[ 3 / biz / cgl / ck / diy / fa / g / ic / jp / lit / sci / tg / vr ] [ index / top / reports / report a bug ] [ 4plebs / archived.moe / rbt ]

Maintenance is complete! We got more disk space.
Become a Patron!

/g/ - Technology

View post   

[ Toggle deleted replies ]
File: 58 KB, 514x524, akari2.jpg [View same] [iqdb] [saucenao] [google] [report]
54012877 No.54012877 [Reply] [Original] [archived.moe] [rbt]

I just finished my CPU-based tripcode searcher!

I don't have any modern hardware to test it on, so please run it and tell me how well it runs on modern CPUs. It's multi-threaded, so it should also serve as a nice benchmark.
My 9 year old C2D maxes out at 314 kTrips/s.
TRIP: '!s6AKARI0gc' -> PASS: '_6U>2'tb' @ 304.57 kTrip/s
TRIP: '!SwJ3takARI' -> PASS: '2:nT``ZZ' @ 308.14 kTrip/s
TRIP: '!Fn6daKarIE' -> PASS: 'Bu_p`"=4' @ 307.80 kTrip/s
TRIP: '!kMAkaRiBhs' -> PASS: 'RK{Xi;:N' @ 307.79 kTrip/s
TRIP: '!kYMXgakArI' -> PASS: 'i:3gF9aW' @ 308.69 kTrip/s
TRIP: '!qGAkArIfT2' -> PASS: 'NrB)S=73' @ 308.28 kTrip/s
TRIP: '!nSSgbAKArI' -> PASS: 'pXtIo30]' @ 309.27 kTrip/s

>> No.54012914

have you considered killing yourself?

>> No.54012933
File: 333 KB, 823x720, 1460212034527.png [View same] [iqdb] [saucenao] [google] [report]

Wow, why bully?

>> No.54012971

Ill give it a try on my i5 750 in a few. Though i hate tripfags myself i see no bash your project since it helps you learn

>> No.54013052

She looks so adorable

>> No.54013086


>> No.54013091


>> No.54013096

Have you?

>> No.54013097


>> No.54013105 [DELETED] 


>> No.54013124
File: 71 KB, 675x574, 1459475695463.jpg [View same] [iqdb] [saucenao] [google] [report]

it's never gonna be very fast on CPU (as opposed to GPU or ASIC) hardware

>> No.54013127

it's open sore, check for botnet yourself

>> No.54013156

>implying its not possible to obfuscate code

gfto with your botnet

worse than win 10 shills

>> No.54013173

into my filter they go.

>> No.54013176

I just reviewed the code, there is no botnet

>> No.54013181


CUDA or GTFO faggot

>> No.54013190


>> No.54013222

Thanks, I don't have a good GPU because those are botnet so I appreciate it

>> No.54013233

CPUs are botnet too you know.

>> No.54013239
File: 36 KB, 417x419, file.png [View same] [iqdb] [saucenao] [google] [report]

I'll take it for a spin...

>> No.54013251

tripforce 0.2.0
Copyright (C) 2016 microsounds <https://github.com/microsounds>
Released under the GNU General Public License v3.
Utilizing 4 threads.
TRIP: '!Sb1gT2h4x.' -> PASS: 'IBylxM.3' @ 411.11 kTrip/s
TRIP: '!k2h4x60t9U' -> PASS: 'q'MD$I"O' @ 410.10 kTrip/s
TRIP: '!bwG5P2h4xw' -> PASS: '-v'2U33M' @ 409.55 kTrip/s

Clang on Linux looks good

>> No.54013269
File: 214 KB, 400x462, 1460395840953.png [View same] [iqdb] [saucenao] [google] [report]

Well done OP. Now we'll get even more tripfag cancer.

>> No.54013280

Has to be done in Linux, I see. I'll start up one of my VMs.

>> No.54013299
File: 95 KB, 558x361, 2016-04-12-205016_558x361.png [View same] [iqdb] [saucenao] [google] [report]


>> No.54013301

How do I convert it to an exe?

>> No.54013316
File: 78 KB, 884x574, 1458871963419.jpg [View same] [iqdb] [saucenao] [google] [report]


>> No.54013323

What does this have over tripcode explorer?

>> No.54013335


Again this time on gcc for giggles
tripforce 0.2.0
Copyright (C) 2016 microsounds <https://github.com/microsounds>
Released under the GNU General Public License v3.
Utilizing 4 threads.
TRIP: '!XyhQV2h4xQ' -> PASS: 'z97LC}[Z' @ 343.89 kTrip/s
TRIP: '!32h4xy5wkg' -> PASS: ']2{gJ:j4' @ 345.57 kTrip/s
TRIP: '!xh9L2h4xrs' -> PASS: 'PH25NBd0' @ 173.55 kTrip/s
TRIP: '!WJG2h4xkqE' -> PASS: ',u6}Ko:%' @ 370.70 kTrip/s
TRIP: '!2h4xXDhzls' -> PASS: ' }D1DpB ' @ 350.46 kTrip/s

>> No.54013344

compile it, duh

>> No.54013354


doesn't this mean we can use this to crack their tripcode and post "I like dicks lol" ?

should mean less tripcodes, especially if you go for the more subtle troll (recommending arch/win10/osx depending on what tripfag hates)

>> No.54013373

compiling with clang yields a binary that only utilizes one care, what am I missing?

>> No.54013374

compile openssl on windows
statically link libssl and libgomp so you can actually run the .exe on other people's machines

i tried doing it and gcc on windows doesn't seem to support thread-local storage, so you have every thread printing the same string matches, and it's a huge mess

also nobody does it because nobody cares, a tripfag can just change their tripcode, and searching 10 characters takes literally forever

>> No.54013375
File: 75 KB, 260x375, freedoms-respecting cute anime12.png [View same] [iqdb] [saucenao] [google] [report]


>> No.54013385

tripforce uses openmp, and if it can't find it, it replaces all openmp code with null statements and dummy mutexes.

