cell libraries and place and route

[gennari]

I wrote a research version of a GDSII viewer that has image capture
capabilities at:
http://cuervo.eecs.berkeley.edu

Click on the "New GDS Viewer by Frank Gennari".

There are a number of know bugs which may or may not be fixed eventually,
but it should be able to generate JPEG, RAW, and Windows BMP images of
arbitrary resolution with stipple patterns, even for large GDSII files.

Frank

"Manuel Koch" <makoc@gmx.de> wrote in message
news:31472812.0410150105.64b257c0@posting.google.com...

I have a problem creating good printouts from Virtuso's layouts for
documentation porposes.

I want vector graphics output OR high resolution bitmap. The problem
with Cadence is that their printing service produces a horrible
mixture of both.
If you have one of the following outputs set up:

- postscript (1,2): you get the outlines of polygons as a vector
graphic and the fill patterns as low-res bitmaps; unfortunately the
fill patterns are not clipped to their corresponding polygons but
merged with all layers and cut in 'larger squares' afterwards. With
the :residentfonts: option text is preserved as text, but each letter
as a separate object, so you cannot change the size or shape of text
afterwards.
- hpgl: you get only the polygon outlines, no fill patterns;
unfortunately shapes are merged, clipped and stacking orders do no
longer correspond to the drawing layers.

The best way seems to be to export a GDSII stream and use a third
party utility.


So far I tried:
__________
sfviewer 3.5 http://www.nathaniyer.com
Does not preserve colour during export or clipboard-copy. No custom
fill patterns.
__________
GDSII Viewer 1.58b (yuda@akina.ne.jp)
Works fine so far, layer stacking order is messed up though. This
could be fixed in Adobe Illustrator (or similar) afterwards, but is
time consuming.
Perhaps I create a script for that.
I have no URL, but I can send you that program (+source) on request.
__________
SoC GDS Seduction
http://www.dolphin.fr/medal/socgds/socgds_free_overview.html
Does not allow you to print or export graphics.




Do you have any suggestions what else I can try?

 

[fogh]

(warning: do not read further unless your immune system can withstand
repeated evocations of "microsoft" and "windows" )

David Cuthbert wrote:

fogh wrote:
I would like to generate EMF files from (simple) schematics or
waveforms. How could I do this in skill ? (I hope it would be cleaner
than from eps to emf).
Hm. Not easily. The problem is that EMF (Windows Enhanced Metafiles)
is a binary format, and SKILL generally does not handle binary strings
well (specifically, it won't pass NUL characters through).

What platform are you running on? Let me see if I can't get something
written this weekend. There have been more than a few times when I've
needed to get a picture of a schematic into a paper, and it's always
been a bit hairy.

I can use HPUX/HPPA, various linuces, various windows, solaris 8. But
only monday. At home I can use windows XPpro and linux. Concurrently,
thanks to vmware.

I suppose you want to use a windows print queue, some service or RPC,
so that the EMF binary is actually generated by a windows machine. That
would be a fine solution too. You can find on MSDN how to adress the EMF
API, for instance this
http://msdn.microsoft.com/library/default.asp?url=/library/en-us/gdi/wingdistart_9ezp.asp

 

[fogh]

jason wrote:

I could not use:
par('cgsbo(mn1)')
cause cgsbo is zero in level 49 model.
Any one can help me with this?

Sure, here is how you can plot zero:
plot( drCreateWaveform(drCreateVec('double '(1k 1G)) drCreateVec('double '(0 0) ) ) ?expr '("cgsbo(mn1)") )

 

[Andrew Beckett]

pz is new in IC5033.

Andrew.

