ref: http://www.fontplay.com/freephotos/sixt ... 006-26.jpg

Still experimenting with converting bitmaps to topological wireframe renders.

https://www.flickr.com/photos/34520999@ ... 6/sizes/o/

are you using GC new testing filter?

No; still experimenting with the flow. Believe I have a consistent formula for the flow now. For this render, I just used the convolution result (desaturated after the rendering). The trick I found out today is to place a solid white layer on top of your target and adjust the opacity of the white layer down to 80%. Gives better results. Also, of course, you need to start off with a downsized result (target needs to be at least 4 times smaller then final size in order to see a good grid pattern spacing; for this result, the size was 1/5 the final size). May try a tutorial later, but, as many here already know, I hate doing tuts. lol

OK.
When definitely ready for a detailed tutorial, I will try to implement your method in a filter (if you like so).

Would like. Here's the tut:

viewtopic.php?f=10&t=13569

this is the outcome of my filter, Lyle.
I followed your tutorial except for:
- when downsizing applied exactly width/2 and height/2
- when using 3d elevation applied exactly width*2 and height*2
- white layer on top in SCREEN_MODE not applied (bad results for me)
- added a Stretch HSV before applying the final Contrast

If you like, I can post the filter here (when you will wake up again....)

Result looks pretty good, but, again, if you use a smaller image, you will get a wider grid. You could make the downsize amount an option (wouldn't go more than divide by 6).

do you prefer I change the filter or do you want to get it in the current version to test it?

Would be nice, Diego. I like the wider mesh grids.

just now I posted it in its current status (under the other topic).
Now I will update it and post the amended version for when you come back

quicker than expected.
This is the result having used 4 as divisor when downscaling

...and this is the corresponding version2 of the filter

Only test it. I can work with the result (as i did with my Tithonia), but the solid fill screen layer is still at the smaller size started with. Also, the color settings in GIMP aren't reverted back (no big deal to me, but may be to another). Still, quite impressed, Diego. I, again, added some additional embellishments for my taste, but really 99.9% automated is a blessing. Thanks a heap.

Lyle, corrected the missing re-setting of existing FG and BG colours.
I do not post an amended version because I'm waiting on other suggestions to improve the outcome according to your style.
(IMO, when using a larger divisor, the final outcome is too dark; couldn't we add a "dilate" white if the divisor is greater than 2?)

I just dup/screen (result will be dark). Now, my flow may not be perfect. Whatever you come up with will probably be a much better improvement. I've not downloaded your update yet; will soon and play but it would be nice if others would give you some feedback/recommendations too.

will see if someone will do that...I'm here...

My sight might not be as sharp as others, but I like the details light enough or sharply defined enough that you can tell the features of the main image from the surroundings whether it is left black & white or colored. With the Tithonia flower that is true. With the initial doll, it was so dark I could get her face, but her hair and dress looked all one thing. I really admire how you and Lyle can figure all this out! Let's see, we have glass gurus, metal kings. How about "wire Mesh Wizards".

I'm going to bed now (almost midnight).
If you may perform some tests on my filter (version2) and tell me your opinion, thanks in advance

Works ok if sloooow. Like watching through a flyscreen.

Sort of does look like flyscreen, Kimba. lol