Theme
For the final project, I'm going to make a 4k piece that explores the bridge between compact formats and large cinema formats. While large formats push for a higher data rate and larger resolution, technology in many devices and platforms want (Admittedly, many new mobile and portable devices have been released recently with support for better resolutions.) But in reality, quality is something often overlooked by a viewer in exchange for convenience.
In order to create a connection between the standards, I want to take the small and attempt to scale it up to the impossibly large. I'll turn "small" videos into large "objects." adding weight and volume to them, and have them interact physically in space. My main interest, besides using content ironically, is to label the large cinema space as a space of the visual spectral. The commercial film market capitalizes on computer driven visual effects, so I want to pieces to blatantly emulate this. I'm no special effects hater, I just like addressing the intersection between mass media products and art.
Source
Usually content scales to smaller formats. A commercial product is created in generally the highest accessible format, and then scaled down for distribution on DVD, TV, web, and mobile devices. I want to start with content that is at the lowest end of the quality spectrum and scale it up.
I'll use youtube videos that are 480p or lower. I wanted to find videos with very few views that represent popular streaming video genres (we discussed those in class.) Although I think that popular "viral" videos would have worked as well (and there would be some satisfaction in fragmenting them) those videos have been thoroughly remixed and reused already. They also may make the piece seem more of a comment on the videos themselves, and not viewing technologies.
The piece overall should be fairly absurd, will low quality sources being displayed on high resolution tech, in addition to pointless destruction of the physicality of the images and the frame. Chaos for the sake of chaos.
Tuesday, March 13, 2012
Project 2: Practice: Textures
Some objects and materials I'll be using in the animation will include: paint, colored paper, buttons, tissue paper, yarn, Popsicle sticks, pip cleaners, paper clips, and more. Many of these require textures to read properly. In order to get the textures I needed at the proper resolution, I purchased the actual goods from a craft store.
I then took high resolution photographs of the materials for use in texturing:
Monday, February 13, 2012
Project 2: Research: Rendering
Technical Details
I'll be using Autodesk 3ds Max 2012, with Mental Ray for rendering software. (http://usa.autodesk.com/3ds-max/) I'll be using 3d stereo cam script by Louis Marcoux (http://area.autodesk.com/userdata/fckdata/4952/StereoCamModifier_V1_7_3.rar) to generate a camera setup. I'm acquiring materials at high res, so ultimately the project can be scaled anywhere from SD to 4K (2d or 3d.) The higher resolution I want to generate, however, the more CPU time I'll need for rendering. Since I only have a single physical (eight core) Pentium i7 processor at my disposal, I'm going to have to find more machines to render the video in a timely manner.
Rendering Solutions
I looked into possible answers to rendering under a timeline. Ideally, I wanted to get a separate rending machine. I looked into hardware like renderPROs from Boxx (http://www.boxxtech.com/products/renderPRO/pro_overview.asp) and workstations from HP (http://www.hp.com/united-states/campaigns/workstations/z400_features.html#.TznDzVxSQsI) that could serve as dedicated render machines. I was interesting in getting one through a grant, but ultimate found this route too expensive, even if they qualified for a grant.
Another route to go is cloud rendering, where I would send project's files to another server and have their CPUs do the rendering. I found several services that offer this, but the price is also quite steep. Possible services include http://www.renderrocket.com/, http://www.rebusfarm.com/, and http://www.rendernation.com/. Autodesk also offers cloud rendering, for subscribers of certain products. These services charge per CPU usage, and calculating costs is complicated as I haven't done the project yet and am unsure of the render time per frame. I do aim to make the project 1:30-2:00, which equates to 2160-2880 frames. At ~1 hour a frame this method could also be too expensive. If this method is also extremely expensive, I will have to downgrade quality to allow for rendering locally.
