Discussion on Dressups and Tool Controllers

Here's the place for discussion related to CAM/CNC and the development of the Path module.
User avatar
bill
Posts: 376
Joined: Fri Jan 09, 2015 9:25 pm

Discussion on Dressups and Tool Controllers

Postby bill » Thu Mar 01, 2018 4:38 am

Edit: This discussion was split and is a continuation of https://forum.freecadweb.org/viewtopic. ... 17#p217663
sliptonic wrote:
Wed Feb 28, 2018 7:46 pm
I've added a new ticket in Mantis: https://www.freecadweb.org/tracker/view.php?id=3358

This is a hard crash of freecad and fully reproducible. Advice?

The crash appears to be happening because the simulation can't find the associated tool controller. The dress-up doesn't have its own TC but uses the TC of the base operation. Simulation should take this into account.
This sounds more like it is a problem of the Dress-Up; No!
It (Dress-Up) needs to provide/declare a tool-controller just like any-other operation I would think.
chrisb
Posts: 19512
Joined: Tue Mar 17, 2015 9:14 am

Re: CAM simulation for the Path Workbench

Postby chrisb » Thu Mar 01, 2018 7:12 am

bill wrote:
Thu Mar 01, 2018 4:38 am
It (Dress-Up) needs to provide/declare a tool-controller just like any-other operation I would think.
From a technical point: yes. The Dress-Up should be able to tell which tool it uses. From a users point of view: no. So it should always be the tool of the path which is dressed up. Otherwise it's not a Dress-Up, it's an operation on its own.
herbk
Posts: 1570
Joined: Mon Nov 03, 2014 3:45 pm
Location: Windsbach, Bavarya (Germany)

Re: CAM simulation for the Path Workbench

Postby herbk » Thu Mar 01, 2018 7:47 am

chrisb wrote:
Thu Mar 01, 2018 7:12 am
bill wrote:
Thu Mar 01, 2018 4:38 am
It (Dress-Up) needs to provide/declare a tool-controller just like any-other operation I would think.
From a technical point: yes. The Dress-Up should be able to tell which tool it uses. From a users point of view: no. So it should always be the tool of the path which is dressed up. Otherwise it's not a Dress-Up, it's an operation on its own.
+1
Gruß Herbert
schnebeck
Posts: 123
Joined: Thu Jun 22, 2017 8:04 pm

Re: CAM simulation for the Path Workbench

Postby schnebeck » Thu Mar 01, 2018 9:25 am

Hiho,

another Simu-bug I notice:

When you create multiple jobs based on different solids you get a job selector in the simulation tool. If you do not select the first job the stock for setting up the cut material is still taken from the first job.
simu-bug.jpg
simu-bug.jpg (214.14 KiB) Viewed 757 times
Bye

Thorsten
Attachments
Simu-2parts.fcstd
(41.1 KiB) Downloaded 15 times
User avatar
sliptonic
Posts: 1573
Joined: Tue Oct 25, 2011 10:46 pm

Re: CAM simulation for the Path Workbench

Postby sliptonic » Thu Mar 01, 2018 4:04 pm

schnebeck wrote:
Thu Mar 01, 2018 9:25 am
Hiho,

another Simu-bug I notice:

When you create multiple jobs based on different solids you get a job selector in the simulation tool. If you do not select the first job the stock for setting up the cut material is still taken from the first job.
simu-bug.jpg

Bye

Thorsten
Confirmed. added issue #3359
User avatar
bill
Posts: 376
Joined: Fri Jan 09, 2015 9:25 pm

Re: CAM simulation for the Path Workbench

Postby bill » Thu Mar 01, 2018 5:15 pm

chrisb wrote:
Thu Mar 01, 2018 7:12 am
From a technical point: yes. The Dress-Up should be able to tell which tool it uses. From a users point of view: no.
Dont follow the yes and no; hopefully the user knows what they are doing using a dress-up.

Whether the TC is declared, retrieved, inherited, lifted, pinched, or extracted, from the Base-Operation, the Dress-Up Operation needs to look like any other operation; too bad there is no interface specification/design docs; just pure-code I guess.

