26/10/2015-After hitting upon the idea I might be able to extract an image from the video paintings - to create a physical product which could be sold. I experimented and created a number of A4 photographic prints. Two were particularly striking, these were extracted from the 'impressionist' pieces. The two that weren’t as successful were from Industrial Sunset 2. More experiments are needed.
10/10/2015- Today I did some more experiments with frame rate.
Please check quality settings when viewing this content. Although it's hard to tell because of Youtube's re-compression, you should be able to see the difference in the way the waves move between the difference frame rates. This I think shows that frame rate can be used as aesthetic choice. The next experiment will be with light and frame rate.
5/10/2015-I’ve figured out the difference between the two-pass switch, and the one-pass switch -this is in relation to the MPEG 4 encoder. Although it is slight in Impressionist 2 (Pink 1) - at 1:56 the street light turns on; with the one-pass switch selected the light isn’t as bright or beautiful looking. On the version with the two-pass switch on, the light looks much better. Although it is slight the difference is quite noticeable.
4/10/2015-More experiments with MPEG 4 encoder. After doing more research on the subject, I realized I should have been using the two-pass switch. The MPEG encoder when the two-pass switch is connected, reads the video file before writing the video file. In one-pass mode it just writes the file. After looking at the two files - the first being a one pass and the second being a two pass - I can’t really see much difference. The two pass file is about one hundred megabytes more. More research needs to be done.
Impressionist 2 (Pink 1) was born from the need to experiment. After upgrading my video editing program, the need to test out any new functions was too great. Even in filming I have tried to experiment using a different shutter speed to the 'rules', creating a magic light effect -the light looks like the planet Jupiter. This new piece captures the idea of 'light painting' succinctly. By using the layering technique…the colours and light blend to create an effect that is very new.
2/10/2015-Because the film editing took longer than normal, I left the sound till today. I was able to utilize the full effect of 96 kHz in the initial editing and even in the granular synth: Padshop. Because I was able to use 24 bit stereo files, I wanted to experiment more. For this piece, I sliced out some of the low-end via R-Mix by Roland, and exported the 24bit 96 kHz file for further processing. In Reaper, which is another DAW, I used the varispeed function to slow the file down and pitch-shift it to make it sound lower. This I bounced out and re-imported into Sonar. In Sonar I cut out the sections I wanted to use. These were the cricket and cicada sounds - then I placed a granular delay effect over the top.
This created a great feedback sound near the end…in the perfect spot. I’m amazed at how well the sounds seem to match the vision. Particularly, because I haven’t got the vision in Sonar to align the changes and build-ups. Although I am happy with the final sonic result, I still feel I can do better. I think once I have some sort of controller device to play and perform filter sweeps etc…I will take the sound to another level again.
1/10/2015-After purchasing the upgrade to my editing software: Vegas Video - I was keen to experiment. For a while, I’ve been wanting to film a night piece with the lights around the corner from my house as the focus. I thought the best way to get an interesting piece was to film a single-shot piece just as the sun started to sink…unfortunately I did not get out the door fast enough. This meant I had to revise my idea. I liked how capturing the light as it turned on in Impressionist 1 (Purple 2), but again I did not have time to set the shot up. In the end I set-up and hoped for the best. Like Impressionist 1 (Purple 2), I decided to film a number of single-shots and layer them to create the 'impressionist effect'. I also experimented with shutter speed. Instead of the rule of frame rate doubled - near the end of the shoot, I used a shutter speed of 30. In editing I realized this produced a strange but usable effect. The last shot was pitch black apart from the lights which had ballooned into coloured blobs. They actually looked like planets…very cool. In the new version of Vegas, all the familiar ideas and techniques I had used previously were not as effective. This is because parts of the program had been moved around, meaning I was not as familiar with this version. This new version of Vegas is going to take some getting used to and will require a higher level of understanding of various film, photography and editing concepts. The plus is I can record at 96khz and can edit the sound in Vegas before exporting into Sonar for processing. It will also mean I can experiment further with the sound e.g. varispeed, cutting out particular noises, and emphasizing these.
Purple 2 was born in the twilight moments of awake and sleep. After seeing some of Newcastle’s beautiful street lights, I felt that they would make an excellent subject to 'paint', as they turn on and just as the sun sets. The atmosphere they create reminded me of an impressionist painting and in particular Van Gough. The impressionist idea or how to create an impressionist 'light painting' style came to me in that twilight moment. I thought by filming a single continuous shot and layering…like a painter, would do the trick. Each shot would have to have subtle differences, light and focus, but by the end it morphs into an impressionist style.
