PDA

View Full Version : Color mapping without dithering?



James Terry
05-30-2010, 3:35 PM
I like to implement multiple burn layers in my items for relief detail which requires me to color map in the driver. The solid colored layers are taking their power settings but it also looks like they are dithering too which I dont want. And on second look, black is doing the same thing. Is there something I have missed in my driver to disable the dithering or is there some step in Corel that would help this happen like I want? Can I map power settings to layer number instead of color somehow?

It is as if I want all of my layers sent to the table as black, but with different power settings.

So how do I get rid of my weave on solid colors?



http://www.webglider.com/laser/2008/DSC07071.JPG

http://www.webglider.com/laser/2008/200800.jpg


http://www.webglider.com/laser/2008/200801.jpg

http://www.webglider.com/laser/2008/200802.jpg

http://www.webglider.com/laser/2008/200803.jpg

http://www.webglider.com/laser/2008/200804.jpg

James Terry
05-30-2010, 3:55 PM
I just found this in Corel's print output. Perhaps I need to experiment with printing all colors as black and/or output color as RGB perhaps? Of course all lines should be vector, but I am not sure what the softwares are doing with the color fill. That must be my problem.


http://www.webglider.com/laser/2008/200805.jpg

Dan Hintz
05-30-2010, 6:37 PM
You need to set your Corel desktop to use an RGB colorspace. What's most likely happening is you're outputting a color that looks similar to your Mercury colorspace, but it's off by a couple of notches... any time that happens, the driver is going to try and dither the output to match up with your "desired" color.

Rodne Gold
05-31-2010, 2:42 AM
I dont think it's "dithering" as you put it , I often get that pattern when deep engraving using vector fills on stuff like perspex, either that or lines or ridges. But perhaps it is doing something , try other dither patterns and the diffusion mode to see if it changes.

Doug Griffith
05-31-2010, 11:24 AM
Another way to help smooth out multiple passes is to use different DPI/PPI settings for each pass. The different resolutions will have different dither patterns and smooth each other out. It's not ideal but helps.

James Terry
05-31-2010, 3:34 PM
I have still not figured this out. I am applying a solid color in Corel so doesnt that make it a vector fill? Or does the laser raster solid color anyway/always? Print Preview shows colors, not grayscale.

Selecting Manual Color Fill allows you to map colors to speed/power but it does every layer on a separate pass taking much longer, but it also seems to still dither the fill anyway right now. It seems to me that I should be able to find a method where I can etch the entire thing in one pass using varying power.

The first image is what I am trying to duplicate from my original vendor who was using a Mercury table as well. All you can see is what appears to be the grain of the wood. The second is my best attempt on my machine and what you see is squares and what appears to be an X by X dither pattern.

http://www.webglider.com/Laser/2008/DSC07095.JPG

http://www.webglider.com/Laser/2008/DSC07096.JPG


I havent tried this yet, but I wonder if rotating the wood grain parallel to the laser movement axis would change what I am seeing. Heck, perhaps I have some specs of dust on an optic somewhere but that would only mess up a single dot, right, not reproduce a pattern that seems to scale with dpi.

Mike Null
05-31-2010, 4:28 PM
This isn't a solution to your problem but might make your engraving look better--try inverting the image.

Chuck Stone
05-31-2010, 5:03 PM
I have still not figured this out. I am applying a solid color in Corel so doesnt that make it a vector fill?

Is there such a thing as a vector fill? I think the fill would be rastered, but
I admit that my knowledge of Corel is pretty limited. So is my knowledge
of vector drawings. I really don't know the answer to that.

But I just went and set up something for the Mercury. These aren't the
numbers I usually use, I just did it for illustration. My machine is miles from
here so I can't give you the info I'm using.

Using a color scheme like this, I print out grayscale images and the only
lines I get in the engraving are the ones left by the wood grain itself.
It is really nothing more than a gray step wedge with various power
settings applied to it. You will also need to tweak it to start, and each
type of wood may need different numbers.

I saved this as a .MER file and any time I want to do something with
a 3D effect (2.5D, really) I'll desaturate the image, adjust the levels and
contrast and use this .mer profile so it will engrave at different levels.
Also works fine for black and white images, as they'll only laser on full
power or no power (if white is set to 0 power)

don't know if this helped or muddied the waters more..

Scott Shepherd
05-31-2010, 6:02 PM
I agree, the only way to make happen what you want (varying power during a single pass) would be to use the 3D function and use different shades of gray.

Joseph Szegda
05-31-2010, 7:08 PM
Vector fill is done in a true cad program as lines set at a spacing of less than then width of the laser. then change colors and set power or speed to color. do all the time. But not in coral in a cad program, like autocad, designcad or cimcad. send me a copy of file and will do a demo for you
joe

Doug Griffith
05-31-2010, 7:30 PM
Vector fill is done in a true cad program as lines set at a spacing of less than then width of the laser. then change colors and set power or speed to color. do all the time. But not in coral in a cad program, like autocad, designcad or cimcad. send me a copy of file and will do a demo for you
joe

That pretty much removes dithering and even DPI from of the picture. It would also be much slower as the laser is in vector mode and not raster mode.