On 17 Oct 2004 07:28:50 -0700, erikwanta@starband.net (Erik Wanta)
wrote:

Spectre has a pole-zero analysis. I don't know if it is in 4.4.5 however.
---
Erik

CMOSripper@gmail.com (CMOSripper) wrote in message news:<d4253a23.0410170233.415869c5@posting.google.com>...
hi group,

could someone in the group guide me towrds finding the location of
poles and zeros of circuit .i use analog artist and spectre simulator
and the version is IC4.5

thanks for ur time

 

[David Cuthbert]

fogh wrote:

I suppose you want to use a windows print queue, some service or RPC,
so that the EMF binary is actually generated by a windows machine. That
would be a fine solution too. You can find on MSDN how to adress the EMF
API, for instance this
http://msdn.microsoft.com/library/default.asp?url=/library/en-us/gdi/wingdistart_9ezp.asp

Actually, I would prefer to write out the file directly. It's a bit
tricky to talk to machines running different operating systems at N
different sites which all have different IT policies...

I did find the EMF format documentation, and will take a look at the
OpenOffice source you pointed to. Thanks.

Dave
--
David Cuthbert (dacut@cadence.com) Tel: (412) 599-1820
Cadence Design Systems R&D

 

[David Cuthbert]

fogh wrote:

It is a shame there is no documented way to call external C functions
from skill. Could that EMF writer class be turned into some kind of
interpreter daemon that waits for skill to send GDI-like commands to it ?

Right, the C-SKILL interface is internal only -- it would, at present,
be a maintenance nightmare if arbitrary C libraries were allowed to hook
in. This is one area I and a number of others in CIC are keenly
interested in fixing. It will take some time, though.

Unfortunately, there's no easy way to hook arbitrary components together
on Unix platforms (ala Windows COM). We are aware of XPCOM, and did
have an internal project to replicate COM (in the pre-XPCOM days), but
those efforts have not met with much success/enthusiasm/traction/... .
As much as I personally dislike Java, this may be a good application for it.
--
David Cuthbert (dacut@cadence.com) Tel: (412) 599-1820
Cadence Design Systems R&D

 

[FitzDean]

My bet is that you have one hole on top of another. Good luck finding it.

Hey guy's,

im pretty stuck with this one!

I have a design completed Oin rcad Layout and in the Drill Chart it
say's that there are 2200(Quantity) 20mil holes.

I am running the post-processor and generating the gerber files.

In the DTS and the DRD file's for the drill holes it is saying that
there is 2199(Quantity) 20mil holes. There is one missing in these
drill files!

I cant figure out what is wrong.

If any of you could help it would be great

Thanks
Aidan

 

[m]

On 19 Oct 2004 02:42:11 -0700, giohdl@netscape.net (Giovanni) wrote:

Hi everybody!

I'd like to use the new RTL Compiler tool from Cadence,
but i don't know the exact library format to use.
Cadence documentation reports that a "liberty format" is required.
Does anyone know if it's possibile to use the standard synthesys
library formats such as .alf,.db , .tlf?

liberty format is the synopsys .lib format. if you have .tlf for a
library you almost certainly have the .lib too.

 

[fogh]

Frank,

I can see you are using openGL in a less (ahem...)blunt way than me. I
was just making C functions call from the SREFs, and that is quite
annoying if the end-user needs a compiler to just run the gds viewer.
How did you take advantage of the structures in your viewer ? Using
openGL's display lists ?
(BTW: are you going to put yet what can be opened on sourceforge or
savannah ?)

gennari wrote:

I wrote a research version of a GDSII viewer that has image capture
capabilities at:
http://cuervo.eecs.berkeley.edu

Click on the "New GDS Viewer by Frank Gennari".

There are a number of know bugs which may or may not be fixed eventually,
but it should be able to generate JPEG, RAW, and Windows BMP images of
arbitrary resolution with stipple patterns, even for large GDSII files.

Frank

"Manuel Koch" <makoc@gmx.de> wrote in message
news:31472812.0410150105.64b257c0@posting.google.com...

I have a problem creating good printouts from Virtuso's layouts for
documentation porposes.