28/09/2015-I’ve been wanting to experiment with filming at night - outside under some of the lamps which are littered around Newcastle. After going into the CBD last night, I saw some amazing seventies style round globe lamps. By pointing the camera at these lights I should have enough light to film in high quality - perhaps 200 ISO or 400 ISO. Furthermore, I want to experiment with a continuous shoot…no cuts. Five minutes should be enough. I also had an idea, because the lights and the atmosphere reminded me of a Van Gogh style café painting - I was keen to try and replicate this. I think by filming in the same spot, five times and overlaying the film in editing - I should be able to create an 'impressionistic' style of light and sound painting. I will try a couple of techniques, soft-focus and different shutter speeds. Success…the piece Impressionist 1 (Purple 2) turned out very well, although once I got the footage home I discovered that the camera had shifted - the first four sequences were unusable. I filmed this piece at 24fps, which allowed me to keep the ISO between 100 and 400…acceptable and I recorded the audio at 96 kHz. The area outside the Lock-Up in the CBD of Newcastle provided an excellent amphitheater. Whilst I am unsure if it was the higher sample rate or the lack of wind the sound was the best I’ve had to work with so far. It required no real 'doctoring'. In video editing, as I had worked out in my head…by layering a total of nine sequences over one another in a staggered fashion, and using different fades - I was able to create an impressionist 'film painting' style. The only issue I need to work out for the next piece is the sound…whether as I have done on this piece and let the sound double up and fall where it lands, or choose one finished audio clip and stick with it…I will have to discuss this with my supervisor.
21/09/2015-Today on my way to Uni, as is my want…I decided to film another piece…again on the subject of spring. This gives my new film painting practice a wider gamut of subjects - so far I have sunsets, which could be described as narrative, clouds and spring could be described as figurative, and I’ve even experimented with abstraction. I’ve definitely found the best formula in regards to shooting…this is three minute samples which allows enough time to capture interesting footage. The three-minute sample also allows me to edit any wind or other unwanted by-products out, and glue enough footage together for one minute edits - this seems to be a good pattern. One minute allows the viewer to immerse themselves in the footage without becoming bored…however, without becoming too anal for instance - in Yellow Spring 2 there is a beautiful shot of the sun glowing through the tree, I’ve allowed this to go for just on two minutes. It’s important to allow the pieces to create their own structure…if a shot requires only one minute, leave it, but if the shot requires more…again leave it. I will get around to filming a piece where there are no edits soon. Also in today’s shoot, I picked up a strange voice on the audio recording…after some investigating I can only speculate that it is a ghost. Please see sidebar for video: Yellow Spring 2
18/08/2015-Some major breakthroughs last night in editing the sound and in rendering the video pieces. After almost two years of experimenting with RMix, which is a stereo processing plug-in by Roland, I’ve managed to make it work. This plug-in allows the user to clean wind, hiss and breathing from stereo audio files.
And finally using Sonar’s generic tape emulator set to 7.5ips, which dips the top-end of the stereo file, and allowed me to create what I consider a finished master. Although I would prefer to record all the audio at 96khz, and edit at 32 bit fp, which I can once I upgrade to the full version of Vegas Video - the results to my ears are nothing but spectacular. In mastering, now I’ve worked out how to set the compressor correctly - the overall audio level is much more even and this can be heard even on my laptop speakers. Finally, I have been experimenting with a stereo winding plug-in which anchors the right and left channel, and helps clearly define both channels.
I believe I’ve finally figured out the MP4 codec, which compresses the final video into a 'manageable' file for Youtube and general viewing. Perhaps, it should have been obvious, the MP4 codec and compression process was similar to the MP3 codec and compression process - it was because they use different languages to explain the same process. There are also extra processes to take into consideration. What I’ve discovered is by increasing the variable and average bit rate is amazing - the light is denser, and the colours shine. There is also less visible aliasing. Just like an MP3 codec the amount of 'important' data, which the codec believes can be kept and removed is interesting. Until you actually experiment you just don’t know what is being shaved off. Of course the trade-off is the size of the file, which for Abstract Sunset 1 is a file that is almost three gigabytes.
When comparing the two versions for differences go to 4.06, 6.25, 8.10 mins. Although the differences are slight, they are noticeable. Because this project is about High Definition and quality...these subtle differences are important. Please also be aware, Youtube has re-compressed these videos, so it's more noticeable on the actual MP4 files.
