned on the radio without clicking anything, and I let the song finish playing. The RNG advanced 3 frames without any other intervention.

Moreover, the radio script doesn't restart if you change stations. It will start where it left off before you switched stations. I think this pretty much shoots down flipping through stations as a controllable method for RNG advancement. I *could* figure out how far the RNG advances through one playthrough of a song, but that's so time-consuming you might as well keep 5 Pokemon in your party and walk 17152 steps to advance the frame by 670 or whatever.

I'll look into talking to your Pokemon next.

**ΩDonut** · Sep 19th, 2009, 8:48:24 PM

Double-posting because a) it's been a while and b) this is worthy.

I think the responses from talking to your Pokemon are key. I'm pretty certain that each individual response advances the RNG by a set amount. Well, most of them anyway - there's one that seems to increase by a variable amount, a "dancing box" response. I'm still trying to figure it out. I'll post my results once I finish testing all of the responses.

Once we figure out what response we get on which frame, we can have RNG Reporter list the pattern of Pokemon responses to determine if you've hit the correct seed, like with coin flips. It could also say "talk to your Pokemon x times" to get the frame you want.

EDIT: Here's the data on some of the things I've tested. I won't have any more time tonight to do stuff with this, so any other interested people can make use of this.

Code:

Seed		Delay	Frame	Increase Expected	Method
EF1401F2		3	0			open radio only
			3	0			open radio only
41601FC			4	1			switch to upper left
251601FC	499	5	2			switch to upper left, back to lower right
121701FA	497	3	0			switch to upper left, back to lower right, switch to upper left
FE170238	559	4	1			switch to upper left, back to lower right, switch to upper left
20401F4		491	7	4			switch to upper left, back to lower right, switch to upper left
190601FE	501	17	14			speak to Pokémon (… response)
F21501FE	501	17	14			speak to Pokémon (… response)
1501F4		491	22	19			speak to Pokémon (!… response)
41501FE		501	25	22			speak to Pokémon (dancing box)
381501FE	501	36	33			speak to Pokémon (?, yes, item gotten)
501501F6	493	45	42	41		speak to Pokémon (!… response, then dancing box)
F21601F2	489	36	33			speak to Pokémon (?, no, happy dance)
141601FC	499	41	38	38		speak to Pokémon (!… response) x2
41701F0		487	45	42	42		speak to Pokémon (… response) x3
11701F2		489	75	72	60		speak to Pokémon (!… response) x2, then dancing box
31170200	503	76	73			speak to Pokémon (happy dance)
EA0201FA	497	29	26			speak to Pokémon (heart)
F0201F6		493	63	60	60		speak to Pokémon (!… response) x2, then dancing box

That last line makes me think the dancing box response isn't as variable as I thought - maybe I accidentally did something to advance the RNG.

Syberia · Sep 21st, 2009, 7:33:29 PM

Unfortunately, it appears that walking 128 steps no longer works, at least in Suicune's area

EDIT: It's definitely doing something with no syncher, but with a syncher my results may be inconsistent
EDIT2: I see what's happening. With a seed of F30004C2, no syncher, and 3 roamers out, I get a starting frame of 4. With a syncher, playing from the same save state, I get a starting frame of 5. With a syncher, walking 256 steps, I get a frame of 15. Without, I get 14.

EDIT3: Maybe my calculations were off, as now I am getting a frame of 15 after walking 256 steps with 5 pokes in my party both with and without a syncher. However, in previous testing (with different seeds), I was getting an offset of 4 with a syncher, and either 4 or 5 without. This is weird.

***mingot***** · Sep 21st, 2009, 7:34:59 PM

Actually, I can verify that it does work. I have walked in that area under debugger.

Syberia · Sep 21st, 2009, 8:27:16 PM

Yeah, it works, I was being dumb. But the offset thing did happen, I tested 3 times under each case.

***Wild Eep**** · Sep 23rd, 2009, 1:14:55 PM

Quote:

Originally Posted by Fat Viletung

I'm currently trying to do this for the Dogs (the roaming ones). And I'm not sure if you know this: But I'm at the buttom of the tower, I walked 2304 steps with 3 pokemon, that should advance the frame 54, but when I caught the Dog, it was over 2000. Now going to test with 1 Pokémon.

I don't have HGSS, but I've read from other posters who do have it that the area with the legendary beasts contains wild Pokemon, so the RNG will advance with every step and turn.

**ΩDonut** · Sep 23rd, 2009, 5:09:33 PM

Quote:

Originally Posted by Fat Wild Eep

I don't have HGSS, but I've read from other posters who do have it that the area with the legendary beasts contains wild Pokemon, so the RNG will advance with every step and turn.

