sprockets The Snowman is coming! Realistic head model by Dan Skelton Vintage character and mo-cap animation by Joe Williamsen Character animation exercise by Steve Shelton an Animated Puppet Parody by Mark R. Largent Sprite Explosion Effect with PRJ included from johnL3D New Radiosity render of 2004 animation with PRJ. Will Sutton's TAR knocks some heads!
sprockets
Recent Posts | Unread Content
Jump to content
Hash, Inc. - Animation:Master

Happy Families


Recommended Posts

This is a timing and composition check.

Any critical feedback very welcome.

I have a second version going through now with slightly longer pauses at the start ad turn around.

The final version, if timing ok, will have a toon rendered figure in this setting.

simon

 

Nosey.mov

 

Following feedback on Facebook , this is the revised version.Nosey 02.mov

Link to comment
Share on other sites

  • Replies 225
  • Created
  • Last Reply

Top Posters In This Topic

  • Hash Fellow

Not sure if I'll have the chance to redo itfor a while as it was taking 15-20 mins each frame to erase the non toon sections to reveal the toon underneath.

Explain this some more... I think you are doing something that doesn't need to be done manually.

 

Show a picture.

Link to comment
Share on other sites

Robert

Thank you for your feedback once again.

Here are the different layers of the process. The main time consuming part was in erasing the full render sections to allow the toon render to show through.

It was either this way or overlay the Toon on top. Either would result in a lot of erasure work to get the correct alignment (?)

 

This is the final composite look with the three layers, the Clouds were added after render due to an oversight on my part.

 

NHF S8F 000 .png

 

This is the separate Toon layer

 

NHF S8T 000 .png

 

This is the full render with the Toon elements erased

There are some small elements left in place which initially I went back and removed but then kept because it had the effect of making those figures slightly more lively (?).

 

regards

simon

NHF S8 000 .png

Link to comment
Share on other sites

  • Hash Fellow

So you want part of a model to be toon and part regular render, right?

 

You can do that in one pass with groups.

In the surface properties for the model set Toon Lines>Thickness to be zero and the Toon Shading Method to be "Standard"

 

Then make a Group for the parts you want to be toon lined and shaded and and set your desired toon settings for those.

 

When you render, set Toon Render ON, but leave its sub switches OFF

 

Let me know if that doesn't work.

 

If you want to come to Live Answer Time I can walk you though that.

Link to comment
Share on other sites

Robert

Thank you very much indeed for your help with that problem. I tried it overnight and it worked as you say. Rodney suggested it to me a few weeks ago but I misunderstood the process, thank you both. It should cut production time enormously, Although I might have to redo some earlier scenes.

regards

simon

 

Nosey 03.mov

 

noticed some mesh penetration that hadn't been visible before so redoing it now.

Link to comment
Share on other sites

  • 2 weeks later...
  • Hash Fellow

On the "Nosey" clip, I'm not sure of the meaning of him seeing himself in the binoculars. That is what is suggested by the image of himself in the lens? But he's supposed to be seeing someone else, right?

 

The flashing in "NHF" is a tough effect to do. The hard edge of the light cones inside the car seems odd. Do you have a reference of the effect from a live action moviemaybe?

 

Typically in shots where a light is shining in there will be lots of edge lighting effects on objects that are almost but not quite silhouetted. I painted some in on the man's face. There would be something like that on almost every light side edge.

NFSPink.jpg

 

NFSCyan.jpg

 

 

 

 

Perhaps UK emergency vehicles have different color lights but I would expect them to be more red and blue, than pink and cyan.

 

NFSRed.jpg

 

NFSBluer.jpg

 

Link to comment
Share on other sites

Robert

Thank you for your feedback. Very much appreciated.

To try and respond in order.

 

The reflections in nosey were a byproduct of the setup but, I liked the result so decided to keep them.

It came about because he is standing in front of a glass window and the reflections off the window go back into the glass of the binoculars.

I liked it because of the claustrophobic impression and, at risk of being pretentious, the visual qualities of it, Shapes within shapes, the eye in the small inset.

