The Mad Professor!
Moderators: Víctor Paredes, Belgarath, slowtiger
Re: The Mad Professor!
Ha ha! Yes the hands. Ick!
Haven't done much with the hands yet. They look kind a funky. I drew them freehand in Anime Studio at the start and never touched since then, was too focused on the body and head. The finger tips are too pointy and claw like... and they don't look natural, kind of stiff and weird. I need to do individual shapes for fingers with some kind of shading to show finger overlap and delineation. Haven't decided if I want to do simple switch hands or use bones for fingers. Bones are nice but.... simple switches are faster and easier to animate. What I really need to do is go back to photoshop and sketch a bunch of hand designs with the wacom. I can never get hands right just drawing in Anime Studio from scratch. I usually sketch on paper first. I am thinking of a specific "hand style" to match the character. A sort of "squared stubby" type of finger, something fairly simple.
I also still have to do phonemes! I am hoping I can simply do a front and side view of the phonemes and move and scale during turns. The simple less extreme turns just past 3/4 will work with just front view and some layer and point tweaking. I want to add some additional smart bone tweaks to add additional emotions onto the canned phoneme switches. I've seen this done on the forum with smart bones.
The tough part is full side views. For example on the head turn the hair on either side has to go behind the head depending on the turn direction. I hated the "jump" you get when something suddenly goes behind another layer, so for the hair I would squish one side to "line up" with the side of the head over a few frames THEN have the hair go behind the head layer. I scale points of one side of the hair horizontally to zero, then use the "Bend Points" tool to "curve" the "line" of squished points to match up with the side of the head before moving the layer behind the head. This gives a nice animated effect of the hair turning without a horrific "jump" when layer order changes. Works great but is kind of a pain in the arse. I also hate squishing points to 0 for some reason. It works but I just don't like doing it for some reason... feels like a lazy technique, like there should be a better way.
Thanks for the comments on the turn working okay. I have to go back in and make that process easier to animate. It takes a lot of crazy keys to go from "side front' to "side back" to have him turn around completely. I think I could probably put that into a smart bone action. It may not be used a lot. The reason it's a pain is that the side to side turn for the front and back is "opposite" directions (switch layer). For about 5-10 frames I have keys on every frame to do the turn.
The frustrating part is not being able to put layer order in the SB actions. SB layer order sort of locks everything down. I can do most of the turning with SB's but layer ordering still needs to be done in the main timeline.
I am thinking maybe I could do layer order with specific smart bones for different configurations. Instead of having layer order in the turn, put the ordering in another SB with specific ordering combinations on set keys. There isn't an unlimited combination of layer ordering needed. It's basically forearms over body, forearm and bicep over body etc etc. I could put those different combos into a smart bone and just use "step" keys to pick the right layer order in specific spots. This would also allow me to turn specific masking layers on and off for the shoulder. I created alternate masking for some of the turns. It works great but I have been keying layer visibility in the main time line.
Anyway, just posting my thought process on this stuff.
Haven't done much with the hands yet. They look kind a funky. I drew them freehand in Anime Studio at the start and never touched since then, was too focused on the body and head. The finger tips are too pointy and claw like... and they don't look natural, kind of stiff and weird. I need to do individual shapes for fingers with some kind of shading to show finger overlap and delineation. Haven't decided if I want to do simple switch hands or use bones for fingers. Bones are nice but.... simple switches are faster and easier to animate. What I really need to do is go back to photoshop and sketch a bunch of hand designs with the wacom. I can never get hands right just drawing in Anime Studio from scratch. I usually sketch on paper first. I am thinking of a specific "hand style" to match the character. A sort of "squared stubby" type of finger, something fairly simple.
I also still have to do phonemes! I am hoping I can simply do a front and side view of the phonemes and move and scale during turns. The simple less extreme turns just past 3/4 will work with just front view and some layer and point tweaking. I want to add some additional smart bone tweaks to add additional emotions onto the canned phoneme switches. I've seen this done on the forum with smart bones.
The tough part is full side views. For example on the head turn the hair on either side has to go behind the head depending on the turn direction. I hated the "jump" you get when something suddenly goes behind another layer, so for the hair I would squish one side to "line up" with the side of the head over a few frames THEN have the hair go behind the head layer. I scale points of one side of the hair horizontally to zero, then use the "Bend Points" tool to "curve" the "line" of squished points to match up with the side of the head before moving the layer behind the head. This gives a nice animated effect of the hair turning without a horrific "jump" when layer order changes. Works great but is kind of a pain in the arse. I also hate squishing points to 0 for some reason. It works but I just don't like doing it for some reason... feels like a lazy technique, like there should be a better way.
