3 ms·

It's a grueling life. You wake up, start to draw, eat, go back to drawing, sleep. Repeat for x months/years. That's exactly what it is. Animation is kind of my

It's a grueling life. You wake up, start to draw, eat, go back to drawing, sleep. Repeat for x months/years. That's exactly what it is. Animation is kind of my field. My friend and colleague is just now finishing his ~12 minute short (hand-drawn with cats (quadrupeds animation ffs)) which I've helped him with on some of the shots and co-directed it. It's now in editing and scoring. It took him around three years. Day in and day out the cycle you've described. It's a tough life, mostly battling yourself. On a semi-related note, most of the animation was drawn in Harmony on a Cintiq... yet, in final stages of animation he decided to give Krita a spin (Harmony had some issues). Looks like Krita is good enough (for him) to be used in place of Harmony for next projects. He says it's better even in most cases.

buovjaga 10y ago
I think the Krita folks would love to interview him for their blog.
- Keyframe 10y ago
  I didn't even know they had a blog! He's a seasoned vet in hand-drawn animation. Now that his short is almost out, it would be a perfect opportunity.
jomkr 10y ago
Reminds me of the grind a professional poker player goes through. Every day, wake up, play poker for 10-15 hours, study for ~5 hours, sleep. There's a reason they call it the grind, and why there are very few overall winning players, despite plenty of smart individuals playing.
- taway_1212 10y ago
  My friend is a poker pro, and his life is nowhere like this. I'd say he plays maybe 100-120 hours per month. He's been a pro for 10 years now. His cost of living are low though (about $1500 per month).
  - nojvek 10y ago
    With the advent of machines beating poker players, do you see pro poker players rethinking their careers?
    - taway_1212 10y ago
      The machines are only starting to beat people in HU (that is, the two player variant of poker). What most people play is 6 or 10 people variant, where complexity goes through the roof. Additionally, those AIs (unless there's some novel development that I'm missing) are about play game-theoretically optimal (GTO) strategy, which won't do that well in real casino setting - because most profits for professional players come from "fishes" (amateurs making terrible plays), whose weaknesses they map and ruthlessly exploit. A GTO bot is unable to do that.
ralfd 10y ago
Parks & Recreation joke about doing 3 weeks of animation work: https://www.youtube.com/watch?v=iE2T-LM2AZg https://www.youtube.com/watch?v=iE2T-LM2AZg
- Keyframe 10y ago
  It's funny because it's true!
wapz 10y ago
I always thought that after photoshop/illustrator a lot of work is done with computers (layers and separating parts for movement etc). For example, if someone is running through a field I presumed the grass/trees would be 10-20 frames repeated, the clouds would just be "shifted", and the sun's change in colors would be "automated" with a computer. Do artists still have to redraw large parts of every image still?
- dagw 10y ago
  Almost all 'traditional' animation you see is put together almost entirely using computers. Even pre-computers people would paint a background once and then paint each part that moved on transparent plastic or glass and layer it all on top of each other (sometimes each body part of each charter would be on its own layer so that you could keep the body still while just moving one arm for example). There is however a movement that rejects all that for entirely artistic reasons, and insist on painting each frame from scratch by hand. Doing so gives the animation a really unique (and in my opinion cool) look that is hard to mimic using computers. However it is extremely labor intensive and is basically only used for short animations done entirely as labors of love.
  - dagw 10y ago
    If you want to see an amazing example of what hand painted frame-by-frame animation can really do take a look at The Old Man and the Sea from 1999 by Aleksandr Petrov [1] Almost 30,000 frames, painted by hand on glass, producing probably some of the most visually striking animation you are likely to see. [1] https://www.youtube.com/watch?v=vnZD6RBt9S0 https://www.youtube.com/watch?v=vnZD6RBt9S0
    - Keyframe 10y ago
      First thing that came to my mind. I remember, when it won an Oscar, seeing it and thinking 'no f way'.
  - talmand 10y ago
    >> Even pre-computers people would paint a background once and then paint each part that moved on transparent plastic or glass and layer it all on top of each other Older cartoons are easy examples of this. The ones where you can spot the parts to be animated because they are lighter in color than the rest of the frame. This is due to the layering of animated over background. Modern workflows in computers has more-or-less eliminated this visual flaw.
bitL 10y ago
As somebody that worked on a bleeding edge animation software and its most complex algorithms, do you think a bit of a workflow adjustment utilizing deep learning in sense of pre-training possible actions you want to do with your characters, then allow shape morphing between hand-drawn character keyframes, and then simply tell the software using voice what you want to end up with would alleviate the pain of producing an animation? Of course, small adjustments would be needed to do manually, but the rough scene building can be made much faster.
- Keyframe 10y ago
  It all depends on the style of animation. For software there are four areas I see as key to animation where it can help. I'm talking about hand-drawn 2D animation here. 1. Workflow - we've seen that since 90s when animation software became available to general public. First we've made paper to screen workflow available with (softimage) Toonz! and alike where workflow was draw on paper, scan and then check, ink, color on computer. Later on everything became digital and workflow is now, more or less, well understood. I'm not sure if you can make the workflow more approachable unless you redefine animation process. Integration with stuff like Shotgun for project tracking and stuff like that could help, but it's not related to animation itself as much as project management. 2. Technique - computers, obviously, helped pioneer new techniques like 3D, 3D with toon shading, and 'rigged' 2D puppets. Along with numerous special effects that stem from compositing world. 3. Utility while drawing - this is where most innovation came about when helping out hand-drawn animation. Now that drawing is digital, we've seen tools that help a lot. Like 'lazy mouse' approach to semi-automatic curve control while drawing, help while coloring (otherwise a boring task), etc. 4. Utility while animating - this is what you are talking about. It depends deeply on whether you want that style or not, I guess. I know a lot of animators are excited about possible advancements in this department. Basically, drawing a character 'set', if you will, in different poses and viewing angles, rigging them and then puppeteering then with manipulating bones and not drawing everything from scratch each time. It limits what you can do, but at the same time opens up volume production on a small resources scale. Now, there's a trend towards innovating tools you're talking about, where computer manipulates the rig for you based on your indirect input and not only by you manipulating the rig. Character studio from Adobe (in beta still) shows what this is about. I'm excited for this and tools like that. Challenge still is to not have everything produced like that have a certain (same) style to it. edit: you've said you were in animation software. My speciality is more 3D - if you remember SEGA's Animanium... that's what's missing these days. I wonder why it never had a success when it came out. Animanium paired with dead easy rigging would make a splash. Also, something like that, but that allows for real time puppeteering on a touch surface (tablet or whatever) would help A LOT. For example, a rig with controls where you can animate in real time multiple control points at the same time... record that... then do another run where you animate only one point, etc. It's amazing something like that doesn't exist yet. I know you can do it in 3dsmax to an extent, but not really.

