yosefk

11 ms·

121.
Comments on the FTC antitrust investigation of Google(danluu.com)
36 points by yosefk 2y ago|1 comments
122.
by yosefk 2y ago
By going full Vulkan, you mean do nothing at all in Qt, or do you mean a particular way to use Qt but not some parts of it?
123.
by yosefk 2y ago
I do not envy a non-programmer stuck with buggy code generated by an LLM.
124.
by yosefk 2y ago
For inference, Nvidia has DLA since 2017-ish if I remember correctly, which is completely separate from the GPU.
125.
by yosefk 2y ago
35% of vulnerabilities in the biggest tech companies being due to use after free bugs are a part of the answer. (More than 90% of severe vulnerabilities are due to memory bugs impossible in memory-safe languages.)
126.
by yosefk 2y ago
You need to modify base pointer in this example, specifically to point 2 bytes before the current base pointer (moving back from the first red pixel value to the first blue value.) I don't think you can do it without ctypes, maybe I&#x
127.
by yosefk 2y ago
The pygame API gives you a 3D RGB array and a separate 2D alpha array, which happen to reference the same RGBA/BGRA memory. Are you suggesting that OpenCV Python bindings (in most of their functions, and likewise other libraries includ
128.
by yosefk 2y ago
I dunno if it's "more than that" since they're not directly comparable? What you describe sounds more like numba or Cython than what TFA describes which is a different use case?..
129.
by yosefk 2y ago
In TFA I wrote that you can remain lucky for a while if you are managing a good team of a double digit size, among some other situations. Luck runs out much more quickly at triple digits. That said, even at single digits, it's good to
130.
by yosefk 2y ago
I just measured this with the np.ascontiguousarray call (inside the loop of course, if you do it outside the loop when i2 is assigned to, then ofc you don't measure the overhead of np.ascontiguousarray) and the runtime is about the sa
131.
by yosefk 2y ago
25 line pull request doing what? Supporting this format efficiently is probably more LOC and unlikely to get merged as few need this and it complicates the code. Doing the thing I did in Python inside OpenCV C++ code instead (reinterpreting
132.
by yosefk 2y ago
What's the max number of people you managed?
133.
by yosefk 2y ago
The question is how much new numeric code is written in Fortran vs C/C++. My guess is way below 10%, certainly if we measure by usage as opposed to LOC and I would guess by LOC as well. Is Fortran legitimately faster than C with the re
134.
by yosefk 2y ago
While this is a valid point, cv2.resize doesn't actually implement color conversion in a way addressing this issue; at least in my testing I get identical results whether I interpret the data as RGB or BGR. So if you want to use cv2.re
135.
by yosefk 2y ago
You can absolutely get direct unbounded access to memory with ctypes, with all the bugs that come from this. I just think/hope the code I show in TFA happens to have no such bugs.
136.
by yosefk 2y ago
You... probably shouldn't, but, very interesting stuff in that link! I didn't know CPython had no bounds checks in the load_const bytecode op
137.
by yosefk 2y ago
I think you'll have a problem with BGR data absent ctypes, since you have a numpy array with the base pointing at the first red channel value, and you want an array with the base pointing 2 bytes before the first red channel value. T
138.
by yosefk 2y ago
Agreed, realistically you're going to either convert the data or give up on using the function rather than reimplementing it to support another data layout efficiently; while converting the code instead of the data, so to speak, is mor
139.
by yosefk 2y ago
From everything I'm seeing, it follows that Matlab & Fortran have been decimated by Python and C/C++ around 1990s and 2010s, respectively. Of course I could be wrong; any evidence of their still strong position will be greatly
140.
by yosefk 2y ago
I dunno if "close enough", but there's no C code in TFA, only Python. ctypes - guilty as charged
141.
by yosefk 2y ago
OTOH when it writes buggy code, you tend to get decent error messages when trying to use it. In this sense Rust + LLM is way better than C++ + LLM
142.
by yosefk 2y ago
I've read Williams' book, and I've studied animation and done some, though I wouldn't call myself an animator yet. I hope to avoid building rigs because they're, well, rigs. Much nicer and more flexible to control t
143.
by yosefk 2y ago
If an AI can't make animators more productive, it's as close to over for hand drawn animation as it has been for the last decade
144.
by yosefk 2y ago
Post author here - would be very interesting to hear more of your thoughts on this! It's not easy to find a pro animator willing to consider the question given the current level reached by AI methods
145.
by yosefk 2y ago
Almost certainly a good idea. I'm about to start trying things in this direction
146.
by yosefk 2y ago
I doubt you can learn much out of tweened flat cutouts beyond fitting polynomials to data points. The difficulty with full animation is rotation & deformation you can't do at all with cutouts. (Puppet warp/DUIK cutouts are muc
147.
by yosefk 2y ago
They're indeed not diffusion models, though they are trained on animation data as well as specifically designed for it (the raster papers at least.) I'm very hopeful wrt diffusion, though I'm looking at it and it's f
148.
by yosefk 2y ago
The 2nd paper actually uses a 3D dataset, though it explicitly doesn't attempt to handle occlusion beyond detecting it. I sort of hope you can handle occlusion based on learning 2D training data similarly to the video interpolation pap
149.
by yosefk 2y ago
It's counterintuitive but less so considering that it's way easier for a human to draw something from scratch than to inbetween 2 key frames as well! (I guess we're used to machines and people struggling at opposite things so
150.
by yosefk 2y ago
Actually inbetweening is really hard (and I think requires talent) and used to be a big way to learn enough to become a key animator. And I would worry about AI eliminating this learning route if classical animation wasn't struggling t

More ›