Custom LDD bricks and fixes

[jester94]

29 minutes ago, M2m said:

Oh ok. Didn't know that. Thanks !

Seems the new parts are missing bricks outline information. I don't thinks its a big issue, just to let you guys know.

You can see in the picture (left part is original from LDD, right is new ones)

Yeah, I know

For now i don't know how to solve it....

[SylvainLS]

18 minutes ago, Calabar said:

How to manage this situation if it should happened?

It’s already taken care of: LDD-4081 is BL/LDraw 4081b, so the “good guys” made LDD-40811 to match BL/LDraw 4081a.  They also did this for part packs (e.g. Batman emblems: numbered a to f for BL/LDraw and 1-6 for LDD).

As for ldraw.xml, when there’s a choice, I generally use the latest variant in LDraw, so that it’ll cost less if you want to make a Wanted List and buy it.

[M2m]

2 hours ago, jester94 said:

Yeah, I know

For now i don't know how to solve it....

I guess you know this document already https://github.com/Polymaker/ldd-modder/blob/v2/Documentation/G (mesh) file structure.md

Seems the Round edge shader data is not that good understood yet :-(

[SylvainLS]

First stats on the data:

Data for active parts: years ranges of usage (in sets) as known by BrickLink.
(Parts used in a set in 1950, then in 1952 is counted as active for 1951 too.)
(Parts with a “p” or “stk” in their ID aren’t included.)

Data for LDD: active parts that don’t have a corresponding ID in LDD
(“Corresponding ID” = the BL ID, or its known alternates IDs, or the LDraw ID as used by Studio.  “Not in LDD”: the ID doesn’t appear in ldraw.xml, Studio doesn’t have a match for LDD.)

Edited May 17, 2020 by SylvainLS

[M2m]

Does that mean that for ex LDD misses approx 300 parts of the 1600 used in 2020 ?

[SylvainLS]

301 out of 1622, yes.

It’s cumulative of course (old parts still used in 2020 but not in LDD).

7024 parts with dates.

Edited May 17, 2020 by SylvainLS

[M2m]

So I guess that LDD misses around 2600 parts (with dates) all together. I think LDD has ~4400. 

Is it possible to gather which ‘most useful’ or most used parts are missing ? I think this would be helpful. 

Edited May 17, 2020 by M2m

[SylvainLS]

4186 xml files in Primitives (so, roughly, the number of parts in LDD 2670)

5002 .dat files referenced in the latest (not yet published) ldraw.xml (old, obsolete files are importable, plus ~100 custom parts)

7738 parts not in LDD total, 3953 with a date range.

1307 parts not in LDD have a corresponding .dat file, of which 1012 have a date range.

(Note: 1161 parts are in LDD and not in LDraw.)

 

The list of BL parts with dates: CSV (separator “;”, with headers, columns: bl_id; alt_ids; first_year, last_year; nbr_sets).

The list of parts not in LDD: CSV (separator “;”, with headers, columns: bl_id; ldraw_file, first_year; last_year; nbr_sets).

Edited May 17, 2020 by SylvainLS
Update columns

[SylvainLS]

Of course, a lot of parts are not yet known to be used after 2020 

[SylvainLS]

1 hour ago, M2m said:

Is it possible to gather which ‘most useful’ or most used parts are missing ? I think this would be helpful. 

Good idea.  I have the number of sets the parts appear in.  I’ve added them to the data (same links as above).

(Not fun: when someone edits a post, you need to reload the entire page to see the modification, even if you post yourself.  I hadn’t seen that remark though I posted twice.)

Edited May 17, 2020 by SylvainLS
Data added

[jester94]

Sooo.... I built this using our new parts))))

[SylvainLS]

“Usefulness” = number of sets the parts introduced that year were used in.

“Lost usefulness” = number of sets we can’t do in LDD because of the missing parts.

Of course, I only have the number of sets each part is used in, not which sets (I’d need to have the entire BL inventories for that), so sets are counted many times.

The peak is 1978 (minifigure parts?).

 

@jester94 Vroom!

Edited May 17, 2020 by SylvainLS

[M2m]

2 hours ago, jester94 said:

Sooo.... I built this using our new parts))))

That looks amazing !

