Editing

In the last week of term we filmed the shots of the video in the woods (see last post) and have attempted to edit them to the audio. (It is strange how I consider the track to be audio rather than 'music') It took us an entire day to 'log and capture' the footage because the computer was being updated, (this time it WAS the technology that held us up, in comparison to when we seemed to waste/spend alot of time trying to find our around Logic. I am still considering whether Carla and I (especially I, having experience with that type of software) are plain stupid or the Logic interface indeed does suck (And it comes with a manual!?)). The following day we were to start editing, unfortuneatly poor Carla was ill but I intended to make a start, after sorting out things like ISP forms and cups of tea I made a start.

Naturally I dived straight in to editing my favourite parts (the hovering legs) then moved around to the sections surrounding the verses. I was hesitant to start editing the dancing sections as I quessed that Carla would have a more valuable aesthetic judgment to editing the dance than me with my mere sound technician background.

The middle 8 shows us in the grandmothers bedroom with wolfy. I am looking forward to seeing what effects (juddery and flickery) would work with this as, (1) the lighting (despite the candles) is bland and, (2) the one camera angle is also bland. The actual scene, beyond these aesthetics, I like- It is pointless and strange, we have no purpose in crawling around the wolf bed and doing strange actions- not even a mischevious one, which is what is expected from the bemused but determined look on our faces (this was not intented when filming, we just found what we were doing amusing but were determined to finish the scene without laughing). The audio just after the middle 8 appears strange when listening back- probably because it just sits there and doesn't modulate or suggest that anything is about to happen (because of this is it a section on its own?) - the middle 8 finishes almost randomly (to no rhythm) and this little section could be said to finish when the sound that increases in frequency comes in. This particular sound we have used in order to link sections together- using the modulating frequency quality of the sounds to suggest modulation from one section to another? (this would be an unconcious judgment and the reason it was chosen was because it sounded good/worked with our intentions). Back to the little static section between this sound and the middle 8, I observed that when editing footage to it there were tiny points that seemed to fit cuts and certain shots that worked with particular sounds (it is my experience that these observations are only made (1) by accident (2) intuition or (3) experience- in editing a previous project or seeing it done in other works) One example I will make is the moving of mine and carla heads when sat together on the bed with the pebbles- from to right the to left, from a previous project I knew that if I put this movement on top of a certain sound that had similar movement (pitch/timbre to time) the viewer would assume one was because of the other/linked. This 'worked' because the sound and image were harmonised. In my opinion, attention to the exact point of cut and the movement of the image to the sound needs further attention in the final edit. I feel that the video looks amateurish because of this disjointed feeling it has- another suggestion I would make to help this problem would be to change the aural landscape between the grandmothers bedroom and the woods, however as it is meant to be a pop video rather than a film this is not appropriate.

The editing I did solo I felt was uninspired , there where a few reasons for this-

• because there were gaps in the timeline there was no sense of continuity I failed to see the 'whole picture' I found that when the sound carried on but the image went to green frustrating
• The aesthetics of the footage I was not pleased with, therefore it was frustrating to edit
• Carla was not there and I hadn't the confidence to make a final edit before she saw it.

The scenes in the wood were dark, the light from the torches produced the brown of the tree trunks and the earth, the reddy orange of our cloaks and the blackness of our surroundings. The result was indeed mysterious and sinister and this completely contrasted to any smiling, like when we leaped into shot after the hovering - smiling looked really bizarre and I liked that. Going back to that shot of us jumping- interestingly it was incredibly important to me that we jumped on the beat, for reasons I do not know! I don't think Carla approved but I had to insist.

The next day Carla was feeling better and we edited the dancing sections. We had alot to choose from in the first verse, we disagreed sometimes and by the end I was completely fed up, mainly because the shots weren't as good as we'd hoped and the experience of filming them was awful in the first place. There were as few shots where there was a huge shadow in the background and we were pleased with them- the shadows created atmosphere and looked effective.

This is only a short review of the editing and I intend to have another (objective) look at the film after the break.

Rushes

Carla and I have completed our filming apart from one shot and are now at the editing stage. The filming went ok- we had a few problems, mainly with lighting. If I was to do the shoots again I would involve more preparation. Like visiting the location and taking a few snaps to get a better idea of whats involved- maybe doing a few test shots at the location, at the correct time of day and in costume just to check the shot worked and to experiment with different lighting effects.




We did one shot at carlas place, and used her bedroom to be the set of the grandmothers house. We had huge lamps to light the room but they were too bright and we wanted it quite dark and shadowy- instead we got all of the candles that carla owns and lit them to form shadows on the bed. We decided to use a trick carla had seen used by david lynch were we would film ourselves doing a scene backwards then reverse it in post-production so that we looked abit strange, we also did quite alot of stop-frame animation, which took hours but the end result worked really well- very juddermanesc!