Thanks for the comments on the turn working okay. I have to go back in and make that process easier to animate. It takes a lot of crazy keys to go from "side front' to "side back" to have him turn around completely. I think I could probably put that into a smart bone action. It may not be used a lot. The reason it's a pain is that the side to side turn for the front and back is "opposite" directions (switch layer). For about 5-10 frames I have keys on every frame to do the turn.
The frustrating part is not being able to put layer order in the SB actions. SB layer order sort of locks everything down. I can do most of the turning with SB's but layer ordering still needs to be done in the main timeline.
I am thinking maybe I could do layer order with specific smart bones for different configurations. Instead of having layer order in the turn, put the ordering in another SB with specific ordering combinations on set keys. There isn't an unlimited combination of layer ordering needed. It's basically forearms over body, forearm and bicep over body etc etc. I could put those different combos into a smart bone and just use "step" keys to pick the right layer order in specific spots. This would also allow me to turn specific masking layers on and off for the shoulder. I created alternate masking for some of the turns. It works great but I have been keying layer visibility in the main time line.
Anyway, just posting my thought process on this stuff.
Re: The Mad Professor!
Heyvern, you have brought the professor to life. The SB effect is amazing. I'm going to dust off my Wacom and sketch my characters in Photoshop. I think I see why it works better than doing it from scratch in AS.
I put my phonemes in a switch layer but I have a separate mouth vector layer for SB 'expressions'. I use visibility to alternate between them. The great part is I only need one switch layer for all my SB head turns.
I put my phonemes in a switch layer but I have a separate mouth vector layer for SB 'expressions'. I use visibility to alternate between them. The great part is I only need one switch layer for all my SB head turns.
Cheers, Larry
Re: The Mad Professor!
It works very well. There is much to be learned here by using switch layers for turns. They look good and are nowhere near as distracting as a badly done one with bones.
~Danimal
Re: The Mad Professor!
Enjoying the Wacom Intuous5!
New hand sketches done in Photoshop...
New hand sketches done in Photoshop...
- funksmaname
- Posts: 3174
- Joined: Tue May 29, 2007 11:31 am
- Location: New Zealand
Re: The Mad Professor!
very nice! I'm always struggling with line thickness in Photoshop
Re: The Mad Professor!
Any respectable mad scientist will need some lab equipement so I stared making some bottles and test tubes. I think it will be cool to have bubbles and particles in the liquid rising up using a particle layer have to try that later.
Re: The Mad Professor!
Awesome work heyvern.
Re: The Mad Professor!
Some updates...
Hands "vectorized" in AS. Had to use masking for each hand to get the shading I wanted. May still tweak these later.
Tweaked the bottles highlights with gradients and soft edge. Had to use masking on the big round bottle due to the "curved" highlight. Couldn't get the irregular shaped highlight fall off with a gradient. Adding smart bones to control the liquid levels and color.
Hands "vectorized" in AS. Had to use masking for each hand to get the shading I wanted. May still tweak these later.
Tweaked the bottles highlights with gradients and soft edge. Had to use masking on the big round bottle due to the "curved" highlight. Couldn't get the irregular shaped highlight fall off with a gradient. Adding smart bones to control the liquid levels and color.
Re: The Mad Professor!
I am always amazed at what is capable in Anime studio.
Do you plan do use this character in any real project? Or is this just to show what is possible and pass on the info?
Do you plan do use this character in any real project? Or is this just to show what is possible and pass on the info?
- synthsin75
- Posts: 10153
- Joined: Mon Jan 14, 2008 11:20 pm
- Location: Oklahoma
- Contact:
Re: The Mad Professor!
Minor matter, but there should probably be a similar thickness between the outermost glass and the liquid as you have between the outermost glass and the stopper.
- Wes
Donations: https://www.paypal.com/paypalme/synthsin75 (Thx, everyone.)
https://www.youtube.com/user/synthsin75
Scripting reference: https://mohoscripting.com/
Donations: https://www.paypal.com/paypalme/synthsin75 (Thx, everyone.)
https://www.youtube.com/user/synthsin75
Scripting reference: https://mohoscripting.com/
- patricia3d
- Posts: 726
- Joined: Thu May 31, 2007 4:38 am
- Location: India
- Contact:
Re: The Mad Professor!
Glass Shading is great.Really great.
Please guide for this shading.
Please guide for this shading.
Re: The Mad Professor!
Here's a quick tutorial. When I get a chance I will do a video.patricia3d wrote:Glass Shading is great.Really great.
