Re: issues with technology editing in version 7

Pierce Keating <[email protected]> Sun, 22 Apr 2007 10:44:24 -0700 (PDT)
Newsgroups gmane.comp.cad.electric.bugs
Message-ID <[email protected]>
--===============0271101957==
Content-Type: multipart/alternative; boundary="0-383363084-1177263864=:3333"
Content-Transfer-Encoding: 8bit

--0-383363084-1177263864=:3333
Content-Type: text/plain; charset=iso-8859-1
Content-Transfer-Encoding: 8bit

Hi Dima,
   
  The reason I reverted to using version 7 is because I ran into issues converting a technology to a library for editing in version 8.4. Below is a response from Steve Rubin when I first reported this problem. So far, version 7 has been behaving fairly well - an occasional crash but I've been careful about backing up libraries.
   
  Thanks again,
  Pierce
   
  At 10:40 PM 4/1/2007, you wrote:
  I'm using version 8.4.
 
Pierce

Steven Rubin <[email protected]> wrote:
    
   At 12:47 AM 4/1/2007, you wrote:
  
   >Hello,
  
   >
  
   >I think I have found what seems to be a bug. I am trying to add a 
  
   >pbase layer and then an npn transistor node to the mocmos 
  
   >technology. I have tried a simple experiment where I converted a 
  
   >technology to library for editing, and chose mocmos. Then I tried to 
  
   >immediately convert the library back to a technology, giving it the 
  
   >name: mocmos_abn. This failed, with Electric having a problem with 
  
   >determining the stretching rules for the metal-1 pin layer. At other 
  
   >times, the problem can be with the node-P-Transistor-Scalable. I've 
  
   >tried several things and Electric always finds a problem with things 
  
   >I haven't touched, which is why this really seems like a bug to me.  
I think this has been fixed since 8.4 was released.  However, it is not a simple fix...you would need a completely new release.

We plan to make one in less than 2 months, but that may be a long wait for you.

Alternatively, you could hand-edit the technology library to make it more useful.  Nodes are described with 4 separate examples plus some text.  Each of these 5 pieces of graphical information must be far enough from the others that they can be easily separated.  The error message that you are getting indicates they are too close.  So you could hand-edit these cells to move them farther apart.

   -Steve 
   
   
   
  

[email protected] wrote:
  Hello Pierce,

Why do you use version 7 ? The latest released
Electric version is 8.04.

It will be easier for me to help you with technology
tuning of Electric 8.04,
because I already forgot many things from C Eelctric.

For now, I see that Electric 8 reads technology
libraries from Electric 7 with errors.
I can look more tomorrow (on my work computer).
However, if you can recreate the technology library
in Electric 8 , then I shall look immediatly
at technology issues in Electric 8 rather than
dealing with incompatibilities of Electric 8 and
Electric 7.

Moreover, I am developing an Xml format for
technology description which will apper in Electric
8.05 .
I beleive that AMI technology example can be a test
of usability of Xml technology description.

-Dima
  Hello,
   
  I've since found that the metal spacing error mentioned below does get detected by the DRC. I've also discovered that connecting to the transistor pins is better than I first thought, if I work with it, but I still haven't been able to avoid OVERLAP DRC errors.
   
  Thank you,
  Pierce

Pierce Keating <[email protected]> wrote:
    Hello,
   
  I have have been attempting to modify the mocmos library for the (MOSIS) AMI 1.5u SCNA ABN process. In the modified technology, I built an NPN transistor using a pbase layer, all created according to the SCNA layer map as described here: http://www.mosis.com/Technical/Layermaps/lm-scmos_scna.html
   
  (I also removed the P-Well in the N-Transistor, since this process does not have a P-Well.)
   
  There are at least two issues I have run into. First I deliberately included a metal1-metal1 spacing error to see if the DRC would detect the error but it doesn't. Also, I can't seem to connect to the pins of the npn transistor Node, when trying to use the node in a layout. Actually I can connect to them but not without a DRC metal1 spacing error. For other nodes, the layout editor seems to know to connect a metal1 arc to a metal1 port, and without DRC errors.
   
  I've attached the libary to this email. If anyone would check it out and let me know what I'm doing wrong I would appreciate it very much.
   
