Summary of changes:
* Pivot snapping bug fixed.
* All user setting file export/import now default to Documents folder.
* Scaling now respects manipulator pivot position.
* Chamfer tool is improved. The outcome is more reasonable.
Thanks!
passerby: take your time.
rooster: The manual pivot is a temporary stuff, but you can save it to the Transform List for later use. It is a global figure and not bound to any object. You can use the button "Pivot" tool under object mode to edit object pivot. To edit mesh pivot, you have to use the Hierarchy mode.
I can't give you a time frame about the Collada format. There are some tech issues need to be sorted out.
As long as I maintain the object library functionality, I have to keep the current material system as is. Because objects in the library can not use scene material.
I can't reproduce the undo bug you mentioned. Can you tell me the exact steps?
nabeelaejaz: ghib is right. In VW, one edge can only connect to one or two faces. The good thing for this is subdivision can be sure working properly.
Animation part is suspended in v 2.4. it is only available in v 2.3. I wonder how you can spot those interface. They are supposed to be hidden. Can you post over a pic?
Pior: Thanks for your generous 2 cents! I'll keep them in my top draw.
I fully unerstand what you mean. The fact is some fundmental rules are very hard and even impossible to change as the program has grown to such extent. They have been applied across the program.
ya i got a idea going around for the mockup, IStonia but im still trying to get some good images too use, so if anyone has any art they made with VW and dosnt mind it being used in the mockup and maybe the site and some point it would be nice to get a hold of it the the proper names to credit it to.
When I had pressed the register button then the animation timeline came up. That was the only time I guess.
Modo has this edge extend thing, it is used quite often in tutorials and has great benefits, as I use it often too. In the Edge menu, there is an Extrude option, I think that does not do anything at the moment, why not use it for this purpose if it is not doing anything. The edge extend is always connected to one face, so it fits the definition as well.
Summary of changes:
* Collada foramt is available in Save/Load Scene operations.
* Close edge extrusion is possible. Please beware that the extrusion will be a seperate sheel and does not connect to the original mesh.
* Close Vertex extrusion is possible.
* A new event for streamline tool is added, "On Tapped". It fires if the tool is activated then deactivated without performing any action.
Thank you for the extrusion. Why can't it be connected to the original object, when it can be connected to the second object as shown in the picture above. Also the extrusion comes out in odd direction and shape. Well its a great start but a little more tweaking needed. Do check out the trial version of Modo from their website to get the idea.
nabeelaejaz: As I have stated before, one edge can only connect to one or two faces. This rule can not be changed unless I rewrite the whole program. In your case, if they are connected together, one edge will connect to three faces.
You can use the Move manipulator to control the extrusion, or use extrude in cursor direction. That way you can control the extrude direction.
Thank you for the extrusion. Why can't it be connected to the original object, when it can be connected to the second object as shown in the picture above. Also the extrusion comes out in odd direction and shape. Well its a great start but a little more tweaking needed. Do check out the trial version of Modo from their website to get the idea.
This is a very bad idea, This will stop the mesh smoothing correctly as 1 of the 3 faces normals will always point in the wrong direction. Its not really supported by any apps, and allowing it introduces the possibility that it can be created by accident.. which is really bad, especially in a dedicated modelling application that absolutely needs to make portable geometry.
Im amazed that modo allows it. Googling it actually demonstrates the point. there are folks wondering why their meshes wont smooth in maya
that's not how it works in 3dsmax either, you wouldn't be able to extrude that edge unless either the top, or the side face were deleted first.
Max doesn't acceot non manifold geometry, I say it's both a good thing and a weakness. While it's very appreciated when doing subd work (you don't have to worry about breaking your mesh), it's a terrible limitation when doing low poly work. Let's say you want to add alpha mapped bits to a mesh: you can't.
In max you can't do that:
Now I'm not asking istonia to consider adding it, that would be silly to ask for a complete core rewrite at this point, but I understand the disapointment of some users, but there's still some workaround (you can always snap vertices when needed)
thanks for the collada update . unfortunately when I import it into the engine I'm using (quest3d) nothing comes out . I don't get an error.. maybe the version of collada is too new or something . I've been using 1.4.1 I think
thanks for the collada update . unfortunately when I import it into the engine I'm using (quest3d) nothing comes out . I don't get an error.. maybe the version of collada is too new or something . I've been using 1.4.1 I think
what formats does quest support.
im thinking it is prolly easier to just do your modeling in VW than use something else for exports to engine.
if your going the free route blender has pretty good fbx and collada support now.
rooster: The unit setting could be the problem. If the object is too big or too small, you may not be able to see it. Open the dae file with Notepad, you can find this line
<unit meter="100" name="meter" /> in those top lines.
Try to change "100" to a smaller or bigger value and see what may happen.
