nsajko

14 ms·

121.
by nsajko 3y ago
What do you mean, the photos seem fine to me?
122.
by nsajko 3y ago
This is still aimed at megacorps, not at your average hacker without a semiconductor fab handy.
123.
by nsajko 3y ago
Wikipedia page: https://en.wikipedia.org/wiki/Infinite_product Infinite product expressions for the trigonometric and hyperbolic functions: https://en.wikipedia.org/wiki/List_of_trigonometric_ident
124.
by nsajko 3y ago
I think GCC also supports this for a long time already, see -fplan9-extensions and -fms-extensions
125.
by nsajko 3y ago
The "Julia that can compile" characterization is misleading, because Julia is based on LLVM-backed compilation (a Julia interpreter exists as a user package, however it's used only for debugging). I'd restate that as &
126.
by nsajko 3y ago
Doesn't seem like a general purpose programming language: > Lisp-Stat is a domain specific language (DSL) for statistical analysis and machine learning. It is targeted at statistics practitioners with little or no experience in prog
127.
by nsajko 3y ago
Who's forking Windows 7, though?
128.
by nsajko 3y ago
That's just one possible generalization of the (three-dimensional) cross product. And not a commonly used one, I think?
129.
by nsajko 3y ago
You're both correct. In the context of Julia, where compilation and run time are arbitrarily interleaved, something like re2c can be a library and used at run time.
130.
by nsajko 3y ago
IMO, the Summary of Julia Plotting Packages article undersells Vega/Vega Lite. They're really powerful and look good.
131.
by nsajko 3y ago
Apart from what sibling said, another issue is explosion of trust assumptions. On my system, I obviously have to trust the Archlinux project not to serve me malware. When using Julia, I additionally have to trust the Julia project not to se
132.
by nsajko 3y ago
> since Julia software is not really supposed to be running public facing services the security argument is moot anyways 1. both Julialang.org and Juliahub.com run Julia as far as I understand 2. even if your premise were true, it wouldn
133.
by nsajko 3y ago
Tangential: No Grammar to Rule Them All: A Survey of JSON-style DSLs for Visualization https://arxiv.org/abs/2207.07998 Both Vega and Echarts are covered, among others.
134.
by nsajko 3y ago
One annoying thing about trying to install Makie (although it would be silly to blame Makie for it) is that installing it pulls a ton of compiled C/C++ binary dependencies. I'm talking Fontconfig, Xorg(!?), Cairo, a bunch of aud
135.
by nsajko 3y ago
To clarify the above, precompilation usually only happens when installing or upgrading the packages.
136.
by nsajko 3y ago
> RHEL clearly does not care about the GPL You're obviously confused as to the terms of the license. Red Hat seems to be fully compliant.
137.
by nsajko 3y ago
> One core challenge we aimed to solve for memory on the Framework Laptop 16 was being able to offer the same modules for pre-built systems, DIY Edition, and the Framework Marketplace. Historically, we needed to source separate “OEM” mod
138.
by nsajko 3y ago
One of the big issues with Wayland is the (intentional) fragmentation. Each "Wayland compositor" is really its own window system, with its own peculiarities. This forces applications that would have been trivial to implement on X
139.
by nsajko 3y ago
X is suboptimal, but Wayland is way worse.
140.
by nsajko 3y ago
To add insult to injury, the first sentence is false.
141.
by nsajko 3y ago
Yeah, it's crazy how Wayland is being pushed as a replacement for X.
142.
by nsajko 3y ago
Damore did nothing of the sort.
143.
by nsajko 3y ago
A good reference is the Handbook of Floating-Point Arithmetic. If you can be more specific about what is it that's not clear, I'll give more specific "proof". Even though you're being a bit "sealion"-y. &g
144.
by nsajko 3y ago
Take an interval, for sine it might be, for example [0,π/4], I think it was [0,π] in the video (a mistake, most probably). You're interested in the maximum approximation error over the interval, so you could, for example, sample t
145.
by nsajko 3y ago
True, the author of the video should probably have done even more range reduction.
146.
by nsajko 3y ago
I'm not claiming anything extraordinary, though, most or all of what I said is well known in the field.
147.
by nsajko 3y ago
> I guess it's a fundamental difference between RLIBM and your algorithm---you don't guarantee correctly rounded results, right? Yes, but note that the linked package doesn't even attempt to implement a polynomial approx
148.
by nsajko 3y ago
Yeah, turns out I missed that part of the video. So the author, in fact, knew that minimax approximation is a thing, however he only tried to do it by hand ! It's a bit silly to try to optimize a polynomial for minimizing the maximum
149.
by nsajko 3y ago
One difference is that their approach doesn't scale to 64-bit floats AFAIK. My code and their approach are actually quite different, with different goals, different places where they're smart and different places where they could
150.
by nsajko 3y ago
The video is very well produced and engaging, however (from skimming it), it seems that all the implementation methods it presents are suboptimal. The N64 CPU had an FPU, so it seems obvious that the best method, from both a speed and accur

More ›