Re: G-code---for the new millenium
Posted by
ballendo
on 2002-03-02 03:49:53 UTC
Guy,
I'm enthusiastic about your wish to make machining simpler for
yourself, and others.
Actually, except for your message, and having looked at it again for
inclusion in my book 1, I hadn't thought about APT for awhile. (As
Fred correctly stated, in many parts of the CNC world, APT is
CONSIDERED to be obsolete. Within the accepted machining paradigm;
which is to let the CAD/CAM do everything.) Used in your new way (as
a conversational or native control language standard), it could
regain ground.
What I'm really "for", is a level of standardisation that makes
machining accessible to more people.
CNC tools are now priced so that many can afford them. Next step,
IMO, is to simplify the process of USING the tool. As the CAD/CAM
approach is the direction industry has taken so far, I've been
working/thinking within THAT arena. Trying to help the new controls
being written by our list members (and others) to meet the de facto
standards of commercial controls. And trying to get the existing
variations of gcode documented. But many in industry will not help,
for reasons mentioned earlier in this thread.
Your message (and the ensuing thread) got me thinking. A CNC control
which allows higher level input would decrease the "need" for
expensive CAM systems for many types of parts. Conversational
controls already do this, but they still convert to gcode. Your
excitement and interest in pursuing a "new" way made me think that
APT could become a "native" language for such a cnc control. Provided
at first as another "conversational" option. The difference from
existing copnversational controls would be that there are people
across the range of the cnc world who already know APT. And
books/programs to support its use. So it could become the BASIC of
CNC.
What this really does is introduce another intermediate level. CAD
will still be the graphic technique, but The CAM portion could use
APT, which would be simpler to understand than gcode (especially if
supported by the CAD graphics)
As someone pointed out to me today in an offlist message, you
can "see" more of a "C" program on one page, than a
comparable "BASIC" listing. "C" is often a less "wordy" language.
(With wordy meaning number of words, not USE of words.) APT is
less "wordy" than Gcode(it can take MULTIPLE blocks of gcode to
achieve one APT statement), so you could "see" more of what is
happening when looking at the APT code, than by looking at the Gcode
for the same part.
Since APT already can be post-processed to Gcode, there is a clear
path for development. First write "conversational" modes based on APT
for existing controls. These could even be separate programs! (Which
would/could "compete" with cad/cam.) Next, get programmers to allow
APT statements to be generated by inexpensive CAD/CAM programs.
(Expensive CAD/CAM already can generate APT output.) Begin training
operators to use APT as the native language instead of Gcode. (Gcode
will be the "assembly" level of machine control. APT could become the
BASIC of machine control)
The hard part will be the paradigm shift required. But paradigm
shifts are often brought about by dedicated individuals...
So the rest is up to you,
Ballendo
P.S. Since I already understand gcode, I'm kind of like a "C"
programmer saying, "You want me to use BASIC?!?!" But for beginners,
having a "machinists' BASIC" available could help. The CAD/CAM guys
will say, "We already have that! It's US!" But you have shown that an
additional approach may have merit...
I'm enthusiastic about your wish to make machining simpler for
yourself, and others.
Actually, except for your message, and having looked at it again for
inclusion in my book 1, I hadn't thought about APT for awhile. (As
Fred correctly stated, in many parts of the CNC world, APT is
CONSIDERED to be obsolete. Within the accepted machining paradigm;
which is to let the CAD/CAM do everything.) Used in your new way (as
a conversational or native control language standard), it could
regain ground.
What I'm really "for", is a level of standardisation that makes
machining accessible to more people.
CNC tools are now priced so that many can afford them. Next step,
IMO, is to simplify the process of USING the tool. As the CAD/CAM
approach is the direction industry has taken so far, I've been
working/thinking within THAT arena. Trying to help the new controls
being written by our list members (and others) to meet the de facto
standards of commercial controls. And trying to get the existing
variations of gcode documented. But many in industry will not help,
for reasons mentioned earlier in this thread.
Your message (and the ensuing thread) got me thinking. A CNC control
which allows higher level input would decrease the "need" for
expensive CAM systems for many types of parts. Conversational
controls already do this, but they still convert to gcode. Your
excitement and interest in pursuing a "new" way made me think that
APT could become a "native" language for such a cnc control. Provided
at first as another "conversational" option. The difference from
existing copnversational controls would be that there are people
across the range of the cnc world who already know APT. And
books/programs to support its use. So it could become the BASIC of
CNC.
