I can see value in a standalone, GDS-II DRC checker.
However I imagine that the operation of DRC here, has
to take advantage of the GDS-II layout loaded-database
infrastructure, and to "bust out" the DRC you'd probably
end up duplicatin…
In the example, LVS should produce a layout device with
no schematic match. So the problem would be caught.
Does it matter -how- you catch an error? More than just
time spent getting there (DRC precedes LVS usually)?
I wonder whether what's need…
Might start off with inspecting the (presumably) loaded DB
and see whether it still contains multiple topcells.
And be sure that those $args had predefined values that
work; were there preceding error yips, about that?
I'd guess that you have to find a DRC-only means of
1) finding same-named (or swappable, more fun) terminals
of same-named devices, and put them in orderly lists
2) create some sort of polygon that represents a contiguous
connection and "kno…
Maybe your macro search path then needs to have
~/.klayout/pymacros
included as well as the default .klayout?
What happens if you move the Python scripts to ~/.klayout level?
Not sure about the bit regarding
"you need to define a Spice reader delegate to capture model subcircuits".
To me, all that it wants is to spit out an "X" element card,
rather than M/Q/D/R/C/? netlist lines. The subcircuit
de…
I think there may be another "layer" behind this - what
of the case that the resistor's isolation is not "bulk" (i.e.
the P- handle / epi) but some other isolation region like
(say) a P+ resistor in NWell, or my above example of …
Layer properties can be saved in .lyp and reloaded as-saved,
at any time.
Technology setup usually points at such a .lyp, but if you are
saving by another means (the layer display details) maybe
the technology save-as, only saves the .lyp file po…
There is a "netgen" which performs any-to-any netlist compare.
Not easy to find although it's part of the efabless / opencircuitdesign
MAGIC-based toolchain (there are other "netgen" developments,
newer and unrelated, which Boga…
I am still interested to see a "select mask" capability based on a
layer / polygon.
My thinkiing goes to the old Cadence behavior (seemingly broken
in IC6 layout, but remaining in schematic / symbol editors) that
keyed on the instance/dr…
Is the data already local on that SSD? Or coming from some other
source? I imagine that the load-rate could be dependent on other things
such as media data rate (transferring from an older USB2.0 thumb
drive at 10MBPS, or over a 10Mbps Ethernet bra…
This doesn't look like the install / build instructions I followed the
last time I built from source. I don't recognize 'build' command in
all that.
I recommend you find and follow the basic instructions, and look
at the log for complaints about d…
I tried something out.
A multi-vertex polygon can have a single vertex selected by
Partial tool:
(Image)
There you can see an "innie" vertex selected.
Now if you delete that vertex, the shape becomes fully convex.
So seems like, i…
My impression is that the problem involves a convex shape,
and vertex-snapping perhaps is making local concavities that
become "notch" errors?
I guess there "could" be logic that makes all vertex snapping
"radially outwar…
A cursory check's impressions:
1) Tight and pretty, more than just a html-stack "paste job".
Prettier than anything I've done by hand.
2) Cross-links and index seem to work perfectly
3) Got to a "Classes" section which was all…
I guess my first advice would be "Know your audience". Or
audiences.
Are you trying to document the PDK for users, maintainers,
those who might build upon it elsewhere? All of these, some
of these? And do you believe they all want the s…
On the note of marker layers, you might like to take a dive into
the mainstream foundry PDK (if this is meant to get real) and
look for these kinds of layers ("capacitor/drawing"?) and also
see if there's a multifinger MOM cap already dr…
That's going to be tricker because it's a 3D problem (vertical and lateral fringing dominates). An extractor -could- handle it but whether this one does, I do not know. Would be some more elaborate code than a vertical only, plate-plate capacitance …
Maybe a version incompatibility, or a sub-dependency of 'xcb'
that is unsatisfied? There's a jumble of Qt4. Qt5 out there as I
make it out, and if you've pulled from various sources, may have
some sorting out to do?
What if you extended your script to add (bearing in mind that
my advice on scripting is not knowledge based) a 'foreach' that
runs through all the "not_the_target" layers and outputs them all
as-was? Maybe it needs a "null operation&q…
Would "print to PDF" (using one of the PDF viewers) do the job
if you set the DPI option appropriately? And maybe paper size
as well, if it's breaking into multiple pages?
You should tell us more about why you "need to" use PDF…
I prefer a (well constructed) PDF manual for reasons such as
a "search window" and linear document that you don't have to
bounce between pages up / down to navigate, the navigation
pane is handy if headings are harvested to support that…
I have found the move, duplicate commands' action awkward,
most especially the loss of selection if you don't click exactly on
the selected item(s) - then that point is taken as reslect / deselect.
It would be preferable to me if the selection befo…
To me the "Manage Technologies" and dependents, are the PDK more or
less, only "disaggregated", more of a list of pointers and settings.
I could see some value to making a flat-file image from a golden
technology-settings, buil…
In my experience (plenty, but largely trailing-edge) the layers don't
change much (or at all) even when there's a lot of PDK "churn".
I believe there's some facility for importing Cadence PDKs already
(how completely, do not know, but be…
I have never seen such a rule, although it's in the category of
"why wouldn't you?". In my world however the design of fat
bussing proceeds from analysis of current density for
reliability, and ohmic drops for parametric performance
lea…
If you have set up a "Technology" then the name /
path for the .lyp is one of the entries in the forms.
Technology choice is "sticky" so should autoload
until you change technologies or somehow lose
the remembered settings.
In my opinion the answer is to insist on simplicity,
let it be stupid as long as it has the functions that
are needed to start, stop and scale the vertical
axis (would side-diffusion be out of the question?)
to build the third dimension and slice …
Is GDS-II even capable of dealing with circles (in form
of center, radius)? I thought circles were all N-sided
polygons, once you get to GDS-II data (or in some CAD
universes, even in the native data structure).