In a message dated 5/26/2001 7:19:42 AM Eastern Daylight Time,
[EMAIL PROTECTED] writes:
> Wanted to get your opinion on the translator....how accurate was it on large
> designs, do you have a good confidence level with the translation?
Wasn't bad...no, I never have any confidence in anything translated, but the
inspection of 2500 of the 6000 or nets was correct.
The only problem I had stemmed from the fact that the PADS designer decided
to use routing layers for planes via copper pour (4mil trace 4 mil space). I
verified this by looking at the gerbers he generated. He had over 40 pours on
4 of the routing layers. They were the size of the board, he should have used
split planes.
The netlist looked good, the rules were a bit funky 'cause Protel had
interpreted two rules as global (board). This did cause some issues with
manual editing.
I'm doing a reduced version from scratch (capture and all), probably have
10-12 layers and hopefully no more than 1000 components. If it gets sticky, I
may need to purchase a seat of CCT; I guess I'll get a second mortage on my
house so I can miter corners...
BTW: I got my firmware geek (we call hin the Code Wookie - he's one of the
OpenBSD coders) doing the schematics. I showed him how I generate the
schematic parts from the Acrobat datasheets. He's improved on it (for one,
xPDF has some nice features for this) and wrote some Perl scripts. He got the
901BGA scheme symbol completed in a few hours with NO TYPING!!
What we do is to translate the PDF file's pin lists into a Protel ASCII lib
part. The thing comes in as rows and rows of pins. This considerably reduces
the chance of us tranposing pin names/numbers. If it's screwed, it's screwed
in the data the vendor provided.
Why doesn't anyone compete with Specctra? I personally can't stand either
Orcad or Cadence. Someone should take something like the Bartels engine or
the Protel code, fix it. add the control and make it work with Specctra
files. Their patent has to be expired by now (17 years max back then; 20 for
newer patents filed recently).
As to PADS, I did work for them 12 years ago and made most of the discrete
and RF parts for their SCH/PCB libraries. I was to get an upgrade for the
work - never saw a cent or software. But they used the parts for their lib's;
>From what I saw a year ago, they're still using some of the parts I did.
Back to Specctra, do you think its' worth the extra $$$ for all the
options??? I say no but then again for RF I use Genesys. Heck, the whole
suite cost us a fraction of the Specctra base price and it does the design
capture/simulations. Not EEsoft but well worth it.
I did find an online specctra service that charges 5-15c per pin. Just upload
the dsn file and a dolist and away it goes. Gives you back the routes.
Call me if you want to hook up, I''m at 703-392-8575. I can show you the
board from hell. I'm here in Manasshole VA.
Wayne
WAM
* * * * * * * * * * * * * * * * * * * * * * * * * * * * * *
* To post a message: mailto:[EMAIL PROTECTED]
*
* To leave this list visit:
* http://www.techservinc.com/protelusers/subscrib.html
* - or email -
* mailto:[EMAIL PROTECTED]?body=leave%20proteledaforum
*
* Contact the list manager:
* mailto:[EMAIL PROTECTED]
*
* Browse or Search previous postings:
* http://www.mail-archive.com/[email protected]
* * * * * * * * * * * * * * * * * * * * * * * * * * * * * *
