|
|||||||
![]() |
|
|
Thread Tools |
|
|
#401 | ||
|
blatant Nintendo fanboy
Join Date: Mar 2009
Posts: 4,328
your mom sucked at e3
|
Thank you for doing that! I couldn't possibly check the starters on my cart. So, so far we've found:
Quote:
Quote:
Zekrom/Reshiram, Kyumeru, Victini, Musharna, Zoroark, Ulgamoth, The Gifted Monkey Trio, Zorua, That Magikarp you can buy (lol). But I don't really see a reason why most of them wouldn't have their IVs predicted by the IV MTRNG, because most "fit the categories" we've already looked at. I am curious about other "Gift" mons though.
__________________
|
||
|
|
|
|
|
#402 |
|
blatant Nintendo fanboy
Join Date: Mar 2009
Posts: 4,328
your mom sucked at e3
|
heee heee ^^^
Well, I found a decently good IV set while arbitrarily SRing the other day for a completely different purpose. It only appeared twice out of 15 times while Resetting on a certain second, so I figured it was related to some kind of human error and I was accidentally Resetting on some other near by second. So I decided to go back to that second, and do some Resets, adjusting my timer by 1/2 a second up and then later down until I started hitting it consistently. I keep getting really spotty results. Are we 100% sure that the Date/Time to the second is close enough? Could it be like 1/2 a second, or an 1/8th of a second or something? Just throwing it out there because the only proof I've seen so far is that Japanese blog.
...
I do realize this could all just be human error, but I'm using a timer which has given me results consistent to within 4 delay in the past. Could my DS Lite Clock just be that imprecise??
__________________
|
|
|
|
|
|
#403 |
|
Remodeling Kitchens
![]()
Join Date: Apr 2009
Posts: 758
Berkeley, CA
|
Could be, considering the AR is also in the way.
|
|
|
|
|
|
#404 | |
|
blatant Nintendo fanboy
Join Date: Mar 2009
Posts: 4,328
your mom sucked at e3
|
Okay, more on my suspicion that there's something weird going on with the "seed being generated based only down to the second".
I expanded my Soft Resets. I set my DS Date/Time to 10-25-2010 00:01:00 and started Emloop at that moment. I set emloop first to 33.0, reset 5 times, recorded the IVRNG and or IVs, then dropped emloop to 32.5, repeat, etc, etc... I covered 33.0 - 27.0 (6 seconds, 70 Resets over 2 days). Found a weird, but steady pattern:
...
Summed, up, I'm seeing 2 common, distinct IVRNG Seeds for each second, with about 1 "tangent" seed, that just shows up every now and then. This is pretty consistent in these 70 resets, and looking back over my old notes is actually consistent with everything I've seen so far. I expected that maybe the game was determining the IVRNG Seed by the 1/2 a second instead of whole seconds, but these 70 resets don't seem to support that in my view. For an anecdote:Quote:
Time Common Seeds 33.0 A B 32.5 C D E 32.0 C D 31.5 F G 31.0 F G H 30.5 I J 30.0 I J 29.5 K L M 29.0 K M 28.5 N O P 28.0 N O 27.5 Q R 27.0 Q S If this was a matter of human error, or the Action Replay, or the DS Clock inconsistency "throwing the timing off" by a few seconds, I would think that after 70 resets spanning 6 seconds I would find one of those tangent seeds within 3 seconds or so (3 seconds is a pretty dang noticeable amount of lag for the startup screen) but I didn't see a single instance of that. I would also think that I would say (for example from the list I drew up) see an "I" up at 32.5 seconds, or an "M" at 27.5, but nope, not once out of 70 resets. They were always within .5 seconds of each other, and never any farther, always grouped together like that. It's late, and my brain is fried from resets, but I seem to recall OmegaDonut mentioning something somewhere about "the game randomly advancing some seed by varying amounts at game boot". Could this be where the 2 common and 1 uncommon IVRNG Seeds within any given second are coming from? It could still just be human error, but 1) I'm pretty dang good at hitting 4th Gen Delays with a Timer (which I'm using) and 2) I did 70 Resets! x_x I just think there's something more than human error, AR inconsistency, or clock inconsistency going on here.
