popwfx

Hang on a second guys, Unless I don't understand how Andrew fixed this or what the original problem is, there is a good reason to address this. It arose from LWO objects of mine (and others - there was nothing special about my objects except they had lots of uv maps, some of which happened to share the same texture) that could not import into 3DCoat at all. The only workaround was to import the plain object by itself with no textures (or all of them set to 128) and then once inside 3DC, change the Resolution back to 4096, then spend the next several hours creating new layers one by one and manually importing the existing textures onto new layers one at a time -- doing this for each color, spec and normal or displacement map. These changes that Andrew made are CRITICAL because now, I can import the same LWO into 3DC with all of its UV maps intact and all textures at 4096 in one click. Andrew reported that a mistake in importing before was creating 3 times as many internal textures as the thing needed by accident during the import (if I understand him right) - which made the imports crash 3DC. Now large LWOs import just fine into 3DC without problems. I am not keen on a reversal of this fix which makes LWOs with many uvs and many external textures not to load and which require this manual, laborious workaround. If there is another problem, fix that, but don't take away this fix. And also, don't take away the other fix he did with imports which is importing each LWO referenced texture as a separate layer. Before that the import process collapsed/merged all textures into a flattened, single layer, which does not allow for good editing. With importing them as separate layers, it means I can adjust or edit portions of the spec or depth or color which are just not possible with a flattened series of layers. If you guys are having a problem, be sure that we determine the problem first, I really don't want to see a necessary fix like the LWO importing working finally get reversed. There is probably another reason or perhaps this importing fix needs tweaking/debugging, but don't kill the feature please. thanks EDIT: Another reason this was added was sometimes I might have many UV maps which all reference the same texture. So under the hood the texture should only be loaded once but applied to each uv set separately - I think that was the problem - whatever it was, it was making importing many LWOs crash and not load, unless you manually imported them by hand and added surfaces and layers one by one, which takes forever. I agree it needs to be looked at if you are having a problem, but this importing fix to stop crashing and multi-layer creation of textures on import stuff *needs* to be there.

are the WD larger hybrid drives just as good (perceptually? not benchmark-wise) They are blazing with premiere & photoshop but havent tried them on my 3DC machine. Though I heard Crucial just released a $600 1TB drive ( http://www.extremetech.com/computing/152969-crucial-m500-the-first-1tb-ssd-priced-at-just-0-60gig )

I'm asking my basic questions in this section (even though I'm very comfortable with many parts of 3DC it is comprehensive enough of an app that there are some holes in my knowledge).... so How does one projection paint with an image that automatically aligns to the direction of your stroke? so that the image you choose aligns/warps with your stroke, but essentially lets you paint through that image and recycles/continues with the same image from the top when you hit the bottom of the image - so effectively like a vertically seamless image that paints along your stroke. Like the image spline painting, but freehand - is this possible? what tools/settings do I use? thanks

Also, "Get your Coat on!" is a fantastic tag line!

Yeah, so I was like "how good could a new features video be??" and I finally watched it and was like damn that IS a good promo video. And I too really dig the music. It gets you exicted about 3DCoat and your timelapses with some nice editing choices that fit the music was great. I also like that it showed a variety of types of art and wasn't too heavy in the fantasy beast thing. Didn't show any of the new importing, painting or baking features - maybe those are bug fixes? But this is a cool video nonetheless. Makes me realize I need to pay for this beta version we've been using for months now. I want the upgrade to be released so we can do that, but I'm holding out for a few more painting or UI features... And you are welcome, for me not resorting to posting a "first I was like" meme pic illustrating my 3DCoat joy :p

Thank you for this post and thank you to Taros for making the video, it was very helpful as I hadn't used the navi-setup screen before to adjust painting starting outside the mesh!

oh ok thanks. I was thinking you needed to hold the v key down and click while you were in the tool. That sort of works just the v key when in the brush tool and pressing the key while hovering over a surface, that part wasn't clear. But the actual eyedropper tool itself is still a bit odd and doesn't work all the time for depth and spec.

Say I'm touching up a spec layer that has already been painted, and say I want to "erase" part of that or fill an area with the same level spec that is in the layer somewhere. If I don't know the % specularity, then other than exporting the layer to photoshop and painting it there and reimporting, then what is the best way from within 3DCoat to pick the grey level of an existing spec pixel and use that spec level to paint spec with? i.e. I'd like to pick spec with the eyedropper tool, but that doesn't seem to work -- Oh wait... I just noticed there is a "Pick Depth too" check box (but no "Pick Spec too") though I just tried, picking depth doesn't work, only picking color - and yes I've tried both with and without "Choose color from all layers" on and off... any ideas? EDIT: I just noticed, that picking depth picks the color of the depth from a normal map as opposed to setting the Depth % for your brush to the right percentage for that color., Do we need a new tool that is a Depth Eydropper and a Spec Eyedropper too?