15/09/2015-Today I wanted to experiment and shoot in the midday sun. As I expand my knowledge about visual arts concepts, it becomes very important to be able to create pieces which reflect that knowledge. I’ve also started to experiment more with abstraction rather than 'film narrative'. This will be important if I am going to make an argument about this new style of 'film painting'. I also realized today, I should have been working on the sound first…or I should have been cleaning the sound up before editing. This allows me to clean up all the unwanted wind and breathing artefacts that have been annoying me - possibly ruining my work! Success after cleaning up the audio first, I’ve managed to create a much better sounding stereo master. This was achieved by using Roland’s RMix plug-in, which is a stereo effects processer that can clean wind and other unwanted sounds from stereo audio files. I’ve also worked out I wasn’t quite using the mastering compressor properly. After realizing I could set the attack and release time using transients and calculating the correct settings using the time line ruler…I was able to raise the volume closer to industry standard. I’ve also worked out a cleaner way to jump edit shots-where I have cut the wind from. Please see sidebar for video: Yellow Spring
11/09/2015-I raced out the door today after seeing some magnificent cloud formations. Unfortunately, this was after I had seen these clouds fifteen minutes earlier. I need to stop being so anal and drop everything I am doing when this occurs. Even though I got some nice footage it could have been better. I wanted to shoot this piece high up and believed the rail crossing near my house would be a good locale…unfortunately I forgot about the high fences. After my last meeting with supervisor, where he stated he would like a piece with nothing but clouds, I decided to oblige - although I did add a couple of tiny framing shots. By sheer coincidence some RAF planes shot overhead, and I managed to capture this both sonically and visually. In editing, I did the previous evenings Purple Haze 1 first. This was a five shot sequence using high angle shots to obscure the location - this was outside the town hall in the CBD of Newcastle. I think there is a difference in 24fps and 30fps - you can see it in some of the detail. More experiments are needed though. There is also the water fountain shot where I experimented with different shutter speeds, and again although it’s subtle there is some difference in the overall aesthetic. More experiments are needed. Although I keep saying I have mastered the granular process…I keep discovering new ideas and techniques, I tend to have the manual open whilst I sculpt the sound at the moment.
Please see sidebar for videos: Purple Haze 1 and Clouds 6
10/09/2015-Today I wanted to experiment with some new ideas that have been forming. First I wanted to obscure the location where I was filming...only giving sonic clues. I achieved this by using low-down shots or by pointing the camera at another object, and by composition i.e. something in the foreground, middle ground and background. I also wanted to experiment with filming just before dark - this time at 5:45pm or thereabouts. This was because I wanted to experiment with pinks and purples, which this time of the evening yields. I also used 24fps on this piece to test my hypothesis of the picture being slightly fuzzy, but the fps helps enhance the colours of this time of evening. Unfortunately, I forgot to change the shutter speed so I don’t know how useful this data will be. Finally, on the second piece, I’ve tried to use four or five shots no more to try and keep my ideas under control. Please see sidebar for video: Abstract Industrial Sunset 1
5/09/2015-After a few days rest, and because there were some excellent sky conditions i.e. clouds, greys and blues - I decided to do some filming experiments. I’m starting to understand…I think, the idea of frames per second and when to use certain frames per second. My theory, which still needs to be tested is - the more light, the higher fps, less light, lower fps. It all has to do with the shutter speed and double the frame rate formula - this is if I have a frame rate of 24fps, I need to double the shutter speed to 50. The lower the shutter speed, the more light that gets let into the sensor. However, this also means the picture quality falls slightly, so when choosing the frame rate, other considerations need to be taken into account. For example, using a lower frame rate is possibly good for orange sunsets and higher frame rates are good for blues. Again this needs to be tested properly. Please see sidebar for video: Industry Long Shot 1
31/08/2015-I’ve been wanting to do a piece down on the Newcastle foreshore - today this is exactly what I did. The cloud formations were reasonable, and I’ve added another colour to the palate…sea blue. Although I’ve used this colour before, but not as a focus. It works really well with the purples and blues of the setting sky-with the buildings in the background used as abstract shapes. I don’t know how, but some of the audio got mangled and I had to botch two or three shots to make the piece. It’s hard to tell where this has happened, but I can. I’ve also got to mix the audio better…it’s been hard to hear the granular layer in the last couple of pieces.
