LDCad 1.5 Beta 1 (win+linux)

[Milan Vančura]

6126: I'm not sure about the hole in the 'flame', shoudn't it be a clip and if so doesn't it need one on the other side too? I also moved this info to the subfile.

Ah, I missed there is a subfile, sorry for that. I need better glasses
The hole should be a hole, not a clip. I tried it with the real part and it's near to impossible to use that as a clip - you can break the part this way. And about the other side (of the flame shape, you mean, right?): the shape is too open there so it cannot handle the pin. The only place working as "a snap point" is the one I marked.

BTW: While working on this I've found that holes in caps and hair parts fit to flower stem and vice versa: hair/helmet decoration pins fit to the hole in the middle of flower, and sausage ends fit there as well. Interesting. Maybe I use that fact in my next models

I redid both flower parts as the perfectionist in me couldn't help be bothered by the tiny offsets. There are some tricks you can use in the editor to obtain orientations from primitive usage (4-4cyli.dat in this case), maybe I'll should make a tutorial for this.

Thanks for your empathy You can imagine what I was crying for while working on stem parts... At least some words of hint, if you do not have time to make a video, would be helpful. LDCad always tells me "The selection must have editable positional information for this kind of action." when I try to select 4-4cyli.dat line in the source window and then press 'o' key in the edit window.

And can you also add a possibility to create a grid with step lesser than 1?

[Philippe Hurbain]

Ah, I missed there is a subfile, sorry for that. I need better glasses

Mikeheide's LDStructure can be a great help to navigate the tree of part dependancies.

[Jason McReynolds]

Don't feel bad/discouraged about the above notes, all info was still quite usable and is very much appreciated.

Not discouraged at all, this information is much appreciated!

• All the traffic signs can be done by only adding the clip in the used primitive, so I've added that instead.
  

And to think of the time I could have saved had I known that (not blaming anyone but myself, I've really only known about LDraw for a couple weeks)! ;-) I keep meaning to dig into the file formats a little more to better understand them, I had thought that this might be possible on other parts but the thought didn't cross my mind with these signs. I'll be diving into that a little more before I edit some more parts.

Ah, I missed there is a subfile, sorry for that. I need better glasses

Mikeheide's LDStructure can be a great help to navigate the tree of part dependancies.

Thanks Philippe! I'm still learning about the files and how they work/relate to each other. This looks like it will help a lot.

• I redid both flower parts as the perfectionist in me couldn't help be bothered by the tiny offsets. There are some tricks you can use in the editor to obtain orientations from primitive usage (4-4cyli.dat in this case), maybe I'll should make a tutorial for this.
  

I realize a tutorial might be a bit time consuming, but I think it would probably be well worth the time. Those that are editing parts would be able to benefit from it and the returns for you would be worth the effort. Keep up the great work!

[Roland Melkert]

* let the user to choose the snap point (e.g. a center + axis of SNAP_CYL) as an origin of coordinates for part rotation
* let the user to turn on (temporarily) the snap function while moving or rotating the part using keyboard

This sounds very doable, I'll take a look at it when I solved the current stability issues. Thanks.

[Roland Melkert]

Ah, I missed there is a subfile, sorry for that. I need better glasses
The hole should be a hole, not a clip. I tried it with the real part and it's near to impossible to use that as a clip - you can break the part this way. And about the other side (of the flame shape, you mean, right?): the shape is too open there so it cannot handle the pin. The only place working as "a snap point" is the one I marked.

I'll leave it the way it is then. You can use the new 'all dependencies' overview group to see all subparts/primitives used by a part currently being edited.

BTW: While working on this I've found that holes in caps and hair parts fit to flower stem and vice versa: hair/helmet decoration pins fit to the hole in the middle of flower, and sausage ends fit there as well. Interesting. Maybe I use that fact in my next models

Yes all it needs to snap is the same diameter, this is why I usually define all even weird ones you never know what might stick to it later on

Thanks for your empathy You can imagine what I was crying for while working on stem parts... At least some words of hint, if you do not have time to make a video, would be helpful. LDCad always tells me "The selection must have editable positional information for this kind of action." when I try to select 4-4cyli.dat line in the source window and then press 'o' key in the edit window.

