Jump to content
Hash, Inc. Forums

Blockheads


Recommended Posts

Looks good to me!

 

 

Why not make some happy new years toasting?

And contribute with a toasting exercise to this topic:

http://www.hash.com/forums/index.php?showtopic=45734

 

a_toast_happy_new_year.gif

 

 

Esoic

Thank you for your reply and comment.

I usually do my new year message on the day and its often a bad pun around the "Happy New Year' message. Theres one from a couple of years ago in this thread somewhere ( http://www.hash.com/forums/index.php?act=a...t&id=64425).

regards

simon

Link to comment
Share on other sites

  • Replies 190
  • Created
  • Last Reply

Top Posters In This Topic

Part two of this weeks brief. Not having done a run before, it seemed like a good idea to try.

So, after the stretch, the run...

Any critical feedback welcome.

The brick wall was because, originally the idea was to animate him running on the spot and have the wall move in the background ( that was to be a test for another project ) and it needed something on it to get the movement. That proved too difficult for the moment, so went for a straight run from the standing position after the stretch. Thought it might be a bit too quick ?

simon

 

ps

This is a VGA from 1080 conversion

Run_A.mov

Link to comment
Share on other sites

First part of new idea.

The intention is to do a 100 metre sprint race with 5 characters and use a different style run on each figure.

This is the first one using the natural style reference in the Richard Williams book.

Any Feedback welcome.

simon

 

A from DV conversion.

 

Ps

Didn't want to use a cycle action so keyed each move separately, 4 keys in a 6 frame sequence x 13 seconds @25 fps, is a slightly mind numbing task...

Sprint_1.mov

Link to comment
Share on other sites

  • Admin

Nice one. That's a great test.

I really like that.

 

This is a hard one to critique without drawing over frames.

Two immediate things come to mind... perhaps three... but I need to study what I think I'm seeing.

I'll mention them and see what you think.

 

1. Full extension of the legs. This guy is really running fast and yet we don't much full extension of the legs (the knees remain bent throughout most of the sequence)

Note that upon reviewing again I do see at least one section where his legs fully extend so you do have it in there.

It would be good to see how much full extension there was in the original reference.

 

2. The guy seems to move backward at several points.

This may be more of an effect created by the camera and relates to number 3 below.

It's not so much that it looks off... it looks pretty good from here... but it may be stealing some smoothness of motion from the totality of the frames as they play together.

To test this we might look for and track a specific shape on the guy and note how it changes shape and position from frame to frame throughout the entire sequence.

For the sake of animation a little reality might need to be sacrificed for the greater good of visual clarity.

This is one reason why fast moving shots in cinema are often presented in slow motion... to allow the viewer time to read the imagery.

 

3. Visual cues (references) that orient the viewer to where the character is in the scene.

If this is a 50 yard dash then it might be good to have some lines and/or numbers on the track to clue us in to the runners location and speed.

The start and finish is apparent enough but there is potential for loss of orientation as the camera tracks around.

 

Something I really like about this sequence is how you've got that 'banana' shape that allows the perspective to change.

I'm not entirely sure you planned it that way but you've got it going on in this scene!

If you are curious it's similar to what Bobby was doing in these sequences.

 

All in all I'm very impressed with this test.

If it doesn't defeat the purpose of your test you should render it with motion blur on to see how that plays.

 

Nicely done!

Link to comment
Share on other sites

Rodney

Thank you for your kind words and feedback.

 

I think the backward motion might be an effect of the camera work ? as , I'll check but, part of the tedium involved in doing it was moving forward in three metres for every two full steps. Each position was keyed over 52 times I found myself whistling or singing after a while to try to retain motivation.

 

The banana effect you mention was brought about by the camera placing and zoom in and out as he moves along the track. It starts on a focal length of 500, goes to 250 in the middle as he runs past the position, then out again to 500 as he reaches the end. I didn't know you could do this until trying it and will probably use it in the final piece.