If this is not the problem, if you can get me a simple sample file that works in quest3D, I may be able to find a way.
passerby: yeah I did look at blender but I find that exporting from a seperate program really slows things down in the long run, and adds an extra step to debugging any problems- eg. was it the first or second export causing an issue?
also, it seems blender's unit setup is kind of confusing, like the grid settings for the viewport you're in determine the scale? if you change the grid settings you get a different size cube when you create one, but they're both a size of 2? I was confused anyway :P
IStonia- I think it's the file, because I'm looking at the data not the viewport- normally I'd see a matrix node, mesh node, material node etc appear. here's a file exported from maya: http://www.redprodukt.com/misc/bullet.dae
one thing I'm just noticing, in other 3d programs when I open a voidworld exported cube dae, is that the cube is parented under a bone called bone_root. I wasn't expecting to see that, it sounds like a good lead? I'm used to seeing the box on its own
thanks!
EDIT:
the blender stuff is relating to stuff I read about 2.4x.. I still run into problems with blender being z-up though
ah on blender 2.5x you can make it automatically rotate to fix the z up on import and do the same on export.
and the grid problems your having is that things move in increments the size of the grid space you actually have to hit shift+s to snap it to the grid first to have it move on grid point.
but like i said it is just s suggestion, it is the apprache i used for bring stuff into udk.
i do my modeling in VW or silo than bring it into maya or blender for the fbx export and/or adding any bones or deformations.
rooster: Please try this new update for the Collada export.
There is a new export option, make sure it is checked.
Edit > Options > File Format Options > dae > Flatten hierarchy
hey IStonia, I only did a basic test so far but it appears to work like a charm
thanks for sparing me blender!! :]
oo, quick thing I just wondered about: I did a save as, and now at the top it shows I'm working on box.dae. I think I'd like a way to keep working on the original vws after I export instead of 'switching' to the exported file. I guess like doing an 'export' instead of 'save as', where export just saves to the destination file then continues work on the original
heya erm, I may have spoken too soon :P
it works great with one object, but with two they get merged together- I guess that's the flatten heirarchy checkbox? hmm
wait! hold the phone! user error on my part, I messed up my import.. I get two objects if I don't import like an idiot
I still wonder about there being bones, when other programs collada saves don't create them by default, but right now it doesn't bother me
I had one other thing I thought was worth mentioning though- I brought a silo obj in, and the seperate objects were brought in as one VW object, but seperate meshes. I began using split to seperate the meshes to objects, but soon realised that the geometry was being duplicated- I end up with a new object from the mesh I split off (split in mesh mode), but the mesh stays part of the original object too. I guess this is a bug?
the other thing that occurred to me, is I'd like one button which just splits up a multi-mesh object into seperate objects (it could be next to combine, called seperate- it just does the opposite)
edit: maybe also an option to split obj import into objects rather than meshes? this is only a small idea, a one button seperate would make doing it manually easy anyway
ya a nice separate would be nice. maybe have it separate out the selection if in anything but object mode and separate out all separate meshes when in object mode.
Summary of changes:
* File > Export Scene As. It does the same thing as "Save Scene As" except it won't change the original file definition.
* Mesh "Split" Command is improved. The original mesh will be removed.
* Object "Split" Command is added. It will split multi-mesh object into single mesh objects.
Thanks!
rooster: I couldn't figure out what's the reason why multi-objects got merged in quest3d importing. I created one cube and one plane, exported it into a dae file. I open the dae file and can see there are two parallel nodes. That's the same as your maya dae file.
In this case, the importer can either treat it as two object or merge them into one object base on import option settings if there is any? or base on information from the dae file, but that does not sound like logical.
Anyway, there is a special element in the maya dae file:
<comments>ColladaMaya export options: bakeTransforms=0;exportPolygonMeshes=1;bakeLighting=0; isSampling=0;curveConstrainSampling=0;removeStaticCurves=1; exportCameras=0;exportCameraAsLookat=0;exportLights=0;exportJointsAndSkin=0; exportAnimations=0;exportTriangles=1;exportInvisibleNodes=0; exportNormals=1;exportTexCoords=1; exportVertexColors=0;exportVertexColorsAnimation=0;exportTangents=0;exportTexTangents=0; exportXRefs=0;dereferenceXRefs=1; cameraXFov=0;cameraYFov=1doublePrecision=0</comments>
Can you open a dae file created by VW with Notepad, past this element just below the <authoring_tool> element, save it as dae file, then try to import it into quest3d and see what may happen?
you're a machine!
about the import: I had left it on an incorrect import setting -flatten heirarchy- after trying them all to see if i could get the original collada export to work; so when i set it back to keep heirarchy the objects came in as expected, albiet renamed with 'bone' prefix. sorry for the mistake there!
btw if you remember the chamfer undo bug I mentioned a little while back, here's how I recreate it:
1. create 1x1 box
2. in edge mode chamfer all the edges
3. in object mode duplicate the object and move it away
4. select the original object
5. now keep hitting ctrl-z. for me, the chamfer operation is not undone. I can keep hitting ctrl-z until the object dissapears, and it retains its chamfer all the way
rooster: I still can't reproduce it. Here is what I tried
1. create a box.
