howto use stats processor with oclHashcat+ v0.8?

[ntk]

1. why when I feed in large amount of combinations the status window comes up without my involvement?

because it always does each 10 seconds or so plus when its finished as a form of final status. this is required in stdin mode since the program gets the data via stdin. so there is no way for the program to read the keypress from stdin.

at least I know that is not my fault to activate the status screen.

2. does it happen on other linux system, or window when you use combination like this.

its not dependant on the input (notice the final status)

3. why when status screen comes up GPU load immediately goes down to NULL. system stop working, i.e not find this simple password

typical driver problem. use cat 12.6

I am reluctant to take your advise here with 12.6 because it won't solve this problem. Firstly I had use 12.6. It has bad performance even on pyrit so I go back to 12.4. Secondly I have read posts from 3 users on this forum, reporting seemingly various problems but I could make reduction & see this 'stop calculation' every where, If you have trace in the code, we can activate debug and show it.

in a sec SP64 -pw-min 8 ?d?d?d?d?d?d?d?d produce a file of 900MB large dict, it cant be not enough data for oclHashcat+

4 How can I produce some traces for Atom to check this STDIN behaviour.

i dont think its a bug. i think its some problem with driver, or maybe with the hccap.

Hakonen is a standard example, aircrack-ng can solve it, pyrit can solve it, ochHashcat-plus can solve it with direct generation of combinations. it s not problem with the cap or hccap Atom.

let experiment and see failed with stdin, then run ocl with direct mask processor
./oclHashcat-plus64.bin Hakonen.hccap -o foundhakonen.txt ?d?d?d?d?d?d?d?d
you see it pass.
run again with stdin to see it fails, do again with ocl with direct mask processor to see it pass
run again with stdin to see it fails, do again with ocl with direct mask processor to see it pass
run again with stdin to see it fails, do again with ocl with direct mask processor to see it pass

Consistently ===> that is a problem hang with stdin

5 do you use stdin, in different command format?does it work long hours?

the way you did it is correct.

stdin is very important to me. If possible I would be obliged to run stdin-regression test for you in every release, but please do maintain and opt to optimize it too, even when you'd have completed Markov's implementation into oclHashcat-plus

6 if you can use sp64.bin as replacement for BF, how do you pipe in oclHashcat+

same way, just try it

try it and see it fail.
run again with stdin to see it fails, do again with ocl with direct mask processor to see it pass
do three more time
Consistently ===>reduction ===> that is a problem hang with stdin

is it possible that we all have the debug version SW, we only need to activate the trace, Atom?

---

thank for you reply Atom.