Emerald 100%

TyrantNeptune
Preschooler
Posts: 8
Joined: Thu Apr 02, 2015 10:44 pm

Re: Emerald 100%

Post by TyrantNeptune » Thu Aug 27, 2015 1:03 am

Metarkrai wrote:Well, you won't have seen it, but I made a change to my change during that post, so doing it helped me improving the route.
Since I have now a quite good team for Battle Frontier, I'm up to Pokemon Contest things.

And there, the page is mainly blank : what kind of strategy should be adopted in order to get the artist at the end of the contest ? (In RS, he apparently appeared when you made a certain score on Master, maybe as well as winning)
Someone told be about : One 4 star move of the good category - One 4 star move of another category (not disliked) - One move to get first/last - One move to annoy others / ??.
There is also a nice list of combos that could be achieved, if we are able to include them in the regular strategy (as with two 4 stars moves you make oftenly more stars than with a combo).

But first, if you have any suggestion or idea related to a winning strategy for Master Rank contests and museum painting with a single Pokemon that would have 3 Moves from Seedot/Nuzleaf and Flash, this would help greatly.
For the appeal round the best (that is, the one that scores most points) move combination that I could find is Lock-On (3) - Superpower (6*2) - Lock-On (3) - Superpower (6*2) - Explosion (8), available to the regi's (replacing Lock-On by Mind Reader and Superpower by High Jump Kick for the meditite evolution line works as well), for a total score of 38. a disadvantage of this combination is that Superpower makes you take double the jam from the next pokemon.

The secconed best move combination that I could find is Shadow Punch (if you win the first round)/Spite (if you lose the first round) (2+4) - Curse (3) - Spite (2*2+4) - Curse (3) - Destiny Bond (8*2), available to the ghastly evolution line, for a total score of 36. A disadvantage of this combination is that you are moving last due to Curse, which can cause you to become nervous. This combination also uses only Smart and Tough moves, so that could be a problem as well.

User avatar
ClothNapkin
Bug Catcher
Posts: 15
Joined: Sun Aug 03, 2014 8:07 pm

Re: Emerald 100%

Post by ClothNapkin » Thu Aug 27, 2015 4:57 pm

I'm probably missing something here, but for the Palace, couldn't you just have the move deleter delete "attacking" moves other than 1HKO moves?
Currently doing WR attempts in (Whatever stupid hack I'm running now)

Metarkrai
Bug Catcher
Posts: 14
Joined: Thu May 28, 2015 12:08 am

Re: Emerald 100%

Post by Metarkrai » Thu Aug 27, 2015 6:07 pm

If you go into Palace with just an offensive move, you'll have high chance of not being able to attack during battles if the game selects a Defensive or Sneaky move.
For a Timid nature (the nature of the Pokemon here), you have an Offensive/Defensive/Sneaky repartition of 62/10/28, which means 38/2 = 19% chance of not attacking if you go yolo with a single Offensive move.

If the Pokemon could be something with Mystik Guard or something, I would be on this strategy, but without the risk is too high compared to the use of Defensive and Sneaky moves.

And with the moves I had to chose for Palace, Move 0x0434 can't be deleted by Move Deleter as it freezes the game (all the other moves can be deleted without issues, except the IFG one maybe), and as I need to delete move 0x0505 that appears because of Plusle's contest stats, I need to use Glitch Moves with Sketch's effect (as Sketch itslef would take too many EVs).

However, your idea isn't that bad.
With 2 Double Corruptions, we could get a Glitch Pokemon with Mystik Guard and the OHKO Glitch Move.

However, if would take a longer time to be performed since we would need a second corruption initiator for the second Double Corruption. (A previous route involved making 2 initiators from Horsea in order to easily do any kind of double corruption, but Plusle is a working corruption initiator on in-game traded Pokemon, and does require a very small setup.)
Horsea is the sole Pokemon that can become a corruption initiator in order to get a Glitch Poke with Mystik Guard and an OHKO Glitch Move.
However, doing Horsea just for this would be a loss of time, and I don't know if the time gained by having a Pokemon that attacks a bit more (81% of the time instead of 62%) on Palace would cover up for that.

I'll first have to check if there is an easy to get Mystik Guard Glitch Pokemon that can have its moves deleted (whose sprite doesn't freeze the game).


I'm also thinking about one or two corruptions for event Berries, as the strenght of the Pokeblocks you can do with them would help maximizing the Contest stats of the Pokemon.
However, another strategy that could be done would be cloning the Contest Pokemon and feed the clones with different Pokeblocks in order to cover all the categories in terms of high contest stats. (done with high and rare Berries)

And here, event Berries require 160-170 Def Evs, which means around 15 Cascoon/Silcoon with Macho.Brace.
Giving the percentage of these 2 Pokemon (10% each), I think it's faster to just clone the Contest Pokemon and make Pokeblocks with normal berries for all the clones. (As with the event berries, you would also have had to do some Pokeblocks, which means that it would roughly be "finding fifteen 20% Pokemon" against "making 10 Pokeblocks" for the same result in the end).


And thank you TyrantNeptune for the contest strategies.
I didn't have the time to test them today as I was writing off the procedure to unlock Faway and Birth Island in Emerald, but tomorrow or in 2 days, I'll test them to see how well they go.

Also, it doesn't seem a bad thing to me if Smart or Tough are ignored (it's either 1 ignored 1 diliked or 2 disliked with Beauty/Cool/Cute) since the difference would mainly lie in the 5 bonus hearts and in the ovation.
How many hearts does the ovation give already ?
Because with Seedot/Nuzleaf, certain 4th moves with 4 Hearts / something else that goes for Cute/Beauty/Cool could be obtained, in case the ovation could be grasped (if the total of hearts gained is higher or equal than 8).

