Short Version within the SSSSSSSSSSSSSSSS
Text about the pics PPPPPP

Longer Version within the LLLLLLLLLLLL


SSSSSSSSSSSSSSSS
I am unable to create a solid originating from a pair of non-planar but swept or mirrored lines that have a z distance representing thickness. The end result is supposes to be a symmetrical solid. Using the various tools i get ACIS errors, error messages that lines are not planar, and cannot even sweep one surface that would remain bound by a mirror of the two original lines, and cannot sweep or otherwise extrude a solid bound by a fore or aft boundary line.
SSSSSSSSSSSSSSSS

About the pics i attached

PPPPPP

In image 1, i show the two mirrored (red) curves originating from the two green curves.

In image 2, i show 8 curves to which i could attach 2 surfaces, and if i want to enclose the whole thing, i'll need 6 more surfaces, using either newly-created curves, or by trying to attach surfaces to surface edges...

In image 3, i show where a hopeful expectation resulted in surfaces not desired. I used one of the rail sweep tools but none of them let me grab a surface (between the green curves) and extrude or mirror/extrude in mirror fashion. Having a mirror-extrude tool would eliminate multiple, unnecessary tasks when all that is needed is a single solid or one needs a simple enclosed surface that could provide volume information (without being compelled to create an unneeded solid)

In image 4, i show the non-parallel curves

In image 5, i show the ad-hoc surface i had hoped to mirror-extrude somehow

In image 6, i show an area of deck surfaces/lines preparation from before, which got tedious or can get tedious after a while

In image 7, i show a zoomed out view to put all this tedium into perspective

(post-uploading comment: the images are NOT presented in the order that i upload/attached each in, and, some became thumbnails. But, the images as i've described them have their number in their file names or in the mouse-over name.)

PPPPPP


Longer explanation:

LLLLLLLLLLLL

I have two non-planar lines where the forward x values are equal to each other and the aft x values are equal to each other. Being non-planar, the z values are not equal (and while they need not be parallel, in my case all of mine will be parallel to the baseline or the x axis).

As to the y values, in plan view, the forward y value of a supplied line/curve is not equal to the after y value, and this is because the final shape is to be a deck plate fitting in a faired hull. (Visualize a ship or a canoe, if you will.) In short, these two non-planar curves/lines come from pre-defined waterlines given different z values, which represent the thickness of the deck plating.

In the past i have laboriously used intersecting curves from what are the vessel's watertight bulkheads (solids from stitched surfaces originating from dxf 3d lines). I then created two surfaces from waterlines for each of the respective deck thickness lines. Then, i used the intersection tool to get fore and aft boundaries that are known due to the thickness of the transverse bulkhead previously mentioned.

That process is getting old, and it is very demoralizing to face correcting linework because of other larger changes nearby - such as shifting the x location of a bulkhead. Due to the size and large number of items in the model, i really do not think this model is an ideal candidate for use of smart surfaces that redraw according to plate boundaries. I feel this way because if i make a mistake and move something its dependencies may move, and this will create very long waiting times as the shape regenerates. This regeneration would affect or be affected by the presence of multiple other decks in the that area of the model.

Rather than create 14 additional lines representing a simple box (a box to the eye in geometry class, but not to the software, since 24 lines would be needed for all of the six surfaces for just ONE deck in that area of the model), it would be vastly easier if i could start out with just TWO lines, and tell the software to mirror those two about the x-axis while or before extruding a solid simultaneously. This resulting solid should be calculated BEFORE VC/VCP begins sweeping anything. (This would avoid vexatious ACIS complaints i previously endured when some badly-trimmed line didn't create a neatly closed boundary or line. So, it should not be heavily dependent upon ACIS as i'm irritated by ACIS blowing up on what seems to be quite simple constructions. )

So, to summarize, I want to take two curves/lines which in elevation or side view are parallel to the deck, but non-planar (because one is above and somewhat offset and not uniformly offset), and in plan view converging or diverging (depending on where in the hull they change due to lines fairing), and create a solid that is symmetrical about the centerline and which do not requre any subsequent trimming nor any selections of closed planar lines, and so forth.