__________________
|
|
|
|
|
|
|
#405 |
|
Remodeling Kitchens
![]()
Join Date: Apr 2009
Posts: 758
Berkeley, CA
|
Pokemon from the Dream World
PIDs and IVs are unset, you have to catch them in High Link to "obtain" them. Thus they are RNGable like Wild Pokemon and not like Pokewalker Pokemon. Not sure on the pattern for PID generation yet, but there is a pattern. You can't save, and the CGear is ON and you can't turn it off (this or something else is advancing it exactly like the cgear did in the overworld), and I don't have a 64bit seed advancer to get a seed/frame spread to see what is called when. RNGing PIDs is pretty pointless, see latter half of post. IVs are static over a given timespan. They don't change when the cgear ticks change the main RNG. They would be fairly easy to RNG depending on how/if it reseeds upon entering High Link. Also the CGear must be on, so the typical MTRNG prediction fails. What's important to note is that the MTRNG memory location we know and love is not used to predict the IVs just like once the CGear is activated. It's gotta be another memory location. ------ Pokemon from the Dream World caught in the High Link cannot be shiny due to a shiny check. Can't catch eggs, obviously. RNG Process => PID1 PID1 is Shiny -> PID1+1000000 hex -> PID2
proof, this post is getting some publicity on other sites
Not sure how to get PID1 and PID2 to be shiny to see if it checks multiple times, but I'm gonna go out on a limb and say it's infinite like the Mystery Gift one (which is XOR'd with 0x1000000 if shiny). The main point is that shiny checks exist for both of these PID generation methods. Same nature, since that's calculated from a previous call. |
|
|
|
|
|
#406 |
|
Dies, died, will die.
Join Date: Jun 2010
Posts: 302
|
kaph and i were also experimenting with shiny legendaries tonight and we had some interesting results. this is the process i followed:
-start game and note date/time to re-hit seed -approach victini/kyuremu and catch them, noting their PID -use pokegen to generate a matching shiny TID/SID for that PID -change the save file's TID/SID to that shiny TID/SID -hit the same seed(date/time) to get the same PID and guarantee a shiny this works just fine for kyuremu. it appears shiny with no issue and has no problems. victini, however, has issues. encountering/catching it for the first time is no problem. once you change the TID/SID and rehit the seed, however, it will not battle you. if you hit the seed and speak to it, the dialogue box will come up and disappear, but the battle never starts. you can keep talking to it over and over, but the battle will never start. i tried this with multiple shiny TID/SID sets to match its PID, but was never able to get into a battle. kyuremu worked every time, however. |
|
|
|
|
|
#407 |
|
Join Date: Dec 2009
Posts: 141
Sinnoh
|
I realized that the egg seems to change everytime you get it from a save...
I am willing to help, I have white (JPN Cart) also, Omega, you mentioned above that you can manipulate shinyness at this point, do you know how? and does anyone know if the frame advances or changes if you run from High Link pokemon (trying to SR for shinies but its easier if you can jsut run and battle again) thanks
__________________
The master of Garchomp HG FC:1462-8955-0680 Use it for everything unless I say otherwise |
|
|
|
|
|
#408 | |
|
Join Date: Oct 2010
Posts: 123
|
I have the retail cart too (no access to AR though) so if there's anything simple that a noob can help out with, I don't mind.