Testing will tell us if with the available Pokeblocks, the starting contest stats, and that strategy, we are able to score 800 points to get the artist at the end of the contest.


For Metitite/Medicham, what move would replace Explosion ?
Oh, I see, you were planning to get it from Seedot/Nuzleaf, what a brillant idea.
Thus, I would go for Metitite as we need 104 HP EV for him (1 Marill with Macho Brace), whereas we need 146 HP and 1 Atk Evs for Regice (11 Marill @ Macho Brace, 1 Marill, 1 Poochyena) for the same result in the end. (as the 4th move doesn't seem to matter that much, and as Explosion is taught by a move tutor in Emerald).

TyrantNeptune
Preschooler
Posts: 8
Joined: Thu Apr 02, 2015 10:44 pm

Re: Emerald 100%

Post by TyrantNeptune » Thu Aug 27, 2015 8:31 pm

Metarkrai wrote: And thank you TyrantNeptune for the contest strategies.
I didn't have the time to test them today as I was writing off the procedure to unlock Faway and Birth Island in Emerald, but tomorrow or in 2 days, I'll test them to see how well they go.

Also, it doesn't seem a bad thing to me if Smart or Tough are ignored (it's either 1 ignored 1 diliked or 2 disliked with Beauty/Cool/Cute) since the difference would mainly lie in the 5 bonus hearts and in the ovation.
How many hearts does the ovation give already ?
Because with Seedot/Nuzleaf, certain 4th moves with 4 Hearts / something else that goes for Cute/Beauty/Cool could be obtained, in case the ovation could be grasped (if the total of hearts gained is higher or equal than 8).

Testing will tell us if with the available Pokeblocks, the starting contest stats, and that strategy, we are able to score 800 points to get the artist at the end of the contest.


For Metitite/Medicham, what move would replace Explosion ?
Oh, I see, you were planning to get it from Seedot/Nuzleaf, what a brillant idea.
Thus, I would go for Metitite as we need 104 HP EV for him (1 Marill with Macho Brace), whereas we need 146 HP and 1 Atk Evs for Regice (11 Marill @ Macho Brace, 1 Marill, 1 Poochyena) for the same result in the end. (as the 4th move doesn't seem to matter that much, and as Explosion is taught by a move tutor in Emerald).
Each time you use a move which type corresponds with the type of the contest, you will gain one heart bonus. If that moves fills up the applause meter you will gain six hearts bonus instead. However, timing your move so that it will fill up the meter can be very tricky, unless you can predict all of your opponents moves.

Also, I found out that Hitmonlee (Mind Reader and (High) Jump Kick) and Poliwrath (Mind Reader and Submission) can also perform the first strategy.

Metarkrai
Bug Catcher
Posts: 14
Joined: Thu May 28, 2015 12:08 am

Re: Emerald 100%

Post by Metarkrai » Thu Aug 27, 2015 10:10 pm

I made a calculus shift to see how Pokeblock recipes would work on the Voting Score, and how many Hearts would be needed to get at 800 points.

I tried a recipe with 4 Watmel and 4 Belue Berries, all blended with 4 players (they have less Feel like this), and at 90 RPM.
They would have had 1 more points in Contest Stats at 100 RPM but I'm not sure if 100 RPM + can be achieved, so I used 90 RPM since this will always be achieved.
If we choose a late day hour at the start of the speedrun, we should be able to get two rare berries that would yield the best results (I'm currently testing if the berry master wife can give another berry when midnight passes)

Here's the data :
Final Score = 20*Hearts + Voting Score
Voting Score = (Contest Stat B + Contest Stat C + Total Feel)/2 + Contest Stat A.
Contest Stat A is the main contest stat, whereas B and C are auxiliary.
Cool Contest: Beauty, Tough
Beauty Contest: Cute, Cool
Cute Contest: Smart, Beauty
Smart Contest: Tough, Cute
Tough Contest: Cool, Smart

Pokeblocks :
Cool/Beauty/Cute/Smart/Tough//Feel
Belue : 00/00/11/00/50//32 (12 Cute since Timid)
Watmel : 11/00/50/00/00//32 (10 Cool, 55 Cute since Timid)
Metidite Base Contest stats : 66/136/00/128/00 / 00

The Mind Reader - Hi Jump Kick - Mind Reader - Hi Jump Kick - Explosion combo gives 38 Hearts in Cute/Smart, 39 Hearts in Beauty, and 40 Hearts in Cool/Tough if the Pokemon is never troubled and if he never wins any ovation. (only the 1 Heart Crowd bonuses)

I don't know for now how many hearts could be lost in the process due to nervousness, being troubled, and not having Crowd bonuses, but I was thinking of 26-30 hearts.

Here is the amount of Voting Score earned with certain Pokeblock recipes, and the amount of hearts required to get to 800 points or higher :
1 Watmel, 1 Belue : 64 Feel
Voting Score 202 241 231 218 184
Remaining score 598 559 569 582 616
Needed hearts 030 028 029 030 031

2 Watmel, 2 Belue : 128 Feel
Voting Score 269 311 329 308 271
Remaining score 531 489 471 492 529
Needed hearts 027 025 024 025 027

3 Watmel, 3 Belue : 192 Feel
Voting Score 336 381 427 398 358
Remaining score 464 419 373 402 442
Needed hearts 022 021 019 021 023

4 Watmel, 4 Belue : 256 Feel (the last Pokeblock will max out Feel)
Voting Score 403 451 525 488 445
Remaining score 397 349 275 312 355
Needed hearts 020 018 014 016 018

Using 2 different rare berries is more useful than using a single one as you have a main boost in two different contest stats instead of one, and I focused on Cute and Tough as they are at 0 when you get Meditite, and as the Contest Stat related to the contest category is worth the double of the auriliary contest stats.