 

The colours were wrong in NHF S10.mov so I redid them when Tore pointed it out to me on FB. In the UK, and I think in the rest of Europe, the emergency vehicles use Blue lights. The image in my head was really from US films and TV. I tried some further tests, including only blue, but went with Blue and a very red orange in the end because it was more dramatic and added to the drama. I rendered it over night. and changed it to a POV shot from the drivers seat.

 

Some of the files were taking a very long time to render ( nearly three hours with 5 passes ) because there are six lights, all with volumetric settings. All the lights are shadow mapped rather than ray traced, with a map resolution of 4096x.

 

This was the redone version

NHF S10B.mov

 

 

With regard to the hard edge. I don't know what might be causing that ?

Initially it was on a 100% quality setting to try and cut down on the render times. Have just spent several hours trying to find the cause of the hard edges but to no avail.

Screen Shot 2017-05-31 at 13.15.37.png

 

Wondered if it might be the model so tried this too but, as you can see, that was not the cause.

I removed the models and decals.

This is the movie without models, just the lights.

Notice the hard edge is there at the start

A Lights.mov

 

This is the cho file

Ambulance Lights.cho

 

regards

simon

 

 

Edit

 

 

Screen Shot 2017-05-31 at 14.27.38.png

Link to comment
Share on other sites

Start of the final scene.

This is the first pass at the acting so, any critical feedback very welcome indeed.

 

The scenario is, they are sitting in a hospital corridor awaiting news about the Grandmother who has had a motorcycle accident on the way home from the party for her golden wedding anniversary,

( I think 50 years is golden, or is it Ruby ? )

 

NHF S12.mov

 

No real lighting work as yet and the render settings will be mixed properly later.

simon

Link to comment
Share on other sites

  • Hash Fellow

If you made their clothes 0,0,0 black instead of invisible you could simplify the compositing. Is there still a problem combining the toon and regular in one render?

 

 

 

I would experiment with a different treatment of the eye blinks.

 

Right now it seems they go like

 

(open percentage) 100-50-0-50-100

 

Change the middle so they are not the same going down as going up, maybe like this

 

100-75-0-25-100

 

or maybe the reverse of that

 

100-25-0-75-100

 

Adding an extra frame of zero might help them seem more weary and less flickery

 

100-75-0-0-25-100

 

 

 

 

Head turns will usually look better if they move in an arc rather than a straight path. Usually a small dip in the middle. They also will move faster in the middle and slow-in to the end position. It's very rare for anyone to turn their head in even increments from start to stop.

Link to comment
Share on other sites

Robert

Thank you again for your help and feedback.

I tried the files you were kind enough to adapt for me.

I loved the AO effect and may have to redo the whole thing to get visual consistency ( will leave that for later though)

When I rendered them out I got this result.

NHF S12AO  010.png

 

Tried to find out why the Perry figure came out so dark and, in the end decided to render them out as toon to be composited afterwards. Using the time to get on with working out what to do next.

regards

simon

 

Link to comment
Share on other sites

  • Hash Fellow

Tried to find out why the Perry figure came out so dark and, in the end decided to render them out as toon to be composited afterwards. Using the time to get on with working out what to do next.

 

 

 

 

Some sort of toon setting change I would guess.

Link to comment
Share on other sites

 

 

Some sort of toon setting change I would guess.

 

Robert

It seemed to happen when I was trying to do the combined styles in one render, Toon and Standard, when AO was selected, The curious part was that it affected one figure but didn;t seem to affect the other, female figure,.

I'll use the render time length to have another look tomorrow. Its taking about 80 mins per frame at the moment.As AO is set to 100%

regards

simon

Link to comment
Share on other sites

100 is so I get the best I can but,

I can use the time to do other things.

One thing I do like about computers is that, while they are doing the grunt work, you can get on with other things.

A few years back, I had three machines rendering in the garden shed, while I animated in the living room. It was warmer in there than the house. The joke being I was working about 550 hours a week or more !

regards

simon

Link to comment
Share on other sites

AO set to 70 has been good enough for me so far.

David

Thank you for your feedback. I'm still very new to AO so thought to go with best quality for now. I can get on with other things while its working through too !

regards
simon
This is version with Toon rendered added.
Do the wobbly lines distract too much
May have to redo to get it right ?
The wobbly likes are caused by erasing the 'not needed' toon parts in compositing but, its awkward to get consistent line placement over time.
Link to comment
Share on other sites

Robert.

