hinterlufer

joined 2 years ago
[–] [email protected] 31 points 1 week ago (7 children)

Balatro is indie, songs isn't it? Developed by a single dude with probably zero budget

[–] [email protected] 4 points 1 week ago

I think this is a perfect strategy - you can sell code, and if any of it contains issues/bugs/gaping security holes you can just blame your customer for not checking the AI output

[–] [email protected] 2 points 3 weeks ago (4 children)

For length, for an average male one meter is about one large step with extended legs (useful for distances), or the distance between e.g. the left side of your torso to the end of the extended right hand (useful for estimating the length of rope or smth).

For weight, it might be useful that 1 liter (that's 1 dm3 but noone uses that except sometimes in scientific literature) is almost exactly 1 kg, and a typical cup fits 0.25 liter. A shot of alcohol is either 20 or 40 milliliters (0.02 or 0.04 liter) depending on where you are and what you order.

For conversions you just need to remember the base unit (e.g. meter and grams/kilograms) and the decimal prefixes. But you really only need milli (1/1000), centi (1/100) and kilo (1000) in day to day life. Then you simply shift the decimal.

[–] [email protected] 1 points 4 weeks ago (1 children)

In Europe you still need to give way for cars that are crossing straight from the other side when you want to turn left, and take care of pedestrians that also have green on a right hand turn. Granted, not all crossings are like this, but many are.

[–] [email protected] 67 points 4 weeks ago (9 children)

What's wrong honey? You haven't finished your kale cake.

[–] [email protected] 9 points 1 month ago

might also be to teach actually reading the instructions instead of blindly typing pi into the calculator

[–] [email protected] 6 points 1 month ago

I didn't think of that - also for nvim you typically pull plugins from git repositories

[–] [email protected] 1 points 1 month ago

I used it for a bit and it is working quite well with small vaults. But the memory issue is real - by now obsidian always crashes when I try to sync via git on Android as my vault increased in size by quite a bit.

[–] [email protected] 8 points 1 month ago

Not sure what you want to show with that screenshot. It tells you that 700 MB of your installed RAM is reserved for your integrated GPU which doesn't really have to do anything with Windows.

[–] [email protected] 2 points 1 month ago (1 children)

I just didn't plot anything anymore tbh. I originally wanted to make stencils for electro-etching but I realized that I don't really have that much of use for it.

[–] [email protected] 4 points 1 month ago (3 children)

I did it with my ender 3, using a printed bracket to hold the knife. It's a hassle to use and I barely use it because it's such a pita. I managed to make a few nice cutouts though so it's definitely possible. I just wouldn't recommend it.

[–] [email protected] 15 points 2 months ago (1 children)

So, a typical pupil is around 2 mm in diameter in bright conditions. With the Rayleigh limit that results in an angular resolution of 1.22 * 60010^-9 m / 210^-3 m = 3.66*10^-4 rad

At a distance of 5 x 3 mi = 15 mi = 24.1 km this corresponds to a point to point distance of

tan(a/2) = (d/2)/l

d = tan(a/2) * l * 2 = tan(3.66*10^-4) * 24100 * 2 = 8.8 m

So in conclusion, with regular, human-like eyes he could discern points that are at least 8.8 m apart in the best case scenario. Discerning hair color from the color of the clothes would need a much higher resolution, and the horsemen are probably not 10 m apart from each other either. And again, this is a theoretical limit, real-world resolution would probably be significantly lower.

54
submitted 4 months ago* (last edited 4 months ago) by [email protected] to c/[email protected]
 

It seems like with the current progress in ML models, doing OCR should be an easy task. After all, recognizing handwritten numbers was one of the prime benchmarks for image recognition (MNIST was released in 1994).

Yet, when I try to OCR any of my handwritten notes all I ever get is a jumbled mess of nonsense. Am I missing something, is my handwriting really that atrocious or is it the models?

Here's a quick example, a random passage from a scientific article:

I tried EasyOCR, Tesseract, PPOCR and a few online tools. Only PPOCR was able to correctly identify the numbers and the words "J." and "Chem.". The rest is just a random mess of characters.

Edit: thank you all for shitting on my handwriting. That was not asked for, and also not helpful. That sample was intentionally "not nice" but is how I would write a note for myself. (You should see how my notes look like when I don't need to read them again, lol)

chatGPT can transcribe it perfectly, and also works on a slightly larger sample. Deepseek works ok-ish but made some mistakes, and gemini is apparently not available in my country atm. I guess the context awareness is what makes those models better in transcription, and also why I can read it back without problems.

 

So I got my hands on some flower bulbs which are typically meant to be planted in spring and I was wondering what I could do with them now. To be concrete, I have

  • Dahlia
  • Mirabilis jalapa
  • Ixia

I'm in USDA zone 7b/8a and I could either place them on a south facing balcony or inside. I've also read that you can force flowers in a vase with some bulbs such as Hyacinths but I haven't read anything about that with the ones I have. Or should I just keep them in storage until next spring and plant them then?

view more: next ›