2. goto edge mode and inverse selection to select all edges.
3. press down F key and move cursor to chamfer in NMB action.
4. go back to object mode and hit the Duplicate button.
5. move the duplicated box away.
6. select the original box.
7. keep ctrl + z. the chamfer will be undone before the box dissappear.
Btw, if you can reproduce it again, after chamfering, Edit > Show Undo Stack and see if the chamfer undo is captured.
one difference is I'm using the right hand panel, not shortcuts to achieve the chamfer. Just selecting the edges via marquee, chamfer button, dragging on edges. I'm currently looking at a chamfered box and there's no chamfer in the undo stack
I've found something interesting: If I hit the chamfer button and drag on an edge, the chamfer is not added to undo list. If I hit the chamfer button, choose a value and hit 'apply' the chamfer *is added to the list. I guess just dragging on the edge is somehow failing to register the chamfer?
I've got some more ideas in the mean time btw
some more minor suggestions:
- I'd like it if 'reset base' in object mode reset all selected objects- I think it only resets one currently, the first or last selected I guess
- in the object pivot tool, I'd like a 'center pivot' button (centers to object bounding box)
- not sure how best to approach this, but setting soft selection distance is tricky for me because of my scene scale settings. the usable area of the sliding bar is tiny and hard to get the value I want without the entire model just going red- I wonder if there's some scale-independent way you could control the falloff distance? I like modo's where you hold right mouse (it could be a button though) and drag, with a visual circle onscreen showing the size you're adjusting. then if you're zoomed out, the size could adjust in large amounts, and zoomed in just small amounts?
edit: I think double clicking a name in the object list should trigger renaming
request:
- fairly big request here, not sure if they've been mentioned yet: deformers. eg: twist, bend, lattice. I don't know how this would integrate with VW as it stands, since these usually benefit from having persistent gizmos which you can go back and edit, then apply once you're fully happy. I'm not sure I'd go so far as to suggest a modifier stack, as that's quite an undertaking, but I like in maya how the lattice will stay in the scene to edit until you're done (eg, not a one-shot tool)
No, still ok for me. When you are draggin the mouse button, you should see {LMB} top-left. When you release mouse button, LMB should dissappear.
The "Reset Base" can be improved.
"Center Pivot" can be done.
When the tweak streamline tool is on, Ctrl/T key, move the cursor over the object and should see a softselection effect cage in black, scroll WMB to adjust it. Make sure soft selection is enabled.
I am not sure how deformer works. Modifer stack is too big for a one person team.
hey again will try not to keep bugging about this collada stuff but I just tried to export an asset into my game, and notice it's rotated 90 degrees (should be facing z but its facing x). I notice taking the asset into maya that it comes in the correct orientation, but it has a rotation of 90degrees in y. When I use the asset, the code is expecting that with rotation 0,0,0 the object is facing the intended direction (z, as it is in VW)
chamfer issue: well, after trying it many times I find that sometimes it does undo and sometimes it doesnt. I can't find a pattern. I don't think it needs the object duplication step, just chamfer then hit undo straight after, chamfer again then undo
one thing I am noticing, is that occasionally I'm pressing ctrl-z a bunch of times, and sometimes {ctrl} dissapears so just 'z' is being picked up (which I remapped to focus on object). I don't think its relevant to the bug though
also, can I suggest that new selections should be locked from being made when the chamfer tool is entered? reason being several times I clicked near an edge intersection and before the chamfer began, a new edge was selected and I was chamfering more than intended. also, new edges tend to get selected by accident as you drag around the mouse to chamfer the object- sometimes you just want the chamfer result highlighted after the operation
__________
EDIT: ok, I think this last one seems connected somehow. I notice if I chamfer, and while I'm dragging I select a new edge- it seems to be able to undo. If I drag and do not select any more edges- no undo. this seems fairly reliable when I test it. This applies to new edges created by the chamfer too
Summary of changes:
* Object "Reset Base" command now works on all selected objects.
* Two more button functions for object "Pivot" tool, "Align to Bounding Box Center" and "Aling to Geometry Center".
* If Paint_Select is mapped form MMB to LMB, Paint_Select operation may conflict with some operations like chamfer/extrude/etc. This update has fixed those conflicts that have been discovered so far.
* Two more collada export options "Align top node to world origin"/"Align top node to world orientation". Some importers may ignore top node pose settings and align it to zero.
Thanks!
rooster: please test and see if those problems still there.