The second shoot was set in the woods, Carla knew of some woods near her place so we collected all our equipment and costumes to set off. We intended to use the lamps again to light the shoot but were short of a generator, I mustered all of the torches I could find and took about 4 in of which only two would work!!!! Disaster! We set the iris on the camera to the lowest value (F1.6), the shutter speed to as long as we could get away with and the gain to the highest possible setting, anything to let more light in! none the less it was really too dark- the woods were pitch black and the torches formed rings where they shone. In addition our CD player decided to give out in the freezing cold so we had to quess the timings when dancing and mouthing the lyrics. Better preparation should have been done really in the case of the lights and a back up for the CD player should have been brought along. Lesson for next time. The rushes weren't so bad considering and they looked suitably silly as intended.



There is one shot left, which carla is intending to shoot at sunset in the woods so there is enough light to see the landscape (hopefully when there is a full moon! 24th dec), (will form the initial shot which establishes where the film/video is set).

Tomorrow we are going to have a go at editing- we have to choose which takes are the best and where suitable places to cut are. Once the film is edited we still need master the audio and we will perhaps experiment with a performance element to the project (if there is time) in order to do in Jonathons performance with technology elective.

Filming

Now that there is a structure to the pop song we have decided it is time to start the video. We have been thinking about the video whilst composing the song and thought that we would use the Little red riding hood theme. Yesterday we drew up storyboards, most of the shots we had already considered. We had a few influences from the Metz Judderman video and the old Nosferatu film. We also watched a few Svankmajer films and got a few ideas. Carla suggested we do some animation type effects like the Peter Gabriel Sledgehammer video with our clothes. We intend to make the video abit silly using these different techniques and influences. We will film next week so this we are making preparation to our costumes and the set. We have bought red pashminas and red and black stripy socks, we have also planned objects in the 'grandmothers room'. We visited an exhibit at the mac called 'folk art and fairy tales' were we had a few ideas, so today we made paper mache toadstools and carla has an old pillow case that we will make into wolf ears. We also had a chat with Nick about post production effects and intend to try and replicate the handcranked camera effect and masking effects used in the Judderman video this way. We are also interested in fiddling with the colours like dulling everything and highlighting the reds and/or green.  

Japanese Pop Song?

Today Carla and I placed our vocals onto the mix, they sound quite good with us both singing together, very popy! There are still quite a few things we need to do to complete the song.

• Level out vocals
• Create outro
• Pull in more sounds to decorate the second half
• Include a shamisen for a more japanese (and musical) feel

We do have a complete structure to the song which shows we are progressing (if slowly), these things always take twice as long as you think they will. I need to find out how to use general MIDI instruments in ProTools.

The end is near!

Carla and I have made big progress on our composition. We have pulled some more sounds into the mix and extended the rhythm section. It sounds impressive! There are a few quality issues I'm still worried about but feel the creative process is flowing atm so will sort them out later! Today Carla and I recorded the vocals together so they are on tone and in time so the next step is to place them onto the composition to form a vocal line. We will do this on monday and hopefully create a final structure. Hopefully we will start thinking about the filming at least by wednesday.

A composition is formed! (and Logic really does suck)

Carla and I have made some process on the composition!! At last, it seemed that the logic interface really held us back where we were trying to solve problems for hours. Incredibly frustrating! I was VERY pleased when we pulled everything into ProTools. WOHOO! I will make the observation that the logic visual interface was better to compose rhythm to, and we have spent the last few days dragging our location sounds into logic and chopping them up and placing them to create a beat, it sounds quite interesting (if not abit cheesy) We have had to place a rather destructive high pass filter on the samples because of the ambient noise, we could get rid of the noise using Sparks TC Denoise, which I intend to investigate. We bounced the rhythm and pulled it into a ProTools session (HURRAH!) and pulled in all the sounds we've been working on and put it together. I had made a really long sample, containing a rather sinister sounding low frequency rumble, we felt it set the scene rather well and glued the rest together. I have also given the vocal line to a musician friend and she transcribed the vocals and suggested a few chords that would work well. The result was really cheesy, I am really interested in using the MIDI instruments to play around with these chords but I haven't used it for years and can't seem to get it to work. We intend to rerecord the vocals so they fall on a beat (therefore changing the timings of the dance) and we worked on this today. We will rerecord so the notes are on tone aswell in order to place chords and a bass line on the composition. I am to create a good bass sound by manipulating our samples for tomorrow in soundhack or spear. We have also discussed the video a few times and intend to film after have have our composition completed. I am not sure of this arrangement having researched visual and sound in my undergrad, may suggest we start thinking about locations and ideas in some sort of order before we complete the song. However it will be a pop song and I suppose the video is always made after the song is.