I think that visually, this is trivial. It's just a matter of building a tool that captures a few assumptions that may need answers prior to sweeping. It would be nice of a wire-frame PREVIEW is offered to avert minutes-long ACIS cogitations and computations and generation time.

In the same vein, this tool could use the wireframe as the basis of a fully-bounded volume which could then be queried for its volume's value and this volume could be dynamically or manually updated based on the user inserting a solid or a represenatative solid so as to subtract or pseudo-Boolean the volume when fairly close but not necessarily accurate/final volumes estimates are needed.

LLLLLLLLLLLL


Thank you for reading this...

Hi Zumer,

Sorry, this is long. Don't read it all in one go. Or, don't read it at all. It's just very long and sad...and maybe i'm asking for more than i've paid. But, if VCP had what i'm asking for, I'd pay maybe $400 vs $250... If you're inclined, then read on...

Thanks for your suggestions, but the thickness is not something i can mess with. I set the plate (deck) thickness in Delftship, and those lines of intersection with the mesh that makes the ship's sideshell are formed in Delfship. And, the curves that form the watertight bulkheads and which interesect with the deck also are formed in Delftship. Those lines i then import into VCP.

If i had $100,000 to throw around, i could skip all this and buy an industrial-strength seat license and use a suite of tools that are rigorously designed to facilitate ship design. But, part of the beauty of using VCP is in showing that VCP can - even if vastly slower -- help the user create geometry very easily. Then again, since a lot of that high-power software may be subject to many more DFARS concerns than VCP would be, i don't want to touch those apps with a 10-foot pole because they have far too many sensitive modules the DOD would have a hissy fit over if a dilettante like me plays with that expensive stuff, as if other nations don't have it in their own toolbags.

As for using offsets, i probably could use them, but why create extra material and then trim it off? Then i may as well do what i got tired of doing in the past: create a sizable solid and trim it back from surfaces i created between each foot of station line. I had to resort to that ugly option because VC & VCP don't let us do much with the mesh. I cannot boolean the mesh. It sits there, all sultry and sensually tantalizing, and my hands are tied and i'm in a chair going under water with that mesh being the last sights i have but cannot touch. I could use PolyCAD, as was recently suggested, for turning the mesh into a solid i need, but i want to be as organic in VCP as i can. All i can assume is that Spatial or somebody's got a jugular-vein-crushing chokehold on this seemingly simple could-be feature of VC incarnations.

What might be really kewl is if VCP more intuitively allowed us to flag regions of geometry en masse and say "discard material i mark; cut lines i mark here and here, sort of like Sketchup. But, i am not going to fall into using Sketchup and paying $800 or whatever it is to handle DXF entities. But, it is nice to see how smoothly solids can intersect and be moved and regenerate very quickly. I know VCP is supposed to be able to do that, but if i have to move sideshell or somebody in esteem tells me my bulkheads need x feet more separation to 51 ft vice 50 feet, i would do that in Delftship and then reimport the stations, rebuild the affected decks.... (that would be because there is no kind of two-way comms between Delftship and VC/VCP to instantaneously correct for hydros. The $100,000 seat packages would. Maybe the combo of Rhino/Orca do do that -- for maybe $5,000, but i'm in the $250 to $500 price range, hehehehe...


As you might imagine, a ship of this size faired by hand has had quite a bit of non-fairness in the surfacing and i spent many months tweaking that in Delftship. I did manage to clean up a whoolllle lot of crazy kinked lines that only slowed VCP to crawl at times. It's not something VCP can clean up without taking the model away from its original hydrodynamics approximations.

So, since the cut/trimmed lines (green) come from Delftship, and are never originated in VCP, it would be simpler and more faithful in my mind to mirror/extrude those lines so that even with microscopic spaces between the hull and the plate, they would be symmetrical and balanced for CGs reasons.