As for ctlr + z. I hold down ctrl and keep tapping z, can't see a problem.
hmm, with the ctrl-z thing, are you testing it on its own or actually undoing a bunch of stuff? it only seems to happen when I'm continuously undoing a big list of things.
here's what I'm doing:
one after the other, inset 3 sides of a cube, then extrude each inset seperately. If I work my way undoing back to the beginning, it usually loses ctrl somewhere around insetting the third cube face.
it only happens when undoing things though, in a blank document I can tap ctrl-z over and over and it still registers ctrl
I tested with my other control button too to make sure its not the keyboard
edit: what's interesting, is that when I hit the point it does it, if I redo a few steps and begin undoing, it happens in the same place again
edit2: it appears to happen the moment the active tool changes back to inset (during undo steps)
Summary of changes:
* Edit > Preference > Unit Setup
Numerical input is flexible. The following are the rules for inch input. The same rules apply to metric and generic input.
> 2' 2" - 1' 1" + 2' 2" = 3' 3" >White space will be treated as - or + base on the context.
> 3" 2' = 2' 3" >No order required for unit signs.
> 1'2 1/2 0.5" = 1' 3" >Divider and decimal are allowed.
> 1'2"3 "6' = 7' 5" >Unit sign can appear anywhere after a digit.
> 5 = 5" >Deault imperial unit is inch.
Hi IStonia,
First of all. You're app is great. After the last update of max i am looking for a new modeler and for what i have already seen in yours this could be the one. Only the UV mapping tools could be better. Maybe you can take a look at http://www.renderhjs.net/textools/. He has made some great stuff targeting uv mapping.
But, on a other node, can you give me some insides about what the future plans are.
Will this be for free or else what would be the price?
Are there plans for a Mac version?
Hi, new guy here.
Can anyone help me with the boolean component? It works well with 2 different objects, like a cube and a sphere as it should be. But no 2 similar objects, say a cube and another cube. When I click on "Pick mesh B", nothing happens. Am I missing something?
EDIT: nevermind, I found out that if you move one object just a little, will fix the problem.
And I second the question above. It's a great tool with a lot of potential and I'm also wondering what your goals with the program are.
Yeah, there are big rooms to improve the uv tools. But I have to leave it to furture version because it is a big task.
I have been avoiding thinking about the future plans. Thanks for pushing me! It tells me that I should make a decision now.
I think it would be better to make it a shareware. And I am not sure if $79 us is appropriate. Just try to find a point where it is light burden for users and also good enough to sustain the future development of this project.
I don't have a plan to make a Mac version. I don't have any programing experience in Mac.
well I'm glad you don't write any 4 figure numbers. I think you have three options.
1 You can do it the Blender way. Open source and they get government funds for a movie and develop there program a long side it.
2 A fixed price and then i think something like silo. For me the price tag for silo isn't a big burden. So i would buy it.
3 Like untiy a free version fully future d and a payed pro version with some nice extras....
And correct me if I'm wrong but since you develop in c# isn't it possible with mono develop? I ask because a future employer prefers mac....
personally i would pay up to $150 usd for it, the same i paid for silo 2, though to pay money for it i would want some performance issues fixed and some more polish to it.
i find it becomes very slow and non responsive at around 0.5 million faces.
and functionality wise 2 big things, improvements to UV tools and some kinda a scripting preferable python.
edit:
if you made improvements every year i would be willing ot buy a new version every year too.
Free to use -> Pay a license if using commercially -> sell plugins and stuff.
Then you can create a whole economy under your app. Maybe you don't want to code EVERY possible thing people could want - expose a plugin system so people can extend your program. Maybe something is not worth adding to the core package unless the user is actually going to use it - plugin, payed or not.
Require a commercial license to publish plugins which cost money.
Just some thoughts, I personally cannot stand Unity for the simple fact that for it to even be worth using you need to spend 1500 dollars. One should just use UDK for all that. At least that's unlimited and you don't owe money until you use it WITH some success.
Blender is awesome and all, but their community sort of falls into training-only in terms of economy based on the application. Other than that, it's using the program itself commercially.
If you make it possible to create commercial plugins, you've got potential for addon economy and training.
I wouldn't target development firms for volume, either. I'd target end users and hobbyists - the hobbyist is quickly becoming not just indie, but tomorrows big developer. You have a lot more potential clients when you focus on those categories. You can become an alternative to autodesk, for people who need alternatives.
Blender 2.5x and beyond is going to be very, very hard to compete with, though. Being free until commercial helps you here, but really, blender is very, very good nowadays and completely free.
This all matters the most today than it ever has in the past because the ability to create interactive entertainment gets easier and easier to enter by the week.