Yeah, but you'd think Viletung would notice if he had kept running into wild Pokemon or had to use Repel over and over. I think the floor where the dogs are is a dead zone, but the floor above it has wild Pokemon.

Still doesn't explain the >2000 frame, though.

Sep 17th, 2009, 12:43:10 PM	#26
Red_Phoenix Join Date: Jul 2009 Posts: 36 Italy	Ok, I want to help too. I've been testing this new RNG on the eevee that Bill gives you. And I can confirm that this eevee is abusable (I got a shiny one) and is generated as a method 1 pokemon with a starting frame of 3 (not confirmed). Well, if someone could double check the starting frame it would be great since I am not 100% sure, I just did the save state as soon as I entered the game. As ClikC (thank you) said the lower right radio station advance the frame by 1. (that's how I advanced the frame) Also, there is a girl in Bill's house who can advance the frame, but I took the eevee before she moved. I'll post if I find something new, good luck to everyone who is testing. ;) __________________ Diamond FC:2449 4856 0340 Platinum FC (Battles):3523 9484 6104 New Platinum FC:0174 0292 4963

Sep 17th, 2009, 12:46:40 PM	#27
LightningFusion Join Date: Mar 2009 Posts: 1,781 Michigan	Now that you mention a pokemon you are given, I did some tests on the shuckle you receive. The shuckle IVs are preset. I am not sure if it is the same in all the games, or if it is generated when you defeat the gym leader or at some other point in time. __________________ Diamond Name: Fus!on FC: 2321 1627 1150 Nookington's

Sep 17th, 2009, 1:12:34 PM	#30
Dance Soldier Join Date: Jun 2009 Posts: 13 Berlin	Well normally I would say that the Shuckle's IV's are set because it gets the IV's of the Traded Pokemon. That was the case always before. So try trading a Flawless Pokemon to recieve a Flawless Shuckle. Maybe it's different now but from what I know that type of In game Trades never changed so I would say: Traded Pokemons IV's = Recieved Pokemons IV's and btw Thx for the explanation with the method K. I resetted the Mewtwo normally and it has pretty decent IV's so I more than fine with it. __________________ At the beginning of time Freezer was Born and destructed by the Hand of God and after a cleansing sleep he was Reborn as the Dance Soldier :P

Sep 17th, 2009, 3:22:48 PM	#37
Dance Soldier Join Date: Jun 2009 Posts: 13 Berlin	As of what I noticed is that the Method K advances in an unpredictable way so far casue I had differnet frame jumps from what I just caught and stood right in front facing this "K" and the Frame "Jumps" were : 6,8,17,19,36,82,117,171 from my testing phase. So as long as noone tells me how to look into the Gamescript I can't help you further with that. Hey Guys It would be great if someone could make a List of Interactable Pokes/ Gift / Legends and note their respective Method. So a simple Question which Pokemon in the Game are generated by which Method? Thx in advance greets DS __________________ At the beginning of time Freezer was Born and destructed by the Hand of God and after a cleansing sleep he was Reborn as the Dance Soldier :P

Sep 17th, 2009, 3:41:48 PM	#38
DarkLynx Join Date: Apr 2009 Posts: 686 Brazil	Has someone tested that thing you get from the cientist on Unown Ruins? I don't quite remember the name, but each Unown has a 'word' assigned to it. It has two pages, so I was wondering if that would advance the RNG (I just pwned Whitney, so I can't RNG abuse Eevee or Dratini yet, and I can't play right now because my DS is at home). __________________ DarkLynx's Yotube Channel Platinum Friend Code: 0517 3141 5995 [Always use this, unless I state otherwise on my thread] The Big Bang Theory - Tsubasa-kun & DarkLynx joint Thread [I'll be back on trading next month]

Sep 18th, 2009, 7:08:20 PM	#39
*mingot* free agent Administrator** Join Date: Jun 2008 Posts: 3,045	Method K (With Syncher) = seed(0): If even, synchronize If odd: target nature = seed(1) mod 24 And then pairs of RNG calls to generate PIDs to test. Very much the same thing as Method J with slightly different math. I am about 95% sure this is what I am seeing, just need to write some code and then verify. I'll probably release an interim version of RNG reporter to a few folks on IRC tonight who have time to test. Yay me. EDIT: For cases with no synchronizer: target nature = seed(0) mod 24 And then pairs of RNG calls to generate PIDs to test. __________________ Check out the new In-Game section! RNG Reporter/RNG Help Thread (Discuss) RNG Research Thread

