cakeller

Bug Report: BUILD: 3.5.21A(CUDA) x64 OS: Windows 7 x64 HARDWARE: Dell Precision M6400 w/8GB RAM, Quadro FX 3700M w/1024MB Video Memory. BUG DESCRIPTION: The attached files both cause instant hange when opening. STEPS TO REPRODUCE: Open either 3B file from the files from Archive with both files and image-plane, and windows should bring up the: program has stopped working, a problem caused the program to stop working correctly. Windows will close the program and notify you if there is a solution to this problem I thought I was (manually) incrementally saving as, thinking I would have files to go back to - so I ended up losing a lot of work. NOTE: These files are the only ones that don't open and they are the only ones with a saved, but hidden background image?

+10 This is not the first time I've seen this come up. Best idea is simply to EMULATE the same layer behavior as Photoshop. 1) Active Layer is non-paintable if hidden. 2) Active Layer doesn't change by hiding. 3) OPTIONALLY Activate layer by picking object on said layer - like add an optional key combination that allows you to PICK the active layer by touching an object! that'd be a SWEET speed upper. this would not default behavior. you would have to hold down key combo + click on an object to activate layers

Yup the feature set is workable at this point. There are features I'd like to see, and improvements to tools - however bugz be gone first and foremost. Kill them all and make the software PRODUCTION PIPELINE Qaulity ROCK SOLID, and guaranteed - you'll be respected for it - SERIOUSLY!

thanks - that is a funny expression - and it makes me happy Wonderful improvement has been made, and not noted, that I could see, in the change-log. It appears Andrew has made the adjustment to panning so that when you pan, it's like you're grabbing the object, not some random point in space. THIS IS FANTASTICO! I look forward to the day when, finally, the zoom follows this paradigm. however - it does appear that zoom no longer flies past the object, so maybe we've made a step in the right direction on that ... zooming up to the plane, on which lives, the current pick point. that's not PERFECT, but it's about 1,000,000 times better than flying through and past your object. I LOVE when stuff improves! really important fundamental stuff like this - it's like being able to breath!

That's great - so - how to inspire that guy to do this? probably not by whining that it's not getting done. I bet that's just a way to get it NOT done

mmm.. you're right - that's true. - I think it's a profile thing relating to IP.Board. Maybe Vladimir or someone could update it for a look more inline with the forum. and actually the blue text on white background IS a little harder to read. But as philnolan3d said - at least it's working!

thanks Vladimir... it appears that finally the forum is, not only viewable on iPhone, but fully functioning! nice!

you know - more than anything (kidding with smiles) I LOVE THE TITTLE of this post... W T Heck does "Potty" mean? cracks me up every time I read it. so - with that, I will repeat myself: PLEASE PLEASE PLEASE! zoom to picked point zoom to arbitrary point zoom current layer - zoom so the current' voxel object is centerred and fills the screen. pan so picked point moves in screen space (not faster or slower)... it should be as if you've just grabed the object with the mouse cursor. I don't want any new features until this is done! seriously - this is a time-sucking annoyance. think about it - zooming and panning is something you do almost as much as the strokes... so if you improve the user-efficiency by orders of magnitude, and that's what you would do by implementing these things... think of the improvement in output from the users!! --- more output = more gets done with less effort = users happier! ok ok - I won't repeat myself again, unless this goes on another 2 months being ignored.

Funny I've been gone for 2 months or so - and now, back and looking - and nothing much seems to have changed - except PTex support... seems like most of the requests I would covet have just fallen through the cracks. would be nice if there was a visible todo list. google has a todo list, and a recently implemented "requests" list... might be useful to acknowledge the requests that will be addressed, and have a list of - "we don't know if these ever will be, if you want these vote for them (with the push-button)" and log ip addresses for the votes, so we can't repeat vote not that anyone would ... sorry - this is off this topic, but really - I feel like none of these important things are even looked at any more... but that might just be my incorrect view for lack of knowing what IS being looked at.