One could argue that blender ain't getting anyone any jobs, but this program isn't either, and blender is very quickly becoming something nice to have around even in large companies, even if it isn't a major part of anyone's pipeline, it has its uses even for people who have a bunch of paid seats in other software.
In the end, I'm saying try to appeal to those people using unreal for free. All those people are either already forming teams or will be in the next few years, as that tool and others become easier to use and easier to get a hold of.
free + paid add ons + commercial licenses available (which perhaps cover some addons as well) = maximum word of mouth and exposure and maximum adoption rate, unless I've overlooked some stuff. Like, you can begin small and snowball pretty quickly if you target the new breed, as the old breed is not changing their ways any time soon, for more reasons than just being set in them - a lot of the people who buy volume from autodesk work closely with them, for example, just like they work closely with drivers vendors.
If you don't have that same luxury and friendly ties with them, then they don't really have the reason to come to you. But everyone out there learning or already aspiring has plenty of reason to give your app a look - and to buy a license when they feel that what they are doing with it has potential to be commercial.
Just a loose set of ideas, I wouldn't even consider this advice, just some things to think about. You're entering at a time where you can really capitalize on an all new market that wasn't there just a couple years ago.
@mortalhuman this app isn't competing head and head with blender, it has a focus on poly modeling which is a big weak spot in blender(blenders modeling and rendering tools are lagging way behind), VW is more competing with things like Silo, Wings, and maybe Modo.
and addons is already covered by people who wanted scripting.
and the reason why i and others brought money into it is because with funding the app can improve more, it should at least allow the developer to put more time into developing VW if the app is paying for the time he puts into it, and if it gives a return.
Replies
http://www.digitalfossils.com/Download/VoidWorld-may-5-11.rar
Summary of changes:
* Pivot snapping bug fixed.
* All user setting file export/import now default to Documents folder.
* Scaling now respects manipulator pivot position.
* Chamfer tool is improved. The outcome is more reasonable.
Thanks!
passerby: take your time.
rooster: The manual pivot is a temporary stuff, but you can save it to the Transform List for later use. It is a global figure and not bound to any object. You can use the button "Pivot" tool under object mode to edit object pivot. To edit mesh pivot, you have to use the Hierarchy mode.
I can't give you a time frame about the Collada format. There are some tech issues need to be sorted out.
As long as I maintain the object library functionality, I have to keep the current material system as is. Because objects in the library can not use scene material.
I can't reproduce the undo bug you mentioned. Can you tell me the exact steps?
nabeelaejaz: ghib is right. In VW, one edge can only connect to one or two faces. The good thing for this is subdivision can be sure working properly.
Animation part is suspended in v 2.4. it is only available in v 2.3. I wonder how you can spot those interface. They are supposed to be hidden. Can you post over a pic?
Pior: Thanks for your generous 2 cents! I'll keep them in my top draw.
I fully unerstand what you mean. The fact is some fundmental rules are very hard and even impossible to change as the program has grown to such extent. They have been applied across the program.
Modo has this edge extend thing, it is used quite often in tutorials and has great benefits, as I use it often too. In the Edge menu, there is an Extrude option, I think that does not do anything at the moment, why not use it for this purpose if it is not doing anything. The edge extend is always connected to one face, so it fits the definition as well.
http://www.digitalfossils.com/Download/VoidWorld-may-10-11.rar
Summary of changes:
* Collada foramt is available in Save/Load Scene operations.
* Close edge extrusion is possible. Please beware that the extrusion will be a seperate sheel and does not connect to the original mesh.
* Close Vertex extrusion is possible.
* A new event for streamline tool is added, "On Tapped". It fires if the tool is activated then deactivated without performing any action.
Thanks!
Thank you for the extrusion. Why can't it be connected to the original object, when it can be connected to the second object as shown in the picture above. Also the extrusion comes out in odd direction and shape. Well its a great start but a little more tweaking needed. Do check out the trial version of Modo from their website to get the idea.
You can use the Move manipulator to control the extrusion, or use extrude in cursor direction. That way you can control the extrude direction.
keep up the good work mate
This is a very bad idea, This will stop the mesh smoothing correctly as 1 of the 3 faces normals will always point in the wrong direction. Its not really supported by any apps, and allowing it introduces the possibility that it can be created by accident.. which is really bad, especially in a dedicated modelling application that absolutely needs to make portable geometry.
Im amazed that modo allows it. Googling it actually demonstrates the point. there are folks wondering why their meshes wont smooth in maya
Wow that´s ... i can´t find words to describe it.I mean seriously?
Max doesn't acceot non manifold geometry, I say it's both a good thing and a weakness. While it's very appreciated when doing subd work (you don't have to worry about breaking your mesh), it's a terrible limitation when doing low poly work. Let's say you want to add alpha mapped bits to a mesh: you can't.