I'll be using Autodesk 3ds Max 2012, with Mental Ray for rendering software. (http://usa.autodesk.com/3ds-max/) I'll be using 3d stereo cam script by Louis Marcoux (http://area.autodesk.com/userdata/fckdata/4952/StereoCamModifier_V1_7_3.rar) to generate a camera setup. I'm acquiring materials at high res, so ultimately the project can be scaled anywhere from SD to 4K (2d or 3d.) The higher resolution I want to generate, however, the more CPU time I'll need for rendering. Since I only have a single physical (eight core) Pentium i7 processor at my disposal, I'm going to have to find more machines to render the video in a timely manner.
Rendering Solutions
I looked into possible answers to rendering under a timeline. Ideally, I wanted to get a separate rending machine. I looked into hardware like renderPROs from Boxx (http://www.boxxtech.com/products/renderPRO/pro_overview.asp) and workstations from HP (http://www.hp.com/united-states/campaigns/workstations/z400_features.html#.TznDzVxSQsI) that could serve as dedicated render machines. I was interesting in getting one through a grant, but ultimate found this route too expensive, even if they qualified for a grant.
Another route to go is cloud rendering, where I would send project's files to another server and have their CPUs do the rendering. I found several services that offer this, but the price is also quite steep. Possible services include http://www.renderrocket.com/, http://www.rebusfarm.com/, and http://www.rendernation.com/. Autodesk also offers cloud rendering, for subscribers of certain products. These services charge per CPU usage, and calculating costs is complicated as I haven't done the project yet and am unsure of the render time per frame. I do aim to make the project 1:30-2:00, which equates to 2160-2880 frames. At ~1 hour a frame this method could also be too expensive. If this method is also extremely expensive, I will have to downgrade quality to allow for rendering locally.
Project 2: Concept
Technology
For the stereoscopic project, I wanted to create a video using computer generated graphics. The Sony camera technology available for use is shaky at best. In order to have complete control over the image, I'll use 3d software to produce the video. Software allows the use of virtual cameras, giving the producer full control over the ground plane and the interaxial (positive/negative space.) This control is essential with such a finicky viewing technology.
The abilities that stereo 3d brings to the image comes with several cons. Motion is compromised, framing must be reconsidered, the image is dimmed, and areas of contrast can be seen through the glasses to create ghosting. By conduction as much control as possible over the image, I'll try to work around these visual handicaps.
Form
I wanted to put a twist on 3d by representing it through elements that are, for the most part, two dimensional. I'll use two dimensional bits in a 3d space. I also want to use representations of craft objects. I plan on using elements of paper, crayons, markers, pipe cleaners, popcicle sticks, colored sand, finger paint, and yarn. Through using low tech craft/children's art products, I'll try to pull stereo 3d (a cutting edge technology) back to a simple more humble medium.
Content
For this piece, I want to usurp the point of view of a fictional character, a child of the future. I want to create a motion graphic story about a future where children live in a rapidly diminishing environment, and speak with disdain of their predecessors. There will be narration talking about a fictional future, from the point of view of a child. (This is also in line with the chosen form.) More details about the "story" to come.
For the stereoscopic project, I wanted to create a video using computer generated graphics. The Sony camera technology available for use is shaky at best. In order to have complete control over the image, I'll use 3d software to produce the video. Software allows the use of virtual cameras, giving the producer full control over the ground plane and the interaxial (positive/negative space.) This control is essential with such a finicky viewing technology.
The abilities that stereo 3d brings to the image comes with several cons. Motion is compromised, framing must be reconsidered, the image is dimmed, and areas of contrast can be seen through the glasses to create ghosting. By conduction as much control as possible over the image, I'll try to work around these visual handicaps.
Form
I wanted to put a twist on 3d by representing it through elements that are, for the most part, two dimensional. I'll use two dimensional bits in a 3d space. I also want to use representations of craft objects. I plan on using elements of paper, crayons, markers, pipe cleaners, popcicle sticks, colored sand, finger paint, and yarn. Through using low tech craft/children's art products, I'll try to pull stereo 3d (a cutting edge technology) back to a simple more humble medium.
Content
For this piece, I want to usurp the point of view of a fictional character, a child of the future. I want to create a motion graphic story about a future where children live in a rapidly diminishing environment, and speak with disdain of their predecessors. There will be narration talking about a fictional future, from the point of view of a child. (This is also in line with the chosen form.) More details about the "story" to come.