Quote:
__________________
Pokemon Black Friend Code 3181 1655 5404 Trading DW ability Pokemon Recently added: Caterpie, Slowpoke, Hoothoot, Corphish, Pidgey, Tentacool, Buizel, Sudowoodo, Dratini A Pokemon BW Macdonald's exists? |
|
|
|
|
|
|
#409 | |
|
sometimes experimentation begins with "let's multi battle strip pokémon"
![]() ![]() ![]()
Moderator
Join Date: Aug 2006
Posts: 3,691
|
Quote:
Anyway, I've been looking at the "Mersenne lookup table" generated when you come up with a seed using the C-Gear. Turns out it's calculated very differently than the table generated when seeded normally at startup, which is why we haven't been able to use the number at 022151D4 to predict IVs. Normally, the table is generated by multiplying the previous number by 0x6c078965 and the number of previous numbers before it. This new table is not generated by multiplying a single number and adding another, though. I've also been working on getting some better debugging tools to work, too. |
|
|
|
|
|
|
#410 |
|
Remodeling Kitchens
![]()
Join Date: Apr 2009
Posts: 758
Berkeley, CA
|
Getting Shiny Starters
You can RNG them, the PID is calculated from the 1st frame, ending on frame 2 for a total of 2 advances. IVs predicted by the IV MTRNG seed. They will not appear shiny when you open the box, the first moment you can tell is when they pop out of the pokeball. The reason they won't appear shiny is that only the sprite is displayed. Since the PID is calculated from the frame 1, all would appear shiny if it was shiny, however this is not the case. Not sure on the XOR and why it was different for the different frames, +/-... ------- I'ma go test the High Link PID generation (more for nature manipulation). Harder than usual, can't save game to +1 the RNG. |
|
|
|
|
|
#411 |
|
Remodeling Kitchens
![]()
Join Date: Apr 2009
Posts: 758
Berkeley, CA
|
High Link Pokemon Generation
Pre-Dialogue Advancement Code:
Lower Seed Upper Seed D6B445DE 8924238B Start 0 Just talking the first time BFBBFD59 B95533E0 A2A6F623 198BE052 04B37592 14A8B8DC 0-FR1 B3DD235D 11522C6D 81DD5774 477DE531 0-FR2~=~Start A0 Dialogue stops, go to the yes/no option Engaging? (Yes) Advancement Code:
Lower Seed Upper Seed 81DD5774 477DE531 Start A0 04733387 5A2655BE 0-PIDC C4DE3206 A6EADE7F ... 7ABF9121 392B464F ... Start A1 D64C89C8 7C619F87 0-FC1 1-PIDC Start A2 A68D02AB 1BC4BC66 ... ... 2-PIDC 62E32F3A 02603513 0-FC2 ... ... Start A3 5AA44AA5 B9D1FC96 0-FC3 1-FC1 ... 3-PIDC 556D5EDC 976FF1E8 ... 2-FC1 ... A414C78F E74C212F 1-FC2 ... ... PID Displayed in 0225CCA4 for A1 A40DE12E 910737C7 1-FC3 2-FC2 3-FC1 PID Displayed in 0225CCA4 for A2 B76413E9 34EC0115 2-FC3 ... C0A92AB0 8D0AA6C5 3-FC2 PID Displayed in 0225CCA4 for A3 Start A(#) - Yes/No Starting Frame A(#)+1=Unshown PID Call, changed based on resulting pkm's set gender, and some other factor? A(#)+4=Frame Change 1 Pokemon # Call A(#)+6=Frame Change 2 Gender of Pokemon # Call A(#)+7=Frame Change 3 Battle Flashes Start 0225CCA4 Location for A2 Start-0 022595DC (pointer?) FC-1 B9D1FC96 [A(#)+3] Wonder what this does... Nature!? FC-2 1BC4BC80 male Croagunk 1BC4BC02 female Mareep 1BC4BC01 female Nidoran PIDs and Set Genders Code:
PID Style: (Upper6 hex of PIDC)XY, in which XY is for the gender value which, in general, overwrites Frame 1's Upper Seed's lowest two hex. There may be certain cases in which the first XY modification fails, and it may change the other bits. I didn't see it happen for like 7 tests. The modification likes to play really close to gender ratio lines. 50/50 Female is 0-127, male is 128-225. Female looked like it always was around 0~ decimal Male looked like it was always around 128~ decimal With different ratios it looked like it didn't stick to this, but meh! |
|
|
|
|
|
#412 |
|
Remodeling Kitchens
![]()
Join Date: Apr 2009
Posts: 758
Berkeley, CA
|
The MTRNG location, if changed, will not change the IVs. It's based off of some other calculation that might be for reseeding. Tested with + without modification, same egg and same IV spread after inheritance.
