SW2008 vs Inventor 11

[ I've been away from SolidWorks for almost ten years. At this point I spend the majority of my day on SolidWorks 2007. The program has changed so much in ten years that I'm still way behind where I need to be. ] -Jon Banquer - Aug. 26, 2007

[ He pushes UG as THE CAD/CAM solution. Can't think of worse advice for many small job shops ] - Jon Banquer - [ If I were to guess at this point what system is best it would be UG ] - Jon Banquer-

[ The beautiful thing about working with this 1/2 million dollar horizontal machine with 12 pallets is that I spend very little time running parts and spend the majority of my time on programming, setting up and optimizing the program. ] - Jon Banquer - 12 May 2007

You have a machine designed for high production. You are glad that your machine spends little time running parts while you are programming, setting up and fixing programming errors (AKA: optimizing the program) at the control...........LOL........... only in Banquerland could someone be happy about that.

Oh! Jon, BTW your machine only has 11 pallets, since you can't count check the large sticker on the front of your machine, in large letters it tells you and everyone else how many pallets your machine has.

Are you still working for this company? How long did your employment last before you "left"?

Jon, how does a post processor compensate for something that is not done in the "CAM program"?

Oh! That's right you don't know anything about post processors:

[ Some of our posts do need some work. Fix>"You can do more in MasterCAM to verify tool path and with a good post

So what your saying is you can't program using MasterCAM or do a quick and easy fix to your post processor so it will output good, reliable G-Code for you 100% of the time?

And how do you keep associativity with the solid using NC-Plot? ] -Tom-

[

I know I'm wasting my time on Jon; but there may be others here who actually watch what's posted in order to learn something. (There's another thread about that, so I figure I should practice what I preach.)

Despite Jon's rantings, I use AutoCad because it works for me. That's not an endorsement, it's just a fact. Other stuff might work too; but AC is what I know, and it does what I need.

What it doesn't do, despite Jon's rantings, is create or verify program code. That's because it's a CAD program (Computer Aided Design), and it's not supposed to be a CAM program (Computer Aided Machining) which generates CNC programs. My need, my purpose, and my use of AutoCad all have to do with designing things for the real world.

I build machine tools, so ordinary isometric drawings, or an occasional 3D wireframe to help with visualization of isometric drawings, is what my work requires. Anything with more bells and whistles would just get in my way.

With regard to "virtual simulation software that verifies actual G code": That's mostly nonsense. There was a time, decades ago, when writing CNC code was an unfamiliar task for most machinists. I know because I spent a lot of time in the '70's and early '80's teaching people how to do it. Eventually, however, it became more common, and then very common, and then pretty much a necessary and standard part of what any machinist knew and took for granted. In fact, I'd expect anyone who calls himself a machinist today to be able to write basic code almost without thought, as easily as he'd write English. (No, not fancy shapes or instant calculations; but the basic format of a program for a lathe or 3 axis machining center.)

So, the task of generating code for a machine should be no more difficult than the task of generating text for a newsgroup post. What's NOT so easy, or so common, is knowing WHAT code to generate. Having a fancy piece of software to "verify G Code" doesn't make you a machinist or a programmer anymore than having a spell checker makes you a poet or an accomplished author.

The real skill involved is knowing what to write. That means knowing how to process a workpiece to be machined, how to hold it, what tools are best for which cutting operations, what speeds and feeds and depths of cut will give the best results, and more. Virtual simulation software doesn't do that. It can't, and it's not supposed to.

And it can't make a machinist out of a CAM operator anymore than MS Word could make a functional illiterate into Earnest Hemmingway.

Besides, the machines I currently work with most often are:

1. My own machines for grinding high performance drill points, whose entire user inteface is a custom CAM system I wrote myself for this very specific application.
2. Mazak machines, whose Mazatrol front end is itself a fit-to-the-equipment CAM system, and which doesn't read, or need, code generated off-line except for unusual situations.
3. CNC Swiss type machines, that have so many slides, so many tools, and so many different possible ways of doing things that nobody offers a postprocessor for them anyway, even on the most sophisticated CAM systems.