Thank you for your feedback ,here and on FB.

I tried the combined in one render but, for an unknown reason the Perry figure ( model modified from Jim Talbot's Sir Nigel model ) came out a lot darker than the other figures. I tried lots of things to even it up but, in the end, went with the composite method. I will have another look.

 

You mention turning all the non toon parts to 0:0:0. Is that in the cho or on the model properties and does that include the two figures in standard render?

regards

simon

Link to comment
Share on other sites

  • Hash Fellow

Make two copies of your Project. Of course, all your models must be "embedded" for this.

 

Copy 1: For the models that have some Toon, set the groups that create Toon to black with no specularity. Leave everything else as is and render as usual.

 

 

Copy 2: For the models that have no toon at all, delete all the groups and set the whole model to black with no specularity. For the models that have some toon, delete all the groups that are not creating Toon and set them to black with no spec. Render the scene and you should get all black except the Toon parts. Since this one is only toon surfaces you can save time by not rendering with AO.

 

These changes are made in the objects folder.

 

 

The two renders will composite exactly together with an "add" mode.

 

"Add" literally adds the RGB values from each render together. In this case the colored portions of each render (values greater than zero) will show through the black (zero) portions.

 

Different programs call "add" different names. Photoshop calls it "Screen" but After Effects calls it "Add". A:M calls it "Add" in a "Composite project".

 

We can talk of A:M composite projects if you don't have software that can do the "add"

 

 

Test this out with just one frame from each render first, to make sure you are getting the parts right.

Link to comment
Share on other sites

Robert

Thank you very much again, very much appreciated.

 

I did try it yesterday, and had a lot of trouble with AM crashing and png files not 'parseing'.

I got there eventually, but it was a bit exasperating on the way.

 

I tried it in V18 to try to avoid problems when working normally in V17.

 

Embedded all files as you suggested and followed the instructions you detailed.

 

When I tried to render the test files, using png as the output format, they would render on screen and,when complete, if I closed the render window it would crash AM. Then, when opening the rendered file in photoshop, it flagged a warning saying,

 

it"could not parse the file'.

 

Tried several times, modifying the files accordingly, same result.

 

It eventually worked when the output format was changed to .tga

this is the PS'd composite.

S12.jpg

 

When rendered without OA this was the result

S Two toon 012 copy.jpg

Tried to modify the surface values, but without success.

Tried the flat surface option, it worked but, was not the look I wanted.

Went back to first settings and used AO which seemed to cure the bleached out look. Render times took a big hit though.

 

Going to try another test.

regards

simon

 

Link to comment
Share on other sites

  • Hash Fellow

When you turn off the AO you also need to turn down the Global Ambiance level in the Chor, otherwise you have lots of global light on things with no shading to counteract it.

Link to comment
Share on other sites

"Add" literally adds the RGB values from each render together. In this case the colored portions of each render (values greater than zero) will show through the black (zero) portions.

 

Different programs call "add" different names. Photoshop calls it "Screen" but After Effects calls it "Add". A:M calls it "Add" in a "Composite project".

 

We can talk of A:M composite projects if you don't have software that can do the "add"

 

 

Robert

Thank you. The renders are going through now. While I'm waiting for that, can I ask how you'd go about doing it in AM as a composite project?

I've tried it in PS, and it works but will be a very long job doing it frame by frame for1 30 frames.

Regards

simon

Link to comment
Share on other sites

Hey! Simon

You can import image sequence and (post production) composite, color correction, edit, sound, etc... in Blender. I use Blender as A:M buddy.

attachicon.gifScreenShot_25600617_1045.png

 

 

Bobby

Thank you.

I have Blender but have barely used it.

Robert mentioned a way to do it in AM so I was going to try that.

I have done some composite work in AM before but not a lot.

 

That's the character you used on your bike ride project isn't it, have you made a film with it ?

regards

simon

Link to comment
Share on other sites

  • 1 month later...

Robert.

Thank you.

I'm plugging away at the final scene, its going very slowly, hoping to have it at least blocked out by Sunday.