Wednesday, February 1, 2012
Project 1: Results
http://vimeo.com/36064568
I find that the piece overall achieves in creating a mind-numbing repetition. The increase of tempo also works to compound the repetition. It doesn't quite seem catchy enough to be viral quality, which is something I was shooting for. Perhaps if I was more selective with the clips, or if I cut the clips as short as possible the tempo might be absurd. There is also the element of time warping, but again, I wanted to keep heavy handed decisions out of the process.
It would also be interesting to see how the repetition would play out if I used all eight films, as the duration would be nearly twice as long.
Overall, I think piece carries the idea of brand monotony well, but doesn't read closely as stereotypical internet remix.
I find that the piece overall achieves in creating a mind-numbing repetition. The increase of tempo also works to compound the repetition. It doesn't quite seem catchy enough to be viral quality, which is something I was shooting for. Perhaps if I was more selective with the clips, or if I cut the clips as short as possible the tempo might be absurd. There is also the element of time warping, but again, I wanted to keep heavy handed decisions out of the process.
It would also be interesting to see how the repetition would play out if I used all eight films, as the duration would be nearly twice as long.
Overall, I think piece carries the idea of brand monotony well, but doesn't read closely as stereotypical internet remix.
Project 1: Practice
Download/Conversion
Was able to find the original content easy enough; had to put up with fairly long reencoding times when converting, as well as errors that caused some omission of content from a couple of the films. (Films I later decided I didn't want to use.)
Extracting
As I was extracting the desired clips from the first film, I noticed that the practice was time consuming and extremely unpleasant. Scrubbing through movies of long duration is not very intuitive, and extracting a proper subclip often meant going back over words and sentences several times. Listening for the key words also meant watching the films in a way that doesn't qualify as entertaining, but work.
I also noticed that even with the first film, I was able to extract a large amount of potential content. Because of this, I made the decision early on to only use the content from the first four films. Perhaps at a later date, when I'm feeling especially masochistic, I can expand the project to encompass all eight films.
The extraction process was taking about 1.5-2 times longer than the content duration I was working with, but as I got better at referencing the film transcripts I was able to make subclips for the last film (Harry Potter and the Goblet of Fire) in about the same amount of time it takes to watch.
After extracting "Harry/Potter" from the first four films, I had about 420 subclips that amounted to a 6:20 duration.
Assembling
I decided to arrange the clips from longest duration to shortest duration. This gives the piece an increasing tempo as it goes on, and seems to build tension. It also removes artistic decisions that I would have to make if I placed the clips by choice. Again, I want the system of distillation to be systematic and somewhat controlled. (With as little choice as possible.)
Editing
After creating the order by a predetermined system, I still had to make decisions on what to cut out, as the piece was too long. Purely for flow and aesthetics, I decided to remove the least audible clips. These were, for the most part, clips in which there were loud sound effects or music in the background. I also lowered the volume output for louder than average clips, and raised volume for relatively quieter clips. Fade in/out were also used on most clips to improve overall flow.
I added longer clips on both the beginning and end, to create bookends that went against the rules of the rest of the video, to signify a start stop. I also threw together a title in After Effect: "The Brand Who Lived."
Was able to find the original content easy enough; had to put up with fairly long reencoding times when converting, as well as errors that caused some omission of content from a couple of the films. (Films I later decided I didn't want to use.)
Extracting
As I was extracting the desired clips from the first film, I noticed that the practice was time consuming and extremely unpleasant. Scrubbing through movies of long duration is not very intuitive, and extracting a proper subclip often meant going back over words and sentences several times. Listening for the key words also meant watching the films in a way that doesn't qualify as entertaining, but work.
I also noticed that even with the first film, I was able to extract a large amount of potential content. Because of this, I made the decision early on to only use the content from the first four films. Perhaps at a later date, when I'm feeling especially masochistic, I can expand the project to encompass all eight films.