In max you can't do that:
Now I'm not asking istonia to consider adding it, that would be silly to ask for a complete core rewrite at this point, but I understand the disapointment of some users, but there's still some workaround (you can always snap vertices when needed)
what formats does quest support.
im thinking it is prolly easier to just do your modeling in VW than use something else for exports to engine.
if your going the free route blender has pretty good fbx and collada support now.
<unit meter="100" name="meter" /> in those top lines.
Try to change "100" to a smaller or bigger value and see what may happen.
If this is not the problem, if you can get me a simple sample file that works in quest3D, I may be able to find a way.
also, it seems blender's unit setup is kind of confusing, like the grid settings for the viewport you're in determine the scale? if you change the grid settings you get a different size cube when you create one, but they're both a size of 2? I was confused anyway :P
IStonia- I think it's the file, because I'm looking at the data not the viewport- normally I'd see a matrix node, mesh node, material node etc appear. here's a file exported from maya: http://www.redprodukt.com/misc/bullet.dae
one thing I'm just noticing, in other 3d programs when I open a voidworld exported cube dae, is that the cube is parented under a bone called bone_root. I wasn't expecting to see that, it sounds like a good lead? I'm used to seeing the box on its own
thanks!
EDIT:
the blender stuff is relating to stuff I read about 2.4x.. I still run into problems with blender being z-up though
and the grid problems your having is that things move in increments the size of the grid space you actually have to hit shift+s to snap it to the grid first to have it move on grid point.
but like i said it is just s suggestion, it is the apprache i used for bring stuff into udk.
i do my modeling in VW or silo than bring it into maya or blender for the fbx export and/or adding any bones or deformations.
There is a new export option, make sure it is checked.
Edit > Options > File Format Options > dae > Flatten hierarchy
http://www.digitalfossils.com/Download/VoidWorld-may-12-11.rar
thanks for sparing me blender!! :]
oo, quick thing I just wondered about: I did a save as, and now at the top it shows I'm working on box.dae. I think I'd like a way to keep working on the original vws after I export instead of 'switching' to the exported file. I guess like doing an 'export' instead of 'save as', where export just saves to the destination file then continues work on the original
it works great with one object, but with two they get merged together- I guess that's the flatten heirarchy checkbox? hmm
edit: here's a sample dae with 2 cubes: http://www.redprodukt.com/misc/2cubes.dae
I still wonder about there being bones, when other programs collada saves don't create them by default, but right now it doesn't bother me
I had one other thing I thought was worth mentioning though- I brought a silo obj in, and the seperate objects were brought in as one VW object, but seperate meshes. I began using split to seperate the meshes to objects, but soon realised that the geometry was being duplicated- I end up with a new object from the mesh I split off (split in mesh mode), but the mesh stays part of the original object too. I guess this is a bug?
the other thing that occurred to me, is I'd like one button which just splits up a multi-mesh object into seperate objects
edit: maybe also an option to split obj import into objects rather than meshes? this is only a small idea, a one button seperate would make doing it manually easy anyway
http://www.digitalfossils.com/Download/VoidWorld-may-13-11.rar
Summary of changes:
* File > Export Scene As. It does the same thing as "Save Scene As" except it won't change the original file definition.
* Mesh "Split" Command is improved. The original mesh will be removed.
* Object "Split" Command is added. It will split multi-mesh object into single mesh objects.
Thanks!
rooster: I couldn't figure out what's the reason why multi-objects got merged in quest3d importing. I created one cube and one plane, exported it into a dae file. I open the dae file and can see there are two parallel nodes. That's the same as your maya dae file.
In this case, the importer can either treat it as two object or merge them into one object base on import option settings if there is any? or base on information from the dae file, but that does not sound like logical.
Anyway, there is a special element in the maya dae file:
<comments>ColladaMaya export options: bakeTransforms=0;exportPolygonMeshes=1;bakeLighting=0; isSampling=0;curveConstrainSampling=0;removeStaticCurves=1; exportCameras=0;exportCameraAsLookat=0;exportLights=0;exportJointsAndSkin=0; exportAnimations=0;exportTriangles=1;exportInvisibleNodes=0; exportNormals=1;exportTexCoords=1; exportVertexColors=0;exportVertexColorsAnimation=0;exportTangents=0;exportTexTangents=0; exportXRefs=0;dereferenceXRefs=1; cameraXFov=0;cameraYFov=1doublePrecision=0</comments>
Can you open a dae file created by VW with Notepad, past this element just below the <authoring_tool> element, save it as dae file, then try to import it into quest3d and see what may happen?
about the import: I had left it on an incorrect import setting -flatten heirarchy- after trying them all to see if i could get the original collada export to work; so when i set it back to keep heirarchy the objects came in as expected, albiet renamed with 'bone' prefix. sorry for the mistake there!
btw if you remember the chamfer undo bug I mentioned a little while back, here's how I recreate it:
1. create 1x1 box
2. in edge mode chamfer all the edges
3. in object mode duplicate the object and move it away
4. select the original object
5. now keep hitting ctrl-z. for me, the chamfer operation is not undone. I can keep hitting ctrl-z until the object dissapears, and it retains its chamfer all the way
1. create a box.
