Charlie_P

On the manky mighty mouse problem... Being involved with steel & oil etc. a lot my keyboard and mouse go through the mill. The latest Mac keyboard (titanium thing, looks like a Mac Pro box) is excellent in this regard, utterly bomb proof and easy to clean, much better than the older tall white keys and clear plastic thing which was impossible. The mouse is entertaining. Quick way to reinvigorate the scroll ball is to unplug the rodent, turn it upside down and vigourously rub it's scroll ball on a clean sheet of paper in all directions; keep going until there's no more tracks left on the paper. I find this way is more reliable than the detergent based methods but that's probably because I gum mine up with oil rather than more household grot. Other than the occasional de-clagging the mouse has been no problem whatsoever but the side buttons are virtually unusable - I click them by accident all the time.

It could be handled the way Cheetah 3d works. Very easy to flick between single window and four windows - side front top iso - (or as many as you like). Bit odd at first but much quicker and easier than constantly shifting the one view around as you draw. Cheetah (as the name suggests) is also a seriously quick renderer and animator and fully uses all the processors all of the time...

It could be handled the way Cheetah 3d works. Very easy to flick between single window and four windows - side front top iso - (or as many as you like). Bit odd at first but much quicker and easier than constantly shifting the one view around as you draw. Cheetah (as the name suggests) is also a seriously quick renderer and animator and fully uses all the processors all of the time...

Michael, This has happened to me more than once; I don't know the direct cause but the rogue surfaces (the two or more than have appeared to the left of the extrude) will have something to do with the way you created the path object for the extrude along path. I seem to remember that there was a thread on the rendering part of the forms about a similar problem with sills and walls. I haven't found a direct cure other than recreating the object; extract the profile (double click on the object, select "profile" from the dialog, copy the profile, exit the edit, paste it. Then redraw the path rather than copying the old one and try again. Occasionally you can convert the object to generic solids and the problem goes away, or at least you can remove the rogue surfaces with a solid subtraction. (Duplicate the object and try it on a copy as the object will loose it's history) As you have it at the moment the surfaces are unselectable and only appear under GL rendering. In my case when I have managed to isolate the surface (not always possible) they tun out to be related to the history of the object; for instance the split tool used on a 2d profile before extruding. They are always nurbs surfaces and don't appear to be caused by any one operation. Not a lot of direct help I know but I hope that gets you a little further forward. Perhaps you could copy the object into another file, check it's still misbehaving (in my experience it will be) and email it to bugsubmit as a bug it undoubtedly is. Charlie

Further to the above... Things could be simplified a little if you were building the rings of bricks into a well or column rather than having an array of single courses. In that case the circumference would have to be roughly similar for each course. Out of the 94 different combinations of the three bricks there are some (14 small bricks) which are much smaller than others (14 large bricks) which would make an interesting well. The circumferences of all these combinations forms a normal distribution (bell shaped curve) so presuming that you have a little tolerance and you wanted to maximise the variation for aesthetic reasons then you could pick out the central part of the distribution that gets you the most brick combinations for the least variation in circumference between courses. You could for instance pick out the five most common circumferences which would give you approx 32 different brick combinations (depends on the brick sizes); working through the brick sizes would tell you how much circumference variation you could cope with an still have a column rather than a heap. If however you've got a series of independent rings then revert to the above...

Could use the Model - 3D power pack - create drape surface command? Draw out your rock face lumps and bumps horizontally, use the drape surface command to "lay a tablecloth" over them then flip it to vertical Here's one I did accidentally the other day...

AndiACD Lots of advances since 11.5. If you do decide to go for the upgrade go all the way to 2008 as it's a definite improvement over 12.5.3. I see you're mostly mech design and if that's the case what about Solidworks running through Parallels? It is better than VW for that sort of thing and if you don't need Photoworks (their version of RW) the price isn't totally abhorrent. You do have to live with the SW workflow though which drives me bananas. For instance this morning I've been drawing a rotor with a number of tines arranged in a helix along the length. Dead simple couple of clicks in VW but over 10 minutes in SW - but in SW I can do a pretty picture of the stresses & deformation using cosmos lite (built in FEA-like analysis). No such joy in VW.

