Avatar

ArkyonVeil

ArkyonVeil@lemmy.world
Joined
2 posts • 19 comments
Direct message

Can confirm. I just upgraded to 115, and tried out my own extension Obliterate Curves, which is similarly not monitored by mozilla due to how tiny it is. If the current domain is a “Quarantined Domain.”, all extensions which aren’t monitored will get downright disabled.

Do note, the list was empty by default. 100% troubling but hard to say where they’ll go with it. Might end up as a “tick this website as secure” box later, though I’d personally prefer control over which sites an extension is allowed to run in.

permalink
report
reply

For me, it’s a simple ordeal. I don’t mind paying so long as the product on offer is worth the cost of payment.

Adobe’s pricing model is abusive, so I went with Affinity which is much cheaper and not a subscription. Zynamptic’s Morph sounds sweet and is reasonably price, but it comes bundled with a driver based DRM. So I got it for free without the DRM bollocks.

With games I used to pirate, but games nowadays are dime a dozen. If it looks interesting, I might try out a demo. If the game is shite, refund which is the loudest review you have. Piracy generally isn’t worth the risk for software entertainment in my eyes, yours may differ.

The only thing I still consistently pirate is movies, and that’s because they all have DRM up to where the sun doesn’t shine. I want to support creators, to help fund what they create. But if I have to pay to have what I bought held for ransom. I’d rather have it for free and forever mine.

To my memory the only movies I have bought were DVDs, the movie “Ink” (check it out on GOG, it’s DRM free and its a pretty cool indie movie) and helped fund a S.T.A.L.K.E.R short film on kickstarter.


To wrap it up, Gaben was right. It’s all about the product/service, its cost (not just price, but ease of access, DRM if any, risks, etc) and what it offers the consumers.

If I pay for a license which can be taken away at any time, that is one cost. If I can get the same thing for free and forever, but with the minor risk that it can be bundled with malware, that is another. With how bloated pricing models are and the constant DRM abomination that are forced into everything, it’s no surprise Piracy is still alive and well today.

permalink
report
reply

Thanks for reposting the breakthroughs!

Makes me have to visit Reddit less for news.

It even rhymes, how neat is that.

permalink
report
reply

It would be absolutely awesome, with infinite context length that would mean a much greater ease when it comes to handling models. I can be lazy and instead of creating a LORA, just use an entire book’s style as a reference right there in the prompt.

For programmers, just dump the entire codebase, or Documentation.

Of course, all this is only possible if VRAM is less of a bottleneck than it currently is, as well as the fact that it can reliably reference information on an arbitrarily large context. (Not much use having huge context if performance degrades, it loses its marbles or forgets key pieces of information along the way)

permalink
report
parent
reply

I’ve been here for a couple of days, if anything, I’m pleasantly surprised. This is not Reddit’s first clusterfuck, and although there were alternatives at the time. It’s finally nice to find something that isn’t full of bigots.

permalink
report
parent
reply

I’m actually playing around with EXLlama, IIRC it works with pretty much every model, and it can be a real game changer specially for long conversations, code, or stories.

Unfortunately there is still the unavoidable problem of the context length burning VRAM like no tomorrow. You either get a decent AI with the attention span of a gold fish or an idiot AI which can remember 3 times as much stuff as before.

Handy, progress, but ultimately there is still ground to cover.

permalink
report
reply

No, I salute you! Assembly is miserable, I count myself lucky that all I needed to do is bust out a sheet with the opcodes to figure out what they did. If I had to actually write Assembly instead of just swapping opcodes, this project would be over in an instant.

permalink
report
parent
reply

I may make it sound easy, but when I did this, I already had 8 years worth of C# experience. I definitely recommend that you learn how to program first, then once you know what you’re doing. Read up on some tutorials on reverse-engineering. General practices, that sort of thing.

When you’re ready, pick the right tool for the job. If you can decompile to get the code out (like dnspy for example in .NET programs), that’s much easier, other languages may require other tools, C++ will likely require hardcore programs such as x64dbg, and you don’t wanna touch that until you can understand Assembly to at least a passable degree.

TL DR: It’s hard, but there is a path there, don’t eat more than you can chew.

permalink
report
parent
reply

There’s always a path that leads to where we need to go. For me, I kind of started to get interested when I watched this Excellent video by Exilelord (He did something way harder which was fixing a bug and later adding outright features in a AAA game obfuscated by SECUROM).

My first rodeo was probably cracking the level security of Synthriders, that one wasn’t obfuscated and was modern so it took me about 1-2 hours to get it to spit out the password for the level files.

Then… the only logical way up was cracking an actual game.

permalink
report
parent
reply

Hey! I thought I couldn’t do it until I tried. :3

Sidenote: Background in programming or computer science may be required.

permalink
report
parent
reply