Welcome to TVPaint Community forums !

I touched upon this question in another thread, but I wanted to ask specifically from people who have used TVPaint for a while. I have been playing with the CTG layer for painting and so far, I find it kind of fiddly as it requires a fair bit of adjusting and correcting to get it right. When you create the layer, it creates empty instances which mirror the references layer above. So the idea of using the CTG to paint a number of instances in one stroke is not presented in the first place, although I realize I can simply wipe that and start with a blank slate. But if the default is all separate instances, then how much time does the CTG layer save a user compared to the paint bucket? Is it a case of me just not getting it fully yet, or do some people who have used them both more, just end up using the paint bucket? For me, the bucket is a little more labour intensive, but more accurate. CTG often fills the wrong areas, or doesn't fill, and so you have to exclude or draw bigger or smaller squiggles. That's a number of steps and as I am doing all that, I feel I could have worked quicker with the old bucket.

So anyway, I am interested to know if people love the CTG layer, prefer the old paint bucket, a bit of both, or what?

Thanks!

You can create a preset of ctg layer that don't mimics your layer so you can paint multiple frames at a time and you can break it if needed.

I always use CTG layers simply because I always have holes in my anims, and it's nice because I can have a more rough clean and it works really well. I often need a second referenced layer that i use to add lines to correct CTG bugs, close huge gaps etc, when it does'nt work. That way I can still use my clean layer as it was drawn without having to modify it. And when i'm done, i export the colors and correct little bugs manually.
Try to scribble your whole animation without calculating, you will save a lot of time (anticipate excludes, wrong colors, missinterpreted areas etc). Then you calculate and adjust what remains.

For me it's day and night, I can't use the paint bucket for the kind of lines I work on, the gap option isn't powerfull enough. So I have to use CTG or do it by hand with the filled lasso tool... or close the gaps. But for a japanese-like lines with no gaps, I don't think I would use CTG layers.

So for more traditional line work which is mostly closed, it sounds like you feel that CTG does not really offer any advantages. Is that right? If the CTG has no instances at the start, do you try to use CTG to fill as many instances as possible and then start breaking them as needed to make changes? If a character is moving, it seems impossible for CTG to fill small areas like eyes or mouths. Only the big areas that follow a path that can be designated with a CTG mark. In that case, I'm filling eyes in essentially the same way as using a paint bucket, so what does CTG do for me that the paint bucket doesn't?

(Edit: I made a mistake in this post earlier about referenced layers which I edited out.)

So for more traditional line work which is mostly closed, it sounds like you feel that CTG does not really offer any advantages. Is that right?

In fact when i'm storyboarding for exemple, when I need to fill a character with white, it's really usefull to use ctg even if the drawing is closed. For classic animation it's kind of the same, if you have a lot of zones that are of the same color it's smarter to use CTG. It all depends. But yes for traditional line work I think it's easier and faster to use the paint bucket tool with the right settings (that are really powerfull in TVP)

If the CTG has no instances at the start, do you try to use CTG to fill as many instances as possible and then start breaking them as needed to make changes?

I try to divide my animation in groups of frames that are similar (around my extremes most of the time). Often the breakdown is alone, similar to no other frame. If you color too much frames at once, it's not a good choice because you have to use more paint or erase to have the color you want. So you have to think before using CTG + light table to fill multiple frames at the same time. I guess you have to try and fail to be able to foresee when you are going too far I think. But for big areas it often saves a lot of time, then you always have some precise work to do afterward.

If a character is moving, it seems impossible for CTG to fill small areas like eyes or mouths. Only the big areas that follow a path that can be designated with a CTG mark.

You can try to use a long stroke that covers the eye on all your frames (using the light table) and you put your skin color stroke where there is only the skin, with more paint than you eye stroke. It often does the work and colors all your eye areas on each frame correctly.

Paint bucket all the way for me with border lines (color separation lines). Here's a quick demonstration. Edit: Updated with TVPaint ver.

Process:


AA (Effects>Stylize>Anti-Aliasing):


More info: http://tvpaint.com/forum/viewtopic.php? ... 058#p91074" onclick="window.open(this.href);return false;

I think the paint bucket is quite primitive and easy to pick up. I think if TVP devs could add in a kind of "lasso fill" or "selective fill" it would be perfect, that way small unfilled areas can be dealt with quickly.