Can't get the MTRNG to predict my eggs correctly, hitting 94BDA4EB and yet I get the base IVs starting on frame 8 onwards. Bond got the new version of desmume so we now have the same MAC addresses, but he can't hit the same seeds I can, with the same date/time. 94BDA4EB for me is CF2F251E for him. It won't predict any IVs from that either. If others could elaborate a little more on the circumstances (roamers???) of when they did it and how it would help iron out kinks. |
|
|
|
|
|
#413 | ||
|
sometimes experimentation begins with "let's multi battle strip pokémon"
![]() ![]() ![]()
Moderator
Join Date: Aug 2006
Posts: 3,691
|
Quote:
Quote:
|
||
|
|
|
|
|
#414 |
|
blatant Nintendo fanboy
Join Date: Mar 2009
Posts: 4,328
your mom sucked at e3
|
Well, after a few Resets the other day looking at eggs, I got some promising results:
...
I hit the same egg IVs while Resetting on the same Date/Time, multiple times. Worked fine for me.Then last night, I tried pausing different lengths of times before picking up the egg. Oivey...
...
That looks to me a whole lot like the Egg IVs that are inherited and generated are changing with the amount of time you pause before picking up the egg. I though the Everstone might be screwing with something so I took it off, and no.Can any of you guys with the capability to do so see if there's some value advancing the inherited and generated egg IVs from either time, or the RMNPCs? I hope to God they're not "on the same frame" like 4th Gen Method 1 pokes were.
__________________
|
|
|
|
|
|
#415 |
|
Remodeling Kitchens
![]()
Join Date: Apr 2009
Posts: 758
Berkeley, CA
|
94BDA4EB -> 00000000
0.9.7 |
|
|
|
|
|
#416 |
|
sometimes experimentation begins with "let's multi battle strip pokémon"
![]() ![]() ![]()
Moderator
Join Date: Aug 2006
Posts: 3,691
|
Desmume 0.9.7 assigns a different MAC address to each user, that's why you're getting different results. This was a decision made by the programmers because 0.9.7 can actually connect to Nintendo Wi-Fi*, and they didn't want all the same MAC addresses trying to connect.
* Or so they say. I haven't managed to get it to work. |
|
|
|
|
|
#417 |
|
Remodeling Kitchens
![]()
Join Date: Apr 2009
Posts: 758
Berkeley, CA
|
Interesting, I'd have to ask Bond because he said that he now has the same MAC address as me for some reason...
00-09-BF-12-34-56 is my mac address. Ehh I guess I meant to say 0.9.6, I mistook what he said thinking he got the new version and his mac became like mine |
|
|
|
|
|
#418 | |
|
Dies, died, will die.
Join Date: Jun 2010
Posts: 302
|
Quote:
kaph is on .9.6 which just uses the default mac. i am too and i have the same mac. nothing wrong there. i was just explaining this to him, actually. i had said there was a new SVN version out, but i wasn't using it. there's no official .9.7 yet, right? it's just SVN snapshots. also, you need to install winpcap to get desmume to use your own network hardware as the softAP. yeah though, i hit the same time the same way he did with the same MAC and got a different seed that's nowhere close to his. |
|
|
|
|
|
|
#419 |
|
Remodeling Kitchens
![]()
Join Date: Apr 2009
Posts: 758
Berkeley, CA
|
As per request, checking to see if Zekrom can be shiny or not:
img test results
|
|
|
|
|
|
#420 |
|
Join Date: Nov 2006
Posts: 1,122
"Dashing"
|
Just dropping by to say that I've hit an Egg IV spread that passes 4 IV's from the parents, just to confirm the odd looking spreads mattj's posted. I guess this is a change from 4t gen breeding?