Just figured something out, it's not a perfect way to determine if any stray soft or alpha'ed pixels are on a layer, but if you turn off grid and axis and hide all layers except the one you are interested in by solo'ing the track/layer by Alt+clicking on the eye icon, and then turn on Wireframe, you can zoom around a clear object and try to find any accidental brush strokes or tiny errors you might have inadvertantly added. Manually playing with the light angle in the viewport helps you find stray spec pixels, but depth checking kind of needs to be done in the 2D editor, although depending on how good your eyes and monitor is, you may miss some subtle depth strokes if they are very close to zero. I would still like a way to tell me there are NO pixels on a channel in a layer (or a way to Select All depth from a particular layer and delete it in one go), but this works ok.

I'd like to see better Dual Monitor support in 3DC coat, I use 2 monitors and would like to be able to undock the 2D Texture UV Editor and have that on a different monitor(outside of the app), while having the 3D viewport and tools on the other monitor. Currently the only way to achieve this is to extend the entire 3DC application across the 2 monitors and just size the viewports. This isn't very good for 2 reasons: 1) it is clumsy and hard to line up windows, and 2) I believe CUDA does not work as fast when a single viewport is across 2 monitors, when each viewport is on its own dedicated monitor I believe the OpenGl drivers work faster/better.

there has once in awhile been a USB freakout where inserting a particular kind of USB card reader into another slot killed my mouse. But if I just pulled the logitech dongle out and immeditately reinserted it, then it came back to life. try that. I'm not sure what your original issue with the performancemx was?

thanks - I was mostly looking for a photoshop plugin or addon, and both of these tools seem to be normalmap texture creators, and it's hard to tell from their websites, but I assume they include blur, and that blur is seamless? & do you suggest either over the other?

thanks, I know the link to my Mantis view, was just suggesting a fashion to describe feature requests whether here or in Mantis, most people just put half a sentence fragment and don't thoroughly explain their issue. I was just treating this kind of like a support request format thing... that's all

In the effort to make things more efficient, easier for others here to understand your issue and lend support to it, and ultimately to make things easier for the devs to understand the issues, 3DCoat customers have, and if they choose to, to make it easier for Andrew to develop them, may I suggest the use of a template (as below) for submitting Feature Requests here or in Mantis. Because if the feature request is a sentence and not very clear, it doesn't lend itself to being understood quickly and worked on. The easier you make it to understand your issue and suggestion, the more likely it will get resolved. Plus as a another use, it is hard to see if others have dealt with this request before if everyone describes things in a haphazard fashion. Since we don't have a site for 3DCoat like they do for Unity feature requests ( http://feedback.unit...es/1/hot/active ), perhaps this template with its headings (or something like it) would be helpful: Feature Request Summary: One brief sentence or phrase that covers the request. Basically this is a way to talk about it - i.e. something like "Layer Grouping in the Paint Room" Current Problem or Issue: A short paragraph which describes the problem the feature request is intended to solve. This may be accompanied by pictures, mockups, or videos, that describe your issue, but it is not necessary if your description here is clear. Also, in this section do NOT include your suggested solution. By separating out the problem you are trying to solve, it make it easier to consolidate multiple requests that all need to be combined or possibly all addressed as a group as opposed to as an individual request - this is especially so if the issue you are having is part of a larger inadequacy or fundamental problem that needs solving. Remember to focus just on the problem you are having, and note things like how much time this problem is costing or limitations you are having in working while this problem exists. The problem may also just be an inability to achieve certain tasks without a new feature. Current Workaround (if any): Explain here briefly, any workarounds or halfway attempts at workflow you currently are using to remedy the problem. This section may be merely "none" if no workaround exists. Suggested Feature Request: Now here is where you actually describe your feature request. Try to articulate it as clearly and succintly as possible, with accompanying diagrams or mockups. Even a rough MSPaint scribble over a screenshot is better than nothing if it is not totally apparent what you are talking about. Links to Related Feature Requests or Bugs: Here just put a sentence or two and links to existing forum threads or Mantis requests which are related to the problem you are trying to solve here (not necessarily your suggested solution). If a feature request is so important to you, then 5 minutes of searching the forums and Mantis for existing related issues, and putting links here to make Andrew's life easier, should be worth it. I'm sure it would help staff consolidate requests and see easier what types of requests keep appearing. Just a thought here, but maybe some uniformity in the requests would make things easier. I see a lot of good ideas here, but they are often not very well articulated and because we are all busy, I just gloss over them until finally I see what the original poster meant, and then I can lend my support to it. What do you guys think?

