AmpBenzScientist @AmpBenzScientist@qoto.org

@PawelK @lupyuen Not a problem. One can't go wrong with alcohol and weapons, they are recession proof.

@PawelK @lupyuen The sure shot with codasip. The applications could range from powerful CPUs to CZUB advanced weapons platforms.

@PawelK @pasture The same Nontando Swatch that got owned by an exploit in the GPU of their tablet?

It should be relatively easy as far as these cracks go. Unfortunately the dual wielding screaming eagle imprisons its own people for Japanese profit. This is going to go down in flames and the company should realize that they need worse hardware and more anti consumer policies to fix the problem.

@PawelK Thanks for the link, I'll definitely look into it. Most of what I look into now involves FPGAs.

What crippled the machine learning project was a little gem called VLIW and the research time needed, CISC is causing problems too. It's not that I can't extract the secrets from them but it's the differences in structure of the binaries that cause so much trouble. It is more powerful than I first thought.

If there's enough demand, I'll finish off different parts and put them up as soon as they are functional. There's too much that I want to share to let it fail now.

@freemo I pretended that it was me somehow traveling that fast and jumping vehicles. It was upsetting to be disturbed because it was me.

Did I mention ADHD and extreme sports since childhood? It's probably implied in the above.

@lupyuen It just occurred to me that you could use the Ideal Gas Law with this information. I would not recommend doing this because it's not particularly useful.

@freemo I used to think about that too. It always seemed like a game in the mind, training to respond faster to obstacles and limitations. I still think of this game but I have to pay attention to the road.

@PawelK I think there's a submarine base in Crimea, a big one built into rock. The US used it for a little while but faults were apparent and it's likely going to collapse in the near future. It's not really vulnerable from the air but who knows what could be done?

My project is delayed without a known release date. Progress is still being made but I underestimated the amount of work required for an important portion. I'm not going to limit the scope to meet a deadline. There will be a release of something but the project as a whole is going to need a redesign.

I might make it compatible as a plugin with other existing software. It's difficult to say what the final version will be like. Parts of the work could be useful in some applications where nothing like it is available. It's not the result I wanted but it's not completely unexpected.

@thor Machine Learning can vary in complexity but it is essentially a self correcting statistical model. I used to think it was magic but it's just hamfisted mathematics like Euler's Method. A machine learning the area under the curve with rectangles.

@thor Even putting the MIDI output through a speaker designed for analog will add character to the sound.

A digital signal is pure and that's a big problem. Guitarists will sometimes use active pickups because of the digital signal which sounds terrible. That pure signal is run through audio processors for different effects and sound. It's a system set up for easy replication. The result is that different guitars sound the same and characteristic nuances are lost. They are impurities and inefficiencies are where the character lies.

@DeveloperMemes Might as well leave this here.

https://youtu.be/O6zslvLklyI