buovjaga 10y ago

I think the Krita folks would love to interview him for their blog.

Keyframe 10y ago
I didn't even know they had a blog! He's a seasoned vet in hand-drawn animation. Now that his short is almost out, it would be a perfect opportunity.

jomkr 10y ago

Reminds me of the grind a professional poker player goes through. Every day, wake up, play poker for 10-15 hours, study for ~5 hours, sleep. There's a reason they call it the grind, and why there are very few overall winning players, despite plenty of smart individuals playing.

taway_1212 10y ago
My friend is a poker pro, and his life is nowhere like this. I'd say he plays maybe 100-120 hours per month. He's been a pro for 10 years now. His cost of living are low though (about $1500 per month).
- nojvek 10y ago
  With the advent of machines beating poker players, do you see pro poker players rethinking their careers?
  - taway_1212 10y ago
    The machines are only starting to beat people in HU (that is, the two player variant of poker). What most people play is 6 or 10 people variant, where complexity goes through the roof. Additionally, those AIs (unless there's some novel development that I'm missing) are about play game-theoretically optimal (GTO) strategy, which won't do that well in real casino setting - because most profits for professional players come from "fishes" (amateurs making terrible plays), whose weaknesses they map and ruthlessly exploit. A GTO bot is unable to do that.

ralfd 10y ago

Parks & Recreation joke about doing 3 weeks of animation work: https://www.youtube.com/watch?v=iE2T-LM2AZg https://www.youtube.com/watch?v=iE2T-LM2AZg

Keyframe 10y ago
It's funny because it's true!

