As the title says, it is another one of them. I need to set something up so I can make these on a more regular basis. And actually knowing what I am supposed to animate before I start to stream to an audience of… 1… I think that’s a bug… It’s probably zero viewers.
In an effort of continuing the streak I’ve been having with streaming. (well. I’ve done them at least. I have yet to gain followers.) In that attempt I will now revive the old idea of making a feature length movie using random chance as my sole guide.
The deck of 85 cards I started out with is still present and it’s been complemented with another one with 60 cards.
So, The rules I’m setting out for myself (subject of change if need be):
Choosing of the timecode to work on will be done with two decks of cards each streaming session:
– Deck of 85 cards will select the minute
– Deck of 60 cards will select the second
Once drawn the minute card (from the deck of 85) will not be returned to the deck until the project is done!
Before drawing the seconds card, the timeline will be checked to see that there are no animation already done inside the minute. If it is, then the corresponding cards for those seconds will be set aside until next full draw.
– It is then up to me as a creator to come up with what will happen during this particular point in the narrative and draw frames accordingly. This can be more than a minute worth’s of content but beware of point 3 below.
– The method of animation is all up to the animator at the moment. Again. Beware point 3.
A new timecode to work on will only be drawn from the piles when the last timecode is fully finished. Once finished it will not be adjusted in any way until project is finished in full. This includes all visuals. Soundwork may be done according to point 4
The soundwork for each timecode can extend to outside the animation. But once finished, it too should not be touched.
The project will be created in 2K DCI Cinemascope resolution (2048×854) 24 fps and 5.1 surround sound. The finished timecodes will be uploaded to youtube in full HD Cinemascope (1920×800) 2.0 dolby stereo downmix.
The Finished 4K form will be in two versions. Timeline Corrected where the narrative flows as on the timeline of the NLE. And a second one will be created that keeps the randomised shot-order mainly for fun of the viewer.
Two playlist will be maintained on youtube ( youtube.com/jmalmsten ), the timeline-accurate order and the randomised chronological uploaded order. Gaps in the timeline will have placeholder footage.
Once project is over. A remix of the audio with voice-acting and music will probably ensue before the final directors cut will be released alongside the other two versions.
I’ll probably add something or subtract… we’ll see how things goes. This is a project mainly for fun anyways… (nervous laughter).
I do not claim ownership over After Last Season, El Mariachi, Upstream Color or even A Scanner Darkly. The following article is just a blog overview of what has been on my mind lately and it just happened to be After Last Season
Ok. Some may know that whenever there’s a discussion about “worlds worst movie” and I’m asked about which I think qualifies I always bring up 2009’s epic failure of narrative fiction called After Last Season.
Never heard of it? I do not blame you.
This film stumbled into existence that year. Was distributed to half a dozen american theaters. And then neither its creators or its distributors wanted to touch it ever since. There’s even rumors about how the theater owners were instructed to just burn the 35mm film prints rather than sending them back…
But it doesn’t end there. Oh, no.
Then the rumors started to come in about the production. That the film had a production budget of around $ 5 million. That most of it was basically just a scam to get away with as much money as possible. Because supposedly this wasn’t a Tommy Wiseau level passion project gone awry. No. This was supposedly a hoax all along. And when we see what the unsuspecting moviegoers saw at release… that theory didn’t sound all that far from the truth.
Just as a reminder. Here is what a talented filmmaker can do with roughly $ 7 000 production budget spent wisely in the earliest of the 90’s:
The blank guns barely worked (most of the full auto-fire on screen is looped because the guns would more often than not just fire once and never cycle). Action set-pieces was pieced together with shoestring and Rodriguez himself even became a human guinea pig for medical science to pay the meager bills the production amassed. I mean. The guy should write a book on how it was made.
Ok. You say. Not even the same genre, so how is it compareable?
well. Here’s Upstream Color from 2013. Made for a staggeringly high budget of $ 50 000:
CGI-effects. Thoughtful scifi. Everything.
Now then. Look at what that $5 000 000 got us in 2009:
No. I am certainly not joking. This. This is what was shown as a theatrical feature. And watching the full thing? Even worse. Giant stretches of awkward silence. Barely audible dialogs that go nowhere. Characters brought up and forgotten. Set building that… is just kind of a warehouse when it’s not an obvious home standing in for a top of the tier university hospital. And. Copying paper. You will be fascinated by what they can expect us to accept with copying paper.
The list just goes on and on. Oh and those CGI-sequences. Those ENDLESS CGI SEQUENCES!
The little yellow doton her temple is a “chip” that makes mind-sharing possible.
This is a trashbin
These are birds
These are humans
A woman stalked by a man with a “blur over his face”
I think it’s a human again?
So why do I bring this up with the title of this blog-post? Well. Because a few weeks ago. After a couple of social media conversations I started to think about this film again. And ideas started to brew.
Step 1: First Trimmings
What if… What if one could salvage something from it? At least glean a little about its narrative structure. I have an NLE. I have a legally bought copy of the released film (which wasn’t easy to obtain) on DVD.
And this is bringing me down a rabbit hole I was not expecting to go down.
I ripped the video. Threw it into Premiere Pro and started the dissection. Step one. Simply replicating the edit.
This is the most tedius part. It’s just me going through the film shot by shot and putting in cuts where there already are cuts in the original. It’s the ground-work. And it will help me navigate later.
I sadly have no screen-shots of this step. I forgot about documentation.