>> No.54013386

I would guess openmp?
read the includes in tripforce.c

>> No.54013404

the program info header shoud be output to stderr

>> No.54013407

By becoming the tripfag you help spread tripfag cancer even further. You're literally giving tripfags what they want, attention. The only real solution has been and always will be to ignore them or filter them.

In addition to this should you be lucky enough go trigger a tripfag (0.0000001%) he will just change his trip and you will keep fucking with him giving him more attention, and you will both jerk each other off like the faggots you are until you both kill yourselves.

>> No.54013416

When compiling with gcc it works fine and uses 4 cores, but the binary compiled with clang only uses 1. I installed OpenMP too, mhhh...

>> No.54013423

Hey OP, can you try to make it compatible with pcc? I disabled OpenMP support and tried to run it until I got segmentation fault

tripforce 0.2.0
Copyright (C) 2016 microsounds <https://github.com/microsounds>
Released under the GNU General Public License v3.
Utilizing 1 thread.
TRIP: '!FDg5.anusE' -> PASS: 'vEo4r%:y' @ 25.04 kTrip/s
Segmentation fault

>> No.54013439

>Not secure trip code
>Not GPU based for amd

Aw cmon nigger try harder or at least make arm/MIPS compatible.

>> No.54013445

doesn't omp come with the compiler tho?

I was gonna say add -g -Wall to the makefile CFLAGS, but does pcc even support that?

if this was gcc, valgrind could tell you the very line where it segfaults.

>> No.54013452

>Not secure trip code
It wouldn't be called secure if you could

>> No.54013460

What temps are people getting?

>> No.54013469
File: 378 KB, 1024x1399, transparent.png [View same] [iqdb] [saucenao] [google] [report]

Friendly reminder that the anime tripfags circlejerking over thinly veiled desktop threads are part of the cancer killing this board.

>> No.54013482

You new is showing. Lurk more faggot.

>> No.54013491
File: 294 KB, 1500x1350, blank.png [View same] [iqdb] [saucenao] [google] [report]

Nice job OP.

>> No.54013499

66-68 over here

>> No.54013501

>mfw I cracked it.

>> No.54013515

There are no tripcode explorers on Linux that are GPU based.

>> No.54013545

That's a nice project, How long did it take for you to make it?

Can someone generate the tripcode for "WewLAD" or wewlad or WEWLAD or possibly WeWlAd? Thanks in advance,

>> No.54013553

with the pcc compiled executable, I get
0x00007ffff7a22e6b in fcrypt_body () from /lib64/libcrypto.so.1
when looking at it in gdb.

>> No.54013600
File: 45 KB, 343x1134, file.png [View same] [iqdb] [saucenao] [google] [report]

Xeon E5-2667 V3 with 12 cores under a VM...

>> No.54013604

#[email protected]

>> No.54013620

email me the secure salt pls
[email protected]

>> No.54013622

can someone generate it for pepe and anthony burch?

>> No.54013658

i spent more time rewriting bottlenecks than writing core functionality
i finished most of it in 1 night tho

for CFLAGS, change -lcrypto for -lcrypt
add #include <crypt.h>
and for line 436, replace
char *trip = crypt(password, salt);

it will be much slower, but at least i'll know if it's a problem with your libcrypto

>> No.54013660

TRUMP 2016


>> No.54013678

also, delete line 437
DES_fcrypt(password, salt, trip);

>> No.54013724

Are the generation rates posted in here correct? Because tripcode explorer seems to generate around 3 to 4 million trips a second, even on my low end i3.

>> No.54013753


>> No.54013876
File: 121 KB, 639x436, close.png [View same] [iqdb] [saucenao] [google] [report]


(so close...)

>> No.54014033

i7 4790K here

685.16 kTrip/s


>> No.54014052

This doesn't wear on my CPU quite as much as other trip searching programs do.

>> No.54014096

244.78 kTrip/s on my Core 2 Duo L9400 @ 1.86GHz

>> No.54014124
File: 131 KB, 1284x359, slow.png [View same] [iqdb] [saucenao] [google] [report]

no more than 70C on my T430. Turbo is on. It peaks at 650k Trips/s

>> No.54014129

next time make it utilise the GPU too

>> No.54014162


>> No.54014164
File: 481 KB, 2027x1158, Screen Shot 2016-04-13 at 1.16.58 AM.png [View same] [iqdb] [saucenao] [google] [report]

Am I missing something? This thing is running like dogshit.

>> No.54014179

found the problem
jokes aside, maybe too much threads end slowing down the program. Try with less threads

>> No.54014216

Hi, did you use my OS X fork? Just wondering if it works.

>> No.54014231

akari is molestable desu
got around 6 ktrips on my 2008 celeron 540 kek

>> No.54014239


>> No.54014254

How did you install clang-omp and openssl on OSX?
I'd prefer the build instructions to be reproducible and I don't have a mac to test your changes with.

>> No.54014263

no it just is that slow.

>> No.54014278

Now why would I do that?
I like finding stuff not sharing them.

>> No.54014286


>> No.54014301

thanks akari-chan! *pinches cheeks*

>> No.54014308

And having a tripcode solver means that anyone who has a special snowflake trip can have it broken

>> No.54014318

You should try writing a GPU one too, since you can get some insane speeds from GPUs.

>> No.54014320

This trip looks better

>> No.54014325


>> No.54014334

brew install clang-omp openssl

>> No.54014358
File: 10 KB, 181x171, IMG_6397-1~01~01~01.jpg [View same] [iqdb] [saucenao] [google] [report]

Mfw I can now be called F00tball45

>> No.54014377

No, sorry I didn't use your OS X fork; it was an SSH session.