The intention is to then post it here, and elsewhere, to see if the narrative coheres visually, before putting the sound to it.

regards

simon

Link to comment
Share on other sites

Working on an opening title sequence while a render goes through.

Any critical feedback very welcome indeed.

 

There is a curious problem with the decals. I used png files with a transparent background.

Its not consistent. the same card render differently. As you can see, they render slightly differently during the sequence. In some frames, the background behind the letter is clear, in others the alpha channel shows as a tone ?

Any suggestions ?

simon

Cards.mov

 

Link to comment
Share on other sites

Robert

Thank you for your kind offer of help. I found a way round it by filling in the alpha with the same background colour as the cards. I think some of the anomolies are due to misalignment when applying them. I've got a revised version going through now.

Regards

simon

Link to comment
Share on other sites

Almost done.

If you have the time, could you cast a critical eye at this and let me know what you think.

The final scene is only blocked at the moment as I need to record the dialogue with the right voices. There is only one section with sound. What I would like to find out is, does it hold together visually,without the sound. The sound effects and dialogue will ( I hope ) complete it but I'd like the visuals to do most of the work.

So, two questions

 

a. What is the narrative storyline ?

b. any errors or faults you can spot that I have missed ?

 

Please be as critical as you feel necessary, don't worry about being negative or impolite.

Thank you for your time.

simon

 

 

xhttps://vimeo.com/227532819

 

Password Fred77

Link to comment
Share on other sites

  • Admin

That's a lot of work!

 

I think what I write below is already things you know but... for what it's worth I offer my thoughts:

a. What is the narrative storyline ?

 

I didn't really feel I caught much of a narrative until near the end because the visuals very definitely set the scene for illness/injury at the hospital.

I think I lost track of who was who just a little and some of that might be due to similar coloring on the characters.

For instance, I remember a car driving from right to left but I don't recall who might have been in that car and I don't know what became of it or it's driver.

In hindsight, perhaps it's the guy who runs across the road and hugs the motorcycle driver.

There is a guy with binoculars but... not sure where he was watching from etc.

Is it possible that all of these four are the same four that are in the waiting room at the hospital (I'll have to watch again to confirm).

So, at times I wasn't sure if a new character was being introduced, if an known character was reappearing etc.

 

As you suggest, I do think audio will very likely resolve some of these issues.

Although I think I know who is in the hospital bed... mostly because I've been following this topic... I am not 100% sure in light of the viewing of the video itself so I need to go back and view the video again to be sure. Similarly, I am not entirely sure what put that person in the hospital although I seem to recall the answer to that from the discussion in the forum too. I will guess that an audio cue may be added later that will lock that down.

 

So, "errors"/"faults"? I think taking a good look at the transitions from scene to seen with an eye for where the viewer's eye is most likely to be and then motivating the camera to move/cut accordingly would be ideal. Upon subsequent viewings I should have more to say but I want to capture my initial thoughts on first viewing. The second thing would be to find some bold contrast to differentiate the characters a little more. An example of the sameness in contrast might be that of the character with a beard. The beard didn't initially register and in the process of confirming the character had a beard I may have missed some important narrative beats.

 

any errors or faults you can spot that I have missed ?

 

It seemed to me that a few sequences started from a halted/stopped position rather than 'in progress' which might have detracted a little from the narrative flow.

When the camera 'clicks' started to kick in and the photos were being taken... that sequence read very well although now that I think of it I recall the bearded guy but am not 100% sure who the other person was.

 

P.S. I was NOT prepared for the audio to suddenly start after the prolonged silence. That definitely woke me up!

I need to be more careful with my volume. ;)

 

I will look to see if there is any specific area where I think a transition can be adjusted.

At a guess I'd say several could either be shaved by a few leading frames or refashioned via a cross dissolve.

 

A stray thought: I wonder if there might be a few narrative 'tags' that might help the audience to keep tabs on characters. A few possibilities: color (the 'pretty' girl in purple is a good example of this... she was easy to spot). Having the others wear clothes that do not match that of others might be useful. For instance, one character might wear very dark pants but none of the others wear dark pants. One might wear a tie while none of the others wear one. Certain characters might only (or primarily) enter and exit the screen from the same side or 'own' a specific place on the screen more than the others. This might translate to depth where one character (the grandmother) is free to move in any dimension etc. but others are constrained to move only to the right... only to the left... occupy a specific distance from the camera... just random thoughts here... not necessarily suggestions. I think... but will have to check... that it might be good to have a few more closeups of characters so that we can see and recognize them easily. For instance, in the waiting room we see them together so a closer view from one to another might aid in recognition especially if it 'matches' or echos a scene we've seen earlier. Ah, that that guy... that girl... etc.