30/08/2015-Decided to try a new subject - this time the road at sunset. I chose this particular subject, because I like the lines of symmetry…you can peer down these lines and see an infinite amount of different things…cars, stop lights, people and of course the sunset. Shooting this type of subject can only really be done at night, and on a Sunday because of the amount of traffic. I want a variety of sounds and if I shot this on a weekday, there would probably be nothing but uninterrupted traffic noise-so hardly interesting. I used the NTFS 29FPS standard, although this time I did not achieve the same silkiness as Industrial Sunset 2. More experiments and reading. There wasn’t much cloud cover, so there wasn’t a huge amount of variety in regards to shots, but the greens of the trees with the burnt orange makes up for the lack of clouds. I’ve nailed editing, and am really quick, even whilst putting piece together I am now able to clean-up the stereo master at the same time. This cuts editing time down and allows me to get to the granular process quicker-this is much more fun than the editing I must admit. I’ve almost mastered the granular process, which allows me more room to experiment with both the granular process, and the actual blending of all three layers of audio. Although I’ve noticed if I keep the granular process too low in the mix, it gets washed out in the final video rendering process - more reading and experiments. I also started filming a full-moon piece…I’ve yet to check the footage but I did not quite get enough. This is because I thought I was being cased by a very dodgy looking person. I had the front door open, and was filming the moon from this angle. I will keep the footage and try again next full moon. Please see sidebar for video: Street Sunset 1
29/08/2015-After checking footage from yesterdays’ walk to uni, to make sure the sound was corrupted…it turns out it wasn’t. Sometime software does have glitches, however, when I put the industrial footage together, it turned out there was nothing of real value and use. Although I finished a piece, I will not render because it is plain ugly.
28/08/2015-Took one of my walks to uni and decided to film some stuff on the way. First the canal near the school, again it wasn’t working…I’m not sure if it’s the light or it just has no magic. I’ve seen beautiful cloud formations, and because it has an abundance of bird life, I feel it should yield fantastic results but at this point…nothing. I will try one more time. Next I wanted to film the industrial complex next to uni, it looks magnificent. I feel it doesn’t matter what sort of conditions e.g. light, colour…whatever I shoot it will look great. On my walk I came up with the idea to shoot a long shot from the rail crossing near the Tafe. It’s high up and you have a great 360-degree view of Newcastle. The industrial complex just peaks out from this point. I think this is another 'narrative' I can use, just as I use sunlight and sunset to give my really good pieces 'temporal' placement…the long shot can do a similar thing. When I got home I wasn’t as enthused about the film footage I had captured, and I gave up trying to edit the canal footage…banal footage. I was really disappointed when I viewed the industrial complex footage…the audio seems to have glitched…making me look a bit foolish in regards to the comments I’ve been making about this form of art and computing.
26/08/2015-Today I walked around a little bit to try and find a new place to shoot. I wanted to be near the water, and near where I shot Industrial Sunset 1. I ended up about three hundred meters from that spot. I continued my experiments with frame rate - today I used 29fps or the NTSC standard. I feel that there is definitely a quality increase, but it also might be the conversion rate in Vegas. Conversion rate is something I am continuing to experiment with and more experiments are needed. I will probably go back and render a couple of the older pieces to see if it’s the conversion or the frame rate. Now that I am using sunset as a 'temporal' story telling device, editing is much easier. Shoot a shot, and then take those shots in sequential order - piece together. It’s harder than that. The sound and vision needs some tidying up, and this includes which parts of the segments to cut, and which parts of the sound need further corrective EQ’ing. Sometimes it’s hard to know what bits to cut out too…shot-wise. I’ve feel I’ve discovered the best way to create the sound level. I need three layers…one which is clean, the other two are processed in the granular synthesizer. On the third layer within the granular synth, by not opening up the sample via the speed and time dials - I can create musical notes. All these new developments excite me and only push me further and harder. Please see sidebar for video: Industrial Sunset 2
25/08/2015-Managed to get out the door today…on time, cloud cover wasn’t great. This was even after a thunderstorm earlier in the day. After experimenting with trees, clouds and industry - today I decided to experiment with the storm water canals that wind their way through Newcastle. They have excellent lines of symmetry; this is why I thought they would make an excellent subject. I live near a school with a canal that runs past it, there is a lot of bird activity that I thought would be good for movement and sound. Unfortunately, there was not a lot of great light, but I can cobble together a piece from the footage shot. This piece can be used as a demonstration to show why capturing light in a particular way is important. The wind was an issue, and I forgot to set the shutter speed correctly for the first couple of shots. As I stated in my previous diary entry, today I wanted to experiment with frame rate again. For this piece I used nothing but 25fps or the PAL standard, however, this experiment I am running with frame rate is not going to yield the data I need to make a conclusion, because Vegas Video or the lite version only allows me to render up to 30fps. After viewing the footage, I pieced it together, although there are a number of shots that are 'beautiful' - it’s too dull in others. One of the ideas I am using is the sunset, which creates a temporal time frame for the 'paintings'. The audio, whilst has moments of beauty, I mixed the granular layer too low.
24/08/2015-Tried filming in the previous location for Industrial Sunset 1. Raced out the door, but because there had been a thunderstorm earlier in the day…the clouds returned and I was unable to capture enough footage for a piece.