  Thank you,
  Pierce Keating
    
---------------------------------
  Ahhh...imagining that irresistible "new car" smell?
Check out new cars at Yahoo! Autos. _______________________________________________
Bug-gnu-electric mailing list
[email protected]
http://lists.gnu.org/mailman/listinfo/bug-gnu-electric


    
---------------------------------
  Ahhh...imagining that irresistible "new car" smell?
Check out new cars at Yahoo! Autos. _______________________________________________
Bug-gnu-electric mailing list
[email protected]
http://lists.gnu.org/mailman/listinfo/bug-gnu-electric
_______________________________________________
Bug-gnu-electric mailing list
[email protected]
http://lists.gnu.org/mailman/listinfo/bug-gnu-electric


       
---------------------------------
Ahhh...imagining that irresistible "new car" smell?
 Check outnew cars at Yahoo! Autos.
--0-383363084-1177263864=:3333
Content-Type: text/html; charset=iso-8859-1
Content-Transfer-Encoding: 8bit

<div>Hi Dima,</div>  <div>&nbsp;</div>  <div>The reason I&nbsp;reverted to&nbsp;using version 7 is because I ran into issues converting a technology to a library for editing in version 8.4. Below is a response from Steve Rubin when I first reported this problem. So far, version 7 has been behaving fairly well - an occasional crash but I've been careful about backing up libraries.</div>  <div>&nbsp;</div>  <div>Thanks again,</div>  <div>Pierce</div>  <div>&nbsp;</div>  <DIV id=yiv1849634917>At 10:40 PM 4/1/2007, you wrote:<BR>  <BLOCKQUOTE class=cite cite="" type="cite">I'm using version 8.4.<BR>&nbsp;<BR>Pierce<BR><BR><B><I>Steven Rubin &lt;<SPAN id=lw_1177263441_0 style="CURSOR: hand; BORDER-BOTTOM: #0066cc 1px dashed; HEIGHT: 1em">[email protected]</SPAN>&gt;</I></B> wrote:<BR>  <DL>  <DD>At 12:47 AM 4/1/2007, you wrote:<BR>  <DD>&gt;Hello,<BR>  <DD>&gt;<BR>  <DD>&gt;I think I
  have found what seems to be a bug. I am trying to add a <BR>  <DD>&gt;pbase layer and
 then an npn transistor node to the mocmos <BR>  <DD>&gt;technology. I have tried a simple experiment where I converted a <BR>  <DD>&gt;technology to library for editing, and chose mocmos. Then I tried to <BR>  <DD>&gt;immediately convert the library back to a technology, giving it the <BR>  <DD>&gt;name: mocmos_abn. This failed, with Electric having a problem with <BR>  <DD>&gt;determining the stretching rules for the metal-1 pin layer. At other <BR>  <DD>&gt;times, the problem can be with the node-P-Transistor-Scalable. I've <BR>  <DD>&gt;tried several things and Electric always finds a problem with things <BR>  <DD>&gt;I haven't touched, which is why this really seems like a bug to me.</DD></DL></BLOCKQUOTE>  <DL></DL><BR>I think this has been fixed since 8.4 was released.&nbsp; However, it is not a simple fix...you would need a completely new release.<BR><BR>We plan to make one in le
 ss than 2 months, but that may be a long wait for you.<BR><BR>Alternatively, you could
 hand-edit the technology library to make it more useful.&nbsp; Nodes are described with 4 separate examples plus some text.&nbsp; Each of these 5 pieces of graphical information must be far enough from the others that they can be easily separated.&nbsp; The error message that you are getting indicates they are too close.&nbsp; So you could hand-edit these cells to move them farther apart.<BR><BR>&nbsp;&nbsp; -Steve </DIV>  <div>&nbsp;</div>  <div>&nbsp;</div>  <div>&nbsp;</div>  <div><BR><BR><B><I>[email protected]</I></B> wrote:</div>  <BLOCKQUOTE class=replbq style="PADDING-LEFT: 5px; MARGIN-LEFT: 5px; BORDER-LEFT: #1010ff 2px solid">Hello Pierce,<BR><BR>Why do you use version 7 ? The latest released<BR>Electric version is 8.04.<BR><BR>It will be easier for me to help you with technology<BR>tuning of Electric 8.04,<BR>because I already forgot many things from C Eelctric.<BR><BR>
 For now, I see that Electric 8 reads technology<BR>libraries from Electric 7 with