I want vector graphics output OR high resolution bitmap. The problem
with Cadence is that their printing service produces a horrible
mixture of both.
If you have one of the following outputs set up:

- postscript (1,2): you get the outlines of polygons as a vector
graphic and the fill patterns as low-res bitmaps; unfortunately the
fill patterns are not clipped to their corresponding polygons but
merged with all layers and cut in 'larger squares' afterwards. With
the :residentfonts: option text is preserved as text, but each letter
as a separate object, so you cannot change the size or shape of text
afterwards.
- hpgl: you get only the polygon outlines, no fill patterns;
unfortunately shapes are merged, clipped and stacking orders do no
longer correspond to the drawing layers.

The best way seems to be to export a GDSII stream and use a third
party utility.


So far I tried:
__________
sfviewer 3.5 http://www.nathaniyer.com
Does not preserve colour during export or clipboard-copy. No custom
fill patterns.
__________
GDSII Viewer 1.58b (yuda@akina.ne.jp)
Works fine so far, layer stacking order is messed up though. This
could be fixed in Adobe Illustrator (or similar) afterwards, but is
time consuming.
Perhaps I create a script for that.
I have no URL, but I can send you that program (+source) on request.
__________
SoC GDS Seduction
http://www.dolphin.fr/medal/socgds/socgds_free_overview.html
Does not allow you to print or export graphics.
Do you have any suggestions what else I can try?

 

[fogh]

Are there application notes on using spectre PZ with ocean (or aptivia etc...) to do root-loci ?

Andrew Beckett wrote:

pz is new in IC5033.

Andrew.

On 17 Oct 2004 07:28:50 -0700, erikwanta@starband.net (Erik Wanta)
wrote:


Spectre has a pole-zero analysis. I don't know if it is in 4.4.5 however.
---
Erik

CMOSripper@gmail.com (CMOSripper) wrote in message news:<d4253a23.0410170233.415869c5@posting.google.com>...

hi group,

could someone in the group guide me towrds finding the location of
poles and zeros of circuit .i use analog artist and spectre simulator
and the version is IC4.5

thanks for ur time

 

[David Cuthbert]

Manuel Koch wrote:

How exactly would I do this? The documentation of the display resource
file shows only how to create bitmap patterns.
Could you please give me an example how to define a postscript fill
pattern?

Ah, sorry... I was speculating on how we could fix it (inside Cadence).
There's no way to do this (as of yet).

I am seeing some oddities with the way we draw polygons and conics, but
the resulting output looks correct. Are you seeing the same?

The polygons are correct. But if plot HPGL or remove the fill patterns
from postscript afterwards it is very time consuming to fill these
polygons with nice patterns (in Illustrator).

Ok, now I understand the issue here. Basically, because we're not
outputting something more standard, other tools are having issues
reading the Postscript.

With
the :residentfonts: option text is preserved as text, but each letter
as a separate object, so you cannot change the size or shape of text
afterwards.

Hm... this I'm not seeing. I put a label, "This is a label", on my
layout; I see the following Postscript:


I didn't actually look into the ps file, I just tried to change the
text in Illustrator which failed to work. I will have a look at that
issue again and report back here. Maybe I misspelled the
resident-fonts option.

It's possible that our resident font code is drawing each character
individually in order to make the output text roughly the same size as
the DFII plotter fonts.

Writing a SKILL tool to export to postscript directly would be nice
(and much appreciated by many Cadence customers), but I do not have
any experience with SKILL or postscript.

Heh, ok, I hear you guys loud and clear.

--
David Cuthbert (dacut@cadence.com) Tel: (412) 599-1820
Cadence Design Systems R&D

 

[gennari]

I assume you want to change the size or other parameters of the circles, and
that's why you made them into pcells. If this is the case, then I don't
think you can use a mosaic (array). However, an instance (SREF) in GDSII
takes around 60 bytes, so you either have more like 10 million circles or
the circles themselves are being flattened. There are two things you can try
to reduce file size if you cannot use a mosaic:
1. Make the cellname as short as possible. GDSII stores the cellname in each
instance, and if the cellname is long then that name (duplicated at least a
million times in the GDSII file) probably takes up much of the size. GDSII
is so inefficient.
2. Gzip the GDSII. Gzip typically has a 5.5:1 compression ratio, but because
your file is so regular then the ratio may be 10:1 or more. It's always a
good idea to gzip a lage file such as a GDSII layout before sending it to
someone.