The Next Step

The next step was to think about the backing music to our vocal line. Carla and I discussed the idea of composing something musical but neither of us had much experience in writing japanese pop music. It could possibly been interesting but we decided that we were both more interested in making the pop song more experimental and abstract which works better with our theme. We decided that the composition would include the samples we recorded (most of which we didn't use) to create the backing by altering them to sound as we wished. We were interested in creating the atmosphere of the woods from litte red riding hoods story, so we messed around with the sounds and have so far come up with owls, mouse and bird sound an excellent growl sound from carla and a few other bit and pieces. We decided we wanted some sort of rhythm to the piece (something for the dancing to look for carla suggested), so we thought of footsteps as a kick and leaves on trees to be cymbols. We went out with an Edirol portable recorder to record some sounds in the park. We spent the rest of the week, editing this sounds (using high pass filters to eliminate traffic noise etc). Next week, we will carry on composing sounds and discuss the composition as a whole.

New Post

Carla and I recorded our vocal line by recording it as MIDI data into Logic and using Kontakt as a plug-in to bounce the samples. It was still incredibly difficult to get the samples to trigger from the same movement uniformly. With alot of practise this could be easier, however the following problems are apparent:

• The samples are triggered on a first in first out basis, so if an arm movement accidently triggers the shoulder joint instead of the elbow the wrong sample will be play and the intended sample will be discarded. There is no way around this because we wanted the samples to play one after the other, and so turned note stealing off.
• The rhythm of the dance was constricted because one had to wait for each sample to finish before performing the next movement to trigger the next sample.

We recorded the nearest match to what we intended. We are still deciding whether we should rerecord this vocal line ourselves to make it sound less disjointed.

Progress

Over the last few days and during last week, Carla and I have been tackling getting control from the MIDI suit interface through movement. We rerecorded our samples to more of a melodic tone and listened to the affect of this. We then made the following judgements:

• We would play and control where each sample was mapped using Kontakt as a plug in in Max/MSP as the interface was easy to use and much more intuitive than both Max and Logic. (We overcame the problem of sample stealing by using the script editor (tip from Jonathon)).
• We have decreased the number of sample to use significantly in order to obtain more control over what movement triggers what samples.
• We then studied what movements triggered what range of MIDI values (we are using less than half our original amount of values) in order to map only one sample to each movements range (which was decreased from eight to four)
• We have observed there was less chance of a sample being triggered unintentionally (when meant to be standing still or through other movements) by not mapping the sample over the full range
• We discarded information coming from wrist twist because we observed that it was almost impossible to move the wrist without both twisting and tilting and activating the wrong sample
• Some of the values we were unsure what movement triggered them so we disregarded this aswell.

So at present we are only using 12 samples out of 128! However, in hindsight it would have been next to impossible to control each samples using the MIDI suit using this process and considering the length of the audio being used and that they were to be played linearly.

Over the week we have been thinking about Para Para dancing, a Japanese fad that the Ganguro girls often do. It involves a sequence of moves that are danced together in time to the beat. (We both thought it was absurd because the over the top way the girls dress is to create individuality but because they all do this it they really are just conforming anyway, doing a dance sequence simultaneously also does not promote individuality). Carla and I discussed contemporary dance and decided if we were to dance in a comtemporary style the outcome would be quite a confused one, in the same way as the ganguro girls style and para para dancing. Carla choreographed a short contemporary dance to use to do when in the suit in order to activate the samples. This included a few ideas from the story our samples were from.

We are interesting in making an experimental pop song and video. We have a mulitude of recordings now and we plan to abstract them far enough so to compose a piece of electroacoustic music to form the backing track to our vocal line. This is certainly very interesting to me and I showed Carla a few freeware programs to play with at home (SoundHack and SPEAR) we decided we would try and compose something individually and then bringing our ideas together to compose a completed piece. I am interested in the notion of displaying a transformation of sound or emulating the way it is a fad for Japanese girls to radically change image every few months using sounds. I intend to utilise either Simon Emmersons theory of 'mimetic discourse' or the use of musical motif to portray this.

Tomorrow we will record our vocal line by doing our dance in the suit in order to trigger the samples.

Alteration

On Wednesday Carla and I got our samples to playback using the suit successfully. The resulting sound was really cool but not what we wanted as the vocal line to our song. Mainly because of the lack of melodic range in the samples. We had a go at transposing the samples to create different pitches to each of the samples but the result didn't sound natural. After having a chat with Lamberto we decided to record the words in the samples but to different tones so that if they were played in succession they would form a melody. The interesting thing about it was that we could use the chopped up words or sometimes reversed words and sing it phonetically which would enhance the result with respect to our idea of singing a foreign language.

In our Performance with Technology session today, Jonathon mentioned that we could do a something similar but in a live performance. Where we would have a singer singing into a mic and the words could be played back jumbled up through speakers.

