It looks like you're new here. If you want to get involved, click one of these buttons!

The above algorithm is used to demonstrate the verification of bend curvature based on the eqDRC theory, but the problem is that how can I get x and y(the projection of the shape on X and Y directions)?

In the meantime, how can I write my own drc model in ruby and pack it so that I can use it in drc script?

If there are eqDRC scripts, that will be better.

The DOI of eqDRC paper is 10.1109/VLSI-SoC.2014.7004173, and I am trying to achieve these eqDRC function in klayout, thanks.

## Comments

Thanks for this information.

I'm not a researcher, but the paper seems quite old and I have not retrieved it. But I cannot imagine deriving a curvature is that easy. Could you supply a sketch or any publicly available reference?

The projection of a polygon will basically by the bounding box. I don't see how this can be used to derive the curvature.

Thanks,

Matthias

The title of a paper about how to calculate the curvature is "Silicon Photonics Design Rule Checking: Application of a Programmable Modeling Engine for Non-Manhattan Geometry Verification"

I also cannot image the process is so easy, but the authout of this paper is also the engineer of SIEMENS EDA Mentor Graphic Calibre. So I think the way in this paper is authentic, but the problem is whether it is easy to be achieved in Klayout by using ruby.

I cannot find other public publication or reference.

I think equation-based DRC is really potential.

Thanks.

@Matthias The title of a paper about how to calculate the curvature is "Silicon Photonics Design Rule Checking: Application of a Programmable Modeling Engine for Non-Manhattan Geometry Verification"

I also cannot image the process is so easy, but the authout of this paper is also the engineer of SIEMENS EDA Mentor Graphic Calibre. So I think the way in this paper is authentic, but the problem is whether it is easy to be achieved in Klayout by using ruby.

I cannot find other public publication or reference.

I think equation-based DRC is really potential.

Thanks.

Thanks - but first of all, it is pretty old, not specific and I'm lacking a picture here. To what kind of situations will it apply?

I don't want to request the paper. I think it's a honey pot, not serious research. The term eqDRC is not used anywhere else, but it may be a registered trademark or patented.

Matthias