Now, I'll make the Meditite and Pokeblocks, and see what of the 1/1 or 2/2 or 3/3 combinations is enough to get a good win.


EDIT :
I just read your post TyrantNeptune, and thanks for the reminding on the total amount of won hearts on an ovation.
Well, it could be either Poliwrath or Meditite, as they can all learn their main moves with Levelling (so with Heart scales), and as Poliwrath needs 1 Marill whereas Meditite needs 1 Marill + 1 Pooch.

Also, I was thinking of having a changeable 4th Move that could be able to grab the ovation and get more hearts than Explosion would have (which would require to go first, as we're not protected from other Pokemon being douchebags).
I was thinking of Sunny Day (Beauty) (+60 if Crowd is right before ovation), Rain Dance (Tough) (+60 if Crowd is right before Ovation), Swagger (Cute) (20+40 if the Pokemon moves first), Mega Kick (Cool) (+40).

EDIT II :
It's midnight, and the Berry Master Wife gave me another chance to tell her trendy words.
So yeah, the Watmel/Belue strat is viable, which is nice.
Other Berries like Pomeg, and other EV reducing have quite less effect than these, so it's nice to have a quick access to high quality berries, since else the strategy would have been to clone Ev-reducing berries and do Pokeblocks with them.


EDIT III :
As for the last piece related to the Contest corruption, we have Seedot's Friendship manipulation.
He starts with 70 Friendship, and needs to end at 14/15.
He will need to learn 1/2 Move Tutor Moves, as well as 1/2 Tm/Hms.
He will need 6 HP Up and one Marill kill.
Since he will be at Lv4, the Marill kill will make him gain some levels.

My idea was :
Win League with Seedot in party using IFG.
Give him 3 HP Up (+15) and fight the Marill. (the gain depends on the amount of levels, and if his friendship goes above 100)
Get 1 Energy Root (-10) and 7 Revival Herbs (-15).
Let Seedot be repeatedly killed on a high-levelled Pokemon. Fainting + Revival Herb = -16 Friendship (times 7 it's -112)
Fight with a weak Pokemon to weaken Seedot a bit.
Get in front of the PC.
Use an Energy Root to heal him and to be sure that his Friendship is at 0 (in case you would have completed a 256-step cycle)
Give him the 3 last HP Up. Deposit him in PC until Double Corruption.

Seedot might be also used to get an held item, so he might gain some more levels, which would only take 2-3 more Revival Herbs, so that wouldn't be a bother.

Metarkrai
Bug Catcher
Posts: 14
Joined: Thu May 28, 2015 12:08 am

Re: Emerald 100%

Post by Metarkrai » Sat Aug 29, 2015 2:43 am

Today will have a shorter post as I mainly worked on the Faraway Island unlock. (there were things to change in order for the method to work on every Fr Emerald which took me hours to do)

I made the Poliwrath and to my surprise, 14 or 15 Friendship don't give the Master access to Toughness.
After more calculations, you need 0x36 = 54 Friendship (or 0x37=55) to get Master access to Toughness.

Thus, the strategy will only have to change a little, as it's a matter of :
- Give the 6 PP Ups and 1 HP Up to Seedot.
- Make him gain 2 HP EVs on a Marill
- Make anothe fight and kill and revive him 7-10 times with Revival Herbs (I need to check on the friendship gained from PP Ups and recalculate that) (0 Friendship)
- Kill him for a last time and revive him with a Revive.
- Fly towards Pokemon Center and get to PC. (0/1 Friendship from 256 step cycle)
- Use an Energyroot/powder (0 Friendship)
- Give him the 5 remaining HP Up (25 Friendship, 62 HP EVs)
- Give him 3 EV reducing berries except Pomeg (55 Friendship, 62HP EVs)
Yeah, EV reducing berries give 10 Friendship, and not 2 as it was previously displayed on bulbapedia. (EDIT : They give 10, then less after 100 Friendship, but we don't care about that for now)

I've tested that setup and I got to 0x37 Frienship for a Master access to all contests.

For Pokeblock creation, 90 RPM is easy to do, while 100 seems a bit random on 4 players as they can easily slow the cursor down a bit too much (or I'm not that good at it).

I also forgot that Poliwrath wouldn't be Timid since he's obtained with Seedot and not Plusle. Thus, he's docile, which is a neutral nature.
He'll gain then less Cute Contest stats from Pokeblocks than what was written, but that loss is too tiny to induce a change in the amount of required hearts to win (as they were rounded up) unless we're dealing with the 3/4 Watmel/Belue Blocks, which would add 1 more Required Heart in Beauty, Cute, and Smart. (not always for 3 but I'll approximate it as "always" for now).

Cool :
I was able to win in Cool quite easily with 1 Watmel/Belue. (with a painting in the end)
The first time the Altaria was annoying with its Hyper Beam, but he wasn't there on my second try.

Beauty :
I got owned pretty bad on Beauty due to my placement not being good (right behind that damn Jynx on his Mean Look turn, and right in front of him for his Requiem turn).
On my 3rd attempt, I only was able to get 4 hearts thanks to Sunny Day (Requiem + Hyper Beam turned the 12 Sumbission hearts into -6/-8).
A good amount of Pokemon seem to know moves that induce nervousness/ that scare, so higher contest stats are required for this one.

Cute :
I utterly lost the first time (lost all the combo hearts from a single startle, and was too nervous to pull the second one)
The second time, there was only Slacking who tried to make other nervous, and once the first combo was out, I easily won with Hearts.

Smart :
Barely won on the first try. The Pokemon were all using moves to copy the previous scores, so I lacked a pixel of points to beat the Pokemon with the highest Voting Score.
Utterly won on the second try. Slaking didn't trace my hearts, and I also got Wobbuffet for the second time. Wobbuffet used again Destiny Bond on its 3rd turn, and I don't understand the trick behind that.