 

Keep on keeping on. That is an impressive amount of work.

Link to comment
Share on other sites

  • Admin

On second viewing a lot more is clear.

There are several things that I noticed that I didn't think I did the first time.

For instance.

 

I didn't read the title "Happy Families' is the cards.

I had just seen the cards and photo images of people and thought... that's neat.

 

I caught that the two characters were together but then drove away one in car and one on motorcycle.

I had blinked and missed that they started together.

Not sure where the guy went. (I'm not too sure it's important so... perhaps we don't need to see him driving off etc.?)

In this way the narrative would follow/emphasize the importance of the woman.

 

I'm still not entirely sure who the neighbors are... the fancy woman appears to be Mom of the son mowing the yard.

But not sure why they go in house on other side of road....

Now... that terrorist with binoculars... not sure what he's up to....

 

Okay... accident. How did I miss that the first time?

Got it.

 

I think... grandson, husband, fancy woman/daughter and... a granddaughter? are gathered.

 

Is grandma going to be okay?

Maybe she should stop pulling wheelies and pissing the neighbors off.

Maybe the terrorist guy ran her off the road?

I didn't see him in the hospital but someone better inform the cops that he may visit her.

This has the makings of a thriller!

 

Okay... I'll stop there. :)

Link to comment
Share on other sites

Rodney

Thank you very much indeed for your detailed feedback. Very much appreciated.

I won't answer your points just yet if thats OK (?) as I don't want to colour the thoughts of anyone else who might reply.

I'm going to give it a few days to see what other feedback I get from the forum and elsewhere, before starting to readjust. The final scene will need a lot of work when the voices are rerecorded.

The end credits will be the same house of cards but doing the thing that inevitably happens

regards

simon

Link to comment
Share on other sites

  • Admin

It's more than fine not to answer now (or at all even!) because it's mostly just impulse reactionary marginalia.

 

Sometimes I wonder if critiques are even worth venturing into...

 

After viewing your video multiple times I find that many of the points I raised are in fact there in the video.

For instance, you've put considerable effort into varying the color of the clothes the characters are wearing... transitions from one shot to another do connect better than I originally comprehended... the narrative flows better than I thought.

 

Some of the questions I raised I still don't have any insight into but again, much of that will be supplied by the audio and voice tracks.

 

There is something I read not long ago that may be useful concerning critiques and that is (roughly):

When someone critiques a thing it's not very important to consider the details of what they suggest but rather to consider the areas that those suggestions occupy.

Most of these equate to things you already know and will refine if/when you have the time. For instance, transitions. Each transition is important but they all must work within the context of the greater whole... which only you have the bigger picture of. Another example: Character differentiation is important. No surprise there. Much of that will be brought to light with the recording of voices. Yet another; things that aren't clear at the beginning often crystalize into fuller clarity by the end of the telling.

 

Having said all of this I would therefore amend my thoughts to suggest that anything you can do to direct the viewers attention to those things that drive your narrative of "Happy Families" and what that means from the various perspectives of the characters will be time well spent.

 

I look forward to your next update!

Link to comment
Share on other sites

Some of the feedback received thought that both the Bill and Doris characters didn't look old enough so I've spent the last few days trying to get a mapping setup that will age them. Its proving dificult with a toon setting.

 

I post this as an amusing attempt.

The eyes might have to be adjusted !!!

 

Rogues Gallery.jpg

Link to comment
Share on other sites

  • Hash Fellow

I like the wrinkles on the middle version!

 

To do "aging" on a toon character however I think you'll have to work with larger details... saggier jowls with more prominent folds, likewise with the bags under the eyes.

 

A bigger nose. Everyone's nose gets bigger with age.

 

Basically I think you would need more sagging on the big structures that can cast toon lines.

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.

×
×
  • Create New...