wapz 10y ago

I always thought that after photoshop/illustrator a lot of work is done with computers (layers and separating parts for movement etc). For example, if someone is running through a field I presumed the grass/trees would be 10-20 frames repeated, the clouds would just be "shifted", and the sun's change in colors would be "automated" with a computer. Do artists still have to redraw large parts of every image still?

dagw 10y ago
Almost all 'traditional' animation you see is put together almost entirely using computers. Even pre-computers people would paint a background once and then paint each part that moved on transparent plastic or glass and layer it all on top of each other (sometimes each body part of each charter would be on its own layer so that you could keep the body still while just moving one arm for example). There is however a movement that rejects all that for entirely artistic reasons, and insist on painting each frame from scratch by hand. Doing so gives the animation a really unique (and in my opinion cool) look that is hard to mimic using computers. However it is extremely labor intensive and is basically only used for short animations done entirely as labors of love.
- dagw 10y ago
  If you want to see an amazing example of what hand painted frame-by-frame animation can really do take a look at The Old Man and the Sea from 1999 by Aleksandr Petrov [1] Almost 30,000 frames, painted by hand on glass, producing probably some of the most visually striking animation you are likely to see. [1] https://www.youtube.com/watch?v=vnZD6RBt9S0 https://www.youtube.com/watch?v=vnZD6RBt9S0
  - Keyframe 10y ago
    First thing that came to my mind. I remember, when it won an Oscar, seeing it and thinking 'no f way'.
- talmand 10y ago
  >> Even pre-computers people would paint a background once and then paint each part that moved on transparent plastic or glass and layer it all on top of each other Older cartoons are easy examples of this. The ones where you can spot the parts to be animated because they are lighter in color than the rest of the frame. This is due to the layering of animated over background. Modern workflows in computers has more-or-less eliminated this visual flaw.

bitL 10y ago

As somebody that worked on a bleeding edge animation software and its most complex algorithms, do you think a bit of a workflow adjustment utilizing deep learning in sense of pre-training possible actions you want to do with your characters, then allow shape morphing between hand-drawn character keyframes, and then simply tell the software using voice what you want to end up with would alleviate the pain of producing an animation? Of course, small adjustments would be needed to do manually, but the rough scene building can be made much faster.

Keyframe 10y ago
It all depends on the style of animation. For software there are four areas I see as key to animation where it can help. I'm talking about hand-drawn 2D animation here. 1. Workflow - we've seen that since 90s when animation software became available to general public. First we've made paper to screen workflow available with (softimage) Toonz! and alike where workflow was draw on paper, scan and then check, ink, color on computer. Later on everything became digital and workflow is now, more or less, well understood. I'm not sure if you can make the workflow more approachable unless you redefine animation process. Integration with stuff like Shotgun for project tracking and stuff like that could help, but it's not related to animation itself as much as project management. 2. Technique - computers, obviously, helped pioneer new techniques like 3D, 3D with toon shading, and 'rigged' 2D puppets. Along with numerous special effects that stem from compositing world. 3. Utility while drawing - this is where most innovation came about when helping out hand-drawn animation. Now that drawing is digital, we've seen tools that help a lot. Like 'lazy mouse' approach to semi-automatic curve control while drawing, help while coloring (otherwise a boring task), etc. 4. Utility while animating - this is what you are talking about. It depends deeply on whether you want that style or not, I guess. I know a lot of animators are excited about possible advancements in this department. Basically, drawing a character 'set', if you will, in different poses and viewing angles, rigging them and then puppeteering then with manipulating bones and not drawing everything from scratch each time. It limits what you can do, but at the same time opens up volume production on a small resources scale. Now, there's a trend towards innovating tools you're talking about, where computer manipulates the rig for you based on your indirect input and not only by you manipulating the rig. Character studio from Adobe (in beta still) shows what this is about. I'm excited for this and tools like that. Challenge still is to not have everything produced like that have a certain (same) style to it. edit: you've said you were in animation software. My speciality is more 3D - if you remember SEGA's Animanium... that's what's missing these days. I wonder why it never had a success when it came out. Animanium paired with dead easy rigging would make a splash. Also, something like that, but that allows for real time puppeteering on a touch surface (tablet or whatever) would help A LOT. For example, a rig with controls where you can animate in real time multiple control points at the same time... record that... then do another run where you animate only one point, etc. It's amazing something like that doesn't exist yet. I know you can do it in 3dsmax to an extent, but not really.