On Tue, Sep 06, 2011 at 01:42:42AM +0100, andy pugh wrote:
> I tried a 4-quadrant circle with an AIO change at 12, 3, 6, 9 'o'clock.
> First, M68: Timing is good, but velocity stutters:
> http://imagebin.org/171202
> Then with the M67 patch:
> http://imagebin.org/171203
> The velocity still stutters fractionally, but is much better.
> I don't understand why both have pretty much equally good timing, that
> doesn't fit with my understanding of the motion queue.
May you share gcode used for tests?
> The patched version badly fails the test suite:
> Runtest: 49 tests run, 38 successful, 11 failed + 0 expected
> Failed:
> .
> ./tests/ccomp/lathe-comp
> ./tests/ccomp/mill-g90g91g92
> ./tests/ccomp/mill-line-arc-entry
> ./tests/interp/cam-nisley
> ./tests/interp/g10/g10-l11
> ./tests/interp/g10/g10-l1-l10
> ./tests/interp/g10/g10-l1
> ./tests/interp/g76
> ./tests/m70-m73/m70m72-restore.0
> ./tests/m70-m73/m73autorestore.0
Hm, all tests for me are working fine...
Runtest: 43 tests run, 43 successful, 0 failed + 0 expected
I've tested this on v2.5_branch. What's your branch?
> There is an implicit limit of 32 of each channel as iomask_t is a u32.
> This would need to change as current EMCMOT_MAX_DIO is 64 and
> EMC_MOT_MAX_AIO is 16. (Both seem low to me)
I thought that both are 16 :) So 64 bit mask is needed...
Patches are available in m67-fix branch at my repo [1]
Pavel
--
[1] http://psha.org.ru/cgit/psha/emc2.git/log/?h=m67-fix
------------------------------------------------------------------------------
Special Offer -- Download ArcSight Logger for FREE!
Finally, a world-class log management solution at an even better
price-free! And you'll get a free "Love Thy Logs" t-shirt when you
download Logger. Secure your free ArcSight Logger TODAY!
http://p.sf.net/sfu/arcsisghtdev2dev
_______________________________________________
Emc-developers mailing list
Emc-developers@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/emc-developers