Also, the circles will likely be written as polygons with a large number of
vertices. I believe it is legal to represent circles as singe point paths
with rounded corners, but I don't know if Cadence does/can do this.

Frank

"Roger Light" <roger.lightDOESNTLIKESPAM@nottingham.ac.uk> wrote in message
news:ckj31q$hor$1@oyez.ccc.nottingham.ac.uk...

Sharon Oh wrote:

I am having problems streaming my layout to gds. I have a very simple
unit cell, which is a pcell with just two circles. Within this pcell,
I have set x and y repetition. This generates a lot of circles (almost
a million polygons) in the final layout. When I export to gds, the
file is large (around 600 Meg). Is this normal? The file is a bit
unwieldy to ftp... is there any way to reduce its size (before zipping
it, that is)?

By default, pcells are flattened when they are exported as gds2 (gds2
does not support the notion of pcells). This means your simple pcell
with repetitions becomes a very large layout cell that does not take
advantage of the redundancy of your design.

Would it be possible to simply create a mosaic of your two circle pcell?
Obviously this depends on how your repetition works.

If you can do this, the gds2 file will contain one structure that is
your pcell and another structure that is the mosaic and will hence be
much smaller. Your current gds2 file effectively has a million
structures whereas what I have proposed would have two.

Hope this helps.

Regards,

Roger

 

[S. Badel]

yes, you'll have to register a new view type and create a data.reg
file in your home or modify the global one. all this is explained in
the following solution at sourcelink,

http://sourcelink.cadence.com/docs/db/kdb/2003/May/11010994.html

cheers,

stéphane

Manju Arasaiah wrote:

I want to add a view called "html-description" to my cell where
I can give lot of hyper-linked info about the cell or library
to the user who is viewing the cell from DFII.

Can we start a web-browser when he selects this particular view from
lib-manager? first of all is it possible? I know we can start an ipc process
to bring up the required page, can we link ipc to lib-manager action?

Thanks, Manju

 

[fogh]

Thanks for the background info.

gennari wrote:

My layout viewer is quite complex. It loads the GDSII file into a compressed
hierarchical database and traverses the hierarchy to find visible cells to
display. The shapes are all drawn with glVertex calls and consist of
combinations of points, lines, quads, triangles, triangle strips, and
triangle fans. The polygons are decomposed into my internal display list
format, where level-of-detail algorithms are used to render the primitive
shapes. The screenshot images are generated directly from tiled frame buffer
images, which are produced by OpenGL typically through some hardware support
from the video card.

The source code for the research version is currently available from my
research group to SRC companies only (since my research was sponsored by
SRC), and by next summer will be available to non-SRC companies as well. My
company is currently working on an improved version of this code, but it may
not be officially released. The source code of the new viewer is part of a
commercial product that is under development and therefore cannot be
released.

Frank


"fogh" <adff_at@xs4all_dot.nl> wrote in message
news:41714d58$0$29715$e4fe514c@dreader14.news.xs4all.nl...

Frank,

I can see you are using openGL in a less (ahem...)blunt way than me. I
was just making C functions call from the SREFs, and that is quite
annoying if the end-user needs a compiler to just run the gds viewer.
How did you take advantage of the structures in your viewer ? Using
openGL's display lists ?
(BTW: are you going to put yet what can be opened on sourceforge or
savannah ?)

gennari wrote:

I wrote a research version of a GDSII viewer that has image capture
capabilities at:
http://cuervo.eecs.berkeley.edu

Click on the "New GDS Viewer by Frank Gennari".

There are a number of know bugs which may or may not be fixed

eventually,

but it should be able to generate JPEG, RAW, and Windows BMP images of
arbitrary resolution with stipple patterns, even for large GDSII files.

Frank

"Manuel Koch" <makoc@gmx.de> wrote in message
news:31472812.0410150105.64b257c0@posting.google.com...


