The Azure Heights Forum


Post New Topic  New Poll  Post A Reply
my profile | search | faq | forum home
  next oldest topic   next newest topic
» The Azure Heights Forum   » Main Campus   » Research Lab   » Softboiled\recover

UBBFriend: Email this page to someone!    
Author Topic: Softboiled\recover
Wizzymoto
Farting Nudist
Member # 60

Member Rated:
posted 03-19-2000 07:50 PM      Profile for Wizzymoto   Email Wizzymoto   Send New Private Message      Edit/Delete Post  Reply With Quote 
Are they exactly the same????
I Think the only difference is softboileds only learned by Mew Chansey, and can heal out of battle.

------------------
"I thought you said you were gonna watch a video"
"We did, an video on how to make your own sausge"


From: Irvine, CA, USA | Registered: Feb 2000  |  IP: Logged
Avalanche
Farting Nudist
Member # 38

Member Rated:
posted 03-19-2000 08:22 PM      Profile for Avalanche   Author's Homepage   Email Avalanche   Send New Private Message      Edit/Delete Post  Reply With Quote 
You are correct, there is no difference but the out of battle thing

------------------
"Screw you guys, I'm going home" Cartman


From: Catherdral City, CA, USA | Registered: Feb 2000  |  IP: Logged
cfalcon
OLDNBLD
Member # 19

Member Rated:
posted 03-20-2000 05:18 AM      Profile for cfalcon   Email cfalcon   Send New Private Message      Edit/Delete Post  Reply With Quote 
Less PPs on Softboiled.

There also may be some difference in their "accuracy". One of the biggest unexplained aspects of pokemon battling is the (apparently > 1/256) odds that certain moves (Rest, Recover, etc.) fail when they should succeed.

This factor may be different, if indeed it even exists.

-cfalcon

------------------


From: 39°45' N, 104°52' W | Registered: Feb 2000  |  IP: Logged
Eevee
Farting Nudist
Member # 97

Rate Member
posted 03-20-2000 08:20 PM      Profile for Eevee   Author's Homepage   Email Eevee   Send New Private Message      Edit/Delete Post  Reply With Quote 
quote:

Less PPs on Softboiled.

There also may be some difference in their "accuracy". One of the biggest unexplained aspects of pokemon battling is the (apparently > 1/256) odds that certain moves (Rest, Recover, etc.) fail when they should succeed.

This factor may be different, if indeed it even exists.

-cfalcon



Yup, CFalcon is correct.

------------------
I'm Mew from the Upnetwork

"Desperate times call for desperate measures." -Me

"I do not follow, I lead." -Me

"You can't accomplish anything without making an effort." -Me

"You know it, I know it, so don't deny it, Eevee is so cool since it can evolve into 5 types! (Theres two more types in Gold/Silver)

"I don't care what anyone thinks of what I like or what I do...so shove it!" -hehehe Me, being defensive.


From: Molino, FL | Registered: Mar 2000  |  IP: Logged
Mr. K
Racist
Member # 2

Member Rated:
posted 03-21-2000 10:45 PM      Profile for Mr. K   Author's Homepage   Email Mr. K   Send New Private Message      Edit/Delete Post  Reply With Quote 
I don't believe what causes Rest and Recover to fail is based on the 1/256 accuracy thing.

I have a theory I've mentioned to Toby, but he hasn't had time to test it out, or even to figure if it makes sense or not.

I think if you have certain values of HP left, it will fail. This would explain why it fails multiple times in a row, if your HP doesn't change.

I had Rest fail 3 times in a row, while an enemy was sleeping.

The difference between max HP and current HP was 511, which is suspiciously close to the magic number of 512.

This could have something to do with division or that position-shifting thing cfalcon's always talking about (which I suppose is pretty durn close to the same thing).

Next time any of you experience a Rest/Recover/Softboiled failure, please note your current HP, max HP, and see if you can duplicate the problem if you don't get hit that turn.

