How the frick do you do this? It's difficult even with Ida Pro. I cracked the AMD part of the program of the program just by using logic. pseudocode, and decompiling. I still couldn't read the pyarmor but just used hints based on watching how it executed and strung together. Cracking the Nvidia part is a lot more difficult because I can only get to the nvml check even using zluda which I'm not even sure works with python and definitely can't do 32-bit apps. I just load up the pyd of the pyarmor and that's where the .dll injection starts.
>moron Black person examines binaries by opening them with text editors
>Can examine binaries by opening them with text editors
is this prosible? open it in text editor and be like 'oh yeah it will be harder'
As long as the text editor has a unique visual representation for each byte, yeah.
If it just spews out a bunch of Unicode replacement characters � then no
I keep forgetting these special chars actually represent bytes because I'm just so used to using hex instead.
It just went right over my head for some reason.
dumbass
>is this prosible?
not really. you need to be able to see what the frick you're doing, which just proves op is a larping moron.
nah. I use Ida to read it as hex rays, where the injection starts, where it ends, and the imports. I was just posting a generic edit view.
>to read it as hex rays
Wtf is a "hex ray"
>Wtf is a "hex ray"
Hex-rays: This is a company known for producing IDA Pro, which is a very advanced disassembler used primarily for reverse engineering of binary files. The name "Hex-rays" could be mistaken as "hex ray" if heard in passing.
Wrong, Hex Rays are evil magical curses that cute thicc goth girls send to their prey (innocent anons) in order to acquire their pure virgin semen remotely.
And yet he mentioned IDA pro?
The frick am I reading?
Is he unable to covert it to hex and assembly output?
It's literally "file --> open file" from memory.
>reads the rest of the post
>the picture is just irrelevant
>"Cracking the Nvidia part is a lot more difficult because I can only get to the nvml check even using zluda which I'm not even sure works with python and definitely can't do 32-bit apps. I just load up the pyd of the pyarmor and that's where the .dll injection starts."
The frick is OP cooking?
Not sure if bot.
Ignore this post, I am a huge homosexual and misinterpreted the thread because OP was a little unclear at first.
For some reason the way he put it out just didn't click.
Zluda uses Cuda. I program AMD dropped because they didn't find the usefulness of cuda. Now they're lagging heavily in AI because of it.Nvidia hates zluda. They put recent protection from their programs being executed with zluda last month. Any program before that is fine. I got the "Cuda acceleration Enabled" with the AMD program. That's cool in itself so I thought why not try it on Nvidia but has checks on it. Also, sorry dude. I'm new to this but surprised with some of the success I've had using IDA. Did not know that was an option. Definitely will make it a lot easier.
>They put recent protection from their programs being executed with zluda last month.
They added that language two years ago and some homosexual online pretended it was a new addition because of zluda
I don't think that's true. Anything before last month I can get with zluda unless they decided to remove that protection at some point or zluda has just made quite a bit of progress since then.
ror
rmao
>Line 13
>BS NUL ACK
>ACK
>t. moron who doesn't know ASCII
Much of this proprietary firmware needs actual hardware to talk to.
so not even a spoofer would load it up?
>"A spoofer"
It's called hardware virtualization, tard
I don't want to use the program but just get to the login process so ida reads up to that point.
If you dont want to run the program then your doing static analysis
thanks for the proper term. I looked up the definition. I'm just getting into this and surprised I was able to bypass the login on AMD so easily
Well, microcode is designed to be invisible to the OS entirely, ring -2 kinda shit. There was a cool video recently posted on a defcon speaker exposing CPU secrets. I am almost certain that same tech is included in GPU hardware as well.
Maybe something from this webpage will lead you to an answer or to information that will help.
https://github.com/xoreaxeaxeax?tab=repositories
Good luck, I am not on your level, but I notice things. ;3
>recently posted
it's been years anon, would recommend checking out all his other talks though he does pretty cool shit.
link?
You have RE talent.
>randomly screenshots
>posts
>doesn't say anything
why
>bypass the login on AMD
is this an llm just using tech words?
I don't think it fits the LLM definition but it is a login to a neural network. Anyway, he scams kids out of $25. Promises to offer lifetime support and help. I rarely see him in troubleshooting. I've spent time helping kids who would otherwise not be able to use the program, use it. They don't understand simple python debugging commands that they can just google to fix. A lot of them will never get it running and they just lost $25. He'll occasionally say "pip install ultralytics" and that's as far as I've seen him go. Btw, they already have that because it's in the install_requirements.bat
it's a free neural network btw. Nothing special. I went to the IP. He likely just stole someone else's code and put pyarmor over the code.
>How
Over in /cyb/ we also cover reverse engineering, and there is some pasta there that can help you.
BTW I would start with strings before a text editor.
use a hex editor