I have a problem creating good printouts from Virtuso's layouts for
documentation porposes.

I want vector graphics output OR high resolution bitmap. The problem
with Cadence is that their printing service produces a horrible
mixture of both.
If you have one of the following outputs set up:

- postscript (1,2): you get the outlines of polygons as a vector
graphic and the fill patterns as low-res bitmaps; unfortunately the
fill patterns are not clipped to their corresponding polygons but
merged with all layers and cut in 'larger squares' afterwards. With
the :residentfonts: option text is preserved as text, but each letter
as a separate object, so you cannot change the size or shape of text
afterwards.
- hpgl: you get only the polygon outlines, no fill patterns;
unfortunately shapes are merged, clipped and stacking orders do no
longer correspond to the drawing layers.

The best way seems to be to export a GDSII stream and use a third
party utility.


So far I tried:
__________
sfviewer 3.5 http://www.nathaniyer.com
Does not preserve colour during export or clipboard-copy. No custom
fill patterns.
__________
GDSII Viewer 1.58b (yuda@akina.ne.jp)
Works fine so far, layer stacking order is messed up though. This
could be fixed in Adobe Illustrator (or similar) afterwards, but is
time consuming.
Perhaps I create a script for that.
I have no URL, but I can send you that program (+source) on request.
__________
SoC GDS Seduction
http://www.dolphin.fr/medal/socgds/socgds_free_overview.html
Does not allow you to print or export graphics.
Do you have any suggestions what else I can try?

 

[dakeb]

If its free why does it ask for money?

"DesignAids" <service@designaids.com> wrote in message
news:39a6f273.0410260731.46bbf9cf@posting.google.com...

Free software at

http://www.designaids.com

CAD, FEA, civil, mechanical and electrical engineering.

 

[Richard Griffith]

You need to use the cadence supplied raster image generator called
"raptor". Raptor can output cadence layout and schematics directly in
the RTL format.

Raptor is in tools/plot/bin. The examples and samples directory show
how to use it. Here is the interface I use to send rastorized RTL files
directly to HP printers. This worked even sized up to a full 48" plot
and screamed out of the plotter at full plot speed in full color.

Ripped from some web page. Look for references to the RTL format
-------------------
RTL FIles for Large Format Plotters
The plotters have serious issues dealing with large postscript format
files. They are much more reliable and faster reading RTL files
(especially those that are 'Processed in Computer') so plotter users
should arm themselves with the following information resources:
-------------------



-- CDS plotinit --
hpjetd_c|HP Office Jet D Color: \
:manufacturer=Hewelett Packard PCL:\
:type=intCLR:\
:sh:sf:maximumPages#4: \
:spool=/cadence/config/bin/cdsPCL.sh officejetd:\
:query=lpq -Pofficejetd:\
:remove=lprm -Pofficejetd $3:\
:resolution#300: \
aperSize="A" 3150 2400 0 0:

-- /cadence/config/bin/cdsPCL.sh
#!/bin/sh
# MODFIED from tools/plot/samples/cdsBSD.sh
#
# Cadence Design Systems November 1995
#
# This script is used as a filter for the BSD print spooling system. It is
# used to execute the Cadence raptor raster driver. The raptor driver
requires
# command line arguments and a file name to configure itself properly. This
# method of using a script in the place of a print filter allow the user to
# set these arguments and supply a plot file name to the driver for
processing.
#
# The standard mode of operation of the BSD print spooling system is to pass
# the data file as standard input to the filter. The raptor driver requires
# direct access to the file and cannot operate on a stream such as standard
# input. To achieve this we need to gain access to the data file that is
# stored in the print spool directory. This file is owned by the submitter
# of the plot and is not readable by the owner of the print spooling
system.
# It is important that this filter script be owned by "root" and have SUID
# privelages set. Check this by typing "ls -l". The script line should
# look something like the following:
#
# -rwsr-xr-x 1 root sys 0 Nov 21 09:28 cdsBSD.sh
#
# If the line for cdsBSD.sh does not show privelage settings
"-rwsr-xr-x" and
# is not owned by root then login as root and run the following commands:
#
# chown root cdsBSD.sh
# chmod 4755 cdsBSD.sh
#
# You must also edit the path to the raptor driver to point to the Cadence
# software installation path.
#
# When installing the printer use the following printcap entry.
#
# <printer_name>|Cadence Raptor Print Queue: \
# :mx#0:sh:sf:lp=/dev/<plot device>:lf=/usr/adm/lpd-errs: \
# :sd=/usr/spool/<directory>: \
# :if=<install_path>/tools/plot/bin/cdsBSD.sh:
#