In relation to capturing our melody next week using the rerecorded samples and the MIDI suit, the combination of movements could either be random, or be intended. I personally think that the MIDI suit should be used in the same why as an instrument, so we spend a bit of time familiarising ourselves with what movements activate what samples and make an artistic decision as to what combination of movements should be used to form the melody- and also what of the question about the piece being played live? should probably be discussed.

Logic Sucks

Today Carla and I imported our samples into Logic, it was not easy and I'm finding logic increasingly unintuitive as it gets more advanced. We were shown how to map our samples onto logics sampler so all 128 span across the keyboard interface. We had a few problems with the pitch being altered but managed to sort them out successfully. We had a play around with some effects, however, we decided that the samples were far too small to apply effects individually to as we predicted the end result wouldn't run lyrically if the samples weren't clean. To solve the problem of the samples playing over the top of each other the sampler was set to MONO. Tomorrow we will set up the MIDI suit to input MIDI data into the sampler and attempt to record a vocal line into logic. Carla and I thought about incorporating a Para Para dance routine to generate the information from the MIDI suit.

I forsee that the vocal line could sound like a foreign language, however, perhaps we might rerecord it singing to a melody. This would solve the problem of different voices singing different words in the vocal line (which could sound odd) and the result would sound more 'musical'.

This is the first blog for an entire week. (Broken Computer) I will try and summarise!

On Tuesday Carla and I met at the VRU lab to discuss our collaborative project for the first time, after going through a few things about the assessment with Greg we wanted to get a few ideas down so we could make a start. Carla and I were both keen to get started right away as it is quite impossible to predict what direction a project of this sort would go in, and we thought it would be best if we did much experimentation and exploration within the time we had.

Our ideas started from Gregory suggesting we used Google to search a random word and see what came up. I am going to try and make a mind map to represent these ideas next time I am in the VRU lab and post it here to give a clearer picture how our ideas were developed.

Presently I have made the observation that our ideas have developed further since we started thinking about, and experimenting with, the technology. I am interested in reading into projects of a similar kind and how they were developed, mainly for inspiration. Greg suggests in the reading list: Digital Performance by Steve Dixon.

On Wednesday Carla and I met again at the VRU lab to play around with the motion capture suit. Carla had already worked with it previously so I had a go and I think its brilliant!! The information was in MIDI which could be very useful if we use any audio software like Logic. (of which I am almost familiar with)! We intend to use the MIDI data outputted by the suit to control audio samples etc. To program the information we looked at Modul8 software, however, we didn't find it very intuitive to use and its potential lies mostly with visual programming. We then had a chat with Jonathon and Mike who suggested we looked at using the sampler within Logic or Max/MSP. (Max Session with Jonathon today was very useful and I feel there is a lot we could do with Max for this project once we grasp how to use it properly!)

The audio material we want to use would form a new or foreign language, this idea stems from making a Japanese pop song and pop video. As neither of us can speak in Japanese we thought it would be interesting to use a 'language of our own' using words from a story written by Carla about Little Red Riding Hood (we felt that the story tied in morally with our initial Ganguro girls subject (see mind map (ytbp - yet to be posted!)))

On Thursday we met at the studio to record our material. Carla wrote out the story for me and to begin with we chose what words or series of words represented the story or sounded good phonetically and wrote them down to read in the recording. The recording ran rather smoothly and we got four takes of our words and phrases. Once read through normally and secondly with some crazy articulation (highly amusing!) from both of us. The second take was in order to give the future arrangement of the words and phrases more colour and sound more lyrical- like a Japanese pop song!

On Friday Carla and I aimed to edit the audio material into 128 samples. To do this we used Protools to select certain parts of the audio file and bounced down smaller files. We aimed to select the most interesting combinations of words and strange or alien sounding phonetics from the material. We even reversed a few phrases to create new words. We intend to use these samples as the lyrics for our Japanese pop song, the suit I mentioned towards the beginning of this post would control what order the samples are played - which we will aim to set up on monday. I think it would also be interesting to be able to control the speed of which the samples are played and the rate they were played back, because I can see a few problems if the samples were played back too fast the result wouldn't sound like dialogue. However, I am excited about the resulting sound! Carla expressed an interest in using sounds effects on the audio which could make it even more interesting. I certainly like the idea of being able to mould our samples into something that is likely to be unpredictable because of the nature of the selection from the information produced from the suit. Carla liked the idea of using the movement used to produce the arrangement in a performance that could appear in the video.

As yet we haven't decided whether we want to have a go at composing the backing track for our pop song ourselves or collaboration with someone else. (Carla mentioned she had a friend who might be interested in doing this).

Introduction

First post for my MA in Digital Arts in Performance. The Blog is intended to be used to keep a record of my progress on the programme.

Progress so far: Set up Blog! I have also set up accounts with googles docs and flickr to share photos and documents online.