"lasso fill"

it's already in the software under the "fill inside" name, but I admit you can't choose the color of the lines you want to add. :(
It should be in the next update.

"selective fill"

What do you mean exactly ?

- - - -

Also, I think we have found an other way to manage the coloring of shadow and highlight lines, for those using CTG.
keep an eye on the next update. :)

Very interesting to hear that the CTG layer does not replace the paint bucket, but that both can be useful and sometimes, the bucket is just the simpler way to go. I do find the CTG kind of finicky, but as I say, that might be down to me being new with it. Paul F tells me he paints with two layers, one with the paint bucket for smaller areas like eyes, mouth, etc and a CTG layer for bigger areas.

Jet, how is it that you can fill into a line so that the line becomes the colour and disappears?

Here's a quick video David, sorry for the low quality.

https://vimeo.com/165026220" onclick="window.open(this.href);return false;


I disable both anti-aliasing on my pen and my paint bucket, it feels safer for me but I believe it works with anti-aliasing too. You don't have to paint over the border lines, they can still be on a separate layer as I show in the video.

Put your outlines on their own layer, then make a new layer below for color and a layer below this for border lines. For this to work you need to load in the border line colors you want to function, select the layer with the border lines and then select the paint bucket, make sure source is set to layer to make it easier to find the colors, enable include borders and then click add color, click add all relevant colors to load all colors on the layer or just click the ones you want.

Now select your color layer, select the paint bucket and set source this time to display, toggle your lines as required.

Thanks Fabrice for the heads up on shape fills, I always assumed they acted like Photoshop shapes, my bad! I need to find time to explore some more.

This is excellent. Thank you so much, Jet. So when the source is set to Display, it means that it treats all visible layers as one, which makes this work, so that's great. It seems obvious now, but I hadn't realized this. One other small question. Is it possible to fill an area so that it has a soft edge for part of it, or does this just have to be done by hand? Here's what I am after where the arm connects to the body.

David_Fine wrote:Is it possible to fill an area so that it has a soft edge for part of it, or does this just have to be done by hand? Here's what I am after where the arm connects to the body.

Tricky! I'd perhaps do it like this but there must be a much easier way to do it, since I'm a cel shade man it's just a shot in the dark. Probably smudging by hand is easier.



Create two color levels, one is for the base color and the other is for the arm (or shadow), this allows the arm level to use opacity over the base color.

David_Fine wrote:This is excellent. Thank you so much, Jet. So when the source is set to Display, it means that it treats all visible layers as one, which makes this work, so that's great. It seems obvious now, but I hadn't realized this. One other small question. Is it possible to fill an area so that it has a soft edge for part of it, or does this just have to be done by hand? Here's what I am after where the arm connects to the body.

I would use a basic blur FX maybe you can animate a mask for the area by adding one line as suggested by Jet, so you can paint the mask to blur only that part.

Nathan is right, even if I would like to have better options or FX to do so (not only to blur between two colors, it should be for other kind of results too)
We will think about it in the coming months.

Wow, thanks for the demo, Jet. Much appreciated. Excellent demo.

Now I have one more question! If I choose to paint by hand, with a brush, how can I restrict the brush to within the line, just as the paint bucket does? That way, if I need to fill a small area with a brush, I can do so quickly, without it extending beyond the line? I'm hoping there is a really easy way, without creating masks or multiple layers or using the selection tool, but just a setting the way the paint bucket does.

I would say yes, use a mask... You could set a script to automatically select the area and then begin to paint maybe. But I don't get the point, why don't you use the paint bucket or the lasso fill tool? You'll have to point at the area anyway, the computer will not guess for you the area you want to paint

The reason is that the area needs to be coloured, but it is within an area, but is not an area itself, so I want to draw by hand. I know I can create a mask by making an additional layer, but that takes a few steps and adds another layer, which is extra clutter. Seems it would be easy to tell a pen that it can only draw within the area as defined by the line in the above layer, exactly as the paint bucket does, but with a brush.