Cracking old recovery CDs for the 9x/2000/XP era shines some light into how companies operated and when concepts came to be in that time:

Packard Bell: An EXE checks that you're running on a Packard Bell machine and reboots if it's not. How do we bypass it? Easy: just fucking delete it. The files to reinstall Windows from scratch come from...
...
C:?
Yup. Turns out Packard Bell was doing the recovery partition thing all the way back to the 9x era, maybe even further. Files aren't even on the restore disc so if your partition table got fucked (pretty common because malware and disk corruption) you were totally fucked and needed to repurchase Windows. (My dad, at the time, only charged at-cost OEM prices for a replacement retail copy. He knew it was dumb so he never sold PB machines.)

Compaq:
Computer check? Nope, remove one line from a BATCH file and it's gone.

Six archives, named "WINA.ZIP" through "WINF.ZIP" (plus one or two extras for OEM software) hold Windows. Problematic? Well... only because they never put the password anywhere so the installer can't install them. (Some interesting on-disc technician-only utils, though!)

Dell:
If not a Dell machine, lock up. Cause? CONFIG.SYS driver masquerading as OAK (the common CD driver) doing the check, then chainloading the real OAK driver. Simple fix: replace the fake driver with the real one.

Issues?

Would I mention this one if there weren't?

Disc is mounted on N:. Subdirectories work, but doing anything in them (a DIR, trying to execute something, trying to view shit in EDIT.COM) kicked you back to the disc root.

Installer couldn't find machine manifest in the MAP folder (it wanted your PC's serial before it'd let you install, to make sure you have the correct recovery disc) so it asked for 12-digit alphanumeric serial. The defined serials in the manifest were something like "02884902-01" or similar (8-2, all numbers) and it couldn't read the file so it couldn't show the right format, nor check for the right type.

Bypassing that issue, trying to do the ACTUAL install process caused nothing to happen... as all BATCHes for install think the CD should be on X:.

Welp.

well that was fun. Now to test on-real-PC behavior, as VBOX and VMWare both don't like the special hardware shit it tries to use. (Why does a textmode GUI need GPU acceleration, COMPAQ?????)

Windows 10 updating, decides it would be cool to install gigabytes of sdk, edge, and other bloatware without asking first, on a metered connection i use for work.

Guys, between you there and those fanboy demons in cupertino, one wants to just shut it all off and return to monke.

Sidenote this, because all of this nonsense started on that crap called Windows 8, which was in the end caused to copy that Unholy crap (sold as gold) that is Apple's range of products. It's a company that sells designs nowadays, like Prada, to say, Jobs era is long gone. Everything related to Apple, Mac, Safari, Development, Gaming, UI/UX, productivity and whatever is a

f***ing Nightmare.

We alreay have a global plague, and Apple exists, we dont' need you too making another catastrophe.

All this said,
Use your goddamn trillions to create your own customizable environment that is stable, fast, and WITHOUT BULLSHIT.

I don't give a mindflying F**k of the blurs, i know how to place them with a shell, if i need those. I want control, the shit i decide is going to happen, to happen fast.
This is of Critical importance, because it defines my productivity. And considering we're all sealed indoors since 2019, i want to get away from my pc asap and live my life, instead of spending time(and money, in this moment of emergency) fixing your F**kfests, or else seeing my pc slowing down to death.

First: IF i want stuff on my pc, I know how to install something, thank you.

Second: You can take it, all your Useless - Bugged as Hell - Nonsensical - and of no practical use Bloatware, and shove it deep in your Backdoor.
I'll debloat my pc with batches again, and there's nothing you can do to stop me doing that at every update you force me into.

So please, stop wasting my time, and yours.

I'm beginning to feel like any kind of specific approximation via neural networks is a myth. That if you can't reduce output to simple categorical values that can be broadly interpreted between two points, that it doesn't work.

I have some questions and they don't seem to be getting answered about the design of the net. How many layers should I use ? How many neurons per layer ? How does this relate to the number of desired quantitive scalar outputs I'm looking to create, even if they are normalized, they can vary GREATLY and will if I'm approximating the out of several mathematical expressions. Based on this and the expected error ranges of these numbers and how many possible major digits could be produced within the domain of the variable inputs being introduced, how many neurons per layer ? What does having more layers do ? In pytorch there don't seem to be a lot of layer types per say, but there are a crap ton of activation functions, and should I just be using these at the tail end or should they actually be inserted between layers so the input of the next layer passes through another series of actiavtion functions ? what does this do to the range of output ?

do I need to be a mathematician to do this ?

remembered successes removed quantifiable scalars entirely from output, meaning that I could interpret successful results from ranges of decimal points.

but i've had no success with actual multi variable regression as of yet, even when those input variables are only 2 and on limited value ranges eg [0,100] and [0, 2pi]

and then there are training epochs to avoid overfitting, and reasonable expectation of batches till quality results will start to form.