|
|
|
|
|
|
#421 |
|
Join Date: Jan 2005
Posts: 4,740
Ingerland
|
Don't know how or if this will help, but here are a few runs I've done throughout the day:
Parent A - 2 / 5 / 14 / 20 / 23 / 31 Parent B - 6 / 31 / 31 / 28 / 12 / 28 (not the best parents I know, but all i had to breed with karp >_>) _________________________________________ Synch Time: 15:57:00 Reset Time: 15:57:20 Continue: 15:57:34 Enter game: 15:57:41 Collect Egg: 15:57:47 IVs: 17 / 31 / 14 / 7 / 20-21 / 28 Adamant (F) ---- 17 / B / A / 7 / 20-21 / B ------------------------------------ Synch Time: 15:57:00 Reset Time: 15:57:20 Continue: 15:57:34 Enter game: 15:57:42 Collect Egg: 15:57:47 IVs: 17 / 12-13 / 31 / 28 / 20-21 / 31 Impish (M) ---- 17 / 12-13 / B / B / 20-21 / A ------------------------------------ Synch Time: 15:57:00 Reset Time: 15:57:20 Continue: 15:57:34 Enter game: 15:57:42 Collect Egg: 15:57:42 IVs: 17 / 31 / 15-16 / 7 / 23 / 28 Lax (F) ---- 17 / B / 15-16 / 7 / A / B ------------------------------------ Synch Time: 15:57:00 Reset Time: 15:57:20 Continue: 15:57:34 Enter game: 15:57:42 Collect Egg: 15:57:43 IVs: 17 / 31 / 15-16 / 7 / 23 / 28 Lax (F) ---- 17 / B / 15-16 / 7 / A / B *repeat despite different egg collect times* ------------------------------------ Synch Time: 15:57:00 Reset Time: 15:57:20 Continue: 15:57:34 Enter game: 15:57:42 Collect Egg: 15:57:44 IVs: 17 / 31 / 14 / 7 / 20-21 / 28 Neutral (F) ---- 17 / B / A / 7 / 20-21 / B **same as first one, but different nature..** ------------------------------------ Synch Time: 15:57:00 Reset Time: 15:57:20 Continue: 15:57:40 Enter game: 15:57:48 Collect Egg: 15:57:49 IVs: 17 / 31 / 15-16 / 7 / 23 / 28 Lax (F) ---- 17 / B / 15-16 / 7 / A / B *another repeat, but this was after intentionally waiting longer at the continue screen, and then collecting the egg asap.* ------------------------------------ Synch Time: 15:57:00 Reset Time: 15:57:18 Continue: 15:57:32 Enter game: 15:57: Collect Egg: 15:57:42 IVs: 6 / 5 / 6 / 4-5 / 12 / 14 Naive (F) ---- B / A / 6 / 4-5 / B / 14 ***changed the reset time, and voila different spread altogether.*** ------------------------------------ Notice all the ones reset at 15:57:20 have 17 HP IV. This could just be "luck." I hope this isn't just repeating what is already known -_-
__________________
[13:17:16] <BenVdd> i mean , if she wasn't my sister and i still lived in the same house as her i'd do her in her sleep |
|
|
|
|
|
#422 |
|
Remodeling Kitchens
![]()
Join Date: Apr 2009
Posts: 758
Berkeley, CA
|
Felt a little bored so I wanted to continue with the what can be shiny what can't stuff... this time Manaphy!
image of... manaphy!
|
|
|
|
|
|
#423 |
|
Join Date: Sep 2009
Posts: 739
Italy (GMT+1)
|
I'm using Desmume 0.9.6 in order to test some spreads and shininess with breeds in White, no C-Gear used.
Fist of all, the same frame holds the same PID, regardless of species, but you won't be able to hit the same frame if you change your parents from [pkm & ditto] to [pkmF & pkmM]. This means that, for example, you found a good iv spread in frame 2 with dokkora and ditto, but when you breed dokkoraF and dokkoraM you'll find the same spread on frame 3 with the same PID but most likely with a different nature. Ivs are separated in two "layers", a first layer which is determined by the seed and another layer which corresponds to the IVs passed on by parents. The first layer is static, it doesn't change, while the second layer varies with the frame. An example: Frame 1 has 20/10/30/31/0/2 (nature = fixed brave) Frame 2 has 31/10/22/31/0/2 (nature = everstone) and so on... Last but not least here's shadowhunter's guide for breeding shinies the old fashioned way -> http://www.smogon.com/forums/showpos...8&postcount=12
__________________
|
|
|
|
|
|
#424 |
|
Join Date: Nov 2006
Posts: 1,122
"Dashing"
|
Which means that if you find a time and date that gives a high number of 31's you can SR on that seed and let the frame advance by waiting and by doing such land on a frame where the right stats are inherited, right?
EDIT: I wonder if this is similar to gen 4 ... |
|
|
|
|
|
#425 |
|
Join Date: Sep 2009
Posts: 739
Italy (GMT+1)
|
Yes, you can do that, power items may come in handy too.
__________________
|
|
|
|
![]() |
| Tags |
| NULL |
| Thread Tools | |
|
|