Primitives etc are blocked for normal pin / seleciton editing during part editing because the main editing matrix handling code doesn't play nice with scaled and or mirrored references. As I was asuming the part edting was mainly for myself I opted for the easy way and blocked it all But when you click a reference in the source window it will change the working orientation. This is done by optionally removing the scaling and mirroring (flips the X-axis if I remember correctly). So if you know which reference has the correct orientation click it in the source window and then click the cyl meta in the part bin and press ins. It will insert the new meta using the working orientation. The position can be obtained/applied through copy / paste using the property dialog. This is the main trick I used, but you are right using 'o' would make more sense, I add that to the next version.

And can you also add a possibility to create a grid with step lesser than 1?

Grid and rotation stepping are currently integers internally but it shouldn't be hard to change that, I'll look into it.

[Roland Melkert]

I realize a tutorial might be a bit time consuming, but I think it would probably be well worth the time. Those that are editing parts would be able to benefit from it and the returns for you would be worth the effort. Keep up the great work!

I'll try to do a youtube clip of adding snap info to a medium sized part this weekend. Any part preference?

In the mean time I described my main orientation trick in the post above.

[Philippe Hurbain]

You can use the new 'all dependencies' overview group to see all subparts/primitives used by a part currently being edited.

Where is it located? couldn't find it...

[Roland Melkert]

It's in the overview group (where the mpd file content etc is located).

[Milan Vančura]

I'll try to do a youtube clip of adding snap info to a medium sized part this weekend. Any part preference?

In the mean time I described my main orientation trick in the post above.

Thanks, Roland. The trick works and it makes possible to fix snap info of parts of nonregular shape not only easier but more precise. Yes: 'o' key working would be very nice improvement but I can live with this trick, meanwhile.

About the part for a video tutorial: round plate 1x1. Because this part snap info contains near to all tricks:
1. it shows an inherited snap info (stud.dat)
2. you can show simple antistud snap info creation
3. then you can mention SNAP_CLR and reasoning for that
4. and as the last point show the way to specify SNAP_CYL with more segments
Appendix: 3741a.dat with your trick

[Jason McReynolds]

I'll try to do a youtube clip of adding snap info to a medium sized part this weekend. Any part preference?

In the mean time I described my main orientation trick in the post above.

Thanks, Roland. The trick works and it makes possible to fix snap info of parts of nonregular shape not only easier but more precise. Yes: 'o' key working would be very nice improvement but I can live with this trick, meanwhile.

About the part for a video tutorial: round plate 1x1. Because this part snap info contains near to all tricks:
1. it shows an inherited snap info (stud.dat)
2. you can show simple antistud snap info creation
3. then you can mention SNAP_CLR and reasoning for that
4. and as the last point show the way to specify SNAP_CYL with more segments
Appendix: 3741a.dat with your trick

Great suggestion Milan! I wasn't really sure what a good part would be. The only thing that I might suggest is to add in something with grids. If there's a piece that can do what Milan suggests and grids as well I think that would be perfect (maybe 2x2 round?). Two videos might not be a bad idea either.

[Milan Vančura]

The rest of Plants category is fixed in this archive, plus some bonus parts.
Enjoy.

P.S.: For Sea Grass, even the trick with 'Ins' key was not fast and precise enough. I wanted an automatic solution. But that meant I had needed to remind pieces of my linear algebra knowledge, after more than 20 years. I must say I still feel loud mess in my brain as many neuron paths, freshly woken up, are upset and crying, surprised why they cannot continue to sleep

[Roland Melkert]

The rest of Plants category is fixed in this archive, plus some bonus parts.
Enjoy.

Thanks.

I see you've been editing them externally given the spacing in some of the matrices etc. You also seem to have used the comma for decimal separation which isn't that bad as LDCad doesn't care if . or , is used it will thread both as the decimal separation (as a result thousands sep is not supported though). I'll correct them though to keep things uniform.

P.S.: For Sea Grass, even the trick with 'Ins' key was not fast and precise enough. I wanted an automatic solution. But that meant I had needed to remind pieces of my linear algebra knowledge, after more than 20 years. I must say I still feel loud mess in my brain as many neuron paths, freshly woken up, are upset and crying, surprised why they cannot continue to sleep

LDraw has a way in making those old math skills useful again There might be a number of part authoring tools to help with these kinds of things though.

