Hi Jinja,

Lovely drawings, been playing with Death To Sharpie and my Polograph for a couple of weeks now and I must say I'm impressed with your results. You seem to be having much better luck getting the shades to come out in the drawings than me, Mine tend to end up more black and white. Think it might have something to do with me not scaling the image properly in DTS. I'll enjoy playing with it more and finding out.

I've attached some pictures of my results, please ignore the couple of stray lines in the Easyrider one, had some mechanical issues which I hope to have fixed.

Hope to see more of your drawings.

Regards

Thanks for the comments both, glad you like them. @reconlap, the Easy Rider one in particular looks pretty nice, look forward to seeing how you develop that.

@kongorilla, as you might imagine there were several poor attempts before I started to get things I liked 🙂 The first thing I did was to ditch sharpies - the Tombow pens give consistent results over the life of the pen, particularly in line width and ink flow. I also don't use a black pen - the code that comes out overdraws the dark areas a lot so it will end up saturated so I use a dark grey or medium grey.
On the image prep front, I have a large library of B&W negs from when I had a darkroom so I have been using those, though I do have some from colour images in the works that seem to be coming out ok. Anyway, I desaturate (if colour) then adjust the image to reduce contrast and get the broadest possible range of greys. For both pictures I removed clouds in the sky completely as they just distracted from the image. also, for the Kings cross one I had to manually add a darker line for the very top edge of the building as it just wasn't coming out in D2S.
Then I do literally dozens of D2S renderings, tweaking the parameters constantly.
I adjusted the on screen output to get it as large as possible so I can see the results better, but I found that fiddling with the paper size settings has a dramatic effect (usually detrimental) so I pretty much leave 'em alone.
For the main fiddlable parameters I had these for example for the Church image:
final int squiggle_total = 650;
final int squiggle_length = 700;
final int half_radius = 3;
final int adjustbrightness = 5;
final float sharpie_dry_out = 0.0;
final String pic_path = "Church1a.jpg";

I added a couple of my drawings to the Polargraph Flickr group as you suggested.
I'll post a bit more later when I have more time. Off to work!

Thanks for the detailed write-up, JingaBeardy. I put Death to Sharpie experiments on hold until I finish building plotters that use Koh-I-Noor Rapidograph pens, so I can draw with custom inks; I came to same conclusion you did regarding using black inks. Ballpoint looks good, though. Still, I found it hard to get the plot out of the dark values fast enough without over drawing every mid-value area as well. Very touchy parameters.

It'd be nice to get the output as SVG, so errant squiggles could be deleted. Like, why did it draw those squiggles on the sky borders of your Kings Cross drawing? All my drawings have something like that as well.

Maybe we could coax an appearance by Scott Cooper in this forum. Some recommendations from him would probably save us much time and art supplies.

@Krummrey - COOL! Thanks! I'll try it soon. 🙂

edit: Just exported the default image...WOW! It's cool to get a good look at the exported lines in Inkscape. They have a more angular character than I had previously appreciated.

Hi,
Just wanted to drop by and share a useful feature for preprocessing images in gimp for D2S. I found the "Cartoon" filter in gimp and suggest D2S users give it a try, I find it's a quick and easy way to get D2S to pick up more detail from an image that it often misses. You can find it in in filters->artistic->cartoon .

It really helped D2S show the stonework in the drawing below.

Have fun playing with it and hopefully some will find it helpful.

regards

the pdfs import into inkscape a bit slowly for my liking, so I hacked together svg output in the script. Output is file per pen as per the gcode rather than all in one.
fork is here:
https://github.com/jwcliff/Drawbot_image_to_gcode_v2

Sammie, your description is so vague that I don't think I can help.

I have updated the https://github.com/Scott-Cooper/Drawbot_image_to_gcode_v2 code to output SVG files. I was inspired by the fine work of John Cliff. Each pen is kept as a separate group/layer.
Hope this reduces some of the pain points.

Thanks John for pushing this forward.

Quick help.

So I installed Processor.org, I downloaded both the original

Drawbot_image_to_gcode_v2 Master

also

Death_to_Sharpie_PDF

With the first one I can select a jpg and it draws it on the screen, but I don't see any output file? gcode, svg, or pdf.

How do I get it to save the file?

If I want it to stop the rendering part way through can I stop it? (I find a simpler version half way drawn to be more appealing).

With the Death_to_Sharpie_PDF, it will render and save an output PDF in the directory, but it doesn't ask to select a file or as with the first one, use the URL of a jpg.

So how are you guys selecting your JPG to render?

Thanks!

I think I have all my electronics and firmware together. Just need to build a gondola and get the mechanical together:-)

Max

Very nice work!