>> No.54014402

oy vey

>> No.54014409


>> No.54014410

is adding -L flags manually the usual way to link libraries on OSX or it it just a brew thing?
I mean, if all it takes is installing it from brew, I'll merge.

>> No.54014425


>> No.54014434
File: 142 KB, 722x429, Screenshot from 2016-04-12 22:44:46.png [View same] [iqdb] [saucenao] [google] [report]

same here. t440s w/ i5-4300U. hovering around 60 c

>> No.54014445

It's an openssl thing on OS X, check the pull request convo

>> No.54014450

Nicely done, thanks for sharing!

they responded to the PR, and it seems like it. Seems ugly as fuck desu, but not your fault

>> No.54014462

Oh my

>> No.54014471

(low) test

>> No.54014483

(low) test

>> No.54014496
File: 28 KB, 444x396, dakari.jpg [View same] [iqdb] [saucenao] [google] [report]

>> No.54014500

I wanna try writing my own in a different language OP. Do you have any references on the specifications of a tripcode?

>> No.54014502


>> No.54014504

Sweet! Now we can all be tripfags

>> No.54014523

this is not even my final form

>> No.54014524

Look at his code

>> No.54014528

from the code:
* Generating Futaba-style tripcodes:
* 1. Convert password to Shift-JIS
* 2. Only first 8 characters of password are significant
* Generating salt:
* 3a. Take 2nd and 3rd characters and then append "H." to the end
* Clamp salt string to DES character range './0-9A-Za-z'
* 3b. Replace any characters outside of range '.' to 'z' with '.'
* 3c. Replace any characters in ":;<=>[email protected][\]^_`" with "ABCDEFGabcdef"
* Generating tripcode:
* 4. Call DES_fcrypt() (fast traditional DES) with password and generated salt
* 5. Return the last 10 characters

>> No.54014547

So this is essentially passing through characters through a dual-level lookup table and then hashing it?

>> No.54014553

crypt 3 is not a dual-level lookup table

>> No.54014587
File: 13 KB, 848x334, ss0413161202a.png [View same] [iqdb] [saucenao] [google] [report]


>> No.54014607

I'm talking about
> Clamp salt string to DES character range './0-9A-Za-z'
> * 3b. Replace any characters outside of range '.' to 'z' with '.'
> * 3c. Replace any characters in ":;<=>[email protected][\]^_`" with "ABCDEFGabcdef"
Because lookup tables will be faster than a branching for-loop when you're filtering.

>> No.54014636

oy vey

>> No.54014638

>>Because lookup tables will be faster than a branching for-loop when you're filtering.
Is this meant to be a joke or are you actually attempting to optimize something that doesn't even take 0.00001% of processor time?

If you want to bicker about terminology, replacing characters is a much better description for the task than a dual-level lookup table.

>> No.54014645

it compiles on armhf :^)

>> No.54014658

>are you actually attempting to optimize something that doesn't even take 0.00001% of processor time?
Why the fuck not?

>> No.54014662

Way better ways to spend your programming time.

>> No.54014678

Did you have to anything it at all?

trying gmake instead flips shit about an openssl include but I'm pretty sure I have all the dependencies

>> No.54014695

What the fuck

>did you have to do anything to it at all?

>> No.54014706

Sorry, I don't think I can help much. I happened to have the dependencies so `make' just magically worked.

But if it helps to know, I'm on Debian, not Solaris (right?)

>> No.54014712

Kill yourself faggot

>> No.54014720

Thank you akari, daisuki da yo

>> No.54014820
File: 6 KB, 540x182, ss0413161227a.png [View same] [iqdb] [saucenao] [google] [report]

I checked my dependencies again and realized I didn't have libssl_dev, it seems to try now but still eats shit.

I tried just running it through GCC to see what would happen and got this out, but I'm not really sure what I'm doing. I don't mess with C development as much as I'd like to.

>> No.54014828

Hey OS X guy, I cleaned up the makefile a bit, does it work?

It should detect if you're on OS X when you run make or make install.

>> No.54014853

-lssl -lcrypto

>> No.54014921
File: 17 KB, 652x384, ss0413161241a.png [View same] [iqdb] [saucenao] [google] [report]

Neato, that's more like it.

Thanks, my new favorite tripfriend.

>> No.54015042
File: 40 KB, 600x515, noice.jpg [View same] [iqdb] [saucenao] [google] [report]

You did good, Anon

>> No.54015076

>314 kTrips/s
OpenCL version where? You know, today bruteforcing on a CPU is nothing.

>> No.54015087

>OpenCL version where?
on your github soon, i'm sure

>> No.54015089


>> No.54015109

TRIP: '!jd9test/G2' -> PASS: 'HJoU^La%' @ 863.98 kTrip/s
TRIP: '!zgh4testN2' -> PASS: 'A[|@E,HE' @ 863.60 kTrip/s

i5 2500k

>> No.54015125

what's your secret m8?
you have literally the best speeds itt

>> No.54015142


No secret. Stock speeds i5 2500k on GA-Z68X-UD3-B3.

Kubuntu as fileserver/kodi.

>> No.54015149

Ungimped CPU

>> No.54015159

TRIP: '!pOvHo8cuck' -> PASS: 'UAQ5|![B' @ 875.01 kTrip/s
TRIP: '!fONUEkcuck' -> PASS: '/-{sY=fg' @ 874.27 kTrip/s
TRIP: '!AYsOSPcuck' -> PASS: '-0*!hPoR' @ 874.27 kTrip/s
TRIP: '!RWRcuckGEI' -> PASS: 'dxV]Xegs' @ 873.00 kTrip/s
TRIP: '!2XQafLcuck' -> PASS: ''28|G*$(' @ 873.00 kTrip/s
TRIP: '!zvcuck5Yws' -> PASS: 'Uc7z<N)5' @ 875.22 kTrip/s
TRIP: '!ecuckk8Y8w' -> PASS: '(lB1^84S' @ 874.47 kTrip/s

Goes even higher searching for cucks!