 

The reference used was that for the 'natural' run on page 178 of the Richard Williams book and there were no full extensions in that example. Although the other references do have full extensions and I will use those for the next runs with the other figures. Also tried Youtube refs for 100metre sprints, including this one of the infamous 88 final at the 88 Olympics.

 

Sprint_88_VGA.mov

Its difficult to tell from this but it does look as if the full extension is only glimpsed very rarely as the forward momentum is so rapid. They cover 10 metres in less than a second and do it in six steps. I'm going to start the next one tomorrow and will post it when done. Hope for a noticeable contrast to this one.

 

Have tried a version with motion blur at 15% but not sure if that was a strong enough setting ? This was a VGA conversion of a 1080 render

regards

simon

 

Ps

revised version , this time with 30% blur.

Sprint_30.mov

Sprint_15.mov

Link to comment
Share on other sites

Second sprinter. Different style , with full extension.

Not sure why the strobing effect ( didn't happen with other sprinter ?)

Sprint_002.mov

 

Heres a comparison of the two, side by side. Still working on the lighting and staging.

 

SprintRace_B.mov

 

 

Going to a wedding so may not get the third figure/style done by next week.

Any critical feedback welcome.

simon

Link to comment
Share on other sites

Hi Simon,

Definitely better leg and foot work in the new one!

I see what you mean about the strobing, hard to tell whats going on just looking at the render. Would you be able to post the project file for a closer look?

My first guess is that the forward momentum of the runner is not constant but the camera's motion, more or less, is, so that at times the camera is turning forward slightly faster than the runner is moving forward, the effect of which makes the runner appear to move back slightly every few frames.

One other thing is that at the moment it is quite hard to tell relative speed. The logical part of our brains says that they are moving forward, probably at speed, but our eyes are not getting any real visual confirmation of this fact, which is a bit disorienting.

Adding markings at regular intervals to the running surface would help as well as possibly some sort of texture. And/or by putting markers at the side of the track at regular intervals could also help.

Link to comment
Share on other sites

Hi Simon,

Definitely better leg and foot work in the new one!

I see what you mean about the strobing, hard to tell whats going on just looking at the render. Would you be able to post the project file for a closer look?

My first guess is that the forward momentum of the runner is not constant but the camera's motion, more or less, is, so that at times the camera is turning forward slightly faster than the runner is moving forward, the effect of which makes the runner appear to move back slightly every few frames.

One other thing is that at the moment it is quite hard to tell relative speed. The logical part of our brains says that they are moving forward, probably at speed, but our eyes are not getting any real visual confirmation of this fact, which is a bit disorienting.

Adding markings at regular intervals to the running surface would help as well as possibly some sort of texture. And/or by putting markers at the side of the track at regular intervals could also help.

 

 

Mark

 

Thank you very much for your help and feedback. Much appreciated.

This is the project file.

Sprinting.prj

The two runs were animated as chors first and 001 saved as an action. The Sprint base chor had the extra lane added with action 001 added to RB17.

The camera is aimed at a null and the null was animated to keep pace with the figures as they moved down the track, staying roughly half way between them every 25 frames.

003 will be a lot more cartoon like.

regards

simon

Link to comment
Share on other sites

  • Admin

While it's possible the camera is playing tricks on us I'd say this is more related to fluid motion and movement forward of the man/appendages.

 

Suggestion: Add a Camera into the Character/Model and in Bones Mode place it to the side of the character so you can view (and render) then entire squence from the view of that Camera.

Note that when created, Cameras inside Models will appear at the 0,0,0 coordinate and one of the best ways to get them where you want might be to type in the coordinates into the Properties Panel versus move it around.

Then in the Chor or Action select that Camera as your view and you'll be better able to analyze the character's run as it'll appear as if he's running in place .

You can place additional spline markers within the Character Model as well to establish registration points/marks as additional visual aids (similar to how we can work with stride length in cyclic Actions).

