My name is James. I have a keen interest in the media. My main areas of expertise is in the practical side of media, having made several Machinima films of my own and am currently active in the field of web design with experience in social media.
This blog will chart my progress within the AS media course.
Evaluative Prezi
Having been introduced to the wierd vectoral world of Prezi I have made my evaluation in it. It reflects upon the issues raised and how we stuck to the main principles of our genre and sums up nicely the practical element of this course.
After some market testing we found that our overly elaborate title sequence ruined the subtle effect we had over the rest of the movie. I have therefore cut it out and replaced it with a simple plain-text Title matching the existing credits.
You can see the ditched titles here - minus cutting:
I've spent the past three weeks now constructing the film from the raw footage. This has gone from major cuts and reshuffles to shaving the odd 1-2 seconds off here and there. So I think it's best I explain the construction of the production - through the help of a handy screenshot:
Layer 1 This is the main footage layer. On it are all the scenes from the actual chase portion of our filming. The little puzzle piece icon shows that I have put an effect on. The effects are layered upon the footage to give the desired effect. Helpfully I have saved them as presets which means I need not go through and tweak each one, one by one.
This image shows the raw footage without any additions.
This is the processed footage, showing the heightened contrast, black and white and the de-interlacing (important for the scenes of fast movement). Which are visible on the effects window below:
Layer 2 This is the diegetic audio from the footage on layer one. This allows me to adjust globally the volume for this layer in relation to the others and as it is all from the same source, it stays pretty much constant. In the image you can see it is set to be a bit quieter as we want to hear the voices over the large pattering footsteps as they run. It also helps mask out the rattle of the camera tripod at fast moving scenes (picture 3 teenagers chasing 2 middle aged men with a camera + tripod).
Layer 3 This is an interesting layer. As you can see it seems to be a large green block, this is infact chroma-keyed out and leaves the rain left. The rain was generated in a Special Effects program called "Particle Illusion". The line seen about a quarter of the way up from the bottom is the opacity, as I just wanted the rain to look natural without being overlayed - this lessened opacity makes it far less obvious while still being visible.
Layer 4 This is the replacement audio from the femme-fatale. When we market tested our production to an audience, the feedback we recieved was that the femme fatale's original voice did not seem as authoritive as the character played. This was no fault of hers, simply her voice was not right. We aquired the help of an older woman to give us a louder, more bossy voice and had to sync it over the top. This was relatively straight forward. It involved looking at the wave patterns and matching them:
As you can see the new voice has a much larger amplitude.
Layer 5 This is the old audio. The challenge with the over-dubbed voice is making it sound diegetic and not recorded separately. I managed this by splitting the original audio and looping sections of nothing but the rain sound over the top. This is why you can see many repeating sections.
Layer 6 This is the footage for the flashback scenes. It is seperate from the main track as it allows more control over it. The effects needed here were different - in order to give it the appearance of being in another time, and place.
The dynamic noise reduction created an unexpectedly wanted effect. Which is the subtle frame blending which matched with the echoing voices put in gives the sense that it is in a memory - the past.
Layers 7-12 These make up the non-diegetic audio. The music, which begins slow and cautious in a minor key before slowing building, through the use of layer 9's heartbeat-esque rhythm which gets fast and the rising creak on layer 11. At various points there are sudden extra sounds, for example when 'the goon' appears from seemingly nowhere.
Layer 12 is simply an added soundscape to add more depth to the opening scene.
The other night me and two fellow members of the group went into our locations with a video camera to trial some of the ideas we have discussed, and to find out exactly how dark was too dark for the cameras to work effectively.
I have compiled our footage into one video and my breakdown of each clip is below the video.
