I`ve tried fixing the creation graph just like you described in your earlier post by relinking the connections but my Bevel Poly Node does not have any of those inputs to connect to. Any ideas?
@Swordslayer why did you add a Smooth modifier to the stack first? It over-rides all the smoothing groups I have set up. Maybe just have a check box to turn it On/Off, and maybe an auto smooth angle you can put in instead of always having to stick with 30.
I`ve tried fixing the creation graph just like you described in your earlier post by relinking the connections but my Bevel Poly Node does not have any of those inputs to connect to. Any ideas?
Hmm, let's try this. Open the folder C:\Users\<<userName>>\Autodesk\3ds Max 201x\Max Creation Graph\Tools\Downloads\ There should be the InsetBySmoothingGroup.maxtool file and InsetBySmoothingGroup folder. Delete everything that starts with InsetBySmoothingGroup that you find there. Basically, if you didn't install any other package, the folder would empty. Redownload extract and rename the InsetBySmoothingGroup.mcg file from .mcg to .zip and open it. It should contain the .maxtool file and folder - extract both to the Downloads folder. One file is present twice in the folder, don't worry if it asks you if it's okay to overwrite it, you can confirm it. Restart max, look in the maxscript listener if there are any errors. If not, look for the InsetBySmoothingGroup in the modifier list.
@Swordslayer why did you add a Smooth modifier to the stack first? It over-rides all the smoothing groups I have set up. Maybe just have a check box to turn it On/Off, and maybe an auto smooth angle you can put in instead of always having to stick with 30.
Done, get the file here, it should be updated on scriptspot as well but I'm no longer sure about that. By the way the smooth modifier enabled state doesn't update visually, that's just a redraw issue though - and you can always just click the eye/lightbulb icon next to the smooth modifier instead as well.
Tried your suggestion and redownloaded the zip file with your updated timestamps - still the same error when trying to evauluate / opening the script. Bevel Poly node does not have any inputs
Tried your suggestion and redownloaded the zip file with your updated timestamps - still the same error when trying to evauluate / opening the script. Bevel Poly node does not have any inputs
Well, it's still the same file, I didn't touch it in a day, I'm just kinda getting desperate as I have no idea why it works for me and a few other people and doesn't work for others. Do you by any chance have any other mcg package installed? Though the only one that I think could be conflicting with this one would be my old InsetExtrudeModifier / InsetBySelectionValue...
So, after I've added the smooth switch, I noticed it might sometimes crash max. Better use the other .ms file supplied instead. By now, I could write a book '1001 ways of crashing max' (the editor's choice being (fn''=''())() for its beauty, simplicity and didactive value).
I love what you're working on @Swordslayer - but I'm highly anticipating the work that @PolyHertz is going to release. Could we perhaps split this discussion into a new thread specifically for your script? That way I can keep an eye on both of them without getting confused.
Hi there, i tested the script out but i get direclty some very stange errors and bad results. i downloaded 3ds max 2016 and get the "InsetChamferStack.ms" and copyed it as written. i started 3ds max and made a quick object with some bools. then i added the insetchamferstack and there you go. the pictures will show my problems up. i tested out with the inset and the chamfer parameter but i dunno whats going on:
@mobpapst looks like there's something wrong with the smoothing groups; anyway, the inset-chamfer thread is here so for further comments and to send the link to your scene, please post there. Thank you
Hey I dig this, but I am getting an error -- type error : Call needs function or class, got: undefined. In the script window its highlighting- chamferModifier = Quad_Chamfer "etc." I am in max 2016. The insetbysmooth works fine.
@ZombieWells The QuadChamfer version is there for those who have Quad Chamfer from Marius Silaghi installed. Use the regular Chamfer one if you're not one of them.
I was so looking forward to giving this a go, but Polyhertz hasnt posted here since mid August. If this isn't something that he plans to release then that's fine, but some sort of update would be appreciated. I cant begin to fathom how difficult it must be to iron out any issues, but feel that the initial excitement is going to fade away if there isnt some sort of info forthcoming.