Wherefore, I don't need or care about the virtual reality that Jon lives in. And, in my humble (but extensively experienced opinion), neither does anybody else. ]-KG-

"I would be willing to tutor him if he can learn to treat others with respect by finding a new way to communicate, much like I have by refusing to engage in personal attacks" - Jon Banquer -

"That's why I moved on and that's why I don't post in the SolidWorks newsgroup any longer. I truly like the company." - Jon Banquer

[ Franco Folini

UPDATE -- July 8, 2007 -- I had to close this blog post to further comments and to remove the personal attacks between Jon and some other newsgroups readers. Before the interview, I made an agreement with Jon about the style of the interview and the way to handle it. Jon didn't respect our agreement, posting comments under fake names. Jon's authentic and fake comments are all posted from the same IP address,

72.199.251.224. I can now see that my trust in Jon was misplaced. ]

What does it say about you Jon, forging comments and posting them to your "interview" to make yourself look better, in direct violation of your agreement with the author?

iQ here to respond to the actual question, unlike the babble of others.

I am in the middle of evaluating both Inventor and WildFire over our current CAD SWx. I have had second demos now of both, these were more focused on our product design.

First of all I would like to state that the pop-ups in these other CAD programs become a nuisance.

And WildFire is the worse that I saw. I saw multiple times where there were 3 layers of pop-up windows that opened over the top of each other. A very well trained user did about 27 picks in 3 pop-ups in about 15 seconds. This alone caused me to cringe as I would never become that understanding of a program to be able to go this fast. Just knowing about these other pop-ups would be draining. I would expect that it would take me 3-6 months to become proficient in this program with about a year to understand this program fully. Now their FEA was excellent, a far cry better then the cosmos that we have today. But in all the interface was not up to what I would have expected in this program. They have had extensive time to refine this, and wildfire was a great leap in the right direction. But with all that I have seen it has not progressed since we looked at this program 6+ years ago. Still looked difficult to use.

Inventor was lacking. I do not believe that AutoDesk is applying enough resources to this effort. The demo person was less than superior. He had been using the software for 2 weeks so he did not know enough about the software to be a presenter. Most of this second demo was canned show and tell with regulated bar charts. We had sent them one of our assemblies to open and show us program effectively and all they brought up was a small sub-assembly. The sales man was not sufficient for the technical end and the manager of the VAR had little to help in the situation. It is like they were looking to get a management approval where we leave CAD software decisions up to the actual users. Also this 2 week user was the technical staff that I would be contacting for support, not good. Good news is that they would give us a 50% cost break to use their software. Other items that I liked was how it handles bolt stackups and associated holes, very good. But even their toolbox did not have everything. The first toolbox item we asked for was a torx head cap screw, not there, or any torx fasteners. It was still more than WildFire had.

Overall they were both good CAD packages and are good for mechanical design. Inventor looks like it is easier to use if starting from scratch. Wildfire we will have to see. Both CAD programs demonstrated crashed in the demos.

After what we have seen, SolidWorks is starting to look better and better. Who would have thought that SolidWorks would look so good, just demo the others.

Anyway I am still in evaluation phase, will write more as I get it. iQ

iQ here to respond to the actual question, unlike the babble of others.

I am in the middle of evaluating both Inventor and WildFire over our current CAD SWx. I have had second demos now of both, these were more focused on our product design.

First of all I would like to state that the pop-ups in these other CAD programs become a nuisance.

And WildFire is the worse that I saw. I saw multiple times where there were 3 layers of pop-up windows that opened over the top of each other. A very well trained user did about 27 picks in 3 pop-ups in about 15 seconds. This alone caused me to cringe as I would never become that understanding of a program to be able to go this fast. Just knowing about these other pop-ups would be draining. I would expect that it would take me 3-6 months to become proficient in this program with about a year to understand this program fully. Now their FEA was excellent, a far cry better then the cosmos that we have today. But in all the interface was not up to what I would have expected in this program. They have had extensive time to refine this, and wildfire was a great leap in the right direction. But with all that I have seen it has not progressed since we looked at this program 6+ years ago. Still looked difficult to use.

Inventor was lacking. I do not believe that AutoDesk is applying enough resources to this effort. The demo person was less than superior. He had been using the software for 2 weeks so he did not know enough about the software to be a presenter. Most of this second demo was canned show and tell with regulated bar charts. We had sent them one of our assemblies to open and show us program effectively and all they brought up was a small sub-assembly. The sales man was not sufficient for the technical end and the manager of the VAR had little to help in the situation. It is like they were looking to get a management approval where we leave CAD software decisions up to the actual users. Also this 2 week user was the technical staff that I would be contacting for support, not good. Good news is that they would give us a 50% cost break to use their software. Other items that I liked was how it handles bolt stackups and associated holes, very good. But even their toolbox did not have everything. The first toolbox item we asked for was a torx head cap screw, not there, or any torx fasteners. It was still more than WildFire had.