The extraction process was taking about 1.5-2 times longer than the content duration I was working with, but as I got better at referencing the film transcripts I was able to make subclips for the last film (Harry Potter and the Goblet of Fire) in about the same amount of time it takes to watch.
After extracting "Harry/Potter" from the first four films, I had about 420 subclips that amounted to a 6:20 duration.
Assembling
I decided to arrange the clips from longest duration to shortest duration. This gives the piece an increasing tempo as it goes on, and seems to build tension. It also removes artistic decisions that I would have to make if I placed the clips by choice. Again, I want the system of distillation to be systematic and somewhat controlled. (With as little choice as possible.)
Editing
After creating the order by a predetermined system, I still had to make decisions on what to cut out, as the piece was too long. Purely for flow and aesthetics, I decided to remove the least audible clips. These were, for the most part, clips in which there were loud sound effects or music in the background. I also lowered the volume output for louder than average clips, and raised volume for relatively quieter clips. Fade in/out were also used on most clips to improve overall flow.
I added longer clips on both the beginning and end, to create bookends that went against the rules of the rest of the video, to signify a start stop. I also threw together a title in After Effect: "The Brand Who Lived."
Project 1: Technique
When creating this project, I will be using a decent amount of source material. Each feature film runs about 120-150 minutes. Because I am working with so much material, I need a system to manage what qualifies as usable material, and the best way to get it.
Step 1: Download Content
I don't have any of the films on physical media, so I acquired them through bit torrent. I will be using 720p source videos at a decent quality bit rate. Since the final output is streaming video, I don't need to be too concerned with the original quality of the content.
Step 2: Convert to Editable Content
I used AVS Video Converter (http://www.avs4you.com/AVS-Video-converter.aspx) to reencode the .mkv files to a more editable friendly format, .mp4. I kept the bitrate, resolution and FPS the same, but downgraded the 5.1 sound to stereo, as streaming video sites only support two channel sound.
Step 3: Extract Audio/Video
This is the biggest step in the process. I go through the entire films, creating subclips for each time a character says "Harry," "Potter," or "Harry Potter." I will also cross reference the movie transcripts (http://thepensieve.org/movieframemain.html).
Step 4: Refine/Remove Subclips
Go back through the subclips, refining them to only include the desired audio. This also allows for removal of bad or problematic clips, where the audio can't easily be read out of context.
Step 5: Place in Timeline
This involves placing the refined and "ready" subclips to the timeline. This step is where artistic choice really comes into the project for the first time, as I can greatly alter the final piece by how I decide to place the order of the clips.
Step 6: Refine Timeline
Once the clips can be viewed continuously next to each other, another pass is required to equalize audio and/or soften the attack or fade of the audio. Also allows for experimentation.
Step 1: Download Content
I don't have any of the films on physical media, so I acquired them through bit torrent. I will be using 720p source videos at a decent quality bit rate. Since the final output is streaming video, I don't need to be too concerned with the original quality of the content.
Step 2: Convert to Editable Content
I used AVS Video Converter (http://www.avs4you.com/AVS-Video-converter.aspx) to reencode the .mkv files to a more editable friendly format, .mp4. I kept the bitrate, resolution and FPS the same, but downgraded the 5.1 sound to stereo, as streaming video sites only support two channel sound.
Step 3: Extract Audio/Video
This is the biggest step in the process. I go through the entire films, creating subclips for each time a character says "Harry," "Potter," or "Harry Potter." I will also cross reference the movie transcripts (http://thepensieve.org/movieframemain.html).
Step 4: Refine/Remove Subclips
Go back through the subclips, refining them to only include the desired audio. This also allows for removal of bad or problematic clips, where the audio can't easily be read out of context.
Step 5: Place in Timeline
This involves placing the refined and "ready" subclips to the timeline. This step is where artistic choice really comes into the project for the first time, as I can greatly alter the final piece by how I decide to place the order of the clips.
Step 6: Refine Timeline
Once the clips can be viewed continuously next to each other, another pass is required to equalize audio and/or soften the attack or fade of the audio. Also allows for experimentation.
Subscribe to:
Posts (Atom)