Tough :
I placed all the combos and got startled once (getting 4 hearts instead of 12 on turn 2), and I lost.
Wobbuffet was here again to help me by not doing anything dangerous, but that didn't suffice.
My Voting Points were really low this time, so I think more Pokeblocks are needed.

I tried with a 4 Watmel/Belue recipe, and that gave Poliwrath more Voting Points than other Pokemon, which made the victories quite easy as long as I was able to cast a combo.

I'll need to write down the Master classes Pokemon pools as there is a good amount of Pokemon that won't be aggressive (maybe 1 small attack/stress that you could avoid and wouldn't be too much of a bother), and it seems quite easy to get 2 passive and 1 semi-passive Pokemon with you in order to grab the win as long as you have enough Voting points to get the final lead.

I'll try out tomorrow with only 2 Watmel/Belue and see how much of lead I can have on the contests where I barely lost with 1 Watmel/Belue.

Metarkrai
Bug Catcher
Posts: 14
Joined: Thu May 28, 2015 12:08 am

Re: Emerald 100%

Post by Metarkrai » Sat Aug 29, 2015 2:00 pm

I made a save file on vba with Pokemon for Battle Frontier and Contests.

You have Poliwrath as it would be obtained without any added Contest stats, and 4 Watmel/Belue Pokeblocks.

-Battle Palace Team : 0x010E, Chansey, Blastoise
-Battle Pike Team : Blastoise, Chansey, Smeargle
Use Stealing Glitch Move to heal from triple Sleep/Freeze, and put Smeargle in lead for wild Pokemon sections in order to use Fluffy Tails and run. (faster than OHKOing the Pokes)
-Battle Arena and Tower Team : Blastoise and Chansey.
-Battle Pyramid and Dome Team : IFG Smeargle and Chansey.
Use IFG everywhere except on Leaders.

The IFG I used is one that could be obtained via double corruption. It isn't Sketchable like this and can't be forgotten, so I think I'll search for another one for more conveniency.
The rest of the moves are as I mentioned before.
These Pokemon also have 128 Speed and 100 Def EVs from item 0x8064 that could be obtained with the first double corruption round (from 128 Speed and 100 Def EVs on a Horsea). We could pu less Speed as something like 104 or 108 would be enough for 0x10E to have 362, but I haven't refined that yet.

Don't look at the first Blastoise and at 0x010E's summary, they have Glitch Move 0x0434 that freezes the game when seen (in battle too).

There's also a cloning glitch Pokemon in box2, and a Banette to perform Pomeg Glitch.

sav file :
http://www.petit-fichier.fr/2015/08/29/ ... -names.sav

sgm file :
http://www.petit-fichier.fr/2015/08/29/ ... names4.sgm

sps file :
http://www.petit-fichier.fr/2015/08/29/ ... -tests.sps


Tell me if anything doesn't go as planned in Battle Frontier.

Metarkrai
Bug Catcher
Posts: 14
Joined: Thu May 28, 2015 12:08 am

Re: Emerald 100%

Post by Metarkrai » Sun Aug 30, 2015 1:20 pm

I started focusing on the Item duplication today.

First Item Duplication :
Duplicate a Bag / PC Item during Sootopolis Gym Door.
Try to duplicate a Rare Candy to gain time.
The main goal here is to sell duplicated items for cash.
Store Berries in PC to restaure the Berry pouch.
Buy TMs at Lilycove to get back TM pouch.

PC Items for First duplication :
1st item : Poison Heal x1 / Super Ball x2
2nd item : Something
3rd item : / Something
4th item : Great Ball x1 / Ice Heal x3
5th item : Something
6th item : Something
7th item : Poke Ball x1 / Burn Heal x1
8th item : Something
9th item : Something
10th item : Net Ball x1 / Dive Ball x1
11th item : Something
12th-18th items : Something
19th item : Rare Candy

Two other duplications will need to be made for Master Balls and Rare Candies after that.
There's another PC setup to do in order to try and corrupt 2 items (trying to corrupt both of them gives higher success chance for the first try).
Also, with an Instant Pomeg Glitch Pokemon and load of money, setting up the PC will be way faster and way more easier, so I'm not worrying that much about the second corruption, since it can be done after all the corruptions for Pokemon.

But for the first corruption, I wanted to scratch a bit the item combination that could be used, as all the last battles will be done with IFG, which won't grant any money.
We should already have a certain amount of spare items like some Repel, Repeat Ball, Fluffy Tail,... as well as other items that would be required for further corruptions like Macho Brace, Exp Share, certain Tms, Heart Scales, PP Ups,...
We would also need more berries in order to be able to decorrupt the Berry pouch. 6-7 Berries will be enough.
And since we would go to Lilycove, Mossdeep, beat Mossdeep Gym and Spatial Center, Pacifildog then Sootopolis, it adds items that could be looted in the way. (Depending on Repel amounts, this could be done before League or after League.)(Rival 4 at Lilycove doesn't need to be beaten, he'll disappear once League is beaten.)(We need to Sketch IFG or buff Swampert to beat Levy&Tatia and Magma Team as you can't fight them with a fully KO team due to double battles.)

Thus, in addition to that, I listed some items that could be quickly looted / cheap to buy :
Easy/Cheap Items :
Super Ball
Great Ball
Poke Ball
Antidote
Ice Heal
Burn Heal
Awakening
X Special
X Speed
Potion
Wave Mail (Lilycove) (Mails are really cheap)
Mech Mail (Lilycove)
Wood Mail (Fortree)
Harbor Mail (Slateport)
Orange Mail (Petalburg City)

Persim Berry
Lum Berry
Pecha Berry
Oran Berry ?
Nanab Berry
Aspear Berry (R 114)
Chesto Berry (R 116)
Pinap Berry (R 116)
Sitrus Berry (R 118)

This gives more than enough items for PC stuffing, which will be discussed later on, after the Pokemon for Battle Frontier, Contests, and catches are validated to be fully working and don't seem to be upgradable in terms ot time improvements.