Hey guys, sorry for the lack of updates. Due to a bunch of things (health issues that have since been taken care of, GPU problems with my home PC that I just got resolved yesterday, and the fact that someone else already made an equivalent) nothing has really happened with this. For those that were interested, I'll at least let you guys have the stand-alone Maxscript version I did previously (the one @musashidan was testing). You can grab that here.
As for the C++ version, with both the above options available, are you guys really still interested in it? I haven't tried the one by @Swordslayer , but it looks like it does an ok job?
Hey Polyhertz, thanks for responding, glad to hear you are feeling better. I would love to see this go a little further, or possibly a joint effort between you and Swordslayer. I cant speak for others but I would pay for a more polished and feature rich version. I think something like what we are seeing for blender from masterxeon would be awesome, hardops and boxcutter leave my jaw on the ground.
thanks for sharing this, and wish you the best with your health and PC issues.
As for the C++ version, with both the above options available, are you guys really still interested in it? I haven't tried the one by @Swordslayer , but it looks like it does an ok job?
Mine is a maxscript modifier with all the limitations maxscript brings about. Or rather several modifiers - one does a better job at the cost of speed, the other one is faster but with all the same shortcomings that your tool faces as well. If you are comfortable with 3ds Max C++ SDK and want to implement some computationally intensive fixes (topology cleanup, edge turning, corner chamfer radius fixes, inner corners support edges etc.), definitely go ahead, there's a lot to gain speedwise and I haven't taken the leap yet (anytime I look at the Shell modifier source, it gives me the creeps ).
This would be great with C++ / Python? I really hope this does not slowly fade away... :-( Especially the slower insetfixquadchamfer would be awesome to have in a faster version. @PolyHertz If you have some time, I am also still very interested!
are you guys really still interested in it? - Yeah I am screwing around with this right now, and have been showing it off to folks at work. There is a huge interest in this, so please keep working on it
@polyhertz Hey mate, good to see you back. Yes, this is the data channel mod. A brilliant feature added. Dynamic chamfering like this is just one of the many examples of what it can do.
this is basically the same as chamfer by smoothing groups only you have more control over the selection.
inset chamfer will deal far better with weird shaped ngons etc.
That is what I thought, I don't really see how data channel could be a replacement for these inset scripts (modifiers).
It isn't supposed to be . I'm guessing polyhertz misjudged. The whole key to these tools is the 'shielding' of n-gon topo through pre-inset, which DC mod can't do. (Yet?)
musashidan said: It isn't supposed to be . I'm guessing polyhertz misjudged. The whole key to these tools is the 'shielding' of n-gon topo through pre-inset, which DC mod can't do. (Yet?)
Since DC doesn't modify topology (it only sets properties of vertices/edges/faces such as position, vertex color, map data, arbitrary data, selection weights, crease weigths, smoothing groups, mat ids etc.), it won't be able to do that (that said, it's a really nice addition to the toolset as it supports polyop methods, unlike scripted modifiers which only support meshop methods).
Moreover, as it currently doesn't support explicit vertex normals, it can't be used to counteract the shading glitches that result from chamfering unconnected verts and complex shapes either.
Buuuut... it could be used to 'pre-process' the underlying mesh and save all the data that only changes when underlying topology changes... Recomputing everything on the fly makes the current modifier slow with complex meshes. Wonder if doing it this way would be faster (in a general case, not some extreme showcase meshes) given that it'd introduce baseobject -> PolyMesh conversion and then -> TriMesh conversion on top... Plus it'd add another piece to the modifier stack.
if you could use the DCmod to identify verts with > 5 attached edges, convert that selection to edges and add it to an edge selection based off smoothing group boundaries? you'd get the same basic algorithm I used in my own version of quick high poly to select edges for chamfering. Outsourcing that work to a proper plugin rather than doing it in maxscript might speed things up a bit - it'd also protect you from accidental deselections etc.
if you could use the DCmod to identify verts with > 5 attached edges, convert that selection to edges and add it to an edge selection based off smoothing group boundaries? you'd get the same basic algorithm I used in my own version of quick high poly to select edges for chamfering. Outsourcing that work to a proper plugin rather than doing it in maxscript might speed things up a bit - it'd also protect you from accidental deselections etc.
