Experts say that is not possible.

I feel better knowing that Chinese military partners get to feel the joy as well. (Curbs can be a bitch sometimes…)

Why not? Those CPUs got perfect scores on Red Star OS.

ur

It was on old 3.5" drives a long time ago, before anything fancy was ever built into the drives. It was in a seriously rough working environment anyway, so we saw a lot of failed drives. If strange experiments didn’t work to get the things working, mainly for lulz, the next option was to see if a sledge hammer would fix the problem. Funny thing… that never worked either.

I used to take failed drives while they were powered on and kinda snap them really with a fast twisting motion in an attempt to get the arm to move or get the platters spinning.

It never worked.

Did you get bad sectors? Weird things can absolutely happen but having sectors marked as bad is on the exceptional side of weird.

Maybe? Bad cables are a thing, so it’s something to be aware of. USB latency, in rare cases, can cause problems but not so much in this application.

I haven’t looked into the exact ways that bad sectors are detected, but it probably hasn’t changed too much over the years. Needless to say, info here is just approximate.

However, marking a sector as bad generally happens at the firmware/controller level. I am guessing that a write is quickly followed by a verification, and if the controller sees an error, it will just remap that particular sector. If HDDs use any kind of parity checks per sector, a write test may not be needed.

Tools like CHKDSK likely step through each sector manually and perform read tests, or just tells the controller to perform whatever test it does on each sector.

OS level interference or bad cables are unlikely to cause the controller to mark a sector as bad, is my point. Now, if bad data gets written to disk because of a bad cable, the controller shouldn’t care. It just sees data and writes data. (That would be rare as well, but possible.)

What you will see is latency. USB can be magnitudes slower than SATA. Buffers and wait states are causing this because of the speed differences. This latency isn’t going to cause physical problems though.

My overall point is that there are several independent software and firmware layers that need to be completely broken for a SATA drive to erroneously mark a sector as bad due to a slow conversion cable. Sure, it could happen and that is why we have software that can attempt to repair bad sectors.

They can be, I suppose. However, the AI libraries that I was tinkering with seemed to all be based around Ubuntu and Nvidia. With Docker, GPU passthrough is much better under Linux and Nvidia.

WSL improved things a bit after I got an older GTX 1650. For my AMD GPU, ROCm support is (was?) garbage under Windows using either Docker or WSL. I don’t remember having much difficulty with Nvidia drivers though… I think there might have been some strange dependency problems I was able to work through though.

AMD GPU passthrough on Windows to Docker containers was a no-go. I remember that fairly clear though.

My apologies. It has been a few months since I messed with this stuff.

I had much better success using WSL, but I haven’t used it or even updated it in a long while. (I have been meaning to see how AMD GPU support has evolved over the last few months. Back in January’ish, AMD support was still bad.)

Anything that even is remotely Linux related is much easier to get working with WSL, btw. Almost all of my personal python stuff is running under it and it works great with VS Code

There is a large collection of poorly written articles/blogs on LinkedIn, actually. They are just bad enough to be good enough for Google.

Strangely enough, LinkedIn is owned by Microsoft. If Microsoft actually let Google use it as a data source, it was to sabotage Google’s AI training.

This was an old feature, before it could be disabled.

So, if my wife was in my contact list first and I was in hers, she would get notified when someone was added to mine.

It was something like, “X has joined Signal on Y’s phone!!” or some bullshit like that.

I was a huge Signal advocate at one time and would try to get everyone to install it and use it. Man, woman or child, I didn’t care who it was. I was worse than a crypto-bro trying to jam BTC down everyone’s throat.

I was chatting with a group of ladies at work and got a few of them to install it. When they did, Signal pushed notifications of them connecting to my wife’s phone.

Needless to say, I got questioned fairly intensely about why there were other girls connecting with me on Signal.

I wasn’t very keen on Signal after that.

would just get frustrated and throw things.

Yeah. Most conservatives I know have short tempers and they get extremely carried away when they are in groups. I mean, it doesn’t matter what they are getting mad at, just as long as everyone is getting pissed off at the same time and about someone else or some other ideology. The key point is they believe they are victims of “the system” or “libruls” and not their own bad choices.

It’s very similar to how Putin is constantly crying about how Russia is a victim of NATO aggression, actually.

Just tell a person that their way of life is being attacked and that you have all the solutions.(Simplified Hermann Goering quote. Allegedly.)

This guy was the real killer. The new RMcD is just the cover story.

Oh cool. I was wondering if it was just a math thing or it was related to hardware used in ternary computers.

What exactly is a half bit? Is a transistor “half open” or something?

Yeah, I would think memory as well due to the screen artifacts in that low res mode. (That depends on how x86 memory is mapped these days, I suppose.)

If parts were generated, copied or acted it seems somewhat irrelevant. They intended to, and did, generate SJh’s likeness, by whatever means, and that is the key point.

But yeah, they probably mixed and matched voice samples to their liking. I wouldn’t doubt that for a second. If actual samples were used in the final product, that would be extremely damning.

OpenAI had an idea of a type of voice they wanted

Yep. It was a very specific voice they wanted.

they hired a voice actor

This made me think a little too, but then I started thinking about how people talk. Even if a person’s tone is similar, the mannerisms are still drastically different. The voice actor had to spoof Scarletts voice well enough to even fool Scarletts friends.

While I haven’t read Altman’s tweets (or tweets from someone else at OpenAI?) personally, rumor has it he knew what he was doing with a specific voice actor. The intent was to spoof her voice. The intent is probably more damning than the actual act, TBH.

To summarize, there are a few nuggets here: They approached SJh first with a specific reason to use her voice; Some asshat bragged about spoofing the voice on Twitter; There was clear intent of generating a likeness of SJh.

(Thinking about the broke actors for a second… /s) Their face and voice are at the core of their career, similar to how company branding is makes a company unique. While I am not a lawyer, it seems there are some parallels with trademark and copyright law here.

“Accidentally” using a voice that sounds like SJh would be a really poor argument now as well.