What this really does is introduce another intermediate level. CAD
will still be the graphic technique, but The CAM portion could use
APT, which would be simpler to understand than gcode (especially if
supported by the CAD graphics)
As someone pointed out to me today in an offlist message, you
can "see" more of a "C" program on one page, than a
comparable "BASIC" listing. "C" is often a less "wordy" language.
(With wordy meaning number of words, not USE of words.) APT is
less "wordy" than Gcode(it can take MULTIPLE blocks of gcode to
achieve one APT statement), so you could "see" more of what is
happening when looking at the APT code, than by looking at the Gcode
for the same part.
Since APT already can be post-processed to Gcode, there is a clear
path for development. First write "conversational" modes based on APT
for existing controls. These could even be separate programs! (Which
would/could "compete" with cad/cam.) Next, get programmers to allow
APT statements to be generated by inexpensive CAD/CAM programs.
(Expensive CAD/CAM already can generate APT output.) Begin training
operators to use APT as the native language instead of Gcode. (Gcode
will be the "assembly" level of machine control. APT could become the
BASIC of machine control)
The hard part will be the paradigm shift required. But paradigm
shifts are often brought about by dedicated individuals...
So the rest is up to you,
Ballendo
P.S. Since I already understand gcode, I'm kind of like a "C"
programmer saying, "You want me to use BASIC?!?!" But for beginners,
having a "machinists' BASIC" available could help. The CAD/CAM guys
will say, "We already have that! It's US!" But you have shown that an
additional approach may have merit...
--- In CAD_CAM_EDM_DRO@y..., "Guy Sirois" <guys@w...> wrote:
>
> Hi Ballendo,
>
> I see you are quite enthusiastic about APT. Or any other
smart language,
> for that matter.
>
> APT is much more advanced and powerful than what I was talking
about.
> in my original post, yet it still seems natural, because it is like
plain
> English.
> It looks like it was invented by somebody who didn't know about G-
code, but
> who was very knowledgeable in machining and CAD.
>
> It just needs a better accessibility to the general public, maybe
by having
> controllers that accept this language. It would also benefit from
being
> "standardized" before it transforms into a multitude of flavors
like G-code.
> That would be the job of an international CNC machining
organization I
> suppose. Is there such a thing already in the world? I suppose yes.
>
> What are your thoughts regarding the implementation of APT at OUR
level? As
> we don't have the smart controllers yet, I suppose it would be the
first
> step ?
>
> It would also be interesting to read opinions of people already
using it.
>
> Thanks
>
> Guy
>
>
> -----Original Message-----
> From: ballendo [mailto:ballendo@y...]
> Sent: Friday, March 01, 2002 1:07 AM
> To: CAD_CAM_EDM_DRO@y...
> Subject: [CAD_CAM_EDM_DRO] Re: G-code---for the new millenium
>
>
> Guy,
>
> Good to see you haven't given up. APT is a "C" of machine control.
> A "portable" language. That is what it was designed to be.
>
> Some thoughts in response to the ORIGINAL thrust of this thread, as
> understood by me:
>
> APT is designed to be used with post-processors to "port" it to the
> various machines. But what if the "post processor" became PART of
the
> CONTROL? Instead of part of the CAD/CAM, where it has been
> traditionally? A paradigm shift to be sure, but one whose time is
> possible, at least technically. This is what got me excited about
> your quest, and prompted my comments about APT in the first place.
> (Others had already mentioned it)
>
> There is going to be strong opposition to this, as many people make
> LOADS of money supporting and writing these "POSTS" (post-processor
> writing and support has been a staple of CAM software income for
> years.
> (I EVEN had one CAM industry person tell me that he didn't want to
> contribute his "considerable" information to the general knowledge
of
> gcode variations between different controls, as this would result in
> his making less money accomodating the differences.)
>
> We're at the point where the CONTROL can become "smarter", so its
> input can be made "dumber" (man readable, as you first suggested).
>
> A similar possibility existed in the PC programming world several
> years back. It "would have been" possible to make the compilers (the
> equivalent of our cnc controls) smarter, so that the Source code
> (input by the programmer) could be simpler. But alas, C took over,
> with its' IMO obtuse(hard to understand) syntax. So now, the
> programming tools have gone "visual", making it easier to implement,
> but the language is still obtuse. We "could have had" a layman-
> understandable native code, like BASIC or even Pascal. Or even
> something simpler, allowing more people to become programmers.