The trouble is that the edge selection is the part that takes the least amount of time (and doesn't break unless there are broken SGs in which case the whole inset-by-SGs concept breaks) - there was basically no difference from making chamfer pick edges by smoothing group and setting my own set of edges (which also include inner support edges but exclude outer support edges). Doing only the inset is (quite) fast as well, what is slow though is building proper corner topology from scratch and I don't know of any shorcuts or tricks to make that part snappier...
Replies
Validation failed for C:/Users/------/Autodesk/3ds Max 2016/Max Creation Graph/Tools/Downloads/InsetBySmoothingGroup.maxtool : Error occurred deserializing node: Can't find operator BevelPoly
Running into the same problem on Max2016:
Validation failed - Error occured, deserializing node: Can`t find operator Bevel Poly.
I`ve tried fixing the creation graph just like you described in your earlier post by relinking the connections but my Bevel Poly Node does not have any of those inputs to connect to. Any ideas?
It over-rides all the smoothing groups I have set up. Maybe just have a check box to turn it On/Off, and maybe an auto smooth angle you can put in instead of always having to stick with 30.
There should be the InsetBySmoothingGroup.maxtool file and InsetBySmoothingGroup folder. Delete everything that starts with InsetBySmoothingGroup that you find there. Basically, if you didn't install any other package, the folder would empty.
Redownload extract and rename the InsetBySmoothingGroup.mcg file from .mcg to .zip and open it. It should contain the .maxtool file and folder - extract both to the Downloads folder. One file is present twice in the folder, don't worry if it asks you if it's okay to overwrite it, you can confirm it.
Restart max, look in the maxscript listener if there are any errors. If not, look for the InsetBySmoothingGroup in the modifier list.
Are timestamps of the compounds the same as these?
Tried your suggestion and redownloaded the zip file with your updated timestamps - still the same error when trying to evauluate / opening the script. Bevel Poly node does not have any inputs
Not that i know of, I have various other scripts installed but that download folder only contains the contents of your zip file
i tested the script out but i get direclty some very stange errors and bad results. i downloaded 3ds max 2016 and get the "InsetChamferStack.ms" and copyed it as written. i started 3ds max and made a quick object with some bools. then i added the insetchamferstack and there you go. the pictures will show my problems up. i tested out with the inset and the chamfer parameter but i dunno whats going on:
As for the C++ version, with both the above options available, are you guys really still interested in it? I haven't tried the one by @Swordslayer , but it looks like it does an ok job?
I would love to see this go a little further, or possibly a joint effort between you and Swordslayer. I cant speak for others but I would pay for a more polished and feature rich version. I think something like what we are seeing for blender from masterxeon would be awesome, hardops and boxcutter leave my jaw on the ground.
thanks for sharing this, and wish you the best with your health and PC issues.
Todd
I really hope this does not slowly fade away... :-( Especially the slower insetfixquadchamfer would be awesome to have in a faster version.
@PolyHertz If you have some time, I am also still very interested!
inset chamfer will deal far better with weird shaped ngons etc.
Moreover, as it currently doesn't support explicit vertex normals, it can't be used to counteract the shading glitches that result from chamfering unconnected verts and complex shapes either.
if you could use the DCmod to identify verts with > 5 attached edges, convert that selection to edges and add it to an edge selection based off smoothing group boundaries? you'd get the same basic algorithm I used in my own version of quick high poly to select edges for chamfering. Outsourcing that work to a proper plugin rather than doing it in maxscript might speed things up a bit - it'd also protect you from accidental deselections etc.