Regarding Berries, I saw that League need to be beaten to get Belue, and that Latias needs to be seen to get Watmel.
I'd really like to use Spelon Berry that can be available before league, as it could be obtained while going at Fortree, but it would be less useful than Belue (as there's already 66 in Cool, and as there would be 0 in Tough), and Watmel would still be required in order to get a good/high Voting score for Cute and Smart.
Thus, in order to be more reliable on contests wins, I think that getting back to Route 123 twice to get Watmel and Belue would be preferable, as the time to get back to Route 123 one more time is shorter than a reset to get different contest opponents.

Seeing Latias isn't that much of a deal, since it would need to be seen for Regional Dex anyways.
Southern Island could be unlocked during attempts to get EGFL, but it's too unreliable. (It might be needed afterwards in order to avoid making a double corruption for Lati@s.)
Thus, the roamer selected after beating the league would need to be Latias.
To see it, take a Lv 15+ Pokemon with Illuminate/Arena Trap (or add a KO Pokemon with that ability), use Super Repel, and go at Route 110 South.
Walk into grass for a certain amount of steps, then go into the Cylcling Road House.
Exit the house and repeat until Latias appears.

By entering the House, Latias has 1/20 chance to be replaced on Route 110, and that entrance is the easiest one to use in order to get back to grass in a minimal amount of time (grass is really near) on a Route where Latias can be (it can't be on Route 104 if you were thinking of using Petalburg Woods entrance).
It doesn't take that many Repels, nor that much time to see Latias, and catch it with a Master Ball (that would have been cloned or duplicated.)
Illuminate is useful to increase as much as possible the chances to see Latias for the amount of steps made, and it's easily available in the form of Chinchou or Staryu.
But if getting Illuminate is too much of a hassle, it's not that much of an issue.
Walking into grass instead of using a bike gives a higher chance to trigger an encounter, which lowers the average time to make the Latias encounter (the door-grass distance being too short to increase that average time by running instead than using a Bike).

Metarkrai
Bug Catcher
Posts: 14
Joined: Thu May 28, 2015 12:08 am

Re: Emerald 100%

Post by Metarkrai » Sun Aug 30, 2015 11:18 pm

Since some things were added to the route, I made a more detailled current route : http://pastebin.com/GjPxMXGu
It took me way more time than I thought to make it, as a lot of approaches and setups for cloning or glitches had to be tweaked.

But that's a good thing, as I now have procedures that work better.
This allowed me to increase the success chance of the first Double Corruption, which is quite nice.
And as you'll see, the setups for a lot of tiny things are a bit different from what I said in my previous posts as it had to be changed in order to properly work.

I haven't finished writing all the stuff planned for now, and there are certain things I need to try, so some things are prone to change.
I also tired a bit to gather the things that were needed to do in a certain place in order to lessen the amount of movements, but I'm sure that some of that are optimizable (I mostly wanted the route written, so only focused a bit on that).

If you have any comment/question/idea regarding the route, please write it, it could help for its improvement.

And as a reminder, the .sav vba file to test the Battle Frontier and Contest Pokemon is upper on that page.


EDIT :
v1.1
Got back to school.
I finished including all the glitches in the Route.
Some things like times for Up holds are missing for now. (some of these Up holds will use visual clues)

Metarkrai
Bug Catcher
Posts: 14
Joined: Thu May 28, 2015 12:08 am

Re: Emerald 100%

Post by Metarkrai » Wed Sep 02, 2015 7:00 pm

I updated the route file, changing certain things and mainly adding a bit more actions for items and Pokemon.
Emerald Speedrun Route Scraps : http://pastebin.com/GjPxMXGu

I started listing the ways to obtain every Pokemon in the dex. I had roughly done that before with 7hryne, so I added more accuracy for obtention methods, like percentages, required EVs for corruptions, locations, location for potential obtention, and routes for trainer steal.
I have to recheck the trainer list to redo the list of sealable Pokemon.
With an Horsea with 41 Atk and 188 HP Evs, having EVs to get an Hoenn Pokemon won't be long. It's only a matter of getting the right digit in HP (0-2 Marill and 0-1 Wurmple), and giving the good amount of Pomeg and Kelpsy Berries.
The only Pokemon that needs more than 188 HP Evs to be obtained is Kindgra. But getting to 230 HP EVs is a matter of 8 Marill @ Macho.Brace + 1 Marill.

Does someone know how Trainer Rematches work on RSE ? (how and when trainers are selected to be rebattled ?)
As rematches would help stealing certain Pokemon/Evolutions.
The timing of a Master Ball steal (with Dex entry) and an evolution is also really important since a good amount of Pokemon can be obtained with both ways, and I don't know which one is faster.

Here is the listing I made (using Bulbapedia's Hoenn Dex list), with additional data for Pokedex completion that I made for the TAS :
http://www.petit-fichier.fr/2015/09/02/ ... number.htm
http://pastebin.com/GjPxMXGu


vba sav file for Battle Frontier and Contests tests :
http://www.petit-fichier.fr/2015/08/29/ ... -names.sav

MKDasher
Schoolkid
Posts: 41
Joined: Fri May 02, 2014 5:31 pm

Re: Emerald 100%

Post by MKDasher » Sat Sep 05, 2015 5:42 pm

I just watched roush's catch em all (212 pokes, only dupping allowed), and checked if there was some consistent way to catch feebas. I see 2 possible ways to RNG Manipulate for feebas, one is save resetting in front of the catchy phrase guy in venandurf town, and change the phrase at a specific time. The other would be just when the game is restarted, and actually, feebas first tiles are generated when SID is generated, so that would give us SID information which I don't know if it's helpful anyway.