Solidworks is OK - I've used both that and VW - but I prefer drawing in VW. There's something about the way you have to work in SW that is restrictive. Mind you I've only used SW2008 and apparently it's been a bit of a retrograde step from 2007. The solid engine is however vastly more stable. Some of that is that SW restricts the initial sketching so that problems don't arise later on and it could be argued that VW is more capable in a wider sense that SW but it's something of a relief that you very rarely get "operation failed" messages with SW. Spaceclaim looks as though it's working in a very similar way to SW2008 but with more freedom to draw (less of the part/assembly business) and similar speed; much quicker than VW. Difficult not to drool whilst watching that vid. What you want from the application will decide how good you think it is. I use VW for a bit of everything; mapping, machine design, building design and it's that flexibility that puts it above anything else. If you've got the money (lots) then dedicated packages for each bit would get you better results but VW will do all things to a reasonably high standard. Maps & SW don't go, steel portal frame buildings in Spaceclaim would be wrong but VW will do both. VW needs precisely what this thread is titled. More power. But also More Stability. If all the tools worked all the time we would be getting somewhere but the lack of pace needs to be addressed too. VW is still a single processor app (apart from bits of RW) and that is barking. If SW and Spaceclaim can be as swift as they are why can't VW? And this is clb's point - get a better solids engine.

Also on a Mac - hold down control and scroll wheel/ball up and down on the mouse to zoom in and out. Use it all the time for that sort of thing.

Morning carpalmer, Sounds like you know your way through this pretty well but I'll go through it in detail anyway... Open the jpeg in Photoshop; duplicate the background layer; turn off visibility on the original background layer; zoom out enough so you've got a bit of a grey border around the picture. Using the magic wand tool click on the sky bit; some of it will be selected but probably not all; click on the grey border to deselect; adjust the tolerance value (just tried it with a similar picture and somewhere around 30 works for me); click on the sky bit again; if you've got most of it then change the magic wand tool mode to "add to selection" and click on the remaining bits; click delete. The sky bit should now be grey checkered (photoshop-ese for "there's nothing there"). Goto file, save as - change the format to something with an alpha channel (like PNG) and save. In VW import the image file selecting PNG for the compression method on the import dialog and all should be well. If there's still some background then more vigorous deleting in Photoshop should clear it; lay the imported image over a black rectangle in VW to see what's actually there. Some qualifications - I'm using a Mac with VW 2008 right now so there may be some slight differences. Also, I'm pretty clunky with Photoshop - there are ways to get smoother, cleaner results using layer masks and more of the photoshop tools but this is fast & easy... It is possible to get JPEG to do alpha using extra channels, TIFF too, but PNG looks more straightforward to me. All the best Charlie

Transparency in a picture file is called it's alpha channel. The tree pictures will need to have the background colours either cropped away or made transparent before you import them into VW - this is because VW sees your picture as a single entity and will vary all of it at once so it's not capable of making the background totally transparent and the foreground partially so. There are are a number of ways to do this before the picture is imported. To crop it (worst way) get it into an editing program, crop it as best you can, reimport it. You'll still have a bit of the background around the tree because the jpeg will be square, and trees are not unless they've become planks. Best and in fact only effective way is to give the original picture a transparent background. You don't say what system you've got (Mac or PC) but you need a half good photo editing program (PC/Mac) or a recent iWork program or the Graphic Converter program that used to ship with OSX (Mac only) Photo editing software such as Photoshop or The Gimp (a free version of Photoshop) allow you to select areas of the picture by the colour of the pixels; you can then select the white background only and delete it. Mac iWork has something called instant alpha which is the same thing but very much easier to do. Finally. Picture file formats. JPEG as a format has no alpha channel which means that JPEG pictures cannot be transparent. Which ever program you use to knock out the background you'll have to save the result as PNG - Vectorworks is perfectly OK with that and will import it happily; all you'll see is the tree bit and not the sky bit. If you want precise how-to's then a few system details will be needed... Good luck, Charlie