I think i even in the past tried to group all those individual 1-foot wide surfaces in lengths of 40 feet to 52 feet (distances between the transverse bulkheads) and use them to cut intersection lines, but that was not very efficient a way to do it.

I know Tim in the past replied that it wasn't Punch's forte, but i feel like crying that that tantalizing mesh is there as an unusable, faithfully-imported would-be surface. It makes me want to reach out and blame somebody outside of Punch! for this. I feel that Tim and team are imminently qualified to GIVE me what i could use, but i feel some force is restraining Tim and team. That mesh should not be such a big deal to be a surface that can be used as a normal VCP surface. For me, all the mesh is useful for is to visually see that i'm still in the boundaries of my model. To do more, i have to break the mesh to surface, and the reassemble them in groups, painstakingly color-coding them and filtering them out but leaving them visible in chunks to make sure i'm not leaving holes in the reassembly. It's tedious, eye-numbing, and makes me sleepy enough to just save the file and stop. Often, i now only do about 1 hour worth of clicking and i just quit before i tire enough to go to sleep.

One psychological trick at diffusing my unsubstantiable frustration and fingerpointing at Spatial or any others with power over the kernel is to reassemble those facets along areas that would likely be steel plating assemblies. But, in doing so, i open myself up to attack of "it's bad enough you think you know how to design a ship-- what makes you qualified to show plate cutting locations?" If i could turn that damned mesh into a contiguous surface or if it simply imported that way, i'd not fear facing that question in future.

I must be quite ignorant. I do appreciate all the mathematical complexity, the psychological upbeatness needed, and the arduous troubleshooting and so on that go into writing even the moderately simple pieces of software, but i just don't see how Delftship's export of the 3D DXF mesh can be so hard to strip of all those facets and just give a surface in coarseness to fineness according to the CPU/GPU capabilities. Failing that, why cannot that surface be touched in other tool modes and just booleaned? If somebody's got a patent on mesh-to-surface conversion then why cannot a sneaky and screw-you-buddy tactic be used to just boolean that surface and trash the offensive mesh and the trimmed-off material. (I'm thinking 1960's Mr. Phelps rubber mask, or today's injection mouding/extrusion, and so on...)

My same frustration peaks when i cannot get a volume measurement from a pseudo box/container unless i solidify it and give it a mass property. In middle school, we never were told to shade and color and assign mass props to a shape before calculating the volume. Yes, visually, intellectually, we know those lines form an enclosed volume, and i swear that a human could tell the computer "Let there be a surface.... but don't make the user build and discard for a simple measurement collection". My frustration there is i cannot figure out how to Boolean out the fuel tanks i'm going to build later. I previously drew interesecting deck plating with the sideshell and the bulkhead. Boy, was THAT mind-numbing. I end up with solids that i know i will discard. Yet, i already have surfaces there that i build to be watertight. I think i used water vice steel as the "filler". I hope my memory of those weeks is not causing me to mischaracterize VCP and VC2D/3D. I really, really like these apps, and i prefer them to ACAD.

I think i'm going to get a bowl of ice cfream and pop in a DVD. I'm vanquished now.

ZLC

Hi Zumer,

The attached file might me "missing some properties", something windows said when i copied the file from my working folder to a shared network folder from which i could upload this... I think, though, that the file should still be intact...

The curves to sweep are the green lines in area "F".

In area "G" are decks i created by making six surfaces. I left them in just for effect.

I noticed another thread's suggestion about changing the lines/curves to interpolated splines to get the 1-rail sweep to work. I got sweeps, but both the upper and lower surfaces swept as one. Since the outboard curves are not parallel to the body's centerline, sweeping the surface results in a surface i didn't take all the way to the centerline. So, i did that for each side, figuring i'd just join them. No go. I also had to add surfaces to the right and left. When i tried to stitch to solid, i got "Stitch Error". If i told it enclosed volumes not required, i of course could not get volumes nor assigne properties. Attempting to assign mass props resulted in 3 or 4 error messages until clicking through all the errors.

BTW, thanks for asking for the 2-line file. I left in extra stuff in case you wanted a sizing perspective of a portion of the model.