>> No.54015191

could you run tripcode explorer in wine and compare your speed?

>> No.54015220


Then i'd have to move my ass over to my tv.

But alright, give me a few secs. Going to test the tripcode program on my WS/GAMING RIG!!1one first.

Tripcode Explorer ran at 21M on my WS though.

>> No.54015233

Scratch that.
Have to download 700mb of shit, which goes against my feels right now as i'm busy being a pirate (12mbit connection).

>> No.54015278

Either way, OP's program seems to be pretty slow compared to explorer then.

>> No.54015283
File: 85 KB, 1024x683, Resize.jpg [View same] [iqdb] [saucenao] [google] [report]


>> No.54015291

>using a computer

>> No.54015330

Get raped and kill yourself, you retarded fucking faggot sack of shit with down syndrome.

>> No.54015342

VM on my WS (i7 5820K)
12 threads ate dogshit at 370k/s roughly.
6 threads: 480k/s (so close to blazeit)
4 threads: 938k/s
2 threads: 604k/s

I see a slight pattern forming around 4 threads.

>> No.54015349

wow great b8 m8, tripcode explorer pumps out 14m trips a sec for me

>> No.54015360


>> No.54015362

i5-4440 @ stock
W10 with Explorer gave me ~11.4m/sec
I don't have the insider build installed so I don't have the linux subsystem installed, but based on benchmarks, you're running at about 2%

>> No.54015364

>all these angry impotent /v/irgin retards who have never contributed jack to the board besides shitty blogs/baitposts getting this bootyblasted

>> No.54015372

I get around 500kTrip/s on my i5 ivy bridge laptop

>> No.54015393

Also, seems like the passwords are wrong.

>> No.54015397

Just finished my own quick and dirty java implementation.

Thread Count:4
WNMcUz.:sK8AakaRiM Count:632578 TPS:632578 Runtime:0
4cgZ:AkArIGo4MI Count:1726850 TPS:863425 Runtime:2
i4OSV:6VWInAKaRI Count:5559085 TPS:926514 Runtime:6
uGcGtxX:n6XqakARiQ Count:12795900 TPS:913992 Runtime:14
RU55:1RiZAKAris Count:12852964 TPS:918068 Runtime:14
y73V:wr0aKaRiFE Count:13971390 TPS:873211 Runtime:16
U.nIp:B/JaKarIDU Count:15311354 TPS:900667 Runtime:17
ksIi0Ij:PyAkARIk0g Count:15815729 TPS:878651 Runtime:18
7MoC:hGy87akarI Count:17279577 TPS:909451 Runtime:19
g90a4RR:O24i0akaRI Count:19242605 TPS:874663 Runtime:22

>> No.54015418

I should really have tested if this works

>> No.54015454

> CPU-based
thats pretty useless to be honest

>> No.54015500

i have an FX8320 @ 3.9GHz
defaults (8 threads) gets around 210 kTrip/s
forced to 7 threads gets around 220kTrip/s
forced to 4 threads with affinity locking them to even-numbered cores gets around 400 kTrip/s
so that's interesting

>> No.54015561

>win 7
>Ubuntu 14.04 LTS
>1 core 2 threads
>7.5 kTrips/s at best
no, im not installing gentoo, thx

>> No.54015592

also, it runs faster as a low priority task
default: ~340kT/s
SCHED_IDLEPRIO: ~400-430kT/s

>> No.54015602

Would you mind sharing it?

>> No.54015648
File: 413 KB, 800x1083, 1344290243694.jpg [View same] [iqdb] [saucenao] [google] [report]


This, I want to cuddle Akari.

>> No.54015791

I just tried porting the Crypt I was using to Aparapi so it's like half broken now lol.

>> No.54015799 [DELETED] 

same setup but kali instead of ubuntu: 45 kTrip/s

>> No.54015809

why are all of you posting pictures of empty rooms or a white/transparent background? Is this a new meme I somehow missed?

>> No.54015812
File: 90 KB, 255x255, 1460246745045.png [View same] [iqdb] [saucenao] [google] [report]


He Guys, i know it's offtopic, but would you mind sharing your akari collection with me ?
I feel like i'm missing those.

>> No.54015830

I wanna put her on my lap and teach her programming while touching her no-nos

>> No.54015844

went up to 8 kTrip/s
same setup but kali instead of ubuntu: ~50 kTrip/s

>> No.54015845

Neat as a hobby/personal project, but doesn't hold a candle to GPU based searchers like MTY or that one japanese dude's new one (that was up to 40-50% faster than MTY). For reference, on my 2011 era GPUs I'm able to get 800Mt/s with MTY.

>> No.54015849

Empty room bullying is the spiciest meme lately.

>> No.54015879

All anime posters, avatar faggots and tripfags should pretty much kill themselves. No wonder this place is shit with all of you shitposting here.

>> No.54015882

No worries m8. Might try porting myself then

>> No.54015891
File: 299 KB, 942x848, 1452566691410.png [View same] [iqdb] [saucenao] [google] [report]

here's one for you

>> No.54015898
File: 78 KB, 640x480, 1443639663140.jpg [View same] [iqdb] [saucenao] [google] [report]

>> No.54015902

I'll try it on my i7-2640QM when I get home in an hour.

>> No.54015911

what a luck that we have our quality 16GiB, pro & contra win10 shill, qbittorrent stalling threads
every. single. day.

>> No.54015922

I'm literally retarded and I don't know how to compile. Can someone do it and upload to Mega?

>> No.54015959

This is a cool program, but can you guys please chill with posting empty spaces in the image box?

