Problem with viaGenerateViasAtPoint function

Without using the code, the example layout which has a problem, and the version you're using, it's going to be really hard to give suggestions. I suggest you log onto our support platform at http://ask.cadence.com and then use the Case menu to submit a support case.

Andrew

Thanks for the reply andrew, i will surely put the code very soon and that might help you identify the issue.

Hi andrew, The issue resolved by replacing window id with the cell view id in the viaGenerateViasatPoint() function.Thanks

CG20240827312 said:

The issue resolved by replacing window id with the cell view id in the viaGenerateViasatPoint() function.

If you'd shared the code, that would have made it possible to identify the problem - of course, guessing all possible mistakes in the dark is not easy!

Andrew

No issues, actually i do not get access to my office laptop for me to write whole code, anyways can you please help me with this:As i mentioned earlier i have created some pathhs and vias over a cap and groped them using dbcreatefiggroup() and dbAddgigftofiggroup (), now i want to sync copy this group on all the selected caps (set of abutted rows and columns): 

for each(cap gegetselectedset()

cf=dbCopy(figgroup cv cap~>transform)

}

but for some reason this creates some offset on its x cordinate(2*cap~>x cordinate). I tried debugging but of no use. Thanks

Again, rather hard to know without seeing what you are doing. My guess is that using the transform of the capacitor instance is not giving you the transformation you need. In essence, cap~>transform will give you the transformation of the origin of the cap cell to wherever it's placed. You will be shifting the figGroup by this same transformation - so unless the figGroup is created around the origin it may end up being offset in some way. dbCopyFig is expecting a transformation to describe the relative shift from the current location of the figGroup to where you want it to end up being - and I can only guess that's not what you've got here.

Andrew

I get it,i was not understanding the way transform works, i get it now. The code works fine now with X cordinate as iam passing:

list(car(cap~>xy)-car(figGroup~>origin) ) cadr(cap~>xy)-cadr(figGroup~>origin) cap~>orient  as transform

However the y cordinate is causing an issue now  because the cap rows are placed alternatively flipped and hence doing that is giving weird values for y codinate. Its making it complicated

Is there a function that let me place these fig group at desired location?

I had the same feature need a month ago, But i did not do it for else than "R0" transform. You might need to adapt.

it is only the substraction of the local position of the center of the figGroup before relative displacement with dbMoveFig.

Here is my function :

;ineed('dbMoveFigAbs t)

;xy=8.92:0
defun(dbMoveFigAbs (@optional (inst css()) (xy 0:0 ))
"dbMoveFig is a relative displacement, i want an Absolute positioning."
"For pins, since it is a bBox, i cannot move xy (i guess)"
"not done for transform list(0:0 'R0') for simplicity"
;xy0=centerBox(inst~>bBox);=> (26.465 -0.5)
dbMoveFig(inst nil list(mapcar('difference xy centerBox(inst~>bBox)) "R0"))
)

++

hi , the code you gave works exactly same as what i had done in my previous reply, it is creating some random offset while converting from R0 to MX( might be). The code i wrote earlier works fine if only R0 devices are selected . Thanks

The trick is to compute the correct transformation. If you start by translating the figGroup back to the origin, and then concatenating the transform of the capacitor, you'll get it placed in the right location.

For example, my figGroup is the blue shapes and the other coloured shapes above are instances of a cell with each of the 8 different orientations. 

I then have figGroup assigned to the blue shapes (which are in a figGroup, with it's origin at the lower left - the same as the origin of the capacitor cell), and select the 8 instances, and run this code:

moveToOrigin=list(-xCoord(figGroup~>origin):-yCoord(figGroup~>origin) "R0")
foreach(cap geGetSelSet()
    dbCopyFig(figgroup cv dbConcatTransform(moveToOrigin cap~>transform))
)

then it ends up like this:

As you can see, the copies have the correct orientations.

Andrew