You'll see in my behemoth post about feature request ("Mirror Extrude (a solid) About Modelspace or a User-Defined Centerline") the overall file and my tome of text for a simple suggestion, hehhe.

Hehehe... Thanks, but i experimented with all of that, and until/unless my "feature request" becomes reality, your and my steps will be the same.

I guess i'll just have to make it a game of speed-cutting and speed surfacing and speed solidifying.

Thanks!

Arrrrgggg.. Group?

You know, i almost NEVER consciously think of "group". It would be nice if certain crucial, make-or-break sub-commands were integrated into or made as a pop-up of or some in-stream notice when using commands like this.

I know i just super-wordily exposed my ignorance of the commands available. This mistake will probably forever im-burn this into my skull.

So, it looks like you used my geometry, i'm assuming. One thing i'll need to be satisfied with is that the outboard (left/right) lines/curves need not be planar for the solidification to work, since plausibly someone might be designing jewelry or whatever.

Again, THANKS, Zumer. I'll try this in a few hours when i get lunch time.

---
edit: I think i've read two or 3 times in other threads users begging for finer-grained or expanded suggestions with the tricker of commands. When a user is trying a feature and universally/intellectually the devs and users know the result can be obtained in other ways, a suggestion should pop up. Excessive wishful thinking would be that vid-clips or in-stream preview of the presumed intent pops up and says, "These are some possible outcomes if you persist; choose one or 2 and i will adjust or constrain the tool to help you..." hehehe

ZLC.
I don't think I can contribute to help you in any meaningful way but thought you might find this thread http://moi3d.com/forum/messages.php?webtag=MOI&msg=3002.1 of interest at the MoI site. It's not about your specific (Pacific :-) problem, but could be handy for ideas in getting your hull how you want it.

Martin.

Hi bl,

I will check it out when I get a chance. Thanks!

Hi,

I have these lines that refuse to be cut where told. I tried using cutting surfaces such as a circle with a surface, and a cutting plane.

This is not the first time i've had this type of line-segmenting problem.

I need to segment the larger curves (green and yellow) because the green solid (it sits at the 36' elevation) made from stitched surfaces based on the two curves generates noticable lag when i zoom or pan or whatever. As soon as i turn off the layer of the offending geometry, zooms return to a better speed. the other solids below the 36' level don't have this lagginess problem. However i rebuild the surface, it happens. I haven't yet "simplified" the object.

Interestingly, other curves don't exhibit this problem (well, maybe two others do) to this degree. i cannot resolve the problem by changing them to plylines, and i am reluctant to further mess with the object type of the lines/curves.

When i use the same x-axis value at which to cut/segment these lines, neither ends on the same or correct x value.

Also, the curve start and end x values do not always show the actual points nor distance.

I haven't yet compared this behavior to what might happen in AutoCAD 2008 or 2010, so i don't know whether this issue is VCP importation/interpretation of the 3D DXF polylines or something else.

Also, i realize there is some 'non-parallelness", or waviness (and, therefore, some change in direction) that may be causing the "craziness" in the very red area of the otherwise green solid. I'll resolve the waviness part later. However, all the other lines were 'faired' by my hand, so they too, i would presume, should exhibit this density/laggy-video/redraw issue.

Thanks for any attention/playing with my model.

ZLC.

Thanks! I'll check it out.

For now, i'm just mirroring the green & yellow x-axes lines about the origin, and then start my y-axis lines from one side out to a convenient length from the yellow and from the green (x-asis) lines and then applying the y-axis' opposite sign to avoid excessive zooming and trimming.

That's cutting down some workloading on my mind. I still hope Tim and team really take heart to my tool suggestion and back port it. I really, truly believe it is a massive step-saver when generating solids that could be swept from just two lines when a desired origin or mirror point is known or settable or cannot be altered but must be used.

heheh... Thanks!

I'll try THAT, too. I didn't get to do my laptop drawing during lunch. Well be maybe 9pm my time before i can try, unless i can do it on my train ride home...