>> No.54016044

go get MinGW and figure it out for yourself you lazy shit

>> No.54016124

I re-wrote some/most of my code real quick, it's by no means good but....
Depends upon (I made some changes to make it faster as well):

>> No.54016162

I'm on it but this crap is a pain in the ass which proves that the C language is shit, just like Linux
>covering the fact that you cant do it either

>> No.54016177


>> No.54016223
File: 148 KB, 418x310, cute anime pic 0627.png [View same] [iqdb] [saucenao] [google] [report]

I just rewrote the PRNG and got a 10% performance improvement.
Now I just need to rewrite it as a reentrant PRNG function to accommodate 4+ threads fighting over the same seed value.

>> No.54016341

Tell us when to clone, captain

>> No.54016969
File: 474 KB, 1500x2050, mweeeeeeeee.png [View same] [iqdb] [saucenao] [google] [report]

Version 0.3.0 is out!

Now 10% faster and no longer subject to rand() seed race conditions, especially on windows.

>> No.54017019

Performance improved from 248.37 kTrip/s max to 255.76 kTrip/s on my Core 2 Duo.

>> No.54017025

I don't know what you did, but it was a miracle.

[email protected] /m/D/t/s/tripforce> ./tripforce -i pride
tripforce 0.2.0
Copyright (C) 2016 microsounds <https://github.com/microsounds>
Released under the GNU General Public License v3.
Utilizing 8 threads.
TRIP: '!PriDEh7m66' -> PASS: 'Gm jjE?<' @ 192.49 kTrip/s
TRIP: '!dGpridEk4w' -> PASS: '>W$eqt/G' @ 192.62 kTrip/s
TRIP: '!ZzIaiPrIdE' -> PASS: 'IUc%xT84' @ 192.55 kTrip/s
TRIP: '!cmtLlprIDE' -> PASS: '?1E_Oi$z' @ 192.20 kTrip/s
TRIP: '!priDESv1tk' -> PASS: ' U'gz&R^' @ 192.51 kTrip/s
TRIP: '!ENAqfPRidE' -> PASS: '65KhvW$N' @ 192.71 kTrip/s

[email protected] /m/D/t/s/tripforce> ./tripforce -i pride
tripforce 0.3.0
Copyright (C) 2016 microsounds <https://github.com/microsounds>
Released under the GNU General Public License v3.
Utilizing 8 threads.
TRIP: '!Sv/GyPRIDE' -> PASS: 'Ur& *z$R' @ 1.10 mTrip/s
TRIP: '!tCipRiDeuc' -> PASS: '`g |(7*H' @ 990.64 kTrip/s
TRIP: '!pRide3EPo6' -> PASS: 'Ytyz+D/'' @ 1.32 mTrip/s
TRIP: '!APrIDeUpKM' -> PASS: 'iB7P-[NH' @ 1.45 mTrip/s
TRIP: '!qmVZUpRIdE' -> PASS: '-O5?:o1G' @ 1.45 mTrip/s
TRIP: '!GlCb3PridE' -> PASS: '1U`6;][email protected]' @ 1.45 mTrip/s

AMD 8320

>> No.54017032

>10% faster
Call me when it's 2000% faster and on par with other CPU tripcode finders

>> No.54017041

Apparently the race condition gives more than a 10% boost on multi cores >>54017025

>> No.54017201

she's in middle school you degenerate

>> No.54017315

You're right, if only she was younger

>> No.54017349

Is there yet a tripcode finder that uses bitslice crypt implementation?

>> No.54017400

>not cuda
fuck off to >>>/b/ and end yourself

>> No.54017423


>> No.54017438


>> No.54017442

oh lad, openssl is impossible to compile on windows
>no wonder its written by linux fans

>> No.54017448

>not opencl
How's those freedoms on your desktop, cuckold.

>> No.54017536

ahh I wanna try this out but I'm compiling stuff ;_;

>> No.54017564

thanks based anon

>> No.54017576
File: 240 KB, 760x1057, Screenshot from 2016-04-13 15-20-52.png [View same] [iqdb] [saucenao] [google] [report]

Pretty sugoi. i5-6600K 4.2GHz

>> No.54017639

You guys realise there are tons of good tripcode searchers already existing, and honestly ones better than this yeah?

This one has its strengths in being lightweight, obviously CPU based and native to Linux, but there are ones out there that will make well good use of an AMD GPU and crack trips faster than you've seen in your life, already existing that you can go download right now and have been able to for years.

>> No.54017665

name? link? i'm too lazy for this shit

>> No.54017688
File: 115 KB, 1050x741, akari2.jpg [View same] [iqdb] [saucenao] [google] [report]

You should use pics of the better Akari

>> No.54017692

but why

>> No.54017694

don't thank me, i'm failing. this shit is impossible to achieve

>> No.54017708

There's a shitload of Japanese ones from 2ch, but the only have on hand is http://harry.lu/mty/

I'm sure you could find more on the archives of /g/ and googling

>> No.54017720


>> No.54017736


>> No.54017764

what is love

>> No.54017768

here http://desktopthread.com/tripcode.php

>> No.54017784



took me what? 3 hours?

>> No.54017819

oh and its single threaded for some reason btw

>> No.54017820
File: 1.12 MB, 1600x900, hmmmm.png [View same] [iqdb] [saucenao] [google] [report]


so slow desu

>> No.54017840

how does this work

>> No.54017886

Yeah OP yours is really slow. This one counts in Mtrip/s. Even just on the CPU I get 35.327Mtrip/s, and the GPU adds another 102.375 Mtrip/s.

>> No.54017920

it's literally a single 12k source file, and it even explains what it does in a comment block near the beginning

>> No.54017924


>> No.54017936


>> No.54017966
File: 26 KB, 640x460, 295395713.jpg [View same] [iqdb] [saucenao] [google] [report]

ching chong !