Zoom to 3d-pick point - OR in other words, zoom toward mouse point - whatever it's sitting on. Also Zoom toward a fixed-picked point. These two things would improve usability 10 fold. One of the hardest things to navigate around is inside small spaces (tunnels or eye sockets, etc.) if the zoom center (and pan center for that matter) is not related to your real-world focus, navigating around your WORKING area is a PITA! We've been asking for this for what... over a YEAR? PLEASE this is the biggest thing that, to me feels like it ought to be a trivial addition. Perhaps I am wrong though - perhaps it is enormously challenging? Or more than likely this wheel isn't squeeky enough - and it's just ONE MORE THING. Zoom toward pick-point is critical, a must add! Pan in screen units - current pick-point (so that the mouse movement across the screen is relative to the picked point) PLEASE PLEASE PLEASE!

Your stuff makes the hair on the back of my neck stand up, and kinda gives me the willies... Pretty neat

Yeah, one of those links said something about a setting though - for iPod Touch and iPhones to make it work... and possibly a ?skin? issue?

As Artman said - Some shaders provide ivsualization information about the surface. Shaders with a reflective look provide information about the shape of the surface in a way that is important for developing the surface. I wrote a shader that I believe is now included with the distro - the voxel grid utility shader, provides a way of 3D gridding the surface up to help visualize as well. Anyway the other way they can be used - since there's a rendering view - is for visualization / presentation rendering if you don't have an external 3D app, or if you simply don't want to spend billable time exporting and setting up for rendering. The features like turntable rendering etc provide a fast way to present your models for review and feedback.

I've only had the iPhone for about a month, so as long as I've had it, it hasn't worked. same blank white screen. It would be great if it was just a setting somewhere on the forum... e.g. "allow format for mobile devices" or something silly.... a search of the invision power community boards revealed this: Topic about mobile skins for ip board. http://community.invisionpower.com/topic/300052-mobile-skin-eta/ looks like it's something you have to provide as a possible skin for mobile? anyway...

iPhone 3Gs - no joy! probably one of the myriad techs that can't be used on iPhone/iPod LAME! if you do a google (GOOGLE APP) google search for the 3dc forums, and then under "more options" blue drop-down link underneath the search entry, there is a "formatted for mobile" link. It is almost un-readible, but it IS readible - more than a blank page. if anyone finds a better solution :!

A small technical nit to pick about transparency - because you brought it up - I have been wanting to point this out but couldn't think of where - and this isn't the best place but!!! As I have found in my experimentation - REAL transparency still doesn't exist - voxels are painted in order from top to bottom. If a lower object is transparent, the upper objects will show through... but lower objects will not show through upper objects. The transparency is quite useable for what you described I have ZERO argument there just wanted to point this out. note the ice-cube is below the enclosed object... Here's a better image... notice how you can see one behind the other, but not the other... since there's overlapping regions you should be able to see how one is trans and the other is not (hopefully you don't mind if I hijacked your posts to make a point ... ) NICE!!! your idea here works GREAT!!! you can make a clone of a part, extrude it by some amount, and then use that as the reference to copy from. That's a neat trick of course... once you paint up to the "copy surface" then what... you're stuck??? not if you've saved a second copy of your original... now you can hide the extrueded one, then ctrl+copy and paint BACK DOWN to the original. nice all around. --- bTW, is this what you were talking about or was there something else really cool?

Good process... makes total sense

----------------------------------- NEW VERSION with more controls! Update! - 2009.12.12 ----------------------------------- Material: Gridding Utility Surface: Color relates to Voxel Grid Spacing Color: Grid color is Black, or White, or Colored. if Grid is colored, Red is X grid lines Green is Y, and Blue is Z Grid Background is Black, unless grid is white, then backgroudn is white. Compatibility: ANY SN_Utility_Voxel_Grid_V2.zip ////////////////////////////////////////////// Controls: ////////////////////////////////////////////// Spacing (x, y, and z): Range: 0.0 to 10.0 Notes: values greater than 10.0 ok by manual entry values of 0.0 or lower turn off specific grid-line Sharpness (Exponent): Range: 0.0 to 100.0 Notes: higher values produce thinner grid lines. Grid Brightness: Range: 0.0 to 1.0: Notes: value of 0.0 = black grid lines value of 0.5 = r g b grid lines value of 1.0 = white grid lines other values blend between options values out of range produce other colored effects but must be entered manually. -1.0 inverts color of grid lines - with white background 1.25 has pastel colored lines - with black background 1.5 looks more like neon with inverted colors and white glowing centerline. ! with negative values and high shading values AND low exponent ( < 1.0 ) the background will take on inverted color VERY HOT!! look at green and purple topographic image below. Backgroudn Opacity: Range: -1.0 to 1.0 value of -1 is completely transparent, grid and background value of 0 is transparent background of grid (grid holes) value of 1 is completely opaque. values above 1.0 and below -1.0 are accepted but are simply clamped at 1.0 and -1.0 respectively. Shading: Range: 0.0 to 1.0 value of 0 is flat-no shading. value of 1 has almost black at at the sillouhette. values above 1.0 cause much more sever shading and blooming of the grid-lines. values below 0.0 invert the effect and make the sillouette more glowy. some combinations of Shading with Grid Brightness can be supprising - definitely worth experimenting. ////////////////////////////////////////////// see attached picture samples...

