Texture Side Boundary?

[Joel Schuman]

Brady,

Please give the preaching, invective, personal stuff, and package-dishing a rest. I doubt you know enough of my experience or background to support your assertions, and your tone does you little credit. Instead, let's get back on topic.

Your inlay "solution" dodges the issue - it is only workable where there is no grain to the material or discontinuous grain is desirable or at least acceptable.

As for your "feathering" para, I don't think you grasp what I am after. Examining traditional handcarved work with gouge-textured sunk backgrounds, one rarely finds an obvious attempt at feathering. The apparent strategy is to run each tool stroke along a natural path until it approaches the boundary, then backcut down from the boundary to meet it. That's also what is commonly prescribed in woodcarving instruction texts.

Mimicking this with CNC is not difficult. It borrows from horizontal raster finishing, which is well understood. For the cove (roundnose) bits I use it is easily done with Rhino or any other generalized 3D modeling program: First, generate a set of texture toolpaths of arbitrary size and shape that is everywhere larger than the desired boundary area by at least one stroke. Then sweep the boundary with a vertically mirrored half section of the texturing bit to create the limiting polysurface. Finally, trim the texture toolpaths with the limiting polysurface.

It is not the trimming that is difficult with Rhino, but the parameterized generation of a convincingly varied and "naturalistic" set of texture toolpaths. VCP does it far better than any algorithm I've come up with in RhinoScript, even using Alibre for parameterization. Doing it manually with Rhino can be effective, but it's too tedious and lengthy a chore to be cost effective in my work.

It shouldn't be hard for VCP to generate a "base" texture that does not use the boundary vector as a base for feathering. I think VCP (not to mention Cut3D) must already do something similar to the kind of toolpath trimming I would like to see incorporated into texturing. It seems quite similar to V-carving, an inherently 3D feature in a nominally 2-1/2D program.

Tony? I know you're watching this - perhaps you could favor us with a word or two on this subject? Is it do-able? Is it difficult? Do you think it is worthwhile?

[CRFultz]

Joel, Welcome to Vectric

It goes without saying, Vectric has a very loyal user base.
To some degree, we are Vectric, at least it feels that way.
We tend to be very protective and will defend the software and the company.

Lemons? Maybe not the best choice of words

The added feature of Texturing is simply that, Texture.
If the user base feels that a hand-carved "feature" is something they need, then it will be considered, thats what makes this software stand out.
Vectric listens when enough customers request a feature, if your request has enough votes (and is possible) then I'm sure they will add it to the wish list.

To Z up at the boundary edge would leave a radius of the tool, would that be a backcut look?
Is it possible to trim the boundary with a different texture to give it a "handcarved look".

I hope you find a solution to your request, with a little more experience perhaps you will. One things for certain, there are alot of folks here that are willing to help.

Good Luck
Chuck

[Joel Schuman]

CR,

Thanks for your constructive and helpful reply.

I'm familiar with the impact of loyalty and identification of self with software, and how they combine with the usual BB/email disinhibitions to make people behave in a way they would not face-to-face. This forum is hardly unique in that. In the CAD/CAM area, I've seen threads on the ShopBot and TurboCAD forums that demonstrate something approaching Stockholm Syndrome, where users defend the software against any remark they perceive as critical, however well intentioned and constructive, even if it might lead to improvements that would benefit them, just as Stockholm Syndrome captives resist attempts to free them or nail their captors.

Your question/suggestions are spot on. Yes, a simple Z-up when the tool edge reaches the boundary would be sufficient - in essence that is the unstated goal in many handcarving texts. But since many handcarved backgrounds show some "shoaling" near the cross-stroke boundaries, endgame might be a parameter that controls the spectrum between 100% straight Z-up and 100% taper to boundary and another parameter that controls how far a stroke may extend outside the boundary. The latter parameter would eliminate the need for iterative tinkering with a boundary offset to get the desired edge coverage.

In all cases, though, I think the vector boundary must be respected and handled uniformly in all directions, just as in all other 3D toolpathing I've seen. Your second question is one that occurred to me early on in my experimenting. I think the answer must be "no", and your excellent simulation example shows why. Although more uncritical, more relaxed souls might find it acceptable, I find it unbalanced, apparently uncontrolled, and finally aesthetically unpleasing. The "stray" strokes at the upper left and lower left are just too heinous.

My Dale Carnegie "lemon" quote was not intended to be provocative, but to suggest that I have been putting VCP's texturing implementation to good use in spite of what I perceive to be a design flaw. For decorative "patches", I have developed several strategies.

One is to apply VCP texturing to raised panels. This is conceptually similar to Brady's proposed intarsia solution: it dodges the edge issue by carving the wayward edge in thin air. VCP texturing on raised panel cabinet doors is a slam dunk, and the artistic or sculptural possibilities in this application are very exciting. I've used it on the doors of a large bookcase/hutch to good effect.