Overall they were both good CAD packages and are good for mechanical design. Inventor looks like it is easier to use if starting from scratch. Wildfire we will have to see. Both CAD programs demonstrated crashed in the demos.

After what we have seen, SolidWorks is starting to look better and better. Who would have thought that SolidWorks would look so good, just demo the others.

Anyway I am still in evaluation phase, will write more as I get it. iQ

I'm a longtime user of both Swx and ProE (Wildfire now). ProE has made some strides in user interface in recent years, but is still way behind Swx. Just compare things like "hole wizard", sketching, and error resolution. Swx beats it hands down. Of course, it depends on what you are doing and if you don't mind spending a little extra for the add-ons in ProE (thinking ISDX here). Haven't used Inventor, but it looks half-baked.

Jon,

You where the one not handling change and was complaining that the machine you were running didn't have G12 & G13 canned cycles. I told you to stop complaining about it and program the canned cycle yourself.

I even spent my time to help you by writ>

]

Tom

*

From: snipped-for-privacy@aol.com Newsgroups: alt.mach>Any one know the reason why Fanuc does not have a G12 / G13 like

Fanuc used to have this and for some reason dropped it many years ago.

If it's that valuable a feature for you why don't YOU write one for yourself?

Here is a start for you.

You stated that your company uses a lot of macros so I did not use any #500 variables. I used local variables that are reset every time the macro is called up.

I left you some work if you want it to emulate G12, G13. If you want to make the macro modal like a normal G-Code canned cycle someone who knows how to program macros can easily do it.

If you want the option of CW or CCW that is easy enough to do. I laid it out for you and commented the macro much more than I normally do to help those reading it here in this forum to understand it without having to study it too closely.

There are a lot of ways to go about it, I am showing just one way. There are a lot of ways it can be improved to add more functionality but YOU have to START somewhere and I am not going to do ALL the work for you.

If your machine supports it you can call this 9000 program (macro) two ways.

G65 P9013 X Y Z I D R F Q

or

G113 X Y Z I D R F Q

******************************************

If you want to see how this macro works you can download a trial version of NCPlot at:

***********************************************I just wrote this macro and have not tested it.

***********************************************

O0001 (TEST SUB-PROGRAM MACRO) G40 G17 G80 G90 G49 G20 G54 G94 M6 T1 S2500 M03 G00 X0.0 Y0.0 /M08 G43 H1 Z1.0 G65 P9013 X0.0 Y0.0 Z-1.0 I10.0 D.5 R.2 F10.0 Q3.0 G00 Z1.0 M9 G40 G0 Z4.0 M5 G91 G28 Z.0 M30

O9013 (CCW CIRCLE WITH RADIUS LEAD IN/OUT) (SIMILAR TO G13 BUT NOT THE SAME AS) (RADIUS LEAD IN AND OUT IS HALF THE HOLE RADIUS) IF [#17LT0] THEN #3000=3 (PLUNGE FEEDRATE Q LESS THAN ZERO) IF [#9LE0] THEN #3000=4 (NO FEEDRATE GIVEN) IF [#4LE0] THEN #3000=5 (NO HOLE DIAMETER GIVEN) IF [#7LE0] THEN #3000=6 (NO TOOL DIAMETER GIVEN) IF [#4LE#7] THEN #3000=7 (TOOL DIA. TOO LARGE) IF [#24LT0] THEN #3000=8 (NO X LOCATION GIVEN) IF [#24LT0] THEN #3000=9 (NO Y LOCATION GIVEN) IF [#26EQ#0] THEN #3000=10 (NO Z HOLE DEPTH GIVEN) IF [#18LT0] THEN #3000=11 (NO CLEAR POINT GIVEN) IF [#18LT#26] THEN #3000=12 (RAPID CLEAR LT HOLE DEPTH)

#8=#4003 (G90/G91 STORE CURRENT MODE) #1=#5003 (STORE CURENT Z POS.) #2=[#26-#18] (CALCULATE INCREMENTAL Z DEPTH) #5=[[#4-#7]/2] (SUBTRACT TOOL DIA FROM HOLE DIAMETER CONVERT DIAMETER TO RADIUS) #6=#5/2 (LEAD IN AND OUT 1/2 CIRCLE RADIUS) #100=#17GT0 (IF Q IS LARGER THAN 0 FEED TO DEPTH) IF[#17LT0]THEN#100=1 (IF Q IS 0 RAPID TO DEPTH)

IF[#17LT0]THEN #3000=1(PLUNGE FEED WRONG Q VALUE)