Clip 1: We were aiming for a fast run-by the camera to almost exagerate the speed of movement, with the very fast camera tracking. The problem we found with this is the auto-focus is unhappy at such speeds and therefore we would have to take into account either focusing manually or not using the shot.
clip 2: This demonstrates another problem we may face which is that filming in a live environment means you receive a lot of unwanted background noise. Although the ambient sounds of the city are what we are after, loud motor bikes and such will present significant problems especially over the scenes were there is dialogue or extreme quiet.
clip 3: This clip is testing out the idea of having the camera pan alongside the character (represented by Tom) behind some bars. This too will give a greater sense of speed and could also be used to infer inprisonment - IE prison bars etc. The problem with this is the narrow alley which means that manouvering the camera is difficult and it is highly likely that it would get caught on a wall and either mess up the shot or be ripped from our hands.
Clip 4: In this shot we are experimenting with the appearence of 'the goon'. The plan is to have a news paper that will disguise the goon until the page is turned. This we hope will cause a reaction in the audience. This clip I feel is not going to do that due the fact that 'the goon' (represented by Blaine) is too far away and simply looks like a normal member of the public rather than a shocking entrance.
Clip 5: This time we tried it with 'the goon' moving from behind the tree. This created a far more dramatic effect with 'the goon' appearing more out of the blue and with clearer meaning.
Clip 6: This uses a technique used on the Continuity task whereby holding the camera and tripod upside down and then vertically flipping the footage post-shoot gives a ground tracking shot that follows the feet of an individual. One issue was the camera did keep getting kicked.
Clip 7: Practising the ambiguous close-up mentioned in the presentation regarding our ideas and plans. Clips 8 and 9: These were just experiments of different methods of tracking a character.
Clip 10: This was to try and use the location shown in my previous post. The light was fading at this time and produced what I feel is a nice atmosphere with the sunset and the illumniated facade of the city hall. The problems I said about when scouting for locations, involving masses of people did not come true in such great magnitude but unhelpful members of the public did deliberatly decided to try and get in shot. You can also hear an awful lot of background noise of kids skating and shouting. Another issue is getting a smooth movement with the rudimentary tripods, which as seen in the footage was difficult and still resulted in jerky motion.
Clip 11: This time we aimed to combine the previous with a tracking shot. This did come out quite well, especially with the darking sky and backlighting from the market place creating a silloette.
Clip 12: This is trying to show a Point of View shot (POV) from 'the goon' as he looks through the bars. It also gives a location reference from the Street Sign.
Clip 13: This is another upside down flip it around view to try and simulate a bonnet camera of a car as it comes down the road. It can also be used to give a sense of speed as the proximity to the surface of the cobbles gives a nice guage. The only issue is the difficulty in keeeping the camera steady.
Clip 14: This shot is a low angle looking at the lluminated window of a church. As our plan is to create enigma through the use of shadows, light and sound in order to fit the criteria of 'Film Noir'. I think this came out well, however short it may be of a clip. It would also be easy to incorporate into our production.
Clips 15, 16 and 18 (an error in the counting department it would appear): These are my favourite shots from the night of testing. The dark blue sky and orange, old fashioned lamps shining off the cobles along with the canted angle emphasizing the old architecture and oddness of the buildings creates a beautiful scene giving a great feel of distance and of the darkness. The lights also cast odd shadows off the buildings, cobbles and the character.
Clip 19: is another shot of 'the goon's' appearance. This time we tried it when it was darker and one immediate problem was in setting up the camera in the right position again to avoid the movement of 'the goon' to be seen. Although this was an issue the final appearence coupled with the darkness and face obscuring bright backlight created the creepy feeling we were after. It's also good to note that the sound of a clock chiming really did make the scene come to life and we will be sure to include this in our final one.
Clip 20: Another still, short shot using the orange lighting to produce long shadows in the dark. The sloping street also creates a feeling of uneasyness due to the mismatching angles.
Clip 21: Another tracking shot, testing the ability of the camera user to keep the shot still when running. This came out well but communication errors did cause a lot of near misses and full blown accidents involving the runners.
Complete tinkerer, like to learn new techniques and make interesting stuff. I think of myself as creative, both in web-design and in video production. I'm also a keen photographer.