#
# Update the path to include the installation directory for Cadence Software
#
PATH="/bin:/usr/bin:/usr/ucb:/cadence/ic446/tools/plot/bin:"
export PATH

file=/var/tmp/cadPlot$$
# capture stdin
cat - > $file
raptor -p -f RTL $file | lpr -P$1
rm -f $file


----== Posted via Newsfeeds.Com - Unlimited-Uncensored-Secure Usenet News==----
http://www.newsfeeds.com The #1 Newsgroup Service in the World! >100,000 Newsgroups
---= East/West-Coast Server Farms - Total Privacy via Encryption =---

 

[fogh]

no problemo: CST and HFSS are about as free as cadence IC.

I noticed that some spanish speaking team has revived oregano, but I could not
compile the CVS version. Probably because I don t know how to drive GNU
autotools. And again the manpages tell you nothing useful and redirect you to
the infopages. These are times when I feel like having a bite of GNU tartare...

dakeb wrote:

If its free why does it ask for money?

"DesignAids" <service@designaids.com> wrote in message
news:39a6f273.0410260731.46bbf9cf@posting.google.com...

Free software at

http://www.designaids.com

CAD, FEA, civil, mechanical and electrical engineering.

 

[gennari]

My layout viewer is quite complex. It loads the GDSII file into a compressed
hierarchical database and traverses the hierarchy to find visible cells to
display. The shapes are all drawn with glVertex calls and consist of
combinations of points, lines, quads, triangles, triangle strips, and
triangle fans. The polygons are decomposed into my internal display list
format, where level-of-detail algorithms are used to render the primitive
shapes. The screenshot images are generated directly from tiled frame buffer
images, which are produced by OpenGL typically through some hardware support
from the video card.

The source code for the research version is currently available from my
research group to SRC companies only (since my research was sponsored by
SRC), and by next summer will be available to non-SRC companies as well. My
company is currently working on an improved version of this code, but it may
not be officially released. The source code of the new viewer is part of a
commercial product that is under development and therefore cannot be
released.

Frank


"fogh" <adff_at@xs4all_dot.nl> wrote in message
news:41714d58$0$29715$e4fe514c@dreader14.news.xs4all.nl...

Frank,

I can see you are using openGL in a less (ahem...)blunt way than me. I
was just making C functions call from the SREFs, and that is quite
annoying if the end-user needs a compiler to just run the gds viewer.
How did you take advantage of the structures in your viewer ? Using
openGL's display lists ?
(BTW: are you going to put yet what can be opened on sourceforge or
savannah ?)

gennari wrote:
I wrote a research version of a GDSII viewer that has image capture
capabilities at:
http://cuervo.eecs.berkeley.edu

Click on the "New GDS Viewer by Frank Gennari".

There are a number of know bugs which may or may not be fixed
eventually,
but it should be able to generate JPEG, RAW, and Windows BMP images of
arbitrary resolution with stipple patterns, even for large GDSII files.

Frank

"Manuel Koch" <makoc@gmx.de> wrote in message
news:31472812.0410150105.64b257c0@posting.google.com...

I have a problem creating good printouts from Virtuso's layouts for
documentation porposes.

