开云体育

Looks like I need help again.
I loaded the latest EMC and compiled it. Next I made some changes in the ini and run files for memory and linux versions.
When I try to run EMC I get the following error:

Starting steppermod.o...plat/rtlinux_09J/lib/steppermod.o: unresolved symbol stepperCount can't install it.

Where do I go from here?
Darrell

Darrell Gehlsen wrote:

From: "Darrell Gehlsen" <dgehlsen@...>

Looks like I need help again.
I loaded the latest EMC and compiled it. Next I made some changes in the ini and run files for memory and linux versions.
When I try to run EMC I get the following error:

Starting steppermod.o...plat/rtlinux_09J/lib/steppermod.o: unresolved symbol stepperCount can't install it.

Where do I go from here?
Darrell

Darrell

I don't think steppermod works in this version too well. A couple of
others
and myself were unable to get it to work, but if you use freqmod instead
it
works ok. Also theis version's ini file comes with the rt platform
section
set to sim. Make sure the rt_Plat setting points to your rtlinux_09j.
There
are 2 or 3 areas in the ini file that you need to change.


bill

There is a new build of EMC on the NIST site as of today (Feb 29). The txt
file says it fixes the steppermod problem (But from my experience I would
say that freqmod is the only way to go if your computer has the horse power)
and that backlash is re-enabled in freqmod.
Also some other updates and integration of the work Ray Henry has been doing
on the interface.


Tim
[Denver, CO]

From: "Tim Goldstein" <timg@...>

There is a new build of EMC on the NIST site as of today (Feb 29).

Tim,

Are you still getting less performance from these newer versions than from
the 31 Jan release? I was talking to Fred today and you came up in the
conversation as one of the things that needed attention. If the "leap year"
version doesn't work well (as well as past releases) let me or Fred know and
we'll try to figure out what happened.

Matt

From: Matt Shaver [mailto:mshaver@...]
Tim,

Are you still getting less performance from these newer versions than from
the 31 Jan release? I was talking to Fred today and you came up in the
conversation as one of the things that needed attention. If the
"leap year"
version doesn't work well (as well as past releases) let me or
Fred know and
we'll try to figure out what happened.

Matt

I was wondering why my ears where burning <g>

I just posted a long message on the EMC list about what I have found. The
short version is the problem was self induced. My P value was 1000 because I
missed changing it in generic.ini when I created a new ini file. Seems my
setup wants a value in the 150 - 200 range. Don't quite understand how to
find the optimum setting, but anything in that range is way better than the
1000 in the generic.ini.

Thanks for thinking about me and checking up.


Tim
[Denver, CO]

timg@... <mailto:timg@...>

Well I changed to the freqmod.o and now it loads but I can't get it out of
ESTOP.
I have tried changing all of the defaults from 1 to 0 and still no luck.
Darrell

When you upgrade from one release to the next is there a preferred way to
proceed?
Darrell

Well, I downloaded the 2.2.13 version (29Feb2000) and was able to build
it (almost) out of the box (minor flag changes only) under RH6.0.
That's the first time I was able to get EMC to build under the
2.2.13/rt2.0 kernel without a LOT of work! (Although I have to admit
that >95% of that work was my gross unfamiliarity and newby status with
the EMC source!!) At least the demo works OK - don't have anything
hooked up to that pc to really test it. But hey - its a start! Thanks
Fred Proctor and everyone else!



George Fouse
G F I Consulting


Tim Goldstein wrote:

From: "Tim Goldstein" <timg@...>

There is a new build of EMC on the NIST site as of today (Feb 29). The txt
file says it fixes the steppermod problem (But from my experience I would
say that freqmod is the only way to go if your computer has the horse power)
and that backlash is re-enabled in freqmod.
Also some other updates and integration of the work Ray Henry has been doing
on the interface.

Tim
[Denver, CO]

--