G#8G0X#24Y#25 (G90/G91 RAPID TO HOLE CENTER) G0Z#18 (RAPID TO CLEAR Z POINT) IF[#17GT0] THEN GOTO1 (IF FEEDING TO Z DEPTH GO TO N1) IF[#17EQ0] THEN GOTO2 (IF RAPID TO Z DEPTH GO TO N2) N1 F#17 (PLUNGE FEED RATE) N2 G91 G#100 Z#2 (RAPID OR FEED TO Z END DEPTH) G91G3X#5Y0I#6J0F#9 (RADIUS LEAD IN WITH CIRCLE FEED RATE) G3I-#5 (CIRCLE) G90G3X#24Y#25I-#6J0 (RADIUS LEAD OUT BACK TO START POSITION) G0G90Z[#1] (RAPID TO INITIAL POINT) G[#8] (G CODE BACK TO PREV. G90/91) M99 %

YOU MUST SPECIFY X, Y, Z, I, D, R, F AND Q VALUES IN SUB-PROGRAM CALL OR IT WILL GENERATE AN ERROR.

MAIN PROGRAM CAN BE ABSOLUTE OR INCREMENTAL POSITIONING.

MACRO STORES G90/91 SETTING, RUNS THE MACRO THEN RESETS BACK TO STORED MODE BEFORE RETURNING TO MAIN PROGRAM.

(X Y POSITION TO CENTER OF HOLE) (Z DEPTH/END) (I HOLE DIAMETER) (D TOOL DIAMETER) (R RAPID START/END POSITION) (F FEED RATE FOR CIRCLE) (Q PLUNGE FEED, IF Q=0.0 WILL RAPID TO DEPTH)

Tom

***********************

From: snipped-for-privacy@aol.com Subject: Re: G13 FANUC MACRO , Corrected Macro Date: Mon, 28 May 2007 19:19:33 -0700 Organization: An InterNetNews Site

There was a long line where I thought the line wrapped but it didn't. It was a key stroke error that created an EOB dividing the comment in two.

Corrected this line from the original.

#5=[[#4-#7]/2] (SUBTRACT TOOL DIA FROM HOLE DIAMETER CONVERT DIAMETER TO RADIUS)

********************** Edited program below ***********************

O0001 (TEST SUB-PROGRAM MACRO) G40 G17 G80 G90 G49 G20 G54 G94 M6 T1 S2500 M03 G00 X0.0 Y0.0 /M08 G43 H1 Z1.0 G65 P9013 X0.0 Y0.0 Z-1.0 I10.0 D.5 R.2 F10.0 Q3.0 G00 Z1.0 M9 G40 G0 Z4.0 M5 G91 G28 Z.0 M30

O9013 (CCW CIRCLE WITH RADIUS LEAD IN/OUT) (SIMILAR TO G13 BUT NOT THE SAME AS) (RADIUS LEAD IN AND OUT IS HALF THE HOLE RADIUS) IF [#17LT0] THEN #3000=3 (PLUNGE FEEDRATE Q LESS THAN ZERO) IF [#9LE0] THEN #3000=4 (NO FEEDRATE GIVEN) IF [#4LE0] THEN #3000=5 (NO HOLE DIAMETER GIVEN) IF [#7LE0] THEN #3000=6 (NO TOOL DIAMETER GIVEN) IF [#4LE#7] THEN #3000=7 (TOOL DIA. TOO LARGE) IF [#24LT0] THEN #3000=8 (NO X LOCATION GIVEN) IF [#24LT0] THEN #3000=9 (NO Y LOCATION GIVEN) IF [#26EQ#0] THEN #3000=10 (NO Z HOLE DEPTH GIVEN) IF [#18LT0] THEN #3000=11 (NO CLEAR POINT GIVEN) IF [#18LT#26] THEN #3000=12 (RAPID CLEAR LT HOLE DEPTH)

#8=#4003 (G90/G91 STORE CURRENT MODE) #1=#5003 (STORE CURENT Z POS.) #2=[#26-#18] (CALCULATE INCREMENTAL Z DEPTH) #5=[[#4-#7]/2] (SUBTRACT TOOL DIA FROM HOLE DIAMETER CONVERT DIAMETER TO RADIUS) #6=#5/2 (LEAD IN AND OUT 1/2 CIRCLE RADIUS) #100=#17GT0 (IF Q IS LARGER THAN 0 FEED TO DEPTH) IF[#17LT0]THEN#100=1 (IF Q IS 0 RAPID TO DEPTH)