To take the analysis to the next level you could then create slightly horizontal markers with the distance to give you a reference of the distance traveled.

Ultimately with techiques like this we can replicate setups like Muybridge did at the dawn of animation by compositing all those frames together.

 

Another useful method might be to move to a Right Side View... back out a little and then turn Onion Skin on and play with the frame/keyframe settings.

This would let you see the how well your current movement is being inbetweened manually by you and as automatically interpolated by A:M.

Link to comment
Share on other sites

Hi again Simon,

I've looked at Chor Sprint 002 and I was both right and wrong!

The runners speed is in fact constant but the way you have set up the camera to follow him means that the camera's speed isn't!

In your original Chor 002, go to Top View and select Null1 (which the camera is constrained to with an Aim At...)

As you advance the animation you can see that the relative position of Null1 to the Body Null of the runner regularly decreases and increases as he runs which, from the camera's point of view, makes the runner seem to speed up and slow down in a regular cycle.

 

In the attached here I've removed the keys for moving the camera but it still has an Aim At... to Null1.

Null1 now has a Translate To... with the Body Null of the runner.

See what you think...

Sprinting_MW_adjusted.prj

Link to comment
Share on other sites

Looking at the actual animation, I think maybe you could get a bit more twist into the runner's body motion. At the moment his body is a bit stiff.

Of course twisting the hips will mean you will have to adjust his stride length!

As soon as you twist a hip forward that leg will loose it's full extension/straightness on the contact pose and you will have to move that leg's foot controller slightly forward to compensate and maintain the leg's previous full extension. And so on…

Ah, the joys of animating!

Screen_Shot_2013_12_21_at_16.39.11.png

Link to comment
Share on other sites

Rodney

Thank you for your helpful reply.

I wasn't sure how to follow up your suggestion fully, so I rendered this sideview of the sequence by constraining the camera to the body null of the 2008 rig used. and putting some bars in the background to give a steady level. There are now some markers at track side to denote every 2 meters of track distance.

I'm working on getting a grid at the moment.

One thing that I did notice in this is that the is there is very little fore and aft with the torso. There is a lot of side to side ( x plane ? ) but not a lot of z plane rotation ( pardon me if I get that mixed up.

regards

simonSide_Sprint.mov

Link to comment
Share on other sites

Hi again Simon,

I've looked at Chor Sprint 002 and I was both right and wrong!

The runners speed is in fact constant but the way you have set up the camera to follow him means that the camera's speed isn't!

In your original Chor 002, go to Top View and select Null1 (which the camera is constrained to with an Aim At...)

As you advance the animation you can see that the relative position of Null1 to the Body Null of the runner regularly decreases and increases as he runs which, from the camera's point of view, makes the runner seem to speed up and slow down in a regular cycle.

 

In the attached here I've removed the keys for moving the camera but it still has an Aim At... to Null1.

Null1 now has a Translate To... with the Body Null of the runner.

See what you think...

 

 

Mark

Thank you very much indeed for your help. Much appreciated.

I'm slightly puzzled by the hip rotation as I had tried to include that while setting the keys. Because it was on a 6 frame pattern, the hips were rotated forward with the advancing leg at each extreme position and vertically to get some counter rotation in the torso. Tried to do it on every sixth frame because I didn't want to set too many keys ( a problem in the past ).

I will try to follow up your advice later, domestic and family duties first unfortunately.

regards

simon

Link to comment
Share on other sites

Well, I think I might have to retract my earlier post about the runners speed bing constant!

I just took another longer look at this and the runners forwards motion is not consistent. My apologies Simon!

As you can see in this screenshot with Onion Skin Mode on, there is a clear deceleration during foot contact with the ground.

As I side note there's possibly to much vertical movement of the torso maybe?

Screen_Shot_2013_12_21_at_19.05.36.png

Link to comment
Share on other sites

Well, I think I might have to retract my earlier post about the runners speed bing constant!