2. goto edge mode and inverse selection to select all edges.
3. press down F key and move cursor to chamfer in NMB action.
4. go back to object mode and hit the Duplicate button.
5. move the duplicated box away.
6. select the original box.
7. keep ctrl + z. the chamfer will be undone before the box dissappear.
Btw, if you can reproduce it again, after chamfering, Edit > Show Undo Stack and see if the chamfer undo is captured.
I've got some more ideas in the mean time btw
some more minor suggestions:
- I'd like it if 'reset base' in object mode reset all selected objects- I think it only resets one currently, the first or last selected I guess
- in the object pivot tool, I'd like a 'center pivot' button (centers to object bounding box)
- not sure how best to approach this, but setting soft selection distance is tricky for me because of my scene scale settings. the usable area of the sliding bar is tiny and hard to get the value I want without the entire model just going red- I wonder if there's some scale-independent way you could control the falloff distance? I like modo's where you hold right mouse (it could be a button though) and drag, with a visual circle onscreen showing the size you're adjusting. then if you're zoomed out, the size could adjust in large amounts, and zoomed in just small amounts?
edit: I think double clicking a name in the object list should trigger renaming
request:
- fairly big request here, not sure if they've been mentioned yet: deformers. eg: twist, bend, lattice. I don't know how this would integrate with VW as it stands, since these usually benefit from having persistent gizmos which you can go back and edit, then apply once you're fully happy. I'm not sure I'd go so far as to suggest a modifier stack, as that's quite an undertaking, but I like in maya how the lattice will stay in the scene to edit until you're done (eg, not a one-shot tool)
what do you think about deformers?
The "Reset Base" can be improved.
"Center Pivot" can be done.
When the tweak streamline tool is on, Ctrl/T key, move the cursor over the object and should see a softselection effect cage in black, scroll WMB to adjust it. Make sure soft selection is enabled.
I am not sure how deformer works. Modifer stack is too big for a one person team.
one thing I am noticing, is that occasionally I'm pressing ctrl-z a bunch of times, and sometimes {ctrl} dissapears so just 'z' is being picked up (which I remapped to focus on object). I don't think its relevant to the bug though
also, can I suggest that new selections should be locked from being made when the chamfer tool is entered? reason being several times I clicked near an edge intersection and before the chamfer began, a new edge was selected and I was chamfering more than intended. also, new edges tend to get selected by accident as you drag around the mouse to chamfer the object- sometimes you just want the chamfer result highlighted after the operation
__________
EDIT: ok, I think this last one seems connected somehow. I notice if I chamfer, and while I'm dragging I select a new edge- it seems to be able to undo. If I drag and do not select any more edges- no undo. this seems fairly reliable when I test it. This applies to new edges created by the chamfer too
http://www.digitalfossils.com/Download/VoidWorld-may-13a-11.rar
Summary of changes:
* Object "Reset Base" command now works on all selected objects.
* Two more button functions for object "Pivot" tool, "Align to Bounding Box Center" and "Aling to Geometry Center".
* If Paint_Select is mapped form MMB to LMB, Paint_Select operation may conflict with some operations like chamfer/extrude/etc. This update has fixed those conflicts that have been discovered so far.
* Two more collada export options "Align top node to world origin"/"Align top node to world orientation". Some importers may ignore top node pose settings and align it to zero.
Thanks!
rooster: please test and see if those problems still there.
As for ctlr + z. I hold down ctrl and keep tapping z, can't see a problem.
here's what I'm doing:
one after the other, inset 3 sides of a cube, then extrude each inset seperately. If I work my way undoing back to the beginning, it usually loses ctrl somewhere around insetting the third cube face.
it only happens when undoing things though, in a blank document I can tap ctrl-z over and over and it still registers ctrl
I tested with my other control button too to make sure its not the keyboard
edit: what's interesting, is that when I hit the point it does it, if I redo a few steps and begin undoing, it happens in the same place again
edit2: it appears to happen the moment the active tool changes back to inset (during undo steps)
http://www.digitalfossils.com/Download/VoidWorld-may-13b-11.rar
thanks for the quick updates!
http://www.digitalfossils.com/Download/VoidWorld-may-16-11.rar
Summary of changes:
* Edit > Preference > Unit Setup
Numerical input is flexible. The following are the rules for inch input. The same rules apply to metric and generic input.
> 2' 2" - 1' 1" + 2' 2" = 3' 3" >White space will be treated as - or + base on the context.
