[Author Prev][Author Next][Thread Prev][Thread Next][Author Index][Thread Index]
Re: gEDA-user: Investigating gEDA for commercial use
On Sat, Jun 13, 2009 at 9:00 AM, Chris Smith<cjs94@xxxxxxxxxx> wrote:
> We are looking for a new set of EDA tools at work and have investigated
> both CadStar and Altium so far.
Altium did just finally lower their price to something a normal mortal can
afford. At the previous $~4k and up level, from my usage of it in my day job,
I thought the cost was so high so they could support so many bugs.
Package is bloated (tries to do to much, making anything
hard to find), slow (takes over a minute just to start up) and buggy
(it has eaten files on occasion). No one in the company has ever
gotten the autorouter to work properly,
I've given up trying. A fellow new to the company thought he could do it,
and after a week gave up on it too. There is also the hidden Tax of the yearly
update fee.
To that end I've been working with PCB on Windows. The main thing Altium
has over PCB right now is Altium will do blind and berried vias. Doing those
in PCB has been discussed over the years to no results. If you need
to do those,
for now you can take PCB off your list.
I can't say much about gEDA on Windows as the boss likes his archaic OrCAD v7
package for schematics. I wrote a simple net list converter for that
rather than
fight the up hill battle of 10+ years of existing documents in OrCAD v7, to move
to gEDA.
> I have asked my boss to consider gEDA,
> on the basis that it probably does 95% of what we need it to and it may
> be cheaper to pay for/sponsor development of missing features than the
> cost of the alternative tools.
>
> First, are there any experienced gEDA users around the south coast of
> England
I'm about 90 miles from Pittsburgh, can't help you there.
> Second, are there any developers who would consider taking on paid
> development of features? Obviously I don't know what those features may
> yet be, however I can say that not all of us use Linux, so at the least
> we would need:
>
> 1. a maintained Windows binary installer; and
git clone git://repo.or.cz/minipack.git will build Windows versions.
The installer is a bit 'iffy'. The main reason has nothing to do with
the code itself, but rather with the half dozen licenses it has to
have to install. They keep moving around on the host sites,
if you can find them at all.
"The first thing we do, let's kill all the lawyers." - Dick the Butcher
http://shakespeare.mit.edu/2henryvi/2henryvi.4.2.html
The problem is easy to solve (track down the missing licenses), but
time consuming, since I did not need the installer I stopped trying to
build it. I'll give it an other shot
if you do want to try it out.
As to PCB on Windows which I've been using regularly for a few months
it has two major issues that need addressed, before PCB can be a production
ready product for Windows, that I've been poking at fixing
between other projects.
1) The library footprint dialogs are broken, they simply don't work. It has
to do with what ends up being a recursion problem with the configuration
files under Windows due to differences in the shells and path separators.
I have been working around that issue by using NTFS Junction Points
that are, a virtually unknown, feature of XP:
http://support.microsoft.com/kb/205524
The utility you need is here:
http://technet.microsoft.com/en-us/sysinternals/bb896768.aspx
Don't think it would be hard to fix, but didn't try as I had a work around.
Shipping hardware out the door still pays the bills, not hacking the
tools, for the day job.
2) Printing is broken. I did get PostScript Export to work which does
let you get ink on paper, after applying this patch, and using Windows
GhostView:
http://sourceforge.net/tracker/?func=detail&aid=2700352&group_id=73743&atid=538813
I've taken a look at what it would take to get proper printing working
with the GTK HID
the only one supported by Windows right now. I was appalled at how
complex printing is compared to the wxWidget framework that I'm I
normally use for my Windows development projects. I've been debating
on whether if was a better use of my time to figure out the GTK
printing mess, or just to a full up port of PCB in to a wxWidget based
HID. The latter seems easier to me.
Printing proper still needs to work as there are a few limits with the
PostScript export.
> 2. some simple GUI project/workflow manager -- can't really expect the Windows
> users to manually edit project files and use the command line.
I understand the issues from having to explain to my colleagues how to
set up PCB.
If you can convince your boss that this is a paying proposition I'll
move working
on this Windows stuff to the top of my home project list, rather than
in between other
projects.
--
http://www.wearablesmartsensors.com/
http://www.softwaresafety.net/
http://www.designer-iii.com/
http://www.unusualresearch.com/
_______________________________________________
geda-user mailing list
geda-user@xxxxxxxxxxxxxx
http://www.seul.org/cgi-bin/mailman/listinfo/geda-user