Excluding the data structure, the Dress-Up is simply/only an insertion of path segments

Thus, the Sim process all ops the same.

Havent used Sim for a while and I am not on the latest daily due to past problems with coding/structure/architecure surprises:
Are Drilling ops working nowadays ; Non-Response to button clicks still?, Large-Volume-Objects (yea. your can dial-down the parameters and get by; most testing seems to occur on smaller objects; might be good idea to check initial volume and dial-down automatically and/or alert user with suggestion) blowing up still?; Seems like it runs to completion without relinquishing control to its environment -- not a good thing !

Is the latest daily pretty-much safe on Linux platforms???

All in all, the PATH-Simulation Tool kicks a$$ for as NEW as it is!
chrisb
Posts: 19512
Joined: Tue Mar 17, 2015 9:14 am

Re: CAM simulation for the Path Workbench

Postby chrisb » Thu Mar 01, 2018 5:45 pm

bill wrote:
Thu Mar 01, 2018 5:15 pm
Whether the TC is declared, retrieved, inherited, lifted, pinched, or extracted, from the Base-Operation, the Dress-Up Operation needs to look like any other operation;
+1
User avatar
sliptonic
Posts: 1573
Joined: Tue Oct 25, 2011 10:46 pm

Re: CAM simulation for the Path Workbench

Postby sliptonic » Thu Mar 01, 2018 6:22 pm

bill wrote:
Thu Mar 01, 2018 5:15 pm
chrisb wrote:
Thu Mar 01, 2018 7:12 am
From a technical point: yes. The Dress-Up should be able to tell which tool it uses. From a users point of view: no.

Whether the TC is declared, retrieved, inherited, lifted, pinched, or extracted, from the Base-Operation, the Dress-Up Operation needs to look like any other operation; too bad there is no interface specification/design docs; just pure-code I guess.
Not quite. A dressup modifies an existing path which has a tool controller already. From a user's point of view, it would be wrong to add a TC to the dressup. When it comes to implementation, it might be appropriate to copy the base operation TC but it should not a changeable property of the dressup. In this sense, a dressup is different from an operation.
Excluding the data structure, the Dress-Up is simply/only an insertion of path segments
Insertion/modification/deletion
Thus, the Sim process all ops the same.
Again, from a user's point of view, I agree. The question in my mind is whether we should modify all dressup operations to have a tool controller and then prevent the user from seeing/changing that tool controller just so simulation can treat dressups and operations the same. Or, should simulation be modified to check for the type of operation and if it's a dressup, retrieve the TC from the parent operation.
Havent used Sim for a while and I am not on the latest daily due to past problems with coding/structure/architecure surprises:
Are Drilling ops working nowadays ; Non-Response to button clicks still?, Large-Volume-Objects (yea. your can dial-down the parameters and get by; most testing seems to occur on smaller objects; might be good idea to check initial volume and dial-down automatically and/or alert user with suggestion) blowing up still?; Seems like it runs to completion without relinquishing control to its environment -- not a good thing !
As far as I know, drilling is working perfectly. I don't understand all the other points you raise. We're trying to stabilize Path for 17 release and all bugs in the tracker except the two I just entered, have been closed.
Is the latest daily pretty-much safe on Linux platforms???

All in all, the PATH-Simulation Tool kicks a$$ for as NEW as it is!
I think it's all good on linux (except the sim crash) I agree about the ass-kicking nature of Simulation. :lol:
User avatar
JoshM
Posts: 419
Joined: Thu Oct 05, 2017 5:34 pm
Location: New Hampshire

Re: CAM simulation for the Path Workbench

Postby JoshM » Thu Mar 01, 2018 6:40 pm

On the topic of Dressups, the bigger confusion to my mind--and I've mentioned it before--is that the moment you add a Dressup, it moves to the bottom of the Job workplan. I cannot see any upside to this, and a LOT of potential issues...
User avatar
sliptonic
Posts: 1573
Joined: Tue Oct 25, 2011 10:46 pm

Re: CAM simulation for the Path Workbench

Postby sliptonic » Thu Mar 01, 2018 6:47 pm

I agree. That's a problem. Want to add a bug?