TyrantNeptune
Preschooler
Posts: 8
Joined: Thu Apr 02, 2015 10:44 pm

Re: Emerald 100%

Post by TyrantNeptune » Sat Sep 05, 2015 6:53 pm

So I tested contests today. The good news is that if you get (near) perfect score in appeals (and if you win obviously) you already have enough contest points to get the painting, even without using pokéblocks (I'm assuming the poliwrath in the sav file hasn't have been fed any pokéblocks yet) and even in toughness. Therefore it may be possible to skip berries altogether or to make pokéblocks from easier obtainable berries.

The bad news is that moves like stun spore suck badly. They just remove all your hearts from your combo at once. Combined with other jam moves that also destroy your heart count due to submission and the chance of getting nervous, I think the Poliwrath strat is just too risky.

Therefore, I suggest going with the ghastly family and use Spite (1+5) - Curse (3) - Spite (1*2+5) - Curse (3) - Destiny Bond (8*2) combination for a total of 35 points (my initial calculation was wrong). This way, we can only get jammed on turns 2 and 4, stun spore and the like will only jam for 1, and other jam moves won't double their jam effect. As I pointed out before, this strategy is pretty weak to nervousness, so it's essential to reset if there's more then one pokémon with a nervous move. If there is one pokemon with a nervousness move, it's risky but since the pokemon can opt not to use the nervousness move and the move can fail, it may be worth the try.

User avatar
ClothNapkin
Bug Catcher
Posts: 15
Joined: Sun Aug 03, 2014 8:07 pm

Re: Emerald 100%

Post by ClothNapkin » Mon Sep 07, 2015 4:32 am

Since we're getting all the pokemon in the Hoenn Dex anyway...,

Why are we assuming we need one Pokemon for all the contests? I'm sure that setup time would be fairly small, considering we have to go to most areas anyway. Using Gengar, or Medicham, or Poliwrath might be faster, but it's significantly less riskier just to use a Mud Sport/Water Sport Barboach, for example, for the cute contest.

If we can avoid Glitzer Popping for Feebas, we may as well evolve it and use Milotic for the Beauty contest.

This is more from personal experience but jamming seems to be common enough (and contests are slow enough) that it's worth it to waste a bit of time to get separate Pokemon and make more Pokeblocks to have a fairly high chance of winning.

Also, that was a run-on sentence. I don't care. Kappa

I will now go back to having all this crazy technical stuff fly over my head and will continue mentioning really obvious solutions that don't work for some crazy technical reason. (I define a "crazy technical reason" as "I have no idea what the hell I'm doing")
Currently doing WR attempts in (Whatever stupid hack I'm running now)

Metarkrai
Bug Catcher
Posts: 14
Joined: Thu May 28, 2015 12:08 am

Re: Emerald 100%

Post by Metarkrai » Tue Sep 08, 2015 9:33 pm

I preferred searching for a strategy using a single Pokemon for contests since there wouldn't be any difference in ribbons or contest stats from multiple "Contest" Pokemon as they would be done from a Seedot, since I thought that with a good move combination, we could win regardless the contest type, and since it would take another corruption to get the second contest Pokemon. (The corruption used for Poliwrath corrupts 5 Pokes, so you can't add one more, and every other corruption for Dex Pokemon will invole PID corruptions, and not TID corruptions, whereas we need PID + TID corruption to get the other contest Poke).
By using Seedot's movepool, we also have access to a good movepool list, allowing to change some of the Contest Pokemon moves in order to use different strategies depending on the contest type.

It's true that Poliwrath is really friendly as it only requires one Marill battle and few items and you have it set up, with moves he can learn with Heart Scales, who give a high heart amount if they can be executed without too many scares.

I'll take note of the strategy you're proposing, since every other contest strategy can be interesting as I don't know for now if they can be safer or not.
The issue for some contests types like Cute is that they have more trouble/scare users than in other categories, which means a bit more soft resets to get a decent opponent combination. Depending on the amount of Pokeblocks made, the Pokemon won't start in the same place, which will change a bit the list of risky opponents.
Mind Reader isn't useful to prevent scares/nervousness, but it's useful to block Ovations and reduce their amount, lessening the chance of having another Pokemon winning (as the main issue is to not win the contest, since 800 points are more easily done).
Submission has the issue of increasing the scares danger, even if it gives 6*2 hearts in return.
Mud Sport + Water Sport don't really seem to be more advantageous for me in Cute, as you're lessening your potential heart loss on the combo turn, but as you don't block the ovation to prevent your opponents from grabbing points, and as you're still as exposed to nervousness than before.
It gives a little incrase in success chances, but I think that getting Barboach/Wiscash from Seedot would take a longer time than some resets to get a decent opponent combination and lessen as much as possible the score difference between Poliwrath and Wiscash.

However, if you have a similar strategy using Seedot and Poliwrath moves, I'm more interested into trying it since it wouldn't take that long to get the moves.

