## Search found 1656 matches

### Re: Radon

i hope you find the right tooling to satisfy your needs. my Gen 4 accumulator is done. i now store a decently accurate floating point average in a byte. it's awesome. though in my random ray shooter, i discovered i'd "need" to track write counts per pixel, which Gen 4 "doesn't have en...

### Re: Radon

i often find myself sitting for minutes trying to word something so it's either somewhat clear and hopefully thoughtful. Writer's block. I get that. There are many ways to combat it. What works for me is to save what I have done and then do something else for a while. Very typically when I return, ...

### Re: Radon

it's a thing of a post being invisible, aka some kind of forum bug. well it could be anything. a collusion between gov't and aliens. i have composed offline but when i'm inspired, stuff usually flows. i often find myself sitting for minutes trying to word something so it's either somewhat clear and ...

### Re: Radon

sharpen 1d and coefficients optimizer /' radon transform with sharpen 1d and coefficients optimizer 2024 June 30 by dafhi update new tool - scaled lerp with expon lerp "endpoints" 1. baseline Radon and Inverse 2. 1d-sharpened radon and Inverse for my 1d sharpen i use array "coeffs&quo...

### Re: Radon

unless i'm missing something, i seem to be reconstructing an image nearly perfectly, just not the original image ' ' rd0004a.bas ' ' sciwiseg@gmail.com , 2024 ; lgpl , if applicable ' ' 74% help from ChatGPT ' ' Resize radon() row to closests power of 2, prior to fft . ' '#INCLUDE "../inc/fbmat...

### Re: Radon

i did a search on 1d sharpen, found a post mentioning origin [0 1 0] + sharp [-1 2 -1] but it doesn't do radon justice. i might have an idea of how to proceed next: make a delta of source and inverse radon, then analyze radon of that. [edit] by "analyze radon" i mean compare a radon angle ...

- Jun 24, 2024 2:04
- Forum: Community Discussion
- Topic: UnlockMe #1 - Fun Game
- Replies:
**7** - Views:
**655**

### Re: UnlockMe #1 - Fun Game

Code: Select all

```
dim as any ptr a0
dim as ubyte ptr b0
a0 = @sec
b0 = a0
*b0 = 1 '' all 255 Nope
*(b0+1) = 1
*(b0+2) = 1
*(b0+3) = 1
```

### Re: Radon

my newest post is hidden, like a previous post fxm and i commented on it. i'll reiterate. my latest post is hidden also, it's on page 2 and you'll only see it if you preview reply unable to view your post until i reply Preview Last hidden post from Luxan ! This might be because of how arrays are bei...

### Re: Radon

gen2 accumulator. forget ushort .. i'm using a ubyte might be a possibility for a gen3 #define dbl as double '' reduced text #define sng as single #define min( a, b) iif( (a)<(b), (a), (b) ) #define max( a, b) iif( (a)>(b), (a), (b) ) #macro sw( a, b, tmp ) tmp = a: a = b: b = tmp #endmacro function...

### Re: Radon

gen 4 /' -- Gen 4 accumulator demo - 2024 July 3 by dafhi 1 byte accumulator a hack in _trace_onto() works well enough vs. having to saved a per-pixel iter count. '/ #define dbl as double '' reduced text #define sng as single #define min( a, b) iif( (a)<(b), (a), (b) ) #define max( a, b) iif( (a)>(b...

### Re: Radon

my accumulator thing is useful, if half-baked. for 'simple' accumulation on 1 component (red) for 1 pixel one could intuitively use a ulong for the sum and a ushort for count. my gen 1 accumulator performs well enough - from a ulong, i'm using 10 counting bits, where can rollover. with my gen 2 i mi...

- Jun 12, 2024 12:16
- Forum: General
- Topic: Natural sorting algorithm
- Replies:
**3** - Views:
**337**

### Re: Natural sorting algorithm

big update June 27 u2 /' -- natural sort - 2024 June 27 update 2 - by dafhi rough draft a < A A < a1 1 < 01 empty strings to back update: cleaned demo '/ #define min( a, b) iif( (a)<(b), (a), (b) ) #define max( a, b) iif( (a)>(b), (a), (b) ) #define str_direction < '' ascending vs descending namespa...

### Re: Radon

nice description i'll eventually use full RGB but with integer accumulators. source image data will be standard ubytes for easy test patterns. accumulator const cbits_c = 13 const sum_cbits = 32 - cbits_c const sum_mask = 2 ^ sum_cbits - 1 type accum_int as ulong function pel_read( i as accum_int ) ...

- Jun 09, 2024 13:29
- Forum: General
- Topic: Natural sorting algorithm
- Replies:
**3** - Views:
**337**

### Re: Natural sorting algorithm

you could start with something like [renamed next_num -> find_int] namespace str_compare dim as long pos0, pos1 function find_int( s as string, i as long = 0 ) as longint for j as long = i to len(s) - 1 if s[j] >= 48 andalso s[j] <= 57 then pos0 = j: exit for next for j as long = pos0 to len(s) - 1 ...

### Re: Radon

sure, send your update i doubt i'd be able to get anything to work. atm i'm building code for multi-purpose rasterization: draw, read, measure error. i'm high school algebra with some exposure to higher level math. also one of the slowest coders here :roll: still interested in radon thingy, it's a b...