[Philippe Hurbain]

You also seem to have used the comma for decimal separation which isn't that bad as LDCad doesn't care if . or , is used it will thread both as the decimal separation (as a result thousands sep is not supported though).

I always set up my windows to use '.' instead of ',' as decimal. I have yet to see a drawback of doing this and constantly get the benefit (such as beeing able to cut and paste LDraw source value to/from calculator)!

[Milan Vančura]

I see you've been editing them externally given the spacing in some of the matrices etc. You also seem to have used the comma for decimal separation

Mea culpa. That's, again, about Sea Grass part - I made some calculations in LibreOffice Calc and forgot to set the export LANG to US, sorry.

[Milan Vančura]

The rest of Plants category is fixed in this archive, plus some bonus parts.
Enjoy.

Thanks.

I'm working on Animals category now.

[Philippe Hurbain]

Something that would be useful: in the stepping menu, add a "goto step where this part appears". Useful when you forgot to insert a part at some step: click on a nearby part (hopefully appearing at more or less the same step), use the new goto function and add the missing part.

[Milan Vančura]

This reminds me I want to thank for a feature "locate the selection" in part bin. I have found it couple of days ago and I do not know how long it is there but it is very useful. For example if one needs a similar part and it's not clear in which (sub)category this part is.

[Philippe Hurbain]

I discover this function too... I hoped it could locate a part in the sub bins, but it seems it is not able to do this? My usage scenario would be to search for a part (with search tool), then use the locate function (starting from functional sorting bin, or from category sorting bin) to find the part in one of the sub-bin. This way I could 1- find similar parts 2- learn where this kind of part is stored.

[Roland Melkert]

The find selection in part bin option is not new, it has been a bit buggy though (should be ok now). It locates the selection in the last used / clicked part bin but I will test that to be sure.

It looks for a part like so:

1 Current view's group.
2 Current view's children (exclude ones searched before)
3 Change current view to one level up.
4 Goto 1

Go to selection step shouldn't be hard to add, I'll try to add it to Beta 1a. You could also use the source window for this though as it follows the selection by default.

[Philippe Hurbain]

You could also use the source window for this though as it follows the selection by default.

Yes, but how do I know the step number to move the timeline there and add parts at this step?

[Roland Melkert]

You double click the step meta line to make it the current step.

But I just relized the thing you're asking for is already available in the step menu. It's the 'selected' item.

I don't know my own program anymore

[Philippe Hurbain]

So many possibilities I have yet to discover! Thanks

[Roland Melkert]

Attached is a new shadow library file (rename it to .csl), it includes files donated by Philo and Milan.

I had to make less corrections for this one, I'm impressed

Some mistakes I noticed:

Wrong caps.
Cylinders not covering the whole hole / pin.
SNAP_CYL instead of SNAP_CLP usage.
Not using the grid property when it could.

I did not include the files for unofficial parts as I'm not sure if they should be included or not, I kept them separate so I might add them later of start a second csl for it.

Thanks to all.

[Milan Vančura]

Thanks, Roland.
I checked all your fixes of my parts and I understand them all but one: Plant Sea Grass. Can you explain why did you set caps to "two" or "B" there, if the intention (as I understand it) is to create one big sliding stick?

[Milan Vančura]

commit b199c357cdb409cf24235684815699e05819ad59
Author: Milan Vancura <[email protected]>
Date: Wed Oct 21 17:17:34 2015 +0200

mod04.zip - Animals category

http://forums.ldraw.org/showthread.php?t...1#pid17781

Everything but dinosaur parts (I do not have any)

Open issues:

43744 - rod with hole on top of the head
30478 - antistuds does not fit at the bottom - anybody who knows the
real part? Does it fit at 2x2 studs?

[Roland Melkert]

Can you explain why did you set caps to "two" or "B" there, if the intention (as I understand it) is to create one big sliding stick?

You are right the bottom one should use the 'b' cap. I forgot to rotate it

This part also uncovered a (minor) snapping bug. A length 8 clip shouldn't snap to <8 length cyl sections while using the 'two' cap.

[Roland Melkert]

I've made a quick simple demo/tutorial clip.

https://youtu.be/4xJvdWNgPyw

Hope it clears some of the secrets up

[Milan Vančura]

Can you explain why did you set caps to "two" or "B" there, if the intention (as I understand it) is to create one big sliding stick?