I just took another longer look at this and the runners forwards motion is not consistent. My apologies Simon!

As you can see in this screenshot with Onion Skin Mode on, there is a clear deceleration during foot contact with the ground.

As I side note there's possibly to much vertical movement of the torso maybe?

 

 

Mark

Thank you for your help once again.

I was trying to follow the 'recipe' on the top of page 177 of the Richard Williams book and think I may have over done it with the stride length ? The first one was covering just over 7 metres a second and this one was over 8 metres a second ( 210cm per stride ). I had tried to set as few keys as possible to avoid the jittery quality of earlier work so had allowed the software to interpolate between the contact and the lift. May have to go into the curves to get a more even speed ?

 

Off to my nephew's wedding tomorrow, so may not be able to address this until the return.

Thanks for your help.

regards

simon

Link to comment
Share on other sites

  • 3 weeks later...

Had a few diversions of a practical nature but, the sprint race continues. This is the most recent addition. He's in a much more 'comic' style. There are due to be a further two styles/characters. Because of his running style this one takes 18 seconds to do the 100 metres, the others are 12 and 14 second. The winner will be a lot less. Trying to get the basics before going back and revising it.

Any feedback welcome.

simon

 

Sprint2.mov

Link to comment
Share on other sites

It's an odd run! I think the most missing element is that to get moving he would need to lean into it at the start of the run.

 

 

Robert

Thank you for your feedback. I was trying for odd. Someone suggested a while back, that the way to find a reference point was to go to the extremes then dial back from there. The first three were using the Williams book and this one was a memory of, I think, a Chuck Jones cartoon. I was going to go back into it today to get more forward rotation at the start and hip and torso twist. The next one was going to be in a roadrunner style with multiple legs and feet but haven't worked out how I'll do that yet.

regards

simon.

Link to comment
Share on other sites

It's an odd run! I think the most missing element is that to get moving he would need to lean into it at the start of the run.

 

 

Robert

Thank you for your feedback. I was trying for odd.

regards

simon.

 

Be careful when trying to achieve odd... sometimes it winds up just looking bad.

Link to comment
Share on other sites

  • Hash Fellow

I'll note that real walkers, jogger and runners have a very consistent forward speed of the hips. It is not slowing down and speeding back up with every step. It's not possible to run or jog like that and people who walk like that have some significant mobility problem.

 

Check out this reference footage

 

http://www.hash.com/forums/index.php?s=&am...st&p=376617

 

What little variation there is in the forward progress of the hips is probably due to the hip marker being on the outside edge of the hip rather than in the actual center. The marker is catching the rotation on the axis of the spine as the legs sweep forward and back.

Link to comment
Share on other sites

Nancy, Robert, 3D Artz

 

Thank you for your feedback.

I was hoping that the context its in, would allow it to be odd rather than bad ? This is the first 8 seconds of the race so far, there is one more runner to come. The runs are not finished yet as, following Robert's and Mark's earlier suggestions, I'm going to revise the positioning on the first two runners to get a less stroboscopic, more even movement. One thing I did notice about Robert's reference video was how readily it matched Richard Williams description of runs as having very little height variation in them, he talks of it usually being less than half a head of variation. Two of these runs break that convention but that was intended.

regards

simon

 

Sprint_race.mov

Link to comment
Share on other sites

  • Hash Fellow

I've noticed in anime movies when they are showing someone running from the waist up only and the character is against a panning background, they will jostle the body forward and backward as if that were the running motion happening.

 

That may be where the idea of he stopping and starting motion is coming from, but that is a limited animation convention that doesn't really translate to more 3D representations.

Link to comment
Share on other sites

Have reworked the first version of the sprint in the light of earlier feedback, and lessons learnt since.

Haven't finalised the bit at the end yet, which will be a premature celebration.