Thanks DWorks & Pat, your help is very much appreciated Charlie

Thanks Pat, I've had a few attempts and have a respectable result, considerably easier than my technique! I'm also after another version that I can't produce by that method (see attached) - any ideas? Cheers Charlie

Am I going about this the right way? What I am after is a rotated extrude of a polyline (see attached). As far as I can see the best way to do this is : 1. draw the profile (arcs, lines etc.) and compose to a polyline 2. convert to nurbs 3. duplicate array using a Z offset and a rotation to produce a spiral stack of nurbs 4. loft surface tool in loft with no rail mode, select the curves in order, click the green tick 5. go through each curve setting alignment by percentage to the same value, select create solid This produces the solid OK but when rendered there is a visible ripple along the side of the object. I've tried other methods - create helix/spiral, extrude the rib along the path with the idea of adding it to an extruded circle (rib doesn't retain the same orientation to the axis all the way down) - loft with rail on the rib profile alone with the same purpose in mind (all sorts of strangeness). Essentially this is a cylinder with a rib. I've got a working method but it's not perfect - a little too fussy perhaps? Charlie

Islandmon you are a god, sage, druid, guru, savant and many more. It's not just the design in itself it's the economy of action wherein lies the beauty.

Further to the above... We do this all the time both internally (exporting to a plasma cutter for steel profiles) and externally (laser cutters for steel & plastics or flame cutters for serious steel), so it's a familiar process. Some repetition but bear with me... 1. extract the 2d polys either Pat's or Benson's way. Be aware that both have pitfalls; Pat's way tends to give more vertices which can generate less than perfect cuts and Benson's of course only works with polys that have been extruded - if any solids operations have been done on the part (holes added, bits trimmed off etc.) then it won't work. 2. I always try to export from a fresh VW file to avoid calamities. Copy and paste the polys over from the drawing file and save it as plexiglass profiles or some such so you can have several bites at exporting without having to worry if the file has changed in the meantime. Centre the profiles around the origin so the other chap can find them easily. When we are doing this for our own cutter I'll nest the profiles at this point as it's easier in VW than other programs. If it's going to a cutting service give them a little room between bits. Do a sheet layer showing the profiles as exported with a few salient dimensions and export as PDF and email to the cutting service with the DXF. This avoids costly mistakes (I leant the hard way) 3. Export as DXF. Talk to the cutting service if you like but the best bet is an old version (I tend to use 12 or 13) and text rather than binary - but this is not hard and fast. You don't say which operating system you are using; if it's a Mac then make sure you decheck "hide extension" from the save dialog. 4. Send or take the files to the service and get them to actively confirm the dimensions (rather than "Yeah, OK it imported"). Stay on the phone or look over their shoulder as they add cut paths (the cutter instructions based on your drawing); it's at this point that open loops or double vertices show up (open loops are unclosed polys) 5. Keep your fingers crossed. In general things are OK but it helps to draw with profiling machines in mind; draw polys fully formed in 2d then extrude to 3d (no solids modeling) so you can use Benson's extraction method. Avoid ellipses as most cutters break these down into short lines and make a mess of them. Make sure your polys are closed - I use the paintpot poly creator a lot as a check. Check your vertices; if you've used add/intersect/combine into surface a lot this can leave residual useless vertices that would be better deleted. Above all never forget the dimensioned PDF. Inscribe that rule on a vital and often used part of your anatomy. All the best, Charlie

Unfortunately I'm using a similar workflow... There isn't any other realistic way to do it - seems as though this is part and parcel of the 3d alignment problem discussed on other threads. There was a plug in once on vectordepot to change dimensions to 3d objects which at least lets one dimension a 3d drawing but that doesn't address the main problem. All you can do is make sure that the relevant constraints are turned on - especially "multiple divisions" under "snap to distance" and "snap to edges" under "snap to object" As to third angle projection - does "create multiple views" under the "Machine design" menu do the right sort of thing for you? I tend to use that plus drawing over the top with lines of zero width to get the extra snaps. That 3d dimension tool is at http://www.vectordepot.com/plug-ins/ Cheers Charlie