> Instead we have preserved the "white coat status" of programming,
> tho' the new tools DO make it easier for newbies.
>
> So the same thing in CNC: We have traditionally used
> relatively "dumb" controls. And we "feed" them the output
> from "smart" CAM programs...
>
> So again I ask, "What if the control were made smarter so that the
> language used to feed the control became understandable and
> immediately accessible to MANY more people?" Isn't this what your
> message originally wished for?
>
> Having APT, or something like it, as the NATIVE language of the CNC
> control would promote this. As you say below, you "could almost see
> the toolpath..." Admittedly, so can experienced gcode users, but you
> have to look a lot more gcode, since its notation refers to a lower
> level (like looking at source for assembly, when compared to looking
> at BASIC, or Pascal, or C/C++ source)
>
> We're doing a "version" of this already. EMC, like most modern
> controls, includes expression evaluation. Other controls, like HAAS,
> include "conversational" programming. The controls ARE getting
> smarter. But like the "C++" compilers which still use "C's" "need to
> be trained first to even have an IDEA what you are looking at"
> syntax, the cnc controls are sticking with the similarly
> obtuse "baggage" of gcode, instead of using something higher level,
> like APT. Which WOULD be a better way to proceed. We now "count on"
> the CAD/CAM; what if we instead "counted on" the cnc control? Move
> the "smarts" down a level.
>
> APT can be, and has been, made visual. It's just a language; its
> implementation is up to the programmer using it. And that programmer
> could make a windows interface, if one doesn't exist already in an
> inexpensive version.
>
> I hope this clarifies what I was trying to say.
>
> Ballendo
>
> P.S. For those who are not programmers, "C" and "C++, pronounced SEE
> and SEE-plus-plus, are the primary programming languages for most
> professional computer programmers today. EMC is written in "C". BUT
> BASIC (a simpler to immediately understand programming language) is
> STILL the most "popular" programming language.
> I will explain the other programming terms used in my message above
> to anyone who cares to send me an offlist message.
Discussion Thread
Guy Sirois
2002-02-25 09:01:12 UTC
G-code---for the new millenium
Paul Amaranth
2002-02-25 09:20:42 UTC
Re: [CAD_CAM_EDM_DRO] G-code---for the new millenium
Jon Elson
2002-02-25 09:31:02 UTC
Re: [CAD_CAM_EDM_DRO] G-code---for the new millenium
Kevin P. Martin
2002-02-25 09:33:00 UTC
RE: [CAD_CAM_EDM_DRO] G-code---for the new millenium
stephen_stallings
2002-02-25 09:36:15 UTC
Re: G-code---for the new millenium
Carol & Jerry Jankura
2002-02-25 09:37:11 UTC
RE: [CAD_CAM_EDM_DRO] G-code---for the new millenium
audiomaker2000
2002-02-25 11:17:21 UTC
Re: G-code---for the new millenium
Chris L
2002-02-25 19:46:20 UTC
Re: [CAD_CAM_EDM_DRO] G-code---for the new millenium
ballendo
2002-02-26 03:13:03 UTC
Re: G-code---for the new millenium
ballendo
2002-02-26 03:38:55 UTC
Re: G-code---for the new millenium
wanliker@a...
2002-02-26 07:04:09 UTC
Re: [CAD_CAM_EDM_DRO] Re: G-code---for the new millenium
Tony Jeffree
2002-02-26 09:11:10 UTC
Re: G-code---for the new millenium
Guy Sirois
2002-02-26 12:57:20 UTC
RE: [CAD_CAM_EDM_DRO] Re: G-code---for the new millenium
imserv1
2002-02-26 14:39:32 UTC
Re: G-code---for the new millenium
Guy Sirois
2002-02-26 15:46:38 UTC
RE: [CAD_CAM_EDM_DRO] Re: G-code---for the new millenium
imserv1
2002-02-26 19:05:17 UTC
Re: G-code---for the new millenium
Raymond Heckert
2002-02-26 19:48:36 UTC
Re: [CAD_CAM_EDM_DRO] Re: G-code---for the new millenium
catboat15@a...
