claude boosted

What is best from the audience point of view?

Worked a bit on today. Saving one addition in the perturbation inner loop makes take 6% less CPU time in that phase. Doing some more for the derivatives saves 17% overall. Other formulas show higher savings, but it's slow doing these by hand (KF has around 100 formulas, today I optimized 5 of them...).

code.mathr.co.uk/kalles-frakta

claude boosted

Hi guys! Programming question:

In C, is an array _exactly_ the same thing as a pointer to the first element of the array? How does the program know the size of the array? Where is this information stored?

I just found out the difference between the length of a string and the size of the buffer reserved for that string. But if `char s[]` and `char *s` are the same thing, who controls the size of that?

(boost appreciated)

Went to bed thinking my refactoring of KF had led to severe performance regressions (1.5x - 3.5x slower) and I'd have to reconsider (or investigate link time optimization things, or something else), but this morning when looking at the graphs I realized the old version didn't have some new formulas that took a very long time (because of transcendental functions), in fact the speed is about the same as before.

Really I should fix the bug where an unknown formula number gets reset to power 2 Mandelbrot without any warning or error message...

I got the wall-clock taken to on my desktop down from ~10mins to 4m30s.

I refactored the build system from doing ~8 'd passes on a monolithic C++ file generated from a monolithic XML file with a monolithic XSL stylesheet, into ~100 XML files (one for each formula) and ~8 XSL stylesheets (one for each algorithm), generating ~800 small C++ files which can all be compiled in .

I reorganized the structure of files to avoid including too many things (like multiprecision library) that aren't needed by every component.

To avoid reported "argument list too long" errors on , I split linking into stages with intermediate ar .a archives for each algorithm. This also sped it up a little bit.

Eventually I did the algebra by hand for the visualisation part. The analytic part (root-finding, etc) is still broken at the moment.

Formula is z^2 |z|^a + c, with a fractional and/or negative. I also implemented power 3 and 4 versions. In the animation, a varies between about +/-0.4.

Show thread

New print book (ebook and audio book soon) from Andy Farnell (author of Designing Sound):

> A final print version of Digital Vegan is out!

> digitalvegan.net

> Hopefully a delicious smorgasbord of digital rights, tech minimalism and healthy technology talk for you.

> Enjoy, share, review, but most of all I hope this philosophy changes you toward a healthier, happier relationship with technology.

Need to improve my computer algebra thing. Currently it outputs things with subexpresssions like

```
exp((((2*Ar)*1/(((4*log(((Xi*Xi)+(Xr*Xr))))*log(((Xi*Xi)+(Xr*Xr))))))*log(((Xi*Xi)+(Xr*Xr)))))
```

which can be simplified to

```
exp(Ar/2 * log(Xi*Xi+Xr*Xr)
```

So I should add cancellation to quotients somehow....

claude boosted

What's the typical fractal dimension of a tree? Say, an apple tree? So if it doubles in height, by what factor does its weight increase?

claude boosted

Super vague AI question for MIR 

Super vague AI question:

I have a pile of wav files and I would like some AI to come in and do an unsupervised analysis of them where it comes up with some ways in which they're distinct from each other and give them a score of 1-5 on how close they are to each pole of whatever it decides makes them distinct. So it decides on some pronounced x-factor and then scores them accordingly.

My questions are:
1. What's the terminology to describe this process?

2. Is there code lurking around (preferably in python or SuperCollider) that can do the machine learning/analysis/(see question 1)?

3. Are the results of it likely to be something that humans can perceive? (Or is there a way to encourage that?)

claude boosted

#Call for text line contributions: https://www.researchcatalogue.net/view/1154218/1294325

Send me your notes and thoughts on 'waiting'. What are you waiting for, is that waiting hopeful, or anxious, or incurious?

A hybrid piece with an online Mastodon bot form via @kontakt ; and a physical installation form at Reagenz Graz.

Boosts welcome!

claude boosted
claude boosted

seems something is wrong in Big Internet, 503 errors with varnish cache something on many sites.

claude boosted

I added motion blur as suggested by lycium in the fractal chats discord, not sure if I like it (I think 8 subframes is not enough to avoid steppy appearance, and rendering 8x as many frames takes 8x as long, it took 3 hours to render 1024 images for this 128 frame output, not counting the time taken to trace the rays to the locations etc).

Show thread

UK NHS Digital GP data grab, link to opt out form 

> If you do not want your personally identifiable patient data to be shared outside of your GP practice for purposes except your own care, you can register an opt-out with your GP practice. This is known as a Type 1 Opt-out.

> (NHS Digital GP) collection will start on 1 July 2021 so if you do not want your data to be shared with NHS Digital please register your Type 1 Opt-out with your GP practice by 23 June 2021.

via some way down [1], linked from [2], linked from [3]:

> If you wish to register a Type 1 Opt-out with your GP practice before data sharing starts with NHS Digital, this should be done by returning this form[4] to your GP practice by 23 June 2021 to allow time for processing it. If you have previously registered a Type 1 Opt-out and you would like to withdraw this, you can also use the form to do this. You can send the form by post or email to your GP practice or call 0300 3035678 for a form to be sent out to you.

[1] digital.nhs.uk/data-and-inform

[2] digital.nhs.uk/data-and-inform

[3] bbc.co.uk/news/technology-5738

[3] nhs-prod.global.ssl.fastly.net

`paste` is my new favourite command line tool, along with `parallel -k`. Here I use it to add columns to a table. It's annoying that you can't replace the file you're reading from all in one operation, but I suppose it's good to have copies in case of a crash or power failure etc.

```
#!/bin/bash
prec=1000
cat a.txt | while read ray period node ; do echo "m-exray-in 100 '${ray}' 8 $((2 * 8 * ${period}))" ; done | parallel -k | paste a.txt - | tee b.txt
cat b.txt | while read ray period node re im ; do echo m-nucleus $prec $re $im $period 64 1 ; done | parallel -k | paste b.txt - | tee c.txt
cat c.txt | while read ray period node ere eim nre nim; do echo m-domain-size $prec $nre $nim $period ; done | parallel -k | paste c.txt - | tee d.txt
cat d.txt | while read ray period node ere eim nre nim ds ; do echo $ds ; done | ghc -e "interact $ unlines . map (show . (\s -> 2 / (10 * s**1.125)) . read) . lines" | paste d.txt - | tee e.txt
cat -n e.txt | while read n ray period node ere eim nre nim ds zoom ; do echo -e "Re: $nre\r\nIm: $nim\r\nZoom: $zoom\r" | m-align-nodes ${node} > $(printf %04d.kfr $n) ; done
for i in ????.kfr ; do kf.x86_64.exe -s settings.kfs -l "$i" -c palette.kfp -t "$i.tif" 2>&1 | grep -v fix ; done
```

Show thread
Show more
post.lurk.org

Welcome to post.lurk.org, an instance for discussions around cultural freedom, experimental, new media art, net and computational culture, and things like that.