Thanks Benson, Mostly in house design & manufacture; we sub out some of the larger machining work (some components weigh 3/4 of a ton and my lathes aren't up to that) but the majority is in house. Our lathes & mills are not full CNC (numerical readout but no programming) so turned & milled components are from drawings; we have a Hydefinition Hypertherm plasma cutter with a Burny controller able to cut to 1" plate which we indirectly program from VW - sequence is this - design part in vw (all the way to 3d), extract by various means the 2d version of the face of the plate, export a dxf of that over wireless link to the Windoze lump beside the plasma, import the dxf into the cutting program and add cut paths, nest with other bits, add the settings for the Burny etc. then export from that to the burny and cut the parts. Prolonged but flexible and not too many disasters yet (parts coming off at 1/25.4 size is easily done). Some fab work is subbed out too which means talking to Solidworks, Radarn, Inventor etc. which can be very entertaining; 3d export as SAT or IGES tends to segment arcs when imported into foreign systems which is a real pain - no such problem importing their stuff. I don't make anywhere near so much use of symbols as I should, I shall instill some disciple and make sure I get a firm handle on it. Sorry to cross thread slightly but I was reading another post that was bemoaning the lack of sheet metal tools in VW (fold & unfold etc) - this would be seriously handy in my situation. Charlie

3d positioning is a pain - Benson Shaw's list of workarounds summarises things precisely. I always work/build using the cardinal views and make use of whatever snap points work (take care that the right options are selected on the constraints palette, double click on some to get further options). You need to know the dimensions of what you are drawing fairly intimately so that you can use the 2d move command to achieve the alignments you need; drawing becomes a pretty mathematical process but the positioning is absolutely correct. Frequently I find myself drawing 2d circles/lines/arcs/rectangles over 3d objects to add more effective snap points, shift-clicking to select the underlying 3d object as well, moving the selection too where it needs to go and then deleting the overlying 2d object afterwards. The original lead on this thread was pictures of things designed with VW Mech. I haven't updated to 2008 yet so all my stuff is 12.5.3 drawn. Website : http://www.padcouk.com I make large self propelled potato harvesting machinery; there are a number of pictures scattered over the website but most particularly at http://www.padcouk.com/computerpics/computerpictures.html There are of course many others other than those on the website. Incidentally for animations I use Cheetah as well as VW.

I've also had this; managed to reproduce Petri's experience exactly just now. Changing the fill colour to something (anything) else cures it - you can change the fill colour back again afterwards and the transparency of the applied texture still shows. It only seems to occur the first time VW is run after a restart and only with GL, not RW. Once cured once everything is OK until the next restart. I have no idea why... Charlie

fsung you are of course right, I was being far too parochial. It would be nice to have both though. Charlie

update on the above... according to this Apple tech note "mutithreading" in the case of openGL means "two threading" and mclaugh's comment regarding the GPU is probably bourne out Charlie

Hello again Peter. No, haven't sent anything to NNA yet, but... Progress. I've done all the things outlined in the previous posts, and good suggestions they all were. Believe it or not what I hadn't tried was reducing the load on the graphics card. So, a bit of background to the monitor - 30" Dell (I know, I know, but I couldn't stretch to a ACD and I need pixels.) running at 2560 x 1600 and millions of colours. I changed the colours to thousands, and everything functions again. GL renders as it should under all settings. Question is why? Obvious answer is a memory handling issue but there my knowlege ends. There is also the black magic response that it's because I've got a (whisper) dell (/whisper) plugged into a mac and it took umbrage. Thanks for all the suggestions, Charlie

Islandmon - no change, still chokes on GL, renders ok under FQRW on current drawing (this hasn't always been the case, see earlier, but there's no hard and fast rule) Cheers Charlie

From what I see on activity monitor VW is not multithreading on openGL at the moment. The advantage to attacking that before the radiosity is that it would affect the whole speed of response of the program whilst moving around the wireframe drawing; this can become very slow and clunky currently, makes a pro feel like a plus - there are quite a few other threads on this subject from the past as well as islandmon and Peter's comments above. However if mclaugh is right then we're all barking up the wrong tree anyway, aren't we?