Suspicious Behavior?

Steve Sergev

PRO Member
For the attached file which has a symmetric laminate, the Gen 2 laminate seems reasonable (symmetric above and below the horizontal centerline) up to a Gen 1 cut angle of 38 degrees. But when increasing the Gen 1 cutting angle above 38 degrees and while looking at the Gen 2 laminate, the top/bottom symmetry starts to get lost. Specifically, the outer band above the centerline grows and the outer band below the centerline shrinks. Seems odd. Or am I not understanding correctly what happens to the laminate?

Running version 3.2.018.
 

Attachments

mfisher

Super Moderator
Staff member
Steve:
I downloaded your file and have looked at it in LP. I am sure I am seeing what you are seeing.

When you click on the Laminated view the cut lines showing are for the 1st Gen cuts.
When you select Gen 1 it is showing what the glued up lamination should look like along with the next gen cut lines. I noticed you have the flip selected. If you are referring to the tabs that are forming. That is because of the continued flipping of your cut pieces.
 

Attachments

Lloyd Johnson

Administrator
Staff member
Well, I can see what the problem is but fixing it may be problematic. The key to the graphics is to find the vertical center of the image. The problem is that the number of vertical lines is always an even number so the line I identify as the center line is off by 1/2 of a pixel which rounds off to a full pixel. So the image is too high by one pixel for the first generation and two pixels for the 2nd generation and so on.

It might be as simple as moving all images one pixel lower before I start the image processing. That would be nice, but it is seldom that simple when the code is this complex. :-<

I'll add it to my to-do list.
 

Steve Sergev

PRO Member
Well, it's more than a pixel or scan line. (I'm having trouble uploading another image) If I start with the .lam file and increase the Gen 1 cut angle incrementally up to say 60 degrees and watch the Gen 2 image, the Gen 2 pattern is far from symmetrical top to bottom; yes, the central feature pattern seems OK and stays in the center of the screen, but the "plain wood" margins at the top and bottom differ dramatically in width. The margin at the top increases while the one at the bottom does not.
 

mfisher

Super Moderator
Staff member
Steve:
imo, it is the steep angles being cut. it will tend to reduce the bottom material.
 

Lloyd Johnson

Administrator
Staff member
Steve, I see exactly that the problem is a cumulative error. Because the image is one pixel higher than it should be before I start the imaging process, when I slice the image, the pixels that are closest to the center are fine. As you move farther from then center, the error becomes noticeable and if you've used an extreme angle, you are a lot further from the center so the error continues to grow. When I repeat this process for the second generation, that image is now two pixels too high and the slicing and rotation of that image makes the discrepancy more noticeable. But moving the image down by one pixel will simply make the same error on the bottom of the image. Making a change that will fix this means changing code in lots of places which is why it will not be a simple fix.

As you accurately pointed out, the center of the design is as perfect as it needs to be even if you use extreme angles. So I appreciate that you pointed this out to me because I never noticed it in the past 12 years. I never noticed it because I would ever make a design with an extreme angle because in practice, the same slicing/rotation would make it hard to maintain quality.

Because of this, this issue has a ways to go before it gets to my top priorities list.
 
Top