Thank you, drive thru please.

------------------
KOFFING!
Azure Heights Pokémon Laboratory


From: Cinnabar Island | Registered: Feb 2000  |  IP: Logged
cfalcon
OLDNBLD
Member # 19

Member Rated:
posted 03-22-2000 05:26 PM      Profile for cfalcon   Email cfalcon   Send New Private Message      Edit/Delete Post  Reply With Quote 
Err... I was testing REST today, and couldn't get it to not work... what was your Snorlax's max HP? I set current to 511, and there aren't to many maxed monster stats above that (513,515...523), all of which I tested with curHP set to 511. No lockup.

-cfalcon

------------------


From: 39°45' N, 104°52' W | Registered: Feb 2000  |  IP: Logged
Mr. K
Racist
Member # 2

Member Rated:
posted 03-22-2000 09:25 PM      Profile for Mr. K   Author's Homepage   Email Mr. K   Send New Private Message      Edit/Delete Post  Reply With Quote 
My Snorlax has a max HP of 523 and had (I believe) 12 HP left when it tried to Rest.

I'm sure about the 523, not sure about the 12. No stat changes were in effect, but the enemy (another Snorlax) was sleeping.

------------------
KOFFING!
Azure Heights Pokémon Laboratory


From: Cinnabar Island | Registered: Feb 2000  |  IP: Logged
10,000Lb.Snorlax
loves long time.
Member # 13

Member Rated:
posted 03-23-2000 05:11 PM      Profile for 10,000Lb.Snorlax   Author's Homepage     Send New Private Message      Edit/Delete Post  Reply With Quote 
Could 255 also be a magic number, as seemingly everything in pokémon relates to it (max time on game, max level of pokémon, max number of hall of fame enterers, etc...). What are the odds of rest failing three times in a row when it's pure chance? 1/256 x 1/256 x 1/256 ?

That is a very low probability for pure chance occurances... I think Mr. K is on to something

------------------
If I had a Snorlax, I'd use him as a Bed


From: Denver | Registered: Feb 2000  |  IP: Logged
Mr. K
Racist
Member # 2

Member Rated:
posted 03-24-2000 01:31 AM      Profile for Mr. K   Author's Homepage   Email Mr. K   Send New Private Message      Edit/Delete Post  Reply With Quote 
Yeah, for the same reason 511 is important, 255 is important. (These numbers are 512 and 256 minus 1...since we start counting from 0, not 1.)

The short answer is that computers use binary numbers and for that reason, all exponentials of 2 are important. So in the same way 999 or 9999 are significant to digital thinking, 63, 127, 255, etc. are important to binary thinking.

So, when other people have Rest/Recover fail, perhaps the difference between max and current HP was 255 or 127 or something.

Maybe. Could be total coincidence that the difference in my Snorlax was 511, but it's the only thing that jumped out at me, and it's especially interesting that during the 3 times it failed, the HP difference did not change.

I suspect most people only have it fail once because they're dead on the next turn.

------------------
KOFFING!
Azure Heights Pokémon Laboratory


From: Cinnabar Island | Registered: Feb 2000  |  IP: Logged
MewtwoSama
Asshole
Member # 12

Member Rated:
posted 03-24-2000 02:14 PM      Profile for MewtwoSama   Author's Homepage   Email MewtwoSama   Send New Private Message      Edit/Delete Post  Reply With Quote 
I think I started a topic on this failure deal.

My completely stupid theory is that there is a grey area between 1/4 to 1/2 Max HP where those moves can fail.

Maybe its a certain percentage of the Max HP where it will fail.

I could hack myself a Level one monster with Seismic Toss to find the certain HP wher it fails to test this out....

------------------
MewtwoSama
A viscous,Psychic Pokemon that was created by a scientist after years of horrific gene splicing and DNA engineering experiments. It's DNA is almost the same as MEW's. However, it's size and disposition are vastly different. It's cold, glowing eyes strike fear into it's enemies.