Sep 18th, 2009, 7:20:55 PM	#41
*mingot* free agent Administrator** Join Date: Jun 2008 Posts: 3,045	I strongly suspect it is just going to be the change from the old division to the MOD math, but I need to check into it. This is what happens when I get homw a few hours early and have no family around ;) They are here now, though, so I will probably have to wait until later tonight to finish testing this all. You on IRC would be so nice. All the other researchers are there :P __________________ Check out the new In-Game section! RNG Reporter/RNG Help Thread (Discuss) RNG Research Thread

Sep 18th, 2009, 8:49:36 PM	#42
ΩDonut sometimes experimentation begins with "let's multi battle strip pokémon" Moderator Join Date: Aug 2006 Posts: 3,691	I'm getting some inconsistent results with the radio so far. I'm starting on frame 3 (two runners out), and the radio is preset to start on the bottom-right station. Click the top-left station: frame 4 Click the top-left station, then click bottom-right: frame 5 Click the top-left station, then click bottom-right, then click top-left: frame 3 Click the top-left station, then click bottom-right, then click top-left: frame 4 I'll be doing some more tests tonight. Normally I have a difficult time connecting to IRC at work, which is where I am a lot because I only have dial-up at home. So that's why I'm not on IRC. But I'll see if I can be there tonight. Which room?

Sep 18th, 2009, 9:15:07 PM	#43
*mingot* free agent Administrator** Join Date: Jun 2008 Posts: 3,045	I have had problems getting the radio to do much of anything, but I may not be doing it right. . . #smogonwifi, #smogon, #stark, or the indie scoop if you want to help out. I just finished adding Method K to RNG Reporter, but it needs some serious testing since I am having some issues capturing legendaries. I will PM you the link shortly. __________________ Check out the new In-Game section! RNG Reporter/RNG Help Thread (Discuss) RNG Research Thread

Sep 19th, 2009, 2:42:05 AM	#44
ΩDonut sometimes experimentation begins with "let's multi battle strip pokémon" Moderator Join Date: Aug 2006 Posts: 3,691	After checking my original Silver (which HOLY SHIT still works, most people's internal battery died three years ago), I rediscovered that some of the radio programs change depending on the day of the week. So, for consistency, people should also record that when testing the radio. I think it's the bottom-right station that does. Some of the programs choose Pokemon and phrases at random, which is probably why the RNG advances. It also might mean that the RNG will keep advancing if you leave a station playing. EDIT: Yeah, I just turned on the radio without clicking anything, and I let the song finish playing. The RNG advanced 3 frames without any other intervention. Moreover, the radio script doesn't restart if you change stations. It will start where it left off before you switched stations. I think this pretty much shoots down flipping through stations as a controllable method for RNG advancement. I could figure out how far the RNG advances through one playthrough of a song, but that's so time-consuming you might as well keep 5 Pokemon in your party and walk 17152 steps to advance the frame by 670 or whatever. I'll look into talking to your Pokemon next.

Sep 21st, 2009, 7:33:29 PM	#46
Syberia Join Date: Jun 2005 Posts: 4,905 Irvine, CA	Unfortunately, it appears that walking 128 steps no longer works, at least in Suicune's area EDIT: It's definitely doing something with no syncher, but with a syncher my results may be inconsistent EDIT2: I see what's happening. With a seed of F30004C2, no syncher, and 3 roamers out, I get a starting frame of 4. With a syncher, playing from the same save state, I get a starting frame of 5. With a syncher, walking 256 steps, I get a frame of 15. Without, I get 14. EDIT3: Maybe my calculations were off, as now I am getting a frame of 15 after walking 256 steps with 5 pokes in my party both with and without a syncher. However, in previous testing (with different seeds), I was getting an offset of 4 with a syncher, and either 4 or 5 without. This is weird. __________________ Black/White Friend Code: 1721 2578 4968 My Pokemon \| Free Pokemon \| YouTube \| Wonder Cards (now with Movie Celebis for Platinum and HG/SS!)

Sep 21st, 2009, 7:34:59 PM	#47
*mingot* free agent Administrator** Join Date: Jun 2008 Posts: 3,045	Actually, I can verify that it does work. I have walked in that area under debugger. __________________ Check out the new In-Game section! RNG Reporter/RNG Help Thread (Discuss) RNG Research Thread

Sep 21st, 2009, 8:27:16 PM	#48
Syberia Join Date: Jun 2005 Posts: 4,905 Irvine, CA	Yeah, it works, I was being dumb. But the offset thing did happen, I tested 3 times under each case. __________________ Black/White Friend Code: 1721 2578 4968 My Pokemon \| Free Pokemon \| YouTube \| Wonder Cards (now with Movie Celebis for Platinum and HG/SS!)