On a trivial note, his time for the 100 metres is 8 seconds and 14 frames ( about 8:56 ) @ 25 fps. Which might make Eusain Bolt seem slow ! That wasn't intend, it was a side effect of the method used. Would that it was that quick to do...

simonSprintB001.mov

Link to comment
Share on other sites

  • Hash Fellow

The thing that just doesn't "fly" about that run is that he appears to be halting in mid air for no reason. There is no force applied to him that woulf make his hips stop moving forward in mid stride (and no force to make them restart in mid air) If the slowdown was timed after his foot having hit the ground in front of him... maybe then.

 

Also, he ran all that way, built up that speed, then suddenly halted before he jumped. He wouldn't be able to do that.

Link to comment
Share on other sites

The thing that just doesn't "fly" about that run is that he appears to be halting in mid air for no reason. There is no force applied to him that woulf make his hips stop moving forward in mid stride (and no force to make them restart in mid air) If the slowdown was timed after his foot having hit the ground in front of him... maybe then.

 

Also, he ran all that way, built up that speed, then suddenly halted before he jumped. He wouldn't be able to do that.

 

 

Robert

Thank you once again for your help.

I'm a bit puzzled by the mid stride slowdown. Its a six frame stride, each stride being three metres in length and there are three keys for the body null, frames 1,3, 6 and two for the hip rotation on frames 1, 6, Might it be a case of the curves needing editing ? I'll have a look later today.

regards

Simon

Link to comment
Share on other sites

  • Hash Fellow
This is a test for the next sprinter.

Can anyone please, kindly remind me how to do a shadows only render ?

simon

 

Sprint_test.mov

 

 

Well, that's just silly! :lol:

 

I believe there is something about shadows-only in my "Post in Progress" in the tutorials section

Link to comment
Share on other sites

This is a test for the next sprinter.

Can anyone please, kindly remind me how to do a shadows only render ?

simon

 

Sprint_test.mov

 

 

Well, that's just silly! :lol:

 

I believe there is something about shadows-only in my "Post in Progress" in the tutorials section

 

Robert

Thank you. I've got a print out somewhere but can't find it at the moment...

Here is an update. It will need a bit of refinement but I think it should work for what I want.

Curiously, the track didn't render this time and I got the white. It was all done in a chor using layers.

regards

simon

 

Sprint_test.mov

Link to comment
Share on other sites

This afternoon's tests for the sprint race

The ones called Feet were done by changing the speed of the cycle over the same time frame of movement.

 

Feet_001.mov

Feet_004.mov

 

 

The ones called Offset were done by retaining the cycle length but off setting it within the time line.

 

Offset_001.mov

Offset_Legs_004.mov

 

 

Any feedback welcome

simon

Link to comment
Share on other sites

I like the offset versions best. And in particular the last one with the legs.

Oh yeah... that's the stuff! :)

 

 

Rodney

Thank you for your feed back. I'm trying to work out how to get the legs to cast shadows but leave the multiple feet doing a roadrunner. Think I may have got it but, if successful, will try to post it later.

regards

simon

 

Sort of got the shadows only bit but it will need further work.

This one withe full complement of feet.

 

Shadow_Legs.mov

 

And this is one with only half that amount.

Reduced_Track_Legs.mov

 

Personally I prefer at one ?

 

Did try one with the track but it kept getting in the way of the running and then wouldn't render in the same way. In the interests of foolishness here it is.

Shadow_Track_Legs.mov

 

I shall sleep on the problem and tackle it afresh ( vain hope )

Link to comment
Share on other sites

Join the conversation

You can post now and register later. If you have an account, sign in now to post with your account.
Note: Your post will require moderator approval before it will be visible.

Guest
Reply to this topic...

×   Pasted as rich text.   Paste as plain text instead

  Only 75 emoji are allowed.

×   Your link has been automatically embedded.   Display as a link instead

×   Your previous content has been restored.   Clear editor

×   You cannot paste images directly. Upload or insert images from URL.

Announcements


×
×
  • Create New...