Jamming (you're putting that word back in my mind) is frequent, but highly depends on the Pokemon and on its strategy.
A good amount of Pokemon will prefer nervousness moves (which can be avoided if we always go first) or short range jam, or doing something in their corner. (ex : Slaking uses jamming on a contest type, but prefers using copying moves on another one)
Depending on the jamming type, you can still win a contest if you only have one jammer as an opponent.
And resetting the opponents generation doesn't take that long, so you can afford 4-5 soft resets to get a working opponent combination and run with it.
(Obtaining another Pokemon with Pomeg Glitch and making more Pokeblocks take enough time for you to make enough soft resets to get a working opponent combination, in my mind)


On another matter, Feebas and Milotic will be stolen.
Feebas from a fisherman, and Milotic from Juan.
With the route plans for Hoenn Dex I have now, League will be completed 3 more times in order to steal Pokemon, with :
Elite Four Sidney : Cacturne, Crawdaunt, Absol
Elite Four Phoebe : Dusclops, Banette, Sableye
Elite Four Glacia : Sealeo, Glalie, Walrein
Elite Four Drake : Kingdra, Flygon, Salamence
Champion Wallace : Wailord, Milotic (one of the 3 attempts would be a loss on Wallace)
I want 3 times for now because Kingdra takes too much time to be corrupted, Flygon too and requires around 30 Rare Candies, and Salamence 20 Rare candies.
Glacia also has Glalie and Walrein, allowing us to skip Shoal Cave as we can get the Spheal + Snorunt family elsewhere.
Wailord can be obtained from a high-level Wailmer, so it might be removed.
Milotic can't be stolen elsewhere and takes too long to be obtained with Pokeblocks. But it might be obtained with corruption, as I don't know how long the ending sequence (from Juan defeat to save) takes.

Since I have only started reworking on the Hoenn Dex completion, I'm leaving multiple potential cases for a good amount of Pokemon as I don't know yet which way would be the fastest one (in terms of raw obtention time and accessibility).

A good amount of Pokemon will be obtained from Pomeg Glitch as they can't be stolen or as they can't be easily accessible (n a trainer in Trick House, in a Gym, or on a Trainer Rematch).
Also, since 5 Pokemon are obtained with one corruption, the average time for a single Pokemon with Pomeg Glitch isn't that high (50s without the setup, so around 1min, 1 min 10), and that average time is better than average times for many wild Pokemon who have 5/10/20/30/35% chances to be seen.
Some of these Pokemon can be obtained during the second training phase (the Plusle training), and will be caught if they appear (basically Shroomish, Slakoth, Relicanth)
Pokemon stealing is also quite fast (35s if you're stealing the first Pokemon and talking to the trainer), so even if the trainer if a bit afar, the time you'll take to get to him with repel and steal the Pokemon won't be that high. And since areas that have trainers with rare Pokemon tend to contain a certain amount of them (like Victory Road), you can go across all the area and collect many Pokemon like this, reducing the importance of the time taken to get to the trainer.

Using Repel, some Pokemon can be caught as their percentages can go really high, but in counterpart their encounter frequency is lowered. I need more tests to see who can be done (if the Repel is faster in average than Pomeg Glitch or not).

A good amount of Pokemon are still interesting to be found in the wild with 20/25/30/35% encounter rate as they can quickly evolve with 1 Stone or few Rare Candies. (whereas with Pomeg Glitch, you obtain the Pokemon at Lv 5).
Hustle from a Pomeg Glitch Corsola will be used on some fished and surfed Pokemon in order to increase a lot the chances to get them to a level high enough to evolve them with a single Rare Candy.
Illuminate will also be useable as the Volbeat obtained with Pomeg Glitch learns it.

Friendship evolution is a waste of time, you need too many items to get to 250 friendship.
By not doing Aqua and Magma Hideouts, Crobat can't be stolen. Thus, he'll be obtained with Pomeg Glitch. (Torkoal could have been repelled in Magma Hideout for 37.5%, but he can be stolen).

I haven't looked into the potential of using Pomeg Glitch for Trainer rematches. It doesn't seem useful to me (too much time for the corruption to only get 1 more Pokemon), but I'll check it a bit.

Also, natural rematches aren't worth much as the only first rematches that provide interesting Pokemon to steal are :
Leader Roxanne : Golem
Leader Brawly : Machamp
Wattson : Electrode, Magneton, Manectric
Flannery : Magcargo, Camerupt
Norman : Slaking

Well, one or 2 Victory Road trainers could be interesting to rematch as they have a rare Pokemon and as they were beaten with IFG the first time.
But apart from that, stealing on Trainer Eye trainers will tend to be done last as these would be a bother earlier with the increased chance to receive a message from them.

I'll continue getting data on Hoenn Dex and building things for a while, as it's for now a good mess.

Also, do you think that catching a Sweet Scent Pokémon (38%) and using it on Sootopolis instead of Old Rod would be a gain of average time ?
Yeah, I haven't timed that yet so I don't know, and the Sweet Scent strategy is still in the pastebin, even if I think it's slower.

My last question is about doing RNG with Sweet Scent to get rare Pokemon.
Can this be useable on the speedrun ? (can it be performed with a good accuracy that gives an average time lower than a Pomeg Glitch or Steal ?)

Here is the updated route (I changed the Master Ball and Rare Candy duplication, and 2-3 tiny things were corrected) : http://pastebin.com/GjPxMXGu


ClothNapkin : Please, continue proposing things for the speedrun. Even if it might not help, it still can, and it's by exploring the highest amount of plausible possibilities that we can get a refined route.
And with tricky setups, it's easy to miss tiny improvements. I discovered so many of them on the route I had in mind just by writing it on pokemonspeedruns / as a pastebin that I'm always happy to post something here since it always makes me check things a bit more differently and helps me summing up the things I was doing for now, as well as writing it somewhere with sentences instead of scraps / thoughts.

User avatar
ClothNapkin
Bug Catcher
Posts: 15
Joined: Sun Aug 03, 2014 8:07 pm

Re: Emerald 100%

Post by ClothNapkin » Tue Sep 15, 2015 5:37 pm

I did a couple quick tests:

Normal RNG manipulation (Without sweet scent) should take about 17 seconds per failed attempt, not accounting for RNG manipulation delay.