At this time the shaders are more of a modelling aid. they are not for shading the polygon surfaces... I am not sure if this is planned to change. The shader for your surface is your paint - you paint on the surface (specular, and diffuse colors, as well as normal or displacement etc)... this is rendered in a standardized way - but... nope - shaders don't shade surfaces, only voxels - at this time.

Please bear in mind this is a rather complex shader - and I just finished it, and finished it, and finished it ... Updated! 2009.12.11 ----------------------------------- Name: SN_BETA_SUPERSHADER: SN_Beta4_SUPERSHADER.zip Material: Advanced Dual PicMat Material Surface: it's Comlicated Color: Whatever you want it to be... Compatibility: GL and DirectX ------ Included in 3D Coat Distro - for a long time now. No need to d/l

Update! - 2009.12.11 ----------------------------------- Material: Utility (Shows grid over voxels) Surface: surface color is derived from XYZ position in actual voxel grid. Color: Red is X voxel grid lines, Green is Y, and Blue is Z Compatibility: GL and DX

well... that is a good question: I know laying out for UVs (which having UVs is required for all the mapping) would be more difficult. laying out seams would as well. HOWEVER!!! here's a big whatif? 2 tris = 1 quad... how hard would it be to decimate first, then remove as many EASY edges as possible to make as many quads as possible. Alternatively... if you decimate far enough, every triangle CAN represent 3 quads - just add a point in the center and subdivide the edges once... Just a thought there ... although the latter idea wouldn't be very pretty topology.

lerp was used because most of the GLSL shaders in 3DC use it ... but I did know it was not a GLSL thing... you're right - prolly better to stick to using mix (unless Akira, or Andrew had a reason for using lerp ?) as for the trunc... it's a very different function than floor - however, you're totally right, floor is what I intended - good catch. but you have to remove the sign() function before it, or you'll end up with mirroring, I thinkg, which was something I tested a while ago. XPlanarUV -= XPlanarUV - ( sign(XPlanarUV) * trunc(XPlanarUV) ); // <== dumb not -1.25 - ( -1 * -1 ) which is - 2.25 ??? what was I thinking thanks for catching that. XPlanarUV = XPlanarUV - floor(XPlanarUV); (-1.25) - ( -2.0 ) which is 0.75, which is precisely the coordinate I wanted, XPlanarUV = XPlanarUV - ( sign(XPlanarUV) * floor(XPlanarUV) ) ; (-1.25) - ( -1 * -2.0 ) which is -3.25, which is again, NOT the coord I wante. but then as long as the UVs are accepted as toroidal, then it would still give the correct coordinate. but so would the original term of -1.25... how wasteful would that be anyway.. thanks to JWiede and Akira for your help on this - I've retested it - and it seems to be fine now. ------ YAY! I just found a WIERD bug - that showed up when messing with this that caused the Z Porjection to dissappear when shadows are OFF?!? which is REALLY nutty, because it happened to be that an abs( blah blah blah (and some blah ) * 2.0 / PI; doesn't work when SHADOWS are defined?! hehe... wierd. cuz it works in HLSL just find, and ... well... whatever an extra set of parentheses fixed it...

Thanks Here is the fixed Shader... hopefully it takes care of the problem that was reported on mac (but probably is due to ATI graphics) SN_Beta2_SUPERSHADER.zip fixed a bug! (fixed some bugz... see below)

NICE! thanks Akira...EXCELLENT I guess that's an nVidia vs ATI thing... nVidia spec says pow will affect component-wise if you use float on float3 or 4... I guess ATI doesn't. I will be careful not to take that for granted! I will fix this ASAP, and hopefully it will fix it for anyone who has ATI card. thanks Akira!