IF[#17LT0]THEN #3000=1(PLUNGE FEED WRONG Q VALUE)

G#8G0X#24Y#25 (G90/G91 RAPID TO HOLE CENTER) G0Z#18 (RAPID TO CLEAR Z POINT) IF[#17GT0] THEN GOTO1 (IF FEEDING TO Z DEPTH GO TO N1) IF[#17EQ0] THEN GOTO2 (IF RAPID TO Z DEPTH GO TO N2) N1 F#17 (PLUNGE FEED RATE) N2 G91 G#100 Z#2 (RAPID OR FEED TO Z END DEPTH) G91G3X#5Y0I#6J0F#9 (RADIUS LEAD IN WITH CIRCLE FEED RATE) G3I-#5 (CIRCLE) G90G3X#24Y#25I-#6J0 (RADIUS LEAD OUT BACK TO START POSITION) G0G90Z[#1] (RAPID TO INITIAL POINT) G[#8] (G CODE BACK TO PREV. G90/91) M99 %

***************************

From: snipped-for-privacy@aol.com Newsgroups: alt.mach>Next priority (when we finally get the old stuff cleared up) will be

You have been running second shift lights out with no probing? Wow, gutsy move.

Have you written Fanuc macros or Probing macros before?

Well I did give you a start.....lol.

I will be editing the macro for my own use to make it emulate G13/12 options, will make it modal and add a loop so it can run multiple DOC passes if needs be.

Tom

*********************************

From: snipped-for-privacy@aol.com Newsgroups: alt.mach>Tom:

Bob,

Well you went to a good source, Tim is a great programmer.

If Parameter #6053 = 13 G13 will execute the macro (Program Number O9013)

*****************************

Make #6054=13 Change macro program number to O9014. G13 will execute the macro pgm O9014.

This is great stuff and way under utilized IMO.

It's a simple thing to do, you want G13, G12 and don't have it? No need to complain just program it. Got a good idea for a custom G Code, program it.

Family of parts shit with a macro you can even have the controller prompt the operator for inputs if needs be.

Want to know the real spindle utilization, not just machine on/off, program start/end? You can tie custom macros to M codes as well. Turn the spindle on with S code (timer starts) M5 (timer stops) macro calculates time and adds it to the variable that stores the total. Dprint that collected data to your desktop computer.

Tom

AutoDesk offered me to trade in my SWx serial munber to get this cost advantage. now you can continue to use SWx as the license cannot be transferred in this manor. you may even be able to upgrade it, i am not too sure. the license to use the software is what we all have. it is not allowed to be sold in most circumstances. but i believe that the intent of Adesk is to get users on their software and stop upgrading SWx. if this was the path that we would go down then i would have kept using the current version for history CAD maintenance. all new CAD endevors would have been done in Adesk.

here is the bad news. my efforts have been stopped on this project. my core engineering group that was in the review with me were so dissapointed in both softwares, PTC & Adesk, that this project has been discontinued, primarily by the chief engineer and design lead engineer. i really wanted to continue as my next step was to train on both softwares and do an internal project on them and see how effective they were. you know it is a real shame that you can look at

10 different CAD software programs and find out that you may have the best in class out there. even with the issues that SWx has, and i have with it, it is pretty darn good. so this is the end of this round of CAD program research. i am very dissapointed.

still working on PDM tools review and WIndChill still looks to be best in class as a full fledged PDM/PLM. iQ

Thanks for the update!

AutoDesk offered me to trade in my SWx serial munber to get this cost advantage. now you can continue to use SWx as the license cannot be transferred in this manor. you may even be able to upgrade it, i am not too sure. the license to use the software is what we all have. it is not allowed to be sold in most circumstances. but i believe that the intent of Adesk is to get users on their software and stop upgrading SWx. if this was the path that we would go down then i would have kept using the current version for history CAD maintenance. all new CAD endevors would have been done in Adesk.

here is the bad news. my efforts have been stopped on this project. my core engineering group that was in the review with me were so dissapointed in both softwares, PTC & Adesk, that this project has been discontinued, primarily by the chief engineer and design lead engineer. i really wanted to continue as my next step was to train on both softwares and do an internal project on them and see how effective they were. you know it is a real shame that you can look at

10 different CAD software programs and find out that you may have the best in class out there. even with the issues that SWx has, and i have with it, it is pretty darn good. so this is the end of this round of CAD program research. i am very dissapointed.

still working on PDM tools review and WIndChill still looks to be best in class as a full fledged PDM/PLM. iQ