 errors.<BR>I can look more tomorrow (on my work computer).<BR>However, if you can recreate the technology library<BR>in Electric 8 , then I shall look immediatly<BR>at technology issues in Electric 8 rather than<BR>dealing with incompatibilities of Electric 8 and<BR>Electric 7.<BR><BR>Moreover, I am developing an Xml format for<BR>technology description which will apper in Electric<BR>8.05 .<BR>I beleive that AMI technology example can be a test<BR>of usability of Xml technology description.<BR><BR>-Dima<BR>  <DIV>Hello,</DIV>  <DIV>&nbsp;</DIV>  <DIV>I've since found that the metal spacing&nbsp;error mentioned below does get detected by the DRC. I've also discovered that&nbsp;connecting to the transistor pins is better than I first thought, if I work with it, but I still haven't been able to avoid OVERLAP DRC errors.</DIV>  <DIV>&nbsp;</DIV>  <DIV>Thank you,</DIV>  <DIV>Pierce<BR><BR><
 B><I>Pierce Keating &lt;[email protected]&gt;</I></B> wrote:</DIV>  <BLOCKQUOTE
 class=replbq style="PADDING-LEFT: 5px; MARGIN-LEFT: 5px; BORDER-LEFT: #1010ff 2px solid">  <DIV>Hello,</DIV>  <DIV>&nbsp;</DIV>  <DIV>I have have been attempting to&nbsp;modify&nbsp;the mocmos library for the (MOSIS) AMI 1.5u&nbsp;SCNA ABN&nbsp;process. In the&nbsp;modified technology, I&nbsp;built an NPN transistor using a pbase layer,&nbsp;all created according&nbsp;to the SCNA layer map as described here: <A href="http://www.mosis.com/Technical/Layermaps/lm-scmos_scna.html">http://www.mosis.com/Technical/Layermaps/lm-scmos_scna.html</A></DIV>  <DIV>&nbsp;</DIV>  <DIV>(I also removed the P-Well&nbsp;in the N-Transistor, since this process does not have a P-Well.)</DIV>  <DIV>&nbsp;</DIV>  <DIV>There are at least two issues I have run into. First I deliberately included a metal1-metal1 spacing error to see if the DRC would detect the error but it doesn't. Also, I can't seem to connect 
 to the pins of the npn transistor Node, when trying to use the node in a layout.
 Actually I can connect to them but not without a DRC metal1 spacing error. For other nodes, the layout editor seems to know to connect a metal1 arc to a metal1 port, and without DRC errors.</DIV>  <DIV>&nbsp;</DIV>  <DIV>I've attached the libary to this email. If anyone would check it out and let me know what I'm doing wrong I would appreciate it very much.</DIV>  <DIV>&nbsp;</DIV>  <DIV>Thank you,</DIV>  <DIV>Pierce Keating</DIV>  <DIV>  <HR SIZE=1>  Ahhh...imagining that irresistible "new car" smell?<BR>Check out <A href="http://us.rd.yahoo.com/evt=48245/*http://autos.yahoo.com/new_cars.html;_ylc=X3oDMTE1YW1jcXJ2BF9TAzk3MTA3MDc2BHNlYwNtYWlsdGFncwRzbGsDbmV3LWNhcnM-">new cars at Yahoo! Autos.</A> _______________________________________________<BR>Bug-gnu-electric mailing list<BR>[email protected]<BR>http://lists.gnu.org/mailman/listinfo/bug-gnu-electric<BR></DIV></BLOCKQUOTE><BR>
   <div>  <HR SIZE=1>  Ahhh...imagining that irresistible "new car" smell?<BR>Check out
 <A href="http://us.rd.yahoo.com/evt=48245/*http://autos.yahoo.com/new_cars.html;_ylc=X3oDMTE1YW1jcXJ2BF9TAzk3MTA3MDc2BHNlYwNtYWlsdGFncwRzbGsDbmV3LWNhcnM-">new cars at Yahoo! Autos.</A> _______________________________________________<BR>Bug-gnu-electric mailing list<BR>[email protected]<BR>http://lists.gnu.org/mailman/listinfo/bug-gnu-electric<BR>_______________________________________________<BR>Bug-gnu-electric mailing list<BR>[email protected]<BR>http://lists.gnu.org/mailman/listinfo/bug-gnu-electric<BR></BLOCKQUOTE><BR><p>&#32;

      <hr size=1>Ahhh...imagining that irresistible "new car" smell?<br> Check out
<a href="http://us.rd.yahoo.com/evt=48245/*http://autos.yahoo.com/new_cars.html;_ylc=X3oDMTE1YW1jcXJ2BF9TAzk3MTA3MDc2BHNlYwNtYWlsdGFncwRzbGsDbmV3LWNhcnM-">new cars at Yahoo! Autos.</a>

--0-383363084-1177263864=:3333--



--===============0271101957==
Content-Type: text/plain; charset="us-ascii"
MIME-Version: 1.0
Content-Transfer-Encoding: 7bit
Content-Disposition: inline

_______________________________________________
Bug-gnu-electric mailing list
[email protected]
http://lists.gnu.org/mailman/listinfo/bug-gnu-electric

--===============0271101957==--