James Terry
05-31-2010, 10:13 PM
Eureka!

I am pretty sure that I resolved my problem, but not yet so sure why. What I mean is that I found my improved output but I am not sure why it is a problem to begin with. It suggests that there is more going on with my beam than just a focused micro dot.

I havent yet come across any pertinent reading regarding wood grain and lasers, but simply, I adjusted so that the grain of the wood is parallel to the travel of the laser head. No more boxes.

This tells me that the laser beam is either not perfect or is making some 'lines' deeper than others. When perpendicular to the wood grain, ugly dither-like line patterns emerge. When done with the grain, I guess it is hidden.

So is the dot width actually making a semi-circle crater in the wood such that it is deepest in the center and lightest on the edges? Then for each line, there ends up being a line of material at the edges of the beam that form a high point or a ridge that was noticeable against the grain? Perhaps my beam is out of focus? Perhaps there is dust in the beam path?


The unwanted dither-like problem presenting itself:
http://www.webglider.com/laser/2008/DSC07104.JPG

The "fix" by adjusting the wood grain. Grain hides the error?
http://www.webglider.com/laser/2008/DSC07106.JPG

Rodne Gold
06-01-2010, 2:15 AM
Ive always been of the opinion that ridging is due to slight misregistration cos of material moving , expansion and contraction due to laser heat locally where its being engraved and the re engraving of dross from the last scan pass.
I do moulds for spin casting where surface roughness is an issue - so I use double sided tape to hold down the item , go a little slower and then use a polishing pass or 2 after (much lower power , much higher speed) and this helps but doesnt actually totally stop ridging and patterns.
We have also used patterned postscript fills to make the background pattern a "feature"

Dan Hintz
06-01-2010, 6:49 AM
James,

I actually would have expected the opposite, if anything... patterns when going with the grain, none when going against. I'll have to think on this one a bit when time permits and see if I can come up with a logical reason (and possibly some useful tests)...

Mike Null
06-01-2010, 7:24 AM
Rodney

It was also our practice to use post script fills to improve the appearance of such molds.

If I use a clean up pass I usually do it at a different resolution.

James Stokes
06-02-2010, 10:35 AM
To me it sounds like the laser is just a hair out of alignment. A couple things you can do to change depth in areas, !. Duplicate the part you want deeper and leave it directly on top of the other. 2. For smoothing I raise my lens up and take it out of focas, This will give you a bigger spot and more overlap.

Stanley Waldrup
06-03-2010, 8:32 AM
Terry, Go to your printer Driver select the Raster tab then turn off Dithering. Select Error Diffusion and select Floyd... Works for me..
Stanley

James Terry
06-03-2010, 1:08 PM
That tab disappears when you select manual color fill which is what I am using here. There is no raster control, or rastering from what I can see.

Mike vonBuelow
10-18-2010, 7:06 AM
Nice info!

Mike Null
10-18-2010, 10:38 AM
I would think this would go without saying but I'll say it anyway. Wood is not uniform in hardness nor density therefore it is likely that you will always have variations in the marks made by the laser when using wood materials.

Niklas Bjornestal
10-18-2010, 12:23 PM
I would think this would go without saying but I'll say it anyway. Wood is not uniform in hardness nor density therefore it is likely that you will always have variations in the marks made by the laser when using wood materials.
I think thats the explanation why you get the pattern when engraving perpendicular to the grain since each line will have alot of variable deapth due to the variation in hardness in the grain. When engraving with the grain the lines will be different deapth but that wont show up as much.
Try engraving the same image using acrylic (or other material with uniform density) and see if you get the same result.

When engraving pine the variation in density for the grain is very visible, it varies a about 4-5 mm between the soft/hard spots using 100sp/100pw on my 25W mercury II.

Joe De Medeiros
10-18-2010, 2:47 PM
Eureka!

I am pretty sure that I resolved my problem, but not yet so sure why. What I mean is that I found my improved output but I am not sure why it is a problem to begin with. It suggests that there is more going on with my beam than just a focused micro dot.

I havent yet come across any pertinent reading regarding wood grain and lasers, but simply, I adjusted so that the grain of the wood is parallel to the travel of the laser head. No more boxes.

This tells me that the laser beam is either not perfect or is making some 'lines' deeper than others. When perpendicular to the wood grain, ugly dither-like line patterns emerge. When done with the grain, I guess it is hidden.

So is the dot width actually making a semi-circle crater in the wood such that it is deepest in the center and lightest on the edges? Then for each line, there ends up being a line of material at the edges of the beam that form a high point or a ridge that was noticeable against the grain? Perhaps my beam is out of focus? Perhaps there is dust in the beam path?


The unwanted dither-like problem presenting itself:
The "fix" by adjusting the wood grain. Grain hides the error?


In the epilog manual it mentions that it's better to engrave with the grain