Test 1~ I took out a very unhappy noctowl out of his box after leaving him there over 100 hours. I put the beserk gene on him and out of ten moves heres what i got

1-confused
2-confused
3-confused
4-confused
5-confused
6-confused
7-confused
8-confused
9-hit
10-confused

Now I did the same thing with my happy lapras and this is what I got

1-hit
2-hit
3-hit
4-hit
5-confused
6-hit
7-confused
8-hit
9-hit
10-hit

Conclusion~ Happiness is a factor in hitting yourself with confusion... or at least with the beserk gene

------------------
"When I see something screech across the room and latch onto someone's neck, and that person starts to scream, I have to laugh because...what IS that thing?"

Noctowl was confused 14/20 times~70%

and

Lapras was confused 7/20~35%

quote:

---

Originally posted by The Great VIVI:
on my second test out of twenty

Noctowl was confused 14/20 times~70%

and

Lapras was confused 7/20~35%

---

So, are unhappy Pokemon like, disturbed or suicidal?

------------------
"The fact of the matter is, I'm fucking brilliant. Not was brilliant, am brilliant."-Pete Townsend

Now for more testing...

quote:

---

Originally posted by White Cat:
Now for more testing...

---

Such as...?

------------------
"When I see something screech across the room and latch onto someone's neck, and that person starts to scream, I have to laugh because...what IS that thing?"

quote:

---

Originally posted by Automaton:
So, are unhappy Pokemon like, disturbed or suicidal?

---

This research is inadequate for proving that Confusion is affected by Happiness This would be clear if an objective analysis of all the relevant data known thus far was performed.

Base confusion information
--------------------------
The rate at which a pokemon would hit itself while confused in R/B/Y (as found on the Confusion page right off the front page here): 50%

Identical information on the basic confusion rate has been noted for G/S (specifically in the basic moves guide put up by Darryn, the information therein having been taken from the original Japanese TeamPA site).

Confusion rates for the tested pokemon
--------------------------------------
Test set #1 (sampling size of 10)
Unhappy - 90%
Happy - 20%

Test set #2 (sampling size of 20)
Unhappy - 70%
Happy - 35%

Analysis
--------
When the sampling size was increased, the Unhappy subject's confusion rate dropped by 20%. Conversely, the Happy subject's confusion rate increased by 15%.

Intermediate conclusion
-----------------------
With the change in rates shown between sampling sizes, the pattern that emerges is that both rates are likely to converge to the base value.

Final conclusion
----------------
The sampling sizes in this test are insufficient for trying to prove an assertion of this nature. The values recorded thus far cannot be extrapolated to prove that Happiness does affect Confusion in any way.

White Cat was right in the idea of encouraging you to test for something that others have not considered testing. He was also right in saying Now for more testing... . More testing is needed to come to any significant conclusion on this subject.

No one could be expected to do the same amount of tests that Wintermute used to do (at least not without an emulator running at 10 times normal speed), but this definitely needs a larger sampling size (unless someone can figure out a shortcut that can make a definite conclusion possible).

I also have hunch on that happiness is a factor in attacks that require accuracy

Confusion rate~ 61/107
Hit rate~ 46/107

I believe "hit rate" is when your Pokemon hurts itself. "Confusion rate" is when it's confused, but doesn't hurt itself.

Aren't there Gold and Silver ROMS available?

quote:

---

Originally posted by Mr. K:
Continue: [b]No one could be expected to do the same amount of tests that Wintermute used to do (at least not without an emulator running at 10 times normal speed)...

Aren't there Gold and Silver ROMS available?

That would substantially decrease test time...[/B]

---

Assuming that someone has an emulator that does run at 10x speed, the technical capability does exist.

My expectations are more focused on the limitations imposed by how much time is available to people and their own patience. It takes both qualities to get in that many trials for things.

Most people are going to balk at doing 1400 trials for anything.

Personally, I would also be concerned about the quality of the results from most ROMs. There have been many strange results noted from various G/S ROMs that are out there. The incident of a male Jynx comes to mind.

If the ROM image wasn't taken directly from the tester's own cart, so they could verify that strange results aren't due to a couple of bits accidently flipped when the image is made, I'm not sure how far I would trust the results coming out of that testing.

That's basically my reasoning on the matter.

no$gb runs all of them fine as far as i can see, and it DOES have a 10x speed option. (basically, unlimited mhz.) it's SO hard to pick up my gameboy after playing the emu, cuz it seems so SLOW now!

no$gb's actually the best emu i've seen, and i've tried a few. if you have a registered version, no other emu compares (and crystal runs great!) i saw a downloadable registered version in karp a while back, but i'd suggest you check it GOOD for viruses if you're goin' that route.

Dumping a ROM is just like copying a file, isn't it? When I copy a file from a floppy or Zip disk, I don't work about the data being changed in the process. The ROM should either be an exact copy, or not work at all (if it was a bad dump).

I was going to say something similar. Wintermute did a lot of testing with his emulator thingy and we never found any problems.

quote:

---

Originally posted by White Cat:
The gender-changing bug was caused by earlier versions of the PR Translations patch.

---

Not the case that I'm thinking of. This was when Elwen was going through her ROM. What she had was among the original batch of ROMS, Japanese only. There were bad ROMs from at least two sources that I recall.

I'm not familiar with the state of the English ROMs.

quote:

---

Originally posted by White Cat:
Dumping a ROM is just like copying a file, isn't it? When I copy a file from a floppy or Zip disk, I don't work about the data being changed in the process. The ROM should either be an exact copy, or not work at all (if it was a bad dump).

---

When you move data between devices on your computer, a good manufacturer will include data integrity checks in device drivers for devices that have a significant problem with proper transfer.

For example, the difference between a CD-RW from a good manufacturer and a poor one can easily be seen over the span of ten disks. A friend of mine thought he was getting a real bargain on his CD burner. He ends up using another friend's burner alot.

Data transmission over a network is a source of error. The chances of it happening are slim. The parity bit checksums for standard packetized transmissions will catch most of the very small percentage of bad transmissions that occur.

There is the problem that parity bit checksums can't catch when pairs of bits are flipped.

The chance that things will go bad and such bad packets will also be considered good is exceedingly small.

I see your point, but if the alternative is that it takes 10 times as long to test, I think it's a fair gamble.

Also, if the results are way off, you're going to notice in regular game play.