You are right the bottom one should use the 'b' cap. I forgot to rotate it

Hehe. If I was right about anything there it's just a matter of luck. I more asked for an explanation, to understand caps better. Because, as I understand them, if I want to put several snap_cyl metas in a (curved) row, the correct caps value is None with Slide=True. To simulate one long rod without any interruption. And now, I expect I'm not correct with this understanding as you changed all caps parameters to "both" value. ??

Originally, I wanted to check my new files (Animals) for caps, with this new understanding (your answer) but, in the end, I was too impatient and sent the set of shadow files, already I may confirm I tested those caps values in the editor and they work as intended. Including cases where they come from subparts. So I hope they are OK - please let me know if I still understand anything incorrectly from your POV.

[Roland Melkert]

I more asked for an explanation, to understand caps better.

The caps parameter is about limiting the range of the cylinder. This is done by indicating what side(s) of the cylinder is/are block(ed). Meaning something sliding/snapping on it should never have overlap into outer space on that side.

The 'a' value means the origin side is blocked. (e.g. stud underside)
The 'b' value mean the outer end is blocked. (e.g. anti stud ceiling)
The 'two' value means both ends are blocked (e.g. the curve blocks both sides in your example).
The 'one' value auto selects 'a' or 'b' depended on the gender. For male it's the origin (A) and for female it's the end (B).

Hope this clears it up some more.

I may confirm I tested those caps values in the editor and they work as intended.

If clips snap to 'two' capped of a lesser length this is the result of the bug I mentioned above. If 'none' is selected you will be able to slide from the outer rims but the clip will be intersecting the mesh of the plant curve.

In the end it is just a helper value as you can still end up moving stuff inside parts as they might use a different snap point or are using very complicated shapes (e.g. a pin inside a larger hole).

[Milan Vančura]

Hello Roland.

I hope I just found at least some hint about occasional crashes, finally: LDCad crashed two times today and it was, both cases, when I clicked on the "up" arrow in the part bin. Last log contains at least one line about that so I attach the whole log file.
Additional detail: I had "show snap points" turned on in both cases.

I hope this report helps at least somehow.

EDIT: new version of the archive, without the relative path of the file inside.

[Milan Vančura]

As I finished Animals category, I continue with "Containers" one.

More snap info: containers, first part.
I've found barrel and crate containers need their own snap and I had to define all matching "female" parts/positions. That's why these two containers "generated" so big commit:

commit 611a22911ef5aec2092cb9157afa5a10231eb7b0
Author: Milan Vancura <[email protected]>
Date: Fri Oct 23 00:04:17 2015 +0200

Barrel and Crate containers plus matching parts

Two new snap objects introduced: R 16 and S 16.
They are used in situations where there are no studs but the part
periphery matches 2x2 studs. Barrel and Crate container tops are good
examples. Important point: not all 2x2 antistuds match S 16 or R 16.
Usually, tubes on the part bottom collide with barrel periphery shape.
So I marked all mathing ("female") parts/places with female R 16 and/or
S 16.

Sent as: mod05.zip
http://forums.ldraw.org/showthread.php?t...0#pid17800

[Philippe Hurbain]

Your file is not empty, but I can't find anything inside... might be on my side, but I think your zip file is somehow corrupted.

[Stephen]

The file is okay, but it's been zipped with the file inside using a ..\..\ relative path, so some extractors might throw a fit.

[Milan Vančura]

Thanks for a note. I re-zipped the archive so it does not make any problems for you, regardless of the unzip program.

[Philippe Hurbain]

Ah indeed. 7zip goes through it but I lost reference (and patience) after clicking 3 times on .. ! Window XP file manager sees nothing. Anyway Milan updated file

[Roland Melkert]

LDCad crashed two times today and it was, both cases, when I clicked on the "up" arrow in the part bin. Last log contains at least one line about that so I attach the whole log file.
Additional detail: I had "show snap points" turned on in both cases.

Thanks Milan, this is probably the same bug I fixed a couple of days ago.

It tries to render a part bin cell (the mascot one) using a recently reloaded LDraw file. Problem being it tries to (re)use the old data which has been freed due to the reloading, it even says in the log 'should never happen' .

This might also be the reason for (some) of the closing bugs, A new beta is due this weekend, hope you didn't loose work.