>> No.54018013

Pretty much the entirety of tripcode generation is in crypt3 and the 12k file does not mention at all how that works.

>> No.54018067

1.93 Mtrip/s max on i7-4720K OC'd to 3.5Ghz
66 degrees was max I saw

>> No.54018505
File: 53 KB, 500x647, 52701656_p0.jpg [View same] [iqdb] [saucenao] [google] [report]

You probably won't end up reading this but for anyone else who wants sum fuk:

>> No.54018581


>4 cores

Don't you mean 4 threads?

>> No.54018597

When you say want sum fuk, you mean she wants sum fuk in this archive, right?

>> No.54018624

Get your mind out of the gutter, anon.

>> No.54018636

S-so I'm wrong?

>> No.54018656

Could always download it and see.

>> No.54019045
File: 302 KB, 800x800, 1455982244144.jpg [View same] [iqdb] [saucenao] [google] [report]

Don't worry, you made me >>54015812 very happy.

>> No.54019083

Holy fuck your code is verbose. Also OpenMP? Why not use like, I don't know, standard threads?

Or are you thinking of creating a cluster for the sole purpose of cracking tripcodes?

>> No.54019130

Testing this out.

>> No.54019194

I'd upload more but that took nearly two hours on my connection.

>> No.54019499

testing my trip.
Should be !fAGGOtEvZc

>> No.54019538

Holy shit you actually got a GPL license for this

>> No.54019561

Eh... Do you think you buy these at a license store or something?

>> No.54019588

OP, put this on app store, and you'll swim in money.
Add a banner or something inside.

>> No.54019678

has anyone found !Fish2Mlhz2 yet?

>> No.54020563

[email protected]:/home/*****/Downloads/tripforce-master# make install
gcc -O3 -ansi -march=native -o tripforce tripforce.c -lcrypto -fopenmp
tripforce.c:7:72: fatal error: openssl/des.h: No such file or directory
#include <openssl/des.h> /* Requires OpenSSL via libssl or libcrypto */
compilation terminated.
makefile:24: recipe for target 'all' failed
make: *** [all] Error 1
[email protected]:/home/*****/Downloads/tripforce-master# apt install openssl
Reading package lists... Done
Building dependency tree
Reading state information... Done
openssl is already the newest version.
0 upgraded, 0 newly installed, 0 to remove and 0 not upgraded.

What do?

>> No.54020574

openssl-devel desu

>> No.54020766
File: 1001 KB, 500x281, thankyou.gif [View same] [iqdb] [saucenao] [google] [report]


>> No.54020887

>i7 4790K here
>685.16 kTrip/s
How? I have an AMD Phenom 9550 (quad core 2.2 Ghz) from 2009 that's getting 643 kTrip/s.

>> No.54020953

op updated it and it's much quicker now than it was when that was posted


>> No.54020961

If you're interested I got a few GPL licenses you can buy for cheap m8 if you pay with bitcoin.

>> No.54021020
File: 42 KB, 609x454, tripcode explorer.jpg [View same] [iqdb] [saucenao] [google] [report]

I don't want to talk it down, OP, the little program you made sure is something nice, but windows user had this for a loooong time now.
>pic related

But anyways thank you, now I can use your tool on commandline, which really is nice.

>> No.54021025

Okay, that makes sense then. I was wondering how the hell my old as processor was doing better than the anons with Xeons.

>> No.54021077


>> No.54021250

Hello everyone, if you're running Gentoo and want to install tripforce via layman I've now included stable and git ebuilds for it. Just add my personal overlay, sync it, and you should be able to emerge tripforce easy-peasy. Please pardon the fucked metadata by the way, I'm working on it.


>> No.54021369

Seriously though, why not share?

>> No.54021415

Post some Akari porn

>> No.54021981

Now if only I actually had a reason to use a tripcode.

>> No.54022010

>Please share the the secure salts that make trips safe.

U wot?

>> No.54022037
File: 14 KB, 238x42, Screen Shot 2016-04-13 at 19.18.29.png [View same] [iqdb] [saucenao] [google] [report]

nigga wat

this is 8 threads using asm AVX. still waiting for an opencl implementation that isn't wangblows only

>> No.54022210
File: 318 KB, 480x270, 1460337813275.gif [View same] [iqdb] [saucenao] [google] [report]

I added build instructions for Gentoo and a precompiled binary Microsoft™ Windows®!

It's checksummed, don't worry.

>> No.54022333

I get 1.16 mt/s with this.
MTY_CL gets 9Mt/s on CPU on a phenom II x6, and 207 Mt/s on a 7990. Not only is your implementation shit, it doesn't even leverage something stupidly faster.

>> No.54022372

MTY_CL has builds for linux.

>> No.54022480

ayyy lmao

>> No.54022534

nice bitcoin miner

>> No.54022574

Rolling for bbbenis.

>> No.54022589

363.23 kTrip/s, redirecting output to a file in /tmp because xfce4-terminal is basically garbage

i3 masterrace (not)

>> No.54022601

This one should work.

>> No.54022678

630 Trip/s for windows? cmon man, gimme more

>> No.54022731

Why don't you try running a search?
I just got 570 kTrips/s on some guy's windows i5.

>> No.54022763

TRIP: '!Llcuck3Tkw' -> PASS: 'brN<>%!:' @ 1.66 mTrip/s
TRIP: '!tscuckMSnc' -> PASS: 'a]5B<zvL' @ 1.77 mTrip/s
TRIP: '!uNxZmYcuck' -> PASS: '4B %^Tm ' @ 1.85 mTrip/s
TRIP: '!AMOYKucuck' -> PASS: '@hKj-!<{' @ 1.85 mTrip/s
TRIP: '!E1Ui9Qcuck' -> PASS: '[email protected]`i{x' @ 1.87 mTrip/s
TRIP: '!ocuckzptMs' -> PASS: 'aMYGHJAD' @ 1.87 mTrip/s
TRIP: '!suGzcuckoo' -> PASS: 'Oh%g&miq' @ 1.86 mTrip/s
TRIP: '!njZGqucuck' -> PASS: 'FlKMF;:0' @ 1.86 mTrip/s

Phenom II X6 1090 at 4GHz.