ICQ: 45770183
AIM: MewtwoSama
E-mail: mewtwo@ureach.com


From: Abyss of Evil | Registered: Feb 2000  |  IP: Logged
Phoenix
Farting Nudist
Member # 143

Rate Member
posted 03-26-2000 12:58 PM      Profile for Phoenix     Send New Private Message      Edit/Delete Post  Reply With Quote 
Hmm....Rest failing? Havn't done that before. 255 is a special number in the game though.

------------------
I am Phoenix. Rising from the ashes to destroy all who stand in my way...


From: Houston, Texas, U.S.A. | Registered: Mar 2000  |  IP: Logged
Mr. K
Racist
Member # 2

Member Rated:
posted 03-27-2000 01:48 AM      Profile for Mr. K   Author's Homepage   Email Mr. K   Send New Private Message      Edit/Delete Post  Reply With Quote 
Toby's just reported an instance of a Mewtwo failing to Recover 3 times in a row with Max-Current HP difference of 255!

Woo hoo!

Now all we have to do is test the reliability of these failures and see if they happen at other multiples of 2.

Mystery (possibly) solved!

------------------
KOFFING!
Azure Heights Pokémon Laboratory


From: Cinnabar Island | Registered: Feb 2000  |  IP: Logged
cfalcon
OLDNBLD
Member # 19

Member Rated:
posted 03-27-2000 03:37 PM      Profile for cfalcon   Email cfalcon   Send New Private Message      Edit/Delete Post  Reply With Quote 

Oh, a *difference* of 511 HP. Sorry, I misunderstood.

[cfalcon whips out Gameshark plus Red]
>for all tests, Snorlax has Spore.
>for all tests, Snorlax HP=MHP-(2^n)-1

Please note that all powers of two in binary look like 10000...000, just like all powers of ten in decimal. This means that a power of two minus 1 will be 0111....111 in binary, just like a power of ten minus one in decimal is 9999...999.
Here we go:

Snorlax MHP=523, HP=12, Diff= 511 = 1FF (hex)
I ran REST out of PPs (10). It did nothing.

Snorlax MHP=2000, HP=977, Diff= 1023 = 3FF (hex)
I ran REST out of PPs (10). It did nothing.

Snorlax MHP=999, HP=936, Diff= 63 = 3F (hex)
Went to sleep, healed.

Snorlax MHP=256, HP=1, Diff= 255 = FF (hex)
I ran REST out of PPs (10). It did nothing.

Snorlax MHP=256, HP=2, Diff= 254 = FE (hex)
Went to sleep, healed.


I'm gonna go ahead and concur with Toby's research and Mr. K's experience.

Basically, what they said was that if the difference between MHP and HP has FF as it's least signifigant byte, REST won't work. If he tested with Recover, it's almost assuredly the same check for all three of the moves in question. Since this doesn't seem to affect other powers of 2 -1 (for instance, 15 wouldn't do it, only ones greater than or equal to 255, such as:

255
511
1023
2047

etc.

Only the first two are even valid for general (high level) pokémon playing.


Does that sound right?


This should definitely be added to the Rest/Recover/Softboiled pages.

-cfalcon

------------------


From: 39°45' N, 104°52' W | Registered: Feb 2000  |  IP: Logged
Mr. K
Racist
Member # 2

Member Rated:
posted 03-28-2000 12:57 AM      Profile for Mr. K   Author's Homepage   Email Mr. K   Send New Private Message      Edit/Delete Post  Reply With Quote 
Cool!

Toby didn't perform any actual testing, he just ran into a situation where Recover failed and noticed the difference was 255.

I'm sure he'll want to be extra thorough and test Softboiled and multiples of 2 (minus 1) even lower than 63, but I think we've just about wrapped up this case. Kick ass.