2002-02-27 13:28:23 UTC
Re: [CAD_CAM_EDM_DRO] Re: G-code---for the new millenium
ballendo
2002-02-28 00:22:59 UTC
APT shareware was Re: G-code---for the new millenium
ballendo
2002-02-28 00:33:14 UTC
OT re: Dragon was Re: G-code---for the new millenium
ballendo
2002-02-28 01:35:02 UTC
Re: G-code---for the new millenium
imserv1
2002-02-28 07:12:59 UTC
Re: G-code---for the new millenium
Guy Sirois
2002-02-28 08:00:05 UTC
RE: [CAD_CAM_EDM_DRO] Re: G-code---for the new millenium
Guy Sirois
2002-02-28 08:15:42 UTC
RE: [CAD_CAM_EDM_DRO] Re: G-code---for the new millenium
Kevin P. Martin
2002-02-28 09:28:38 UTC
RE: [CAD_CAM_EDM_DRO] Re: G-code---for the new millenium
Bill Vance
2002-02-28 09:36:31 UTC
Re: [CAD_CAM_EDM_DRO] Re: G-code---for the new millenium
Carol & Jerry Jankura
2002-02-28 09:58:26 UTC
RE: [CAD_CAM_EDM_DRO] Re: G-code---for the new millenium
Bill Vance
2002-02-28 10:22:05 UTC
Re: [CAD_CAM_EDM_DRO] Re: G-code---for the new millenium
myjakjs
2002-02-28 13:20:31 UTC
Re: G-code---for the new millenium
Guy Sirois
2002-02-28 14:36:45 UTC
RE: [CAD_CAM_EDM_DRO] Re: G-code---for the new millenium
Guy Sirois
2002-02-28 15:17:32 UTC
RE: [CAD_CAM_EDM_DRO] Re: G-code---for the new millenium
ballendo
2002-03-01 01:07:03 UTC
Re: G-code---for the new millenium
Guy Sirois
2002-03-01 07:38:35 UTC
RE: [CAD_CAM_EDM_DRO] Re: G-code---for the new millenium
Hugh Currin
2002-03-01 07:52:50 UTC
Re: G-code---for the new millenium
myjakjs
2002-03-01 10:35:39 UTC
Re: G-code---for the new millenium
ballendo
2002-03-02 03:49:53 UTC
Re: G-code---for the new millenium
ballendo
2002-03-02 04:11:04 UTC
Re: G-code---for the new millenium
Carol & Jerry Jankura
2002-03-02 06:13:53 UTC
RE: [CAD_CAM_EDM_DRO] Re: G-code---for the new millenium
grouchy_old_fred
2002-03-02 15:32:30 UTC
Re: G-code---for the new millenium
catboat15@a...
2002-03-02 21:34:05 UTC
Re: [CAD_CAM_EDM_DRO] Re: G-code---for the new millenium
Chris Clough
2002-03-03 14:02:06 UTC
US Digtal Encoders - Just a follow up
Lew
2002-03-03 15:23:37 UTC
RE: [CAD_CAM_EDM_DRO] US Digtal Encoders - Just a follow up
Chris Clough
2002-03-03 15:26:08 UTC
RE: [CAD_CAM_EDM_DRO] US Digtal Encoders - Just a follow up
Lew
2002-03-03 15:35:41 UTC
RE: [CAD_CAM_EDM_DRO] US Digtal Encoders - Just a follow up
Kevin P. Martin
2002-03-05 21:35:04 UTC
RE: [CAD_CAM_EDM_DRO] Re: G-code---for the new millenium
Carol & Jerry Jankura
2002-03-06 05:47:58 UTC
RE: [CAD_CAM_EDM_DRO] Re: G-code---for the new millenium
Bill Vance
2002-03-06 07:49:29 UTC
Re: [CAD_CAM_EDM_DRO] Re: G-code---for the new millenium
Terry L. Ridder
2002-03-06 08:14:50 UTC
real time extensions for ms windows ( was G-code---for the new millenium )
Bill Vance
2002-03-06 19:02:16 UTC
Re: [CAD_CAM_EDM_DRO] real time extensions for ms windows ( was G-code---for the new
Kevin P. Martin
2002-03-07 00:35:27 UTC
RE: [CAD_CAM_EDM_DRO] Re: G-code---for the new millenium