I want vector graphics output OR high resolution bitmap. The problem
with Cadence is that their printing service produces a horrible
mixture of both.
If you have one of the following outputs set up:

- postscript (1,2): you get the outlines of polygons as a vector
graphic and the fill patterns as low-res bitmaps; unfortunately the
fill patterns are not clipped to their corresponding polygons but
merged with all layers and cut in 'larger squares' afterwards. With
the :residentfonts: option text is preserved as text, but each letter
as a separate object, so you cannot change the size or shape of text
afterwards.
- hpgl: you get only the polygon outlines, no fill patterns;
unfortunately shapes are merged, clipped and stacking orders do no
longer correspond to the drawing layers.

The best way seems to be to export a GDSII stream and use a third
party utility.


So far I tried:
__________
sfviewer 3.5 http://www.nathaniyer.com
Does not preserve colour during export or clipboard-copy. No custom
fill patterns.
__________
GDSII Viewer 1.58b (yuda@akina.ne.jp)
Works fine so far, layer stacking order is messed up though. This
could be fixed in Adobe Illustrator (or similar) afterwards, but is
time consuming.
Perhaps I create a script for that.
I have no URL, but I can send you that program (+source) on request.
__________
SoC GDS Seduction
http://www.dolphin.fr/medal/socgds/socgds_free_overview.html
Does not allow you to print or export graphics.
Do you have any suggestions what else I can try?

 

[David Cuthbert]

Ed Baxter wrote:

Good thing I found this thread...I was having the same problem Michael
is having . I was in a position to easily reinstall Suse 9.1 so I made
root an ext3 filesystem instead of reiser and now cdslmd starts.
Thanks a mil!

Excellent. I've been having issues getting ReiserFS installed (and
VMware back up after some security updates were applied to my desktop
box), so I haven't had a chance to replicate this. But it sounds like
the culprit has been identified.

I'm surprised that this has an effect on file locking code. It's one of
those things that shouldn't affect correctly-written apps (that is,
we're not going in and explicitly doing "Is this filesystem X? Then do
Y" or "Rely on undocumented features that just happen to work on ext2/3").

For now, I'd just say "Don't do that."

--
David Cuthbert (dacut@cadence.com) Tel: (412) 599-1820
Cadence Design Systems R&D

 

[G Vandevalk]

Not sure what you want here.

When you rad in a CDL netlist, do you have the AS AP PS PD info?
(i.e. is the source a layout? )
the proper way is to have the fields populted on Reading them in.

Note that these parameters are highly dependant on layout topology.

I have created a set of callbacks that CAN recalculate the areas &
perimeters of the sources and drains
based on a parameter (called style , i think )
The style would have choices like: Nominal, Minimum, Custom.

When a mosfet was placed in a schematic, the default would be Nominal and
the S/D sizing numbers would be
calculated assuming an isolated mosfet, rectangular layout and minimum
design rules. When a schematic was
backanotated form an extracted rep, we would set the default to custum and
read i the measured S/D sizing numbers
from the extracted rep. If a mosfet was known to be in a series chain, then
the Minimum number would be uses.

(i.e. nominal would calculate using the Design Rules for a square contacted
S/D. Minimum would calculate using 1/2
of the Minimum Gate spacing.)


You say that you only trigger the callback by changing W. Sound like Nominal
values get populated. You could be
in for a surprise if you rely on these numbers ...

-- Gerry


"Szekit" <szekit@gmail.com> wrote in message
news:4d101329.0410291257.3937c654@posting.google.com...

Let me elaborate more...

The AS, AD, PS, PD parameters only get updated when I go into each
schematic and modify the w parameter. In another words, when I CDL in
a netlist, AS, AD, PS, and PD are assigned with default parameters. I
need to modifty the w parameter in order to trigger the CallBack
function to update the AS, AD, PS, PD parameters. Any ways to do this
in an automate steps?

szekit@gmail.com (Szekit) wrote in message
news:<4d101329.0410281512.1bd9f821@posting.google.com>...
Hi,

Somebody has done this for me in the past but I didn't pay attention
on how to do it. Now it's my turn to do it and I am seeking your
help!!!

I have a CDL netlist of various standard cells and I need to import it
for schematic capture and simulation. What is the solution??

Thank you very much for your assistance!!!!

-szekit