Re: Building EMC software
Posted by
Ray
on 2001-04-10 21:07:18 UTC
Keith
Welcome - comments are mixed in.
From: OUTWATER ~ KEITH J /5G3110 <vac4050@...>
Subject: Building EMC software
<s>emc-1.17 dosen't want to build under this setup. I have a couple of
been trashing Red Hat for shipping 7.0 with an experimental compiler. The
simplist solution is to wait a bit and pay $10.00 for the BDI.
install files that purport to find the correct ones for the system being
used is that the evolution of the New Mexico Tech rtlinux has been so rapid
and the changes so great that there is no way to make on size fit all and
still optimize motion control. In the first releases, Fred at NIST had to
modify the rtlinux code in order to get it to work at all. In more recent
releases, they have been able to make adjustments to the EMC code itself to
match up with the rtlinux being used.
Many of the code files have bunches of ifdef ... endif statements for each
possible combination. It's a lot better than it was a year ago.
Now that the real-time OS universe is settling out into about three or four
approaches and Linus and the core developers are deciding what to do to put
real time in the next major release we should see a more universal EMC
release before to long.
suggest is that you repost this to the emc@... list. I believe that I
read something there where it had been done with RH 7.0. They added a more
stable compiler and libraries and told install to use that instead.
I don't know much about flame tables but I service a couple of them. In
machining, we call it tool offset. We move the tool over enough so that
the final part is the size that it was drawn and programmed to. In cutting
they tend to call it kerf. The effect is the same and yes the EMC can do
tool radius compensation. For kerf you would just enter the diameter of
the flame into the tool table and use g41 or g42 depending on which side of
the line you want the flame to follow.
Ray
Welcome - comments are mixed in.
From: OUTWATER ~ KEITH J /5G3110 <vac4050@...>
Subject: Building EMC software
<s>emc-1.17 dosen't want to build under this setup. I have a couple of
>questions:Yep! You've been caught in the Red Hat confusion. Folks everywhere have
been trashing Red Hat for shipping 7.0 with an experimental compiler. The
simplist solution is to wait a bit and pay $10.00 for the BDI.
>1. EMC seems to be heavily dependent on the specific versions of theGood eye. The biggest reason for the many different directories and the
>kernel and rtlinux. Is there any future plan to use the 'standard' GNU
>autoconf/automake/configure tools to make the build process a little
>easier? for example, I see a lot of copying and linking going on to a
>platform specific directory and there appear to be separate spec/definition
>files for each platorm, OS and rtlinx variant.
install files that purport to find the correct ones for the system being
used is that the evolution of the New Mexico Tech rtlinux has been so rapid
and the changes so great that there is no way to make on size fit all and
still optimize motion control. In the first releases, Fred at NIST had to
modify the rtlinux code in order to get it to work at all. In more recent
releases, they have been able to make adjustments to the EMC code itself to
match up with the rtlinux being used.
Many of the code files have bunches of ifdef ... endif statements for each
possible combination. It's a lot better than it was a year ago.
Now that the real-time OS universe is settling out into about three or four
approaches and Linus and the core developers are deciding what to do to put
real time in the next major release we should see a more universal EMC
release before to long.
>2. If not (or in the meantime), is there an EMC package that will copileYes but not easily with the default compiler sent with RH 7.0 What I'd
>with rtlinux-3.0 and kernel 2.2.18?
suggest is that you repost this to the emc@... list. I believe that I
read something there where it had been done with RH 7.0. They added a more
stable compiler and libraries and told install to use that instead.
>3. Finally, for anyone out there who knows a lot about plasma cutters (Idon't):
I don't know much about flame tables but I service a couple of them. In
machining, we call it tool offset. We move the tool over enough so that
the final part is the size that it was drawn and programmed to. In cutting
they tend to call it kerf. The effect is the same and yes the EMC can do
tool radius compensation. For kerf you would just enter the diameter of
the flame into the tool table and use g41 or g42 depending on which side of
the line you want the flame to follow.
Ray
Discussion Thread
OUTWATER ~ KEITH J /5G3110
2001-04-10 12:18:08 UTC
Building EMC software
Joe Vicars
2001-04-10 12:42:28 UTC
Re: [CAD_CAM_EDM_DRO] Building EMC software
OUTWATER ~ KEITH J /5G3110
2001-04-10 12:55:00 UTC
Re: [CAD_CAM_EDM_DRO] Building EMC software
Paul
2001-04-10 13:12:54 UTC
EMC software - UK only
Tom Benedict
2001-04-10 13:39:41 UTC
Re: [CAD_CAM_EDM_DRO] Building EMC software
William Scalione
2001-04-10 14:57:05 UTC
Re: [CAD_CAM_EDM_DRO] Building EMC software
Ray
2001-04-10 21:07:18 UTC
Re: Building EMC software
Tom Benedict
2001-04-11 05:02:57 UTC
Re: [CAD_CAM_EDM_DRO] Re: Building EMC software
OUTWATER ~ KEITH J /5G3110
2001-04-11 07:42:57 UTC
Re: [CAD_CAM_EDM_DRO] Re: Building EMC software
Ian Wright
2001-04-11 09:35:05 UTC
Re: [CAD_CAM_EDM_DRO] Re: Building EMC software
OUTWATER ~ KEITH J /5G3110
2001-04-11 10:50:30 UTC
Re: Building EMC software