m00x

16 ms·

241.
by m00x 3y ago
KiCAD is amazing for simple PCBs, but if you're going into 10+ layers and dealing with high-frequency designs, you're going to have a lot of issues.
242.
by m00x 3y ago
Do you have any data to back up that take or are you just spewing FUD for no reason?
243.
by m00x 3y ago
I don't see what is anti-human in these citations. Your interpretation of it doesn't make it fact.
244.
by m00x 3y ago
I've been in e/acc since the beginning, so I'll add a bit of context of how I interpret the ideology. The core belief of e/acc is that we can solve our problems not by austerity, but by growth. Instead of limiting births
245.
by m00x 3y ago
If this is your take of e/acc, you're wildly misunderstanding it. e/acc is about building fast and making changes that benefit humanity, while pushing useless bureaucracy away. Elon Musk himself is a proponent of e/acc.
246.
by m00x 3y ago
That's the solve. We'll never succeed at stopping progress, so the best we can do is make sure the ones who get it first are the "good guys". It's the same as nuclear weapons.
247.
by m00x 3y ago
You'll be better served with a more capable platform. You can either 3d print it, or you can pick up something like the WaveRover from Waveshare. It's a solid metal frame with geared motors and differential drive. For ~$100 you wo
248.
by m00x 3y ago
I wonder what need they're fulfilling here. It's not really more useful than using llamacpp-python. This is not something old Google would even release.
249.
by m00x 3y ago
Google really has fallen off. This is intern level stuff.
250.
by m00x 3y ago
They could have done layoffs like almost every startup and big tech company did in the past few years. They could have gone there with 50 good employees. Gumroad almost went bankrupt then came back from the ashes.
251.
by m00x 3y ago
CogVLM is also really good at image captioning. I've been using it in the past month and it's shown very good results. I'm excited to see how this works in practice though.
252.
by m00x 3y ago
Untrue, but also incorrect to call it an LLM. This is a VLM.
253.
by m00x 3y ago
Sundar is doing to Google what Steve Ballmer did to MSFT.
254.
by m00x 3y ago
That's mostly true. Nvidia also spends a lot of time internally testing their cards before shipping them out. Nothing would be worse than a massive hardware recall. If anything, Google would have an easier time going live sooner than N
255.
by m00x 3y ago
It's pretty fast, just not as nice to use. You need statically defined tensors, and some functions are just not supported (last time I used it).
256.
by m00x 3y ago
We should let Google know.
257.
by m00x 3y ago
Good context, but not a geopolitical crisis, it was an internal civil conflict.
258.
by m00x 3y ago
Because there's not currently a genocide happening in the region? https://en.wikipedia.org/wiki/Rohingya_genocide
259.
by m00x 3y ago
We were heading into an ice age, which would be just as bad for humans. The issue is that we built a bunch of things that depend on the current climate, but the climate has been changing for a long time and will keep changing. We need to ad
260.
by m00x 3y ago
I think it's mostly a case of software not being optimized for RISC-V. The sipeed website indicates that it's much faster if the benchmarks are built with optimized toolchains. I imagine it has x64 or arm7 emulator for most builds
261.
by m00x 3y ago
You don't need a dedicated vector db, you can use pgvector. You could maybe use Cube for euclidean space search, but you're better off using optimized algorithms for embedding space search. https://github.com/pgvec
262.
by m00x 3y ago
It turns a continuous space into a discrete space. You would first do PQ, then do KNN of the new discrete vector. This way you can compress the vocabulary to a fixed size.
263.
by m00x 3y ago
You don't use vector databases independently, you need to input the embedding from a ML model. For your use-case, it should be pretty simple. You could use a CNN and train it, or use YOLO, Deepface, or other face detection algos, then
264.
by m00x 3y ago
Google has tried to get everyone on board with RCS for years https://www.theverge.com/2023/9/21/23883609/google-rcs-messa...
265.
by m00x 3y ago
Apple is such a shitty organization. They make excellent hardware, but everything they do is just an attack on its customers. They want to keep people in their walled garden so bad, that they'll transparently attack attempts to make th
266.
by m00x 3y ago
Fun experimentation
267.
by m00x 3y ago
A lot of citations just use Fox News, Vox, CNN, etc. as fact. Wikipedia isn't very serious about scientific/objective truth.
268.
by m00x 3y ago
Wikipedia, while still extremely useful, has slowly become a hivemind that remove changes if they disagree with them, even if they're backed by good sources. Wikipedia is not a place for opinions and bias, but for facts. OP should shar
269.
by m00x 3y ago
I'm not sure why there aren't more companies supporting ONNX. It's so nice to use if it's supported by the platform/model.
270.
by m00x 3y ago
Ilya signed the letter saying he would resign if Sam wasn't brought back. Looks like he regretted his decision and ultimately got played by the 2 departing board members. Ilya is also not a developer, he's a founder of OpenAI and

More ›