Another strategy is to design decorative texturing "patches" with adjacent puzzle-pieced vectors, none of wrap around any tight bends or corners, always keeping the raster direction between 60 and 90 degrees to the mean outside edge tangents of each vector. Some might find lower angles produce acceptable results. With careful attention to design and a bit of creative play (T&E), there's a lot that can be done before bumping up against the out-of-bounds issue. Sometimes a small gap between puzzle pieces improves the effect. Dividing a rectangle with diagonals and carving the four resulting triangles this way will demonstrate the benefit.

Where design constraints require a plain one-direction patch, I have found two workarounds. The most useful is drastically shortening the stroke length. Alternatively, I will run the raster direction along the diagonal. Both minimize the bulged hourglass appearance of textured rectangles.

If you have other techniques for using texturing as a featured design element, rather than as unobtrusive low-stakes background, I'm eager to see them.

[TReischl]

You know what? I have always really enjoyed this forum, right up until this thread.

The guys have all been great and willing to help everyone.

They all seem to be able to do it without pointing fingers, telling others that they made bad design decisions, without referring to lemons, ad nauseam.

Tony and Brian work their butts off to help us all and deliver a great package at a reasonable price.

I am surprised you grace us with your presence Joel, you seem to know it all, and are happy to tell everyone else that.

If you want to make these type of comments, you should do them via private email with Vectric. In fact, you should not even make those type of comments in an email.

In short, Joel, you are really annoying a lot of folks.

[Joel Schuman]

TR,

I did not start the personal stuff in this thread, nor have I contributed other than with general (amused, if you couldn't tell - sorry I don't do the smiley thing) observations.

You stepped in with a post that contained both a helpful continuation of the on-topic discussion, and comments about the tone of the thread. I responded to both. If you didn't want responses to both parts, why did you include the off-topic part?

If this forum can't acknowledge even the possibly of anything less than perfection in the products under discussion, or engage in cooperative exploration of flaws, misfeatures, early versions, or whatever term you find comfortable, it's not of much use. I don't have time for or interest in mutual backpatting or a product lovefest.

I'm trying to use VCP for my work, perhaps extending VCP's use for tasks not envisioned by its developers. I think that's a good thing. I'm hoping to meet here other professionals who are encountering the same problems and are working around them in creative ways that they want to share. Denying that there's a problem, or at least a significant opportunity for improvement, doesn't help anyone.

If you've not had much experience of either formal design critiques or engineering discussions, my forthrightness may be jarring and uncomfortable. I apologize for that - it's what I'm used to in dealing with my colleagues and clients. I'll try the pussyfoot approach in future - I sure didn't mean to harsh your mellow.

Once again, can we stay on topic? Anything more to say about texturing?

[Tony Mac]

Hi Joel,

Your feedback has been noted and we understand what your requirements are. Additional texturing options will be considered for future versions of the software but at this point we cannot give any further details.

Tony

PS It would be interesting to see what you can produce using Rhino.

[TReischl]

If you've not had much experience of either formal design critiques or engineering discussions, my forthrightness may be jarring and uncomfortable. I apologize for that - it's what I'm used to in dealing with my colleagues and clients. I'll try the pussyfoot approach in future - I sure didn't mean to harsh your mellow.

Once again, can we stay on topic? Anything more to say about texturing?

Your apology is accepted.

FYI, I have PLENTY of experience with design critiques and engineering discussions. In my career I have designed both CNC laser cutting machines and developed commercial CAD/CAM applications for one of the leaders in the field. You might want to rethink who the audience is here.

Please do not respond.

[HayTay]

Joel,

I'll chime in here with maybe a constructive answer to your request. By making some 'fake' boundaries and then playing around with the Max Cut Depth, Max Cut Length, Max Overlap and Stepover settings in the Create Texturing Toolpath tool I arrived at the following background texturing pattern (see attached image). Is that more of what you're looking to accomplish? Or am I way off base?

I must admit it took several tries and multiple adjustments to the above settings to achieve the look I think you're going for. But it seems to be do-able. All-in-all, I guess I fiddled around with the settings for around 20 - 25 minutes before achieving a satisfactory result. It does seem to be within the realm of possibilities with VCarve Pro though.

Just trying to find a working, maybe not simple, VCarve solution to your request.

I hope this helps,

[Joel Schuman]

HayTay,

The crosswise boundaries are pretty darn close to some of the handcarving I've seen, and way mo' bettah than anything I've achieved by macdinking those same settings.

Would you please shoot me your settings? Screen shot, .crv, or texture file, whatever...

Thanks!

[HayTay]

HayTay,

The crosswise boundaries are pretty darn close to some of the handcarving I've seen, and way mo' bettah than anything I've achieved by macdinking those same settings.

Would you please shoot me your settings? Screen shot, .crv, or texture file, whatever...

Thanks!

No problem, Joel. Here's the CRV file of the Sample Carved Background. I'm glad that approximates what you've been trying to accomplish. I didn't know if I was wasting my time or not. Let me know if you need any additional information.

If you'd like me to try one of the files that you're working on either post it here or PM me for my email address.

Posted fixed CRV file format so it doesn't give everyone an error. Sorry about the confusion.

Glad I could help,