Please guide for this shading.
I am using the round bottle for this because it was the trickiest one due to the irregular shape. For the test tubes it was simpler and didn't require any masking because I just used a gradient for the "highlights". Gradient works because it's "straight". For the irregular shape there weren't any effects that would work right out of the box without masking.
The basic ideas is I used a "larger" shape with soft edge halo effect. I then masked that halo shape so I get a sort of "edge gradient" effect.
This is the highlight set up:
This is the fill shape for the glass. The "white" glass shape is all the way at the bottom of the shape order. It uses a "shaded effect" with "shading only" checked to give that sort of "glowy thickness" to the glass. It has a light gray stroke. This stroke looks bright on a dark background, and still pops on a light background.
The liquid shape uses a gradient for the fill and stroke. I have several liquid shapes with different colors in the gradient. The stroke gradient is the same as the fill gradient but a little bit darker. This creates that dark edge around the liquid filled part of the glass. At the bottom of the whole shape is a "crescent" of a darker color that matches the liquid color. This is just a sort of dark "refraction" shadow to ground the shape. For each liquid color this color is different to match the liquid.
Basically I have saved styles and named shapes for the bottles. I can duplicate a bottle and quickly change the color of the liquid by using a different liquid named style for the 3 liquid shapes.
- patricia3d
- Posts: 726
- Joined: Thu May 31, 2007 4:38 am
- Location: India
- Contact:
Re: The Mad Professor!
I created some smart bones for the test tube.
I can control liquid level, liquid color, pouring and the "pour length" with smart bones. The only draw back of course is that I have to duplicate the bone layer for the test tube to have multiple test tubes along with the with actions. Not a big deal I suppose. (a cool feature would be a way to have sub layer smart bones display in a parent bone layer without having to create and set them up... sort of like smart bone "aliases". So if you place a bone layer with smart bones inside another bone layer, the sub bones smart bones can be accessed in the parent layer automatically.)
It's a bit tricky to do the "pour" because of all the variables involved. There is the angle of the bottle and amount of liquid. Both of these effect when and where the pour starts. I haven't figured out a way to completely automate this with smart bones but still trying.
I had to split the simple one layer test tube into multiple layers and use masking. Most of this was due to the gradients and shape movement and how smart bones work. The gradients needed to be layer rotated rather than point movement because there are "multiple" combined movements. Moving points with smart bones are "absolute" so it was a bit tough to get the mixing of the actions correct. Instead I rotate the entire layer/s with layer binding and then mask the liquid as the main bone rotates the bottle. By using layer movement this works much better.
The liquid mask is just a big square moved by a bone. This mask bone has an angle constraint to the main test tube bone so that the liquid level maintains a straight horizontal orientation as the bottle rotates. The liquid surface is bound to another bone with the same constraint. Instead of point motion I "squish" the surface with layer scaling.
Using layer transformations in the smart bones worked much better than point movement due to the gradients. I had issues with the gradients not "rotating" with point motion. Rotating the entire layer helped with this issue.
I can control liquid level, liquid color, pouring and the "pour length" with smart bones. The only draw back of course is that I have to duplicate the bone layer for the test tube to have multiple test tubes along with the with actions. Not a big deal I suppose. (a cool feature would be a way to have sub layer smart bones display in a parent bone layer without having to create and set them up... sort of like smart bone "aliases". So if you place a bone layer with smart bones inside another bone layer, the sub bones smart bones can be accessed in the parent layer automatically.)
It's a bit tricky to do the "pour" because of all the variables involved. There is the angle of the bottle and amount of liquid. Both of these effect when and where the pour starts. I haven't figured out a way to completely automate this with smart bones but still trying.
I had to split the simple one layer test tube into multiple layers and use masking. Most of this was due to the gradients and shape movement and how smart bones work. The gradients needed to be layer rotated rather than point movement because there are "multiple" combined movements. Moving points with smart bones are "absolute" so it was a bit tough to get the mixing of the actions correct. Instead I rotate the entire layer/s with layer binding and then mask the liquid as the main bone rotates the bottle. By using layer movement this works much better.
The liquid mask is just a big square moved by a bone. This mask bone has an angle constraint to the main test tube bone so that the liquid level maintains a straight horizontal orientation as the bottle rotates. The liquid surface is bound to another bone with the same constraint. Instead of point motion I "squish" the surface with layer scaling.
Using layer transformations in the smart bones worked much better than point movement due to the gradients. I had issues with the gradients not "rotating" with point motion. Rotating the entire layer helped with this issue.