Then. Step 2. Going through and just trimming anything that is dead space without doing obvious edits like jumpcuts and the like. This brought the runtime down from the original 1:30:28 to a pretty lean 01:08:50. I also took the liberty to speed up any footage that need not be as long but had no obvious edit-points (I’m looking at you, CGI-scenes!).
Step 2: Dogme 95
That was a couple of weeks ago and I put the thing away for a while to focus on some things more pressing. But the project still lingered in my mind.
And today I took another stab at it. This time I wasn’t nearly as careful. I brought out the weed-wacker and took out all the empty space even if it resulted in jumpcuts (I managed to hide a few. But it wasn’t a priority). I call this step the “Dogme95 version” because of how ruthless I was even though it technically doesn’t comply with their stringent criteria. And the result was a fairly lean 00:54:25 runtime.
you may also notice that I added a track above with a cinemascope crop. Now, why would I do that? Well. As other viewers have noted. The film is kind of poorly shot. And it’s surprising how many shots are helped by simply cropping it vertically a bit.
Was this an intended step that was abandon in the original Post Process? I don’t know. It just looks like it. The same with the colors. They do look suspiciously flat. Almost like if they shot the film in low contrast or scanned it low contrast to be able to do a real color grade later? We’ll probably never know. But I did go ahead and put both the crop and a layer of FilmConvertPro LUT onto it and it does look a bit nicer to the eyes. Remember though. I am still working with what is essentially a h264 rip of a 480p DVD release that seems to have been made with a soft focus 16mm workprint. So It still looks kind of awful. But just a little less awful makes this all a lot less of an agony.
The actual picture quality that is present on the DVD-disc
Standard FilmConvertPro plugin to regain a bit of the colors and hide some artifacts with 16mm grain.
With LUT and CinemascopeCrop (Notice the disappearance of the ceiling fan)
Original image from the disc
Simple FilmConvertPro 16mm plugin
Again. The cinemascope frame conveniently hides the distracting shadows above.
And it’s still lit with those work-lights. And no amount of LUTs will get that ugliness nice.
Step 3: Now what?
Well. I am afraid to say that no matter what the underpants gnomes would want. The next step is not “PROFIT” no matter how far I take this silly little project. This is as unofficial as it gets. I have made no attempt to contact the makers of this film regarding what I’m doing. And I highly doubt it they would be interested, anyway. I simply do this as an exercise in video-editing because it felt like the released cut feels more like a rough early workprint than anything finished. It’s a challenge I gave myself. If I was handed this raw material. Could I get something out of it using the meager skills I have amassed over the years?
Maybe. Maybe not. Anyway. I can not show the results publicly anyhow. I
The abyss is staring back at me through the rabbit-hole…
The one thing I can say however. Is that forcing myself to watch and listen to the dialogs and the visuals have actually starter to get to me a bit I think. I’m starting to put together pieces of the puzzle I never knew were there.
But until the next update I will just leave you guys and gals with what is so affectionately called a “clock radio” and a video about the Rotoscoping techniques that made A Scanner Darkly (2006) that breeds the idea that maybe what we have here is a film that was meant to be the rotoscoped or at most of the image would be replaced in post.
That’s it, really… There has been such a shortage in electric fans in me local locale. So when I finally saw one at ÖoB (local budget-shop) for roundabout 299 sek (roundabout $35) I had to acquire one.
It’s flimsy as hell and noisy to boot. But in a heat-wave like this one… I really have no choice when this was quite literally the only one available for the last several weeks… It’ll do until they restock on the nice ones.
And… I am doing my best shampoo commercial face ! 😀
Just wanted to animate an abstract shape that loops in a fun way. Only now that I uploaded it do I realize it kind of resembles a logo. But logo for what? I honestly can’t remember…
The inspiration is a few of those loops that Anno (not sure those linked are Anno’s but yeah… ) and Morimoto has done.
The movement may be a bit on the too fast side… but part of the excersize for me was to try animating in full 24fps instead of my beloved 8 fps
The excersises just keep on coming. Here, I left out the live action background and just wanted to build a short animation that lasted 10 seconds. All done in Adobe Animate. I kept it very sparse in the colors and the finish because I set myself a time limit. When the Academy Awards started, I had to be finished. And so I was.
And thank goodness Roger Deakins actually won for once!
Be seeing yah!
Holy Moly! It's been half a year since the last post? Man. I need to get back into this gambit.
Well. Here's a quickie test. It's basically another Itano Circus. But it's done in a bit more unusual way.
I first filmed the background, using my GH4 in 1080p 2 fps video mode. – (In variable framerate mode it actually only records in the framerate that you set it to. It saves it as a standard 24 fps file for playback but the framerate difference can make either super-speedy footage (if recorded with framerate slower than the playback) or slow motion (if recorded with a faster framerate than the playback). And that works great for my experiment as I want it suuuper speedy in this case. 2 fps recording makes it 12 times faster. Playing that back at 8fps makes it… uhhh… ok, I don't know, but it looked like I wanted it to at least!)
Then I took it into after effects.
Time-reversed it (I walked forward during filming to be able to get footage that was traveling backwards without risking tripping over stuff in my messy apartment).
Then. I exported that as a png-sequence (I found a great youtube channel btw that mentions importing video into flash as png-sequences)
And then using that as the background, I drew the whole circus thingy in flash in 8 fps… sorry, Adobe Animate.
Then I exported it (the animation layers, that is) as a png sequence again and…
imported it to After Effects, layered it back on top of the original footage interpreted as 8 fps and finally exported it.
Kind of worked out better than expected in my opinion. I may be using this method more in ongoing projects… because I still suck at drawing backgrounds… 🙂
Be seeing you!