On 1/26/2012 6:32 AM, andy pugh wrote: > On 25 January 2012 21:44, Kenneth Lerman<kenneth.ler...@se-ltd.com> wrote: > >> If this were old style code (XA YB) it would not be valid because it >> contains two letters in a row. By eliminating the early whitespace removal, >> 'XA YB' and 'AXYB' would mean two different things. 'X 123' and 'X123' would >> still be interpreted identically. > Just to throw another idea into the mix, would it make sense to leave > the G-code as it is, and accept a different command language, perhaps > STEP-NC? > Andy:
That's an interesting question. I've often wondered why there weren't folks working on APT interpretation even though it is old technology. I know a few true-believers have been keeping the APT flame alive. I'd think, though, that STEP-NC, officially ISO 10303 Part 238, is a bit rich for our blood. It's much more complicated than G-code and no normal mortal is likely to create or even edit STEP-NC files by hand, at least not after their first one :-) I've been there and done that with test files conforming to other STEP parts. I left the STEP stage before the STEP-NC project had advanced from creating paper to creating software, but at least at a meta-level I know how its technology is constructed. Three things I see arguing for a more serious look at it are 1) it's based on industry-validated conceptual- and data-information models so there is semantic content to its data structures, 2) the information modeling started with interrelated concepts like "operation", "working step", "feature", and "work piece" rather than just raw machine motion-commands so there is a coherent view of machining technology, tool, toolpath, and intended result, and 3) there is a plethora of software tools available for the underlying STEP basics like parsing the STEP part 21 file format, populating STEP class library instances, etc. (see http://www.step-nc.org/index.htm). Unfortunately, that plus a Euro or two are only sufficient to get you a cup of coffee. Is there any LinuxCNC-user in possession of a CAM system with a STEP-NC postprocessor? Without it, considering STEP-NC seems an academic exercise. With it, one could create some interesting CAD/CAM model, produce both a G-code file and a STEP-NC file from it, and by comparison get an inkling of what's likely to be involved in interpreting the latter for LinuxCNC. At one time, Step Tools, Inc., offered demo plug-ins for several CAM systems but I don't see them on their website now. I just searched Google quickly and didn't come up with any hits on notions like extracting G-code from STEP-NC or creating a G-code-equivalent profile for STEP-NC, only some technical papers on the obvious transition problem of creating STEP-NC content from existing G-code files. This doesn't prove such things don't exist, only that I didn't ferret them out on a first pass. As always, just my 2-cents worth. Regards, Kent ------------------------------------------------------------------------------ Keep Your Developer Skills Current with LearnDevNow! The most comprehensive online learning library for Microsoft developers is just $99.99! Visual Studio, SharePoint, SQL - plus HTML5, CSS3, MVC3, Metro Style Apps, more. Free future releases when you subscribe now! http://p.sf.net/sfu/learndevnow-d2d _______________________________________________ Emc-developers mailing list Emc-developers@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/emc-developers
