Success count appears off, reboot speeds up attack

[GoogleCodeExporter]

This isn't as much of an issue as it is a question as to why certain things are 
happening that I've noticed.

First off, whenever I perform attacks, the percentage appears to jump after 
about 15% up to 90%, I've never been there for the jump, but I've never seen 
anything from 20-80%, and I've checked it fairly frequently, why is the reason 
for this?
On a similar note, it seems like it jumps up to 90 after much less successful 
pin attempts than I would expect, shouldn't there be 100 million possible pins? 
Does more than 1 pin get tried per success or something?

And finally, I've noticed that after time my success rate seems to slow down, 
but simply rebooting the computer will show a sudden boost in successes. From 
what I know about the theory behind this, it makes no sense that a restart 
would cause this boost. I'm using an Alfa wireless adapter if that matters.

Anyway, if those questions could be answered, that'd be awesome, or if there's 
a better place for me to post this, I'd like to do that.

Original issue reported on code.google.com by Danny.C....@gmail.com on 26 Sep 2012 at 6:52

[GoogleCodeExporter]

if understand, the jump is because of the number of the pin, so pin very low, 
jump to 90% when found.
then it will slowly find the exact pin.

Original comment by deltomaf...@gmail.com on 26 Sep 2012 at 1:11

[GoogleCodeExporter]

So you're saying that there's a way to tell if you're close to the actual pin? 
I can't see why they would implement it like that

Original comment by Danny.C....@gmail.com on 26 Sep 2012 at 7:37

[GoogleCodeExporter]

is easy to understand, the pin it's between 00000001 and 99999999. How the 
reaver works this order and can not know in what position is the exact pin, 
when he finds the percentage is move to near of 100% to be more real progress.
I would like to implement the method of choosing the initial pin or decreasing. 

Original comment by deltomaf...@gmail.com on 27 Sep 2012 at 12:09

[GoogleCodeExporter]

reaver cracks first one half of the PIN (4 first digits), when it find the 
first half, jumps to 90,xx% to crack the second half of the PIN (3+1 digits).

Example:
0001 567 1
0002 567 2
0003 567 3 (1%)
0003 001 1 (90%)
0003 002 2 (99%)
0003 002 9 (100%)

First half of the PIN: 0003
Second half of the PIN: 002
Last half of the PIN: 9
The PIN: 0003 002 9

Original comment by Mo.elyou...@gmail.com on 1 Oct 2012 at 8:10

[GoogleCodeExporter]

Hi Mo.elyou,
Are you saying that the second half of the pin doesn't matter until the first 
half is found? This is what reaver is doing on my system. If you notice the 
first half is incrementing by 1 every time while the second half is looping the 
same numbers over and over.
3246 5679
3247 5678
3248 5677
3249 5676
3250 5675
3251 5674
3252 5673
3253 5672
3254 5671
3255 5670
3256 5679
3257 5678
3258 5677

Original comment by parkou...@gmail.com on 27 Dec 2012 at 3:08

[GoogleCodeExporter]

#5 I mean that the most important of the PIN and the most hard to find it's the 
first half of the PIN (the first 4 digits), After reaver finds the first half, 
it has just to find another half between 0 and 999, after that it has to find 
the checksum.

First half: between 0 and 9999
First part of the second half: between 0 and 999 (easier than first half)
Second part of the second half: between 0 and 9 (the easiest one to find)

So the most important and the hardest thing that reaver has to discover it's 
the first half of the PIN.

The part1 of the half2 when reaver it's on a percentage less than 90%, doesn't 
matter because reaver doesn't even use it. As you see in your example, the 567 
it's always the same and it changes just the first half and the checksum (the 
last number).

Original comment by Mo.elyou...@gmail.com on 27 Dec 2012 at 5:24

[GoogleCodeExporter]

Thanks that makes a lot more since. I thought I was doing something wrong. 
After 6.5 hours reaver found the correct PIN 3560 0022.

Original comment by parkou...@gmail.com on 27 Dec 2012 at 6:32