I am slightly confused as to why this would be a problem. You'd think having the difference appear to be a zero would cause the problem, but bugs are bugs and who knows how those Fs are being processed.

Thanks for your testing, cfalcon...

I'm going to be really busy for the rest of the week, and might not be able to do much to the site for a while, but I hope to get this info up soon.

This also lends creedence to my long-held suspicion that moves that affect yourself are most likely 100% effective, not 99.6%.

For example, has anyone ever seen Double Team fail (except in the 7th usage)?

I suppose this curiosity with Rest/Recover/Soft-boiled could be by design, to add excitement (it certainly caught me by surprise!), but it smells like a bug to me...

------------------
KOFFING!
Azure Heights Pokémon Laboratory

[This message has been edited by Mr. K (edited 03-28-2000).]


From: Cinnabar Island | Registered: Feb 2000  |  IP: Logged
Avalanche
Farting Nudist
Member # 38

Member Rated:
posted 03-28-2000 07:48 PM      Profile for Avalanche   Author's Homepage   Email Avalanche   Send New Private Message      Edit/Delete Post  Reply With Quote 
When I was first new to the game, I seem to remember Metapod hardening the second time and it said it didn't work. I don't remember any details, it was probably just a fluke I guess.

------------------


From: Catherdral City, CA, USA | Registered: Feb 2000  |  IP: Logged
Hackman
Farting Nudist
Member # 131

Rate Member
posted 03-29-2000 02:22 AM      Profile for Hackman   Author's Homepage   Email Hackman   Send New Private Message      Edit/Delete Post  Reply With Quote 
You know... If Rest/Recover/Softboiled 'always' fails at that HP difference (255, 511), would it do the same (fail/miss) for other attacks with very high accuracies?

- Swift!?!?!?
- Thunderbolt, Surf, and other high accuracy attacks?
- Other non-damaging moves w/ almost 100% accuracy, like Spore?

I have no shark to play with, unfortunately.

------------------
Hackman
pds64@hotmail.com
http://hackmanpokemon.u4l.com


From: Silicon Valley! | Registered: Mar 2000  |  IP: Logged
White Cat
Nobody knows why I'm an admin.
Member # 42

Member Rated:
posted 03-29-2000 02:53 AM      Profile for White Cat   Author's Homepage   Email White Cat      Edit/Delete Post  Reply With Quote 
Doubtful, because those moves don't have any effect on your own HP.
From: Calgary | Registered: Feb 2000  |  IP: Logged
cfalcon
OLDNBLD
Member # 19

Member Rated:
posted 03-29-2000 08:37 PM      Profile for cfalcon   Email cfalcon   Send New Private Message      Edit/Delete Post  Reply With Quote 

There is definitely a check to see if you can restore / softboiled / rest. The check should see if your current hit points are equal to your max, but it may instead check to see if they are greater than or equal. Since a compare is really just a subtract, and hitpoints are a 16 bit number, and the Z80 is a chip made in 1976 (which is older than me by a couple years, and probably most of you), there is probably some mixup involving sign flags.

This is just a guess. I actually went looking for Z80 opcodes over the last few days, and it doesn't look like this should occur unless the programmers were monkeys.

But because the comparision should only be done for these moves, the other moves shouldn't have that problem.

Which doesn't mean it shouldn't be tested...


Any ideas on the probably coding mistake that yields this silly problem?
-cfalcon

------------------


From: 39°45' N, 104°52' W | Registered: Feb 2000  |  IP: Logged


All times are ET (US)  
Post New Topic  New Poll  Post A Reply Close Topic    Move Topic    Delete Topic next oldest topic   next newest topic
Printer-friendly view of this topic
Hop To:


Contact Us | Azure Heights Pokémon Laboratory

© 2000-2016, Maximum Penetration Industries.

Karpe Diem


The views and opinions expressed on this page are strictly those of the author(s). The contents and links have not been reviewed or approved by the University of Miami.