[Roland Melkert]

Barrel and Crate containers plus matching parts

Two new snap objects introduced: R 16 and S 16.
They are used in situations where there are no studs but the part
periphery matches 2x2 studs. Barrel and Crate container tops are good
examples. Important point: not all 2x2 antistuds match S 16 or R 16.
Usually, tubes on the part bottom collide with barrel periphery shape.
So I marked all mathing ("female") parts/places with female R 16 and/or
S 16.

Thanks

I did not realize those should fit 2x2 plate undersides. I willl add these and mod4 to the next beta's shadow.

[Roland Melkert]

Thanks for the two zips

15279.dat seems to be a git diff file or something, i used the lines I thought were best.
59228s02 seems to contain garbage for the id property. Do you know when this happened?

You seemed to mistook the ID property for the group one in many files, I corrected them though.

The ID property is meant to uniquely id a (collection) of snap points so you can clear them in a higher part in order to redo shapes they were part of.
The group property is used to limit matches.

I also make quite some changes to the horse, cow and dragon parts it should snap more natural this way.

But overall nice work, thanks your files will be part of the new beta I'm hoping to release this weekend.

[Milan Vančura]

Thanks for checking my files, Roland.

You are right about 15279.dat - this is my mistake, it shows how your version differs to the previous one of mine. So no change in this file since your last update.

The garbage in 59228s02.dat surprises me as well because I did not edit this file in any external editor. LDCad is what saved this. But why? And was it my mistake? How can I even put so much of garbage to the edit field? I do not know. Please: for the file itself, ignore that garbage. Only it may be interesting about is to check LDCad if it might be caused by any bug. (What I suggest to spent time on only if/after it happens again...)

Also thank you for your explanation about ID vs. Group fields. I'm not 100% sure I understand that but that's the advantage of a written form: I may read it again and again until I start to understand. Meanwhile, I take it as a simple rule: Milan, do not use ID field!

[Owen Dive]

Hi Roland.

Is there a way to selectively unhide parts? As far as I can tell, your options are limited to the last part hidden, or all hidden parts.

Owen.

[Roland Melkert]

Is there a way to selectively unhide parts? As far as I can tell, your options are limited to the last part hidden, or all hidden parts.

No you can indeed only unhide in reverse (like undo) or show all again. I have plans for more grouping options though and as the hidden parts are internally also groups. This is something for version 1.6 or maybe even 1.7 though.

[Owen Dive]

Ah, that is a shame. At least I wasn't missing something.

I did find a strange behaviour this evening. I had many parts hidden, and when I tried to insert a new copy of a submodel, one of the parts of that submodel (but not the whole submodel) got hidden. As I drag it from the part bin, it's fine, but as soon as it was placed in the workspace the part disappeared (and its line in the source window is greyed out). I was not using nested mode. Can you think of any reason why that would happen?
Unfortunately I haven't been able to reproduce this behaviour.

Owen.

[Roland Melkert]

I did find a strange behaviour this evening. I had many parts hidden, and when I tried to insert a new copy of a submodel, one of the parts of that submodel (but not the whole submodel) got hidden.

Might just be a rendering glitch resulting of a wrong tag number, does e.g. a next part insertion make it visible again?

[Owen Dive]

Might just be a rendering glitch resulting of a wrong tag number, does e.g. a next part insertion make it visible again?

To be honest I didn't try. After unhiding everything, it worked as normal, even after re-hiding parts (though admittedly not exactly the same parts), and as I said, I couldn't get it to do it again. I probably shouldn't even have mentioned it.

[Owen Dive]

And can you also add a possibility to create a grid with step lesser than 1?

Sorry this is a bit late to the party, but I just had a bit of an idea how you can place elements with <1LDU precision. Kind-of.
If you add a helper part like the marker meta somewhere close to the element you're placing, you can select the two of them and rotate them one degree at a time, which will result in a translation of <1LDU.

Of course, the technique is not fool-proof: the element you're trying to move will be rotated as well, so you may have to rotate it back (after deselecting the helper part, of course), and the movement is in two dimensions instead of one - you may need to manually clean up the position in one axis manually.

Owen.

[Philippe Hurbain]

you can also use the move tool / rotate tool (shortcuts Ctrl+M / Ctrl+R) to enter any value, even non-integer