2 hours ago, SylvainLS said:

“Usefulness” = number of sets the parts introduced that year were used in.

“Lost usefulness” = number of sets we can’t do in LDD because of the missing parts.

Of course, I only have the number of sets each part is used in, not which sets (I’d need to have the entire BL inventories for that), so sets are counted many times.

The peak is 1978 (minifigure parts?).

 

@jester94 Vroom!

I think the cumulatively number of times a part is used per set until now. (For ex part number 58011 is used 155 times in 70 sets until today) may provide a good indication of its usefulness. 

[TheNerdyOne_]

I just stumbled upon all this progress and I just have to say that I'm amazed at what's been accomplished in such a short time!  Thank you to everyone who's putting in the time and effort to do this!  I've been doing my best to make do without the new parts in LDD for so long, but this just made it so much easier to actually utilize the new pieces!

[Calabar]

It seems the statistics are much more complete than I meant. Great Work SylvainLS.
Usefulness statistic is really very useful to create a priority list of new parts to add.

Is it possible to distinguish designs from variants? I mean... it would be great to have all the variants of 4081 available, but in many cases variants are so similar that have all of them could have low priority. The lack of a design instead could be much more problematic when building a set containing it.

14 hours ago, SylvainLS said:

It’s cumulative of course (old parts still used in 2020 but not in LDD). 

And what about a diagram like the first one but no cumulative?

It is the one I think is more useful: a list of new design IDs by year compared with their availability in LDD.

 

Edited May 18, 2020 by Calabar

[SylvainLS]

22 minutes ago, Calabar said:

Is it possible to distinguish designs from variants?

Well, I could merge entries when IDs only differ by the end letters.
4081a would be merged with 4081b.
Parts pack would be merged too but as they are parts pack, they should already have the same years and sets.
But there are many variants that have different numbers altogether (e.g. 3794(ab) and 15573).
So I need to make a proper list first (the data is there somewhere).

 

32 minutes ago, Calabar said:

And what about a diagram like the first one but no cumulative?

It is the one I think is more useful: a list of new design IDs by year compared with their availability in LDD.  

That would be the “New parts” diagram with the same for “Not in LDD”:

[jester94]

So guys
Can anybody make a list of missing parts (the most necessary)?

 

[SylvainLS]

12 minutes ago, jester94 said:

So guys
Can anybody make a list of missing parts (the most necessary)?

Working on it….

[jester94]

Now i have an opportunity to modify ldd parts and add color surfaces 

 

[jester94]

This is the most necessary modification)

[SylvainLS]

@jester94 You mean you corrected 20460 and 20461 which already are in LDD or did you make new ones?

@Calabar Merging variants (as per BL info) doesn’t change the graphs much.  (They are only 645 groups of variants, we go from 11210¹ parts to 10286.)

(¹ Yeah, not 11169, I used an outdated list of parts IDs.  Everything has been corrected since.)

So I think listing the parts that exist in LDraw and sorting them by the number of sets they appear in will be a good start.

Now, looking at these “useful” parts, I see that there are assemblies that parasite the data (e.g. complete torsos, used in many sets, not found in LDD but they actually ARE in LDD), so I need to cull them and I’ll give jester94 the list.  He will be able to choose from there.

[jester94]

Just now, SylvainLS said:

@jester94 You mean you corrected 20460 and 20461 which already are in LDD or did you make new ones?

 

I mean that it had the problems with coloring
also we have a lot of parts not color able (but should be)

[SylvainLS]

3 hours ago, jester94 said:

I mean that it had the problems with coloring
also we have a lot of parts not color able (but should be)

Okay, correction then.

A question for the list of useful parts: Can you do assemblies?  Like the old hinge plates 3149ac01.  The connector (313.dat) would be a new kind of connection/hinge, no?

[jester94]

3 hours ago, SylvainLS said:

A question for the list of useful parts: Can you do assemblies?  Like the old hinge plates 3149ac01.  The connector (313.dat) would be a new kind of connection/hinge, no?

I think nothing is impossible)

[jester94]

I added 2 new folders to my google drive (information for @Stephan)
1 - Colors - corrected parts are now multicolored

2 - Assemblies - we had Cyborg arm as separate part (34335), i hid it and added assembly 34412 with this arm

Edited May 18, 2020 by jester94