> 3" 2' = 2' 3" >No order required for unit signs.
> 1'2 1/2 0.5" = 1' 3" >Divider and decimal are allowed.
> 1'2"3 "6' = 7' 5" >Unit sign can appear anywhere after a digit.
> 5 = 5" >Deault imperial unit is inch.
Thanks!
First of all. You're app is great. After the last update of max i am looking for a new modeler and for what i have already seen in yours this could be the one. Only the UV mapping tools could be better. Maybe you can take a look at http://www.renderhjs.net/textools/. He has made some great stuff targeting uv mapping.
But, on a other node, can you give me some insides about what the future plans are.
Will this be for free or else what would be the price?
Are there plans for a Mac version?
So, thanx for the great work...
Can anyone help me with the boolean component? It works well with 2 different objects, like a cube and a sphere as it should be. But no 2 similar objects, say a cube and another cube. When I click on "Pick mesh B", nothing happens. Am I missing something?
EDIT: nevermind, I found out that if you move one object just a little, will fix the problem.
And I second the question above. It's a great tool with a lot of potential and I'm also wondering what your goals with the program are.
Thank you and keep up the good work!
Yeah, there are big rooms to improve the uv tools. But I have to leave it to furture version because it is a big task.
I have been avoiding thinking about the future plans. Thanks for pushing me! It tells me that I should make a decision now.
I think it would be better to make it a shareware. And I am not sure if $79 us is appropriate. Just try to find a point where it is light burden for users and also good enough to sustain the future development of this project.
I don't have a plan to make a Mac version. I don't have any programing experience in Mac.
1 You can do it the Blender way. Open source and they get government funds for a movie and develop there program a long side it.
2 A fixed price and then i think something like silo. For me the price tag for silo isn't a big burden. So i would buy it.
3 Like untiy a free version fully future d and a payed pro version with some nice extras....
And correct me if I'm wrong but since you develop in c# isn't it possible with mono develop? I ask because a future employer prefers mac....
sow... thanx again and keep up the good work
i find it becomes very slow and non responsive at around 0.5 million faces.
and functionality wise 2 big things, improvements to UV tools and some kinda a scripting preferable python.
edit:
if you made improvements every year i would be willing ot buy a new version every year too.
Then you can create a whole economy under your app. Maybe you don't want to code EVERY possible thing people could want - expose a plugin system so people can extend your program. Maybe something is not worth adding to the core package unless the user is actually going to use it - plugin, payed or not.
Require a commercial license to publish plugins which cost money.
Just some thoughts, I personally cannot stand Unity for the simple fact that for it to even be worth using you need to spend 1500 dollars. One should just use UDK for all that. At least that's unlimited and you don't owe money until you use it WITH some success.
Blender is awesome and all, but their community sort of falls into training-only in terms of economy based on the application. Other than that, it's using the program itself commercially.
If you make it possible to create commercial plugins, you've got potential for addon economy and training.
I wouldn't target development firms for volume, either. I'd target end users and hobbyists - the hobbyist is quickly becoming not just indie, but tomorrows big developer. You have a lot more potential clients when you focus on those categories. You can become an alternative to autodesk, for people who need alternatives.
Blender 2.5x and beyond is going to be very, very hard to compete with, though. Being free until commercial helps you here, but really, blender is very, very good nowadays and completely free.
This all matters the most today than it ever has in the past because the ability to create interactive entertainment gets easier and easier to enter by the week.
One could argue that blender ain't getting anyone any jobs, but this program isn't either, and blender is very quickly becoming something nice to have around even in large companies, even if it isn't a major part of anyone's pipeline, it has its uses even for people who have a bunch of paid seats in other software.
In the end, I'm saying try to appeal to those people using unreal for free. All those people are either already forming teams or will be in the next few years, as that tool and others become easier to use and easier to get a hold of.
free + paid add ons + commercial licenses available (which perhaps cover some addons as well) = maximum word of mouth and exposure and maximum adoption rate, unless I've overlooked some stuff. Like, you can begin small and snowball pretty quickly if you target the new breed, as the old breed is not changing their ways any time soon, for more reasons than just being set in them - a lot of the people who buy volume from autodesk work closely with them, for example, just like they work closely with drivers vendors.
If you don't have that same luxury and friendly ties with them, then they don't really have the reason to come to you. But everyone out there learning or already aspiring has plenty of reason to give your app a look - and to buy a license when they feel that what they are doing with it has potential to be commercial.
Just a loose set of ideas, I wouldn't even consider this advice, just some things to think about. You're entering at a time where you can really capitalize on an all new market that wasn't there just a couple years ago.
and addons is already covered by people who wanted scripting.
and the reason why i and others brought money into it is because with funding the app can improve more, it should at least allow the developer to put more time into developing VW if the app is paying for the time he puts into it, and if it gives a return.