I'm not exactly sure what you mean carlosa, but if what you're saying is that having the same group ids means layers in a group folder can be acted upon as a group then yes I like that. Just like in photoshop, we should be able to turn the visibility of a group off with 3DC keeping the internal group folder visibility settings whatever you have them so when you turn the group back on, it's however it was before you hid the group. This also should mean that group folders can have their collective depth, spec and opacity and blending (and masking or styles if we ever get that) modified as a group. So if Group1 contains Layer1 set at 50% depth and Layer2 at 100% depth, if you set the depth of Group1 to be 50%, it is 50% of the combined depths within the group. Just the same how opacity works with groups in photoshop. If we are revamping the layers panel, then the layers should also have locks like photoshop (but for each channel) and ideally each layer can be expanded out to show 3 sublayers of color,spec, & depth with 2D thumbnails of the content on a layer. I think someone already made a mockup like this, and if I can't find one I'd happily make one. However, as I said in another thread, I'd hate for the UI to devolve into a "user requests specific UI and it gets added" - I'd rather interaction design and a usability expert plan all permutations for a feature, but I suspect that may not be feasible here. Although there are still a number of fixes to be made to v4, I too think this feature is critical to v4's success. And, at the probable ire of other people here, I will say that I would gladly pay extra for the upgrade if it included this. I mean - this feature alone could save me hours and hours of time managine layers when painting, and my hours = $ so anything that improves workflow a great deal saves me the money of the cost of development of such a feature. I came to 3DCoat because it was easy to dive into - and that says something positive about its usability and power - but that can still be greatly improved with feature requests like this one.

I'm assuming you are talking about the workaround I mentioned above? That may not be necessary anymore with Beta14A1 - I think Andrew fixed an importing issue which caused imported textures to take like 3 times as much space as they needed to - I am successfully able to import LWOs with large textures now without this workaround, however, this is a good technique if you run into trouble with an import. Also, Andrew has not added my suggestions for the import dialog, and I'm not sure if he intends to, so if people like the idea in the mockups above we should open a Mantis or continue to ask for it (or something like that) to make setting texture sizes quicker.

thanks - I'll try it...

Thank you for the response. Must be my imagination then, as I thought the same scene file switched layers faster. I'll take your word for it. But I'll post back if I find anything definitive...

Beta 14A1 is great that it improves LWO import!! (too bad those import dialog enhancements didn't make it in though...) But I'm getting that visibility layer-switching lag that was fixed awhile ago - it is back unfortunately. What I'm talking about is it taking at least 5 seconds to make a layer invisible now in the Paint Room. This was fixed a few version back (can't recall when - maybe beta 11?) and was fine in 13 but came back in the 14 versions. Not sure why this keeps coming and going. Has anyone else noticed this lag in the latest update?

Anyone know of any Photoshop plugins or other such app that lets you do *proper* wraparound blurs while creating seamless textures? The blur filters in Photoshop aren't seamless so the edges don't wrap around and create seams when you try to tile blurred textures. I'm looking for a blur method that takes into account the pixels on the other edge when it blurs the edge pixels. I can't seem to find anything like that. I know Luxology/Modo used to have a standalone seamless tool, but I think they've abandoned it and there is nothing similar that I can find. I know I can do it in multiple steps, Gaussian blur, offset by half, select a feathered region around the seam and content-aware-fill, but I'm looking for something a little more true to what I'm trying to achieve with the underlying pixels. What do you guys use when wanting to paint seamlessly or tweak existing photographs into seamless textures? thanks

Thanks everyone for the replies! I missed this in the dropdown. Thanks! however, what do you mean about being "fixed using the view projecting method"? Good idea, I seem to always forget about the 2D editing ability. I will try the fill tool as well. I think my UVs are good. I find it hard to control this tool. I like the visual feedback it gives, but it is very cylindrical, and it looks like the applied texture is getting pinched over the 90 degree areas (at the "caps" of the "cylinders"). I wish there was a tool like this, but it was a sphere with spline control points not a cylinder. Here are my comparison tests with the generic sphere: Looks like 2D Paint (assuming you have good UVs) and Paint with spline are the winners?? (in the Paint with Spline attempt I just did a closed curve around the "equator" of the sphere - so you can see pinching in the top "poles" of it). What do you guys think? I'll attach the texture in case you can come up with a better method with a more even application. fabric_depth.zip

Thanks, puntoit, but (as usual) in my case, I am not sculpting as the source of my original models, so I only have a polygonal model imported and painted with PPP in the Paint Room, so I really want ambient occlusion I guess for the depth painting I do in the Paint Room only or via normal maps - but transferred to just darkening a duplicated color texture. This gives me an alternate lighter-weight color texture that at a glance can look like it has depth shading, but really is just color - this is for when people who have less powerful graphics cards play the game, they can get a look closer to the the full depth/normal maps, in the case that they turn the depth/normal maps off in our game to increaes performance. thanks carlosa, but the problem is the depth is independent of the color - so say I have depth to show a texture in the sole of a shoe, if I turn off the depth layer the sole looks flat. So to "shade" the sole with color that looks like the painted depth texture, I need to export that shoe sole layer as displacement map, bring it in photoshop, invert it, (and maybe edit it or blur it) and then import it back into 3DC as a color layer to multiply with the existing color layer. Make sense what I'm trying to do? So I'm still not sure how to achieve this easily without a million steps taking the exported depth to photoshop and then importing back again...

This is solved in beta 14A thanks!!

you aren't crazy, I deleted the 2nd one since the 1st one contained the wide screen 10 player and had better audio...

Is there a Mantis for this? This Undo thing is driving me crazy :p