>> No.54022836


>> No.54022840

MTY_CL requires a rather new graphics card.

>No support:
> All compute 1.0-3.0 cards
> Search may or may not work on these cards, if it does it will do so extremely slowly. On windows they are prone to crash the driver, so i cannot claim they are supported, but some cards might be able to run without crashing, albeit very slowly.

That rules out any pre 2013 Nvidia desktop GPU and any pre 2014 Nvidia mobile GPU.

>> No.54022884

Because it doesn't have any output till it finishes?
Also because if I get ~650 trip/s in random mode at 30% cpu load, and assuming its linear to the load, thats still nowhere decent at 100%

>> No.54022902

>Needs a new card
But that's wrong, AMD cards run it fine. Not my problem nVidia cards can't implement open standards correctly.

>> No.54022981

It's not caching anything and printing everything to stdout as it generates it, terminal simply can't keep up.

On your machine, ram usage would grow by about 40MB/s if you were actually caching it and the terminal still wouldn't catch up.
Or I could just throw out every N results to make it look faster?

>> No.54023141

Doesn't change how some people would have to buy a new graphics card just to brute force tripcodes, and it doesn't change how this implementation would run better for those people than MTY_CL.

>> No.54023158

ooooh im a potato, it doesnt have to be 10 char long
>1.24 Mtrip/s

>> No.54023213

Except even in CPU-only mode MTY_CL runs better. Assuming it's only running on one core on the Phenom II for whatever reason, it'll only scale up to about 6.7 Mt/s, whereas MTL_Cl is running at 9 on CPU alone.

>> No.54023276

What does it do?

>> No.54023360

it generates tripcodes so you can tripfag and attention whore even harder

>> No.54024097

Yeah it works! thanks

>> No.54024422

Hi, I'm that anon. With the new version I'm getting 2.10mTrip/s, but with >>54015089 I get 25mTrip/s + 100mTrip/s on the GPU. On average, with the two combined I get 128mTrip/s. To go along with the 4790k is a GTK 970 meme edition.

>> No.54024518

Does it actually feel faster tho?
Or is it just reporting a faster mtrip rate?

>> No.54024639

869.66 kTrip/s

FX-4300 with your win32 build, process priority set to above normal

>> No.54024816

34 seconds to find a screen full of trips for 'benis' (case insensitive) under tripforce.

It took mty_cl 4 seconds, but at that point it hadn't even finished compiling all its OpenCL crap and only had either my CPU or GPU working, not sure which. So I changed the test to hopefully be more accurate.

I gave them both 120 seconds to find as many case-insensitive benis trips as they could.

tripforce found 62.
mty_cl found 1542.
Reported speeds were 2.04mTrip/s vs 130.188mTrip/s (25.495mTrip/s @ CPU, 104.693mTrip/s @ GPU)

>> No.54024871 [DELETED] 

Something important to note is that MTY_CL uses SHIFT_JIS encoding. I don't know what that means beyond half the characters showing up as crap, let's see if they actually work.

>> No.54024936 [DELETED] 

I'm going to take that as a "no", or else I just have no idea what I'm doing, which is the more likely answer.

>> No.54025006
File: 17 KB, 1243x285, 2016-04-13_224759.png [View same] [iqdb] [saucenao] [google] [report]

Wasting digital ocean credit before it expires.

>> No.54025035


>> No.54025116

if everyone had a funny trip 4chan would be much better
prove me wrong

>> No.54025123

>if we were on twitter it will be a better 4chan
prove me wrong

>> No.54025135


>> No.54025204

During the early days of 4chan, everyone was a trip.
anonomoose culture didn't surface till 2005 or so

>> No.54025439

nostalgiafags are always talking about the good old days, let's bring them back to show them how much we love them

>> No.54025560

this is a great stresstest and a great way to fry your system

>> No.54025611
File: 22 KB, 350x233, 33586913.jpg [View same] [iqdb] [saucenao] [google] [report]

Unless pic related is your heat sink you shouldn't have to worry about frying anything.

>> No.54025682 [DELETED] 

Don't mind me I'm just a humble tripfag

>> No.54025695 [DELETED] 

Don't mind me I'm just a humble tripfag

>> No.54025784

you are awesome!

>> No.54025827 [DELETED] 

Don't mind me I'm just a humble tripfag

>> No.54026296

as a human being, I am already cancer. Yet for some reason human beings are still around...
If the dream of someday being a tripfag on 4chan gives my mostly meaningless life some importance in my eyes(long after said "accomplishment" could be said to have any real meaning), what is it really to you?

>> No.54026301
File: 36 KB, 670x697, tf-p6-vs-sparc.png [View same] [iqdb] [saucenao] [google] [report]

>tfw (entry-level) SPARC iron nearly being matched by a 4 year older ThinkPad with a fucking Pentium II
Something doesn't feel right.

>> No.54026340

go fuck urself

>> No.54026358

What do you expect?

>> No.54026408

only if you live in jewrewsilum, otherwise enjoy your freedom to use whichever platform pleases you most

>> No.54026442

I don't think so, a single UltraSPARC II is pretty weak-ass but it's not that weak.

>> No.54026654
File: 328 KB, 1152x864, Clipboard01.png [View same] [iqdb] [saucenao] [google] [report]

needs multi-threaded performance
learn to scroll down retard

>> No.54026691

Unless you're piping the output of tripforce to another program, you should enter a search string.