Sweet Scent manipulation (I don't really know how viable this is, but it seems like it would work) should take about 23 seconds per failed attempt, again not counting for delay.

With practice, you could probably hit the correct frame at least 1/3 of the time. I don't know how consistent Glitzer Popping is now, but if it's at all inconsistent, RNG manip is probably the way to go.
Currently doing WR attempts in (Whatever stupid hack I'm running now)

Metarkrai
Bug Catcher
Posts: 14
Joined: Thu May 28, 2015 12:08 am

Re: Emerald 100%

Post by Metarkrai » Tue Sep 15, 2015 9:31 pm

Thanks for the info.

The Instant Glitzer Popping that will be used for PC Pokemon corrupts 5 Pokemon with a 6-7/32 success chance.
Thus, the average time required for corruption is shared between the 5 Pokemon.
To that, you have to add some time for EV manipulation in order to get the wanted Pokemon, which consists of 1-2 clones and giving Pomeg Berries.
All the Pokemon that will need to be obtained with Pomeg Glitch will be obtained in the same time, after wild catches and stealing are done, as lucky finds for wild Pokemon (like a Relicanth Underwater while EV-ing with Clamperl) or lucky gym leader rematches will change the amount of Pokemon that will need to be obtained with Glitzer Popping, and as the EV-manipulation sequence will be common to all these Pokemon, reducing the time required per Pokemon.
Right now, I listed 30 Pokemon that will always be corrupted, 5 that could be found luckily but will be corrupted if not, and 14 Pokemon where I'm not sure if Corruption is the fastest choice.
This should roughly give 40-45 Pokemon to obtain with Glitzer, which means 9 Instant Glitzer Popping for them.

And added to the preparation + corruption time, you have the hatching sequence.
Well, you made me realize that due to these dumb contest stats, the Eggs don't hatch in one step cycle, but in 5 step cycles.
Thus, hatching 5 Eggs will take 255*5+255+255+255+255 = 255*9 steps (one Egg can hatch at the time), which means an average of 459 steps per Egg. And I have no way in mind to diminish the step cycles or make consecutive hatches.
But it's a relief that it's less than the double of the first average amount of steps I had, which means that the average time for a Corruption won't be that much higher than what I timed, and thus won't change the destiny of many Pokemon.

Glitzer Popping is here quite consistent, the major part of the time is spent into hacthing the Egg.
Oh, yeah, I forgot to mention that again but RNGing Instant Glitzer Popping is a bit out of the question, as you'd need to manipulate your ID and SID and hit the right frame (when you open your party right after loading the save, it seems to depend on the location or something else, so a second opening would be required too). At least, unless using RNG for ID and SID can be consistent.

Here are the times I had :
Giving a Pomeg Berry : 3'12" (the amount of Pomeg Berries will be calculated later, but count 2-3 per Pokemon in average)
Instant Pomeg Glitch setup : (for all the Pokemon)
Instant Pomeg Glitch for PC : 15'03" (once)
Instant Pomeg Glitch average success time : 80'16" (for 5 Pokemon)
Hatching sequence : 12'23" (for 1 Pokemon)
Getting out of Pokemon center and making 255 steps : 22'05" (roughly count 38s for 459 steps)
Majotared average time : 9'36" + 16'032" + 12'23" + 38' = 75'622". (it was 50s with my mistake)

With a practiced accuracy of 1/3 (that seems quite good to me), you'd take in average 69 seconds to meet the right Pokemon, adding 19'14" for the capture + dex entry (starting from the battle menu) + 2-3 seconds for the end of the appearance animation (when the Pokemon appears on screen).
This means 91 seconds in average to get a rare Pokemon.
Thus, unless you're on a certain areas where you can manage to get multiple rare Pokemon (with high chances to fell on them if you miss the Sweet Scent frame), using Sweet Scent (with practice) could be useful, as the failed hits would still induce (with a high chance) captures, and thus not count as raw time wastes.

I wanted to make a more detailed post about the lists I made for Capture/Stealing/Corruption/Evolution for Hoenn Dex, but I'll have to rework it as the 75s average time for Corruption will change a certain amount of things.
Since a Pokemon steal takes 35'28" (starting from trainer talk to the end of the battle, in the case the first opponent's Pokemon is stolen), some faraway trainers or Elite 4 Leaders could become interesting again. (Elite 4 trainers talk a lot, and you're forced to walk, which adds a good amount of time)
For evolutions, giving a Rare Candy requires 4'14", preventing the learning of a new move takes 3'20", and evolving takes 26'07". Thus, some Pokemon that require up 10 Rare Candies to evolve should fall into the "Evolve" category now.
Some Pokemon as starters final form, Flygon, Weezing, Salamence,... will still be corrupted as they require 20 or more Rare Candies (80 seconds to feed them the Rare candies). And the ones requiring 13-15 Rare Candies should also be corrupted too (since there will be 1 or 2 new moves to learn in the way, the average time for evolution will be over 75s).
With the 75s average time, more wild Pokemon could be attempted to be caught too depending on their percentage.

There will also be a good list of timings I'll have to complete in order to be more accurate about the average time taken, and in order to decide what to do on specific cases like Evolutionary stones (I don't know how much time it requires to get a shard and get a stone, even if with the 75s limit, I think evolutionary stones will be obtained).

I also made a list of route catch and route steal priorities, in order to start having an idea of the route we'd need to take to catch wild Pokemon while doing some steals, and complete the steals, while benefitting as much as possible of rare Pokemon encounters (they are numerous, and there aren't alone in areas like Victory Road, Petalburg Woods, Route 115, Safari Zone or Fiery Path.

Thanks for your post, I'll go back to work to correct my notes.

Post Reply

Return to “Pokémon Emerald”

Who is online

Users browsing this forum: No registered users and 1 guest