>> No.54027403 [DELETED] 

Let's do it, let's #Make/g/GreatAgain by tripfagging it up everywhere.

>> No.54027418

Let's do it, let's #Make/g/GreatAgain by tripfagging it up everywhere.

>> No.54027430

Alright screw you OP your retarded program doesn't trip right.

>TRIP: '!FreBeniSM6' -> PASS: '<p*`M9`e' @ 1.98 mTrip/s

But I got !M2bCchL6yc instead.

>> No.54027483

this should work on the new bash for windows10

>> No.54027499 [DELETED] 


>> No.54027509


>> No.54027518

That's weird. No clue why it doesn't work.

>> No.54027558

Do you realise that tripcodes do not belong on 4chan at all? They should be compromised to the point that no one bothers to use them and the tripcunts fuck off back to their respective normalfag attention whoring social media sites.

The ability to use a tripcode or change posting name should have been removed fucking years ago. No one needs a form of self identification aside moderating, and they have access to tools no one else has so they should be able to post, on the extreme rare occasions they need to, using their own system that isn't part of normal posting.

>> No.54027669

Return /g/ back to the gold old days anon. Alternatively, you can Kiss!tyMa0BooTY XDDDD

>> No.54027684

>No one needs a form of self identification aside moderating
It actually helps in some situations, a good example being OPenheimer on /k/ who has quite a bit of knowledge on nukes and to my knowledge doesn't use his trip when discussing other subjects.

>> No.54027701
File: 363 KB, 590x843, 1458801533631.png [View same] [iqdb] [saucenao] [google] [report]

Any fellow tripfriends want to try my banana cookies recipe?
They're really tasty and easy to make.

>> No.54028012
File: 52 KB, 465x498, 1237917779613a.jpg [View same] [iqdb] [saucenao] [google] [report]


>> No.54028237

I prefer gpu based

>> No.54028628

I don't understand the point of this. What is tripcode

>> No.54028683


>> No.54028715

still doesnt work

problem now traces back to this in pcc compiled executable:

Program received signal SIGSEGV, Segmentation fault.
0x00007ffff781f47c in __mpn_divrem () from /lib64/libc.so.6

>> No.54028805

How do you use the search string in the windows variant?

>> No.54028832

run it from the terminal you baka
it's a command line program

>> No.54029016 [DELETED] 


>> No.54029580

>enabling the tripfags
why would you do that

>> No.54029619

is this faster than tripcode explorer?

>> No.54029811
File: 304 KB, 516x567, Akari43.png [View same] [iqdb] [saucenao] [google] [report]


>> No.54029932

No, it's significantly slower.

>> No.54029946

>someone actually uses OpenMP

>> No.54030708


>> No.54030726

well thats not what it should be

>> No.54030774


>> No.54030789

Does your password have more than 2 spaces in it?

>> No.54030798

should be 0 spaces

>> No.54030802

4chan seems to accept one space so long as it's not the first or final character.

>> No.54030830

thats nice and all but the program didn't show any included spaces. do I just randomly add one somewhere?

>> No.54030853

what's your password senapi

>> No.54030869

whats the chance that the win version is miscalculating or the cmd miscodes a character?

>> No.54030882

I wanna touch akari in her no-no's

>> No.54030907

I wonder if >>54027430 is also using windows

Me too anon, me too

>> No.54030915

just use tripcode explorer instead

>> No.54030927

she's in middle school you degenerate

>> No.54030930


>> No.54030931

i like trying new stuff and i dont care much for trips anyway

>> No.54030966

that was ok, kinda no special characters
>testing "|","@","^"

>> No.54030976


>> No.54030980

testing special characters

>> No.54030984

>testing "&"

>> No.54030997

thats it, thats wrong, confirming with another containing "&"

>> No.54031010

So is is the password that contains the & or the generated output that's triggering it?

>> No.54031025

>not enjoying delicious loli
get with the times grandpa

>> No.54031029

thats wrong too, testing fmG=z4&y from the linux version, should contain "win9"
thze password

>> No.54031105

checking " #406I5&'/ " from this tripcode explorer image

>> No.54031151

Interesting, seems like something's broken. Why would

>TRIP: '!FreBeniSM6' -> PASS: '<p*`M9`e' @ 1.98 mTrip/s

Be !M2bCchL6yc with no &?

>> No.54031223

no idea yet, testing "*" now

>> No.54031255

that was ok, now " ` "

>> No.54031269

retry with direct copy

>> No.54031277

same wrong result, seems like ' & ' and ' ` ' in the passcode breaks the tripcode

>> No.54031286

Good to know, and good to know it's not exclusive to this program as you tested it with trip explorer. I guess something has changed server side.

>> No.54031328

one more test, now from another browser, trip should be !wfAiLhZnTs

>> No.54031810
File: 15 KB, 544x352, file.png [View same] [iqdb] [saucenao] [google] [report]


>> No.54032025

that might be a different issue
BTW if the dev is lurking, I could use some help in self compiling this to winblows, atm im getting conflicting types of "omp_lock_t" and I don't get it why wasn't this an issue beforehand elsewhere? Am I missing an ifdef change?

>> No.54032035

yes it is, can't name yourself moot

>> No.54032121

0/10 it's shit

>> No.54032273

!WewLADMM6g #1T6wi=I^@J
!WewLAD69hs #Yta?ilRj!/
!WEWLADGfOE #[email protected](zy=o

>> No.54032288

What keeps people from using these programs to shitpost "problematic" threadruining tripcunts out of a board?

>> No.54032464

tripfags can just switch their tripcode at the drop of a hat
the only way to stop tripfags is to humiliate them with their personal information them or completely ignore them

>> No.54032653

Thanks guys!

See you next week!

>> No.54032654

>what is secure tripcode

Name (leave empty)
Comment (leave empty)
Password [?]Password used for file deletion.