1st Class BSc Music Technology (Hons)

Today I learned that I have been awarded a 1st for my degree and I'm over the moon.

As it has now been marked I am now able to share my dissertation in full, for anybody who is interested please follow the link.
An Investigation Into Stereo Microphone Techniques and Their Applications

Any questions or comments please get in touch.
Thanks for reading.

Tom Rawlinson BSc

Dissertation Complete

Finished. Done. Complete.

As of last Monday my dissertation is over, my final interview seemed to go well and I think the project was well received. I'll know the results around Mid June, and once they've been confirmed I will upload a copy of my dissertation in case anybody is interested in my experiments and conclusions.

The final recordings I created of the Stafford Performing Arts Centre ensembles have now been uploaded to soundcloud so feel free to have a listen, any feedback appreciated.

Dissertation Recordings- An Investigation into Stereo Microphone Techniques by Rawli

CD Cover art for my final CD

I've bought some fancy lightscribe CD-Rs so I can burn fairly professional looking CDs for my finish product. This is my initial idea, it may change before the 3rd May depending if I have time to play around with it any more before then.
Comments welcome

Update

So there are 22 days left until hand-in, a quick update on my progress.
All of my tracks have now been mastered (using K-20 as a reference point, leaving plenty of dynamic range), normalised to make use of all of the available bits when transferred to the 16 bit CD and noise reduction applied to clean up anything left in there from the noisy pre-amps.

The report itself is going well, I have a rough 1st draft pretty much finished. Once I get feedback from my tutor I suspect there will be plenty of redrafting as well as neatening up all the rough edges that are still present- references missing, diagrams that need scanning, audio samples embedded etc.  To get it all done in time I will be spending most days in the library, which has been pretty conducive to a good work ethos so far.

Hopefully I will be finished well before the deadline of the 3rd leaving me time to print and bind well in advance of the queues but no matter what happens, It will be handed in by that date then all I will have left is 1 exam, GradEx and my dissertation presentation and that is it for my university education. Scary, and exciting.

Second Stereo Recording

So last night I completed the second of my final stereo recordings, this time a local string quartet- 2 Violins, Viola and Cello.

AKG C451b's in an ORTF array

Everything with this session ran a lot smoother this time round, having learnt from last weeks mistakes and allowed half an hour to setup before the performers arrived. Also managed to get the second Digidesign 96 interface working this time, so was able to spot mic each of the instruments. I selected AKG C414s for the spot mics due to their high sensitivity, low noise and wide frequency response however due to a shortage of microphones I ended up using an sE4400 for the Cello, basically sE's copy of the 414. Interestingly I hadn't realised that they came as a stereo pair, if I had then perhaps I would have used them for the Blumlein array as the frequency responses of a matched pair should be almost identical.

The recording itself went very well, once again I was impressed by the quality of the performers and the recordings themselves sound impressive for each technique, with noticeable differences between them which will make for interesting comparisons. All that remains to do now is to edit the tracks and apply limited amounts of EQ and reverb to bring the quality of the recordings up to a commercial standard before I conduct my listening tests and write up the results, my discussion, conclusion and recommendations.

Final Recordings

Been very busy since my last post, I will write up in more depth at some point but so far I've conducted one or two more tests in a slightly larger studio which illustrate the stereo effects much more clearly which is good.

Spaced Omni Pair, ORTF and Mid Side Techniques

Last night I completed the first of my final recordings with a local clarinet quartet who were fantastic. I had a few technical problems with only one of the Pro-Tools interfaces seeming to work, so I wasn't able to have spot mics as well as my stereo arrays but the recordings from the arrays sound great. The C414s were clear and quiet while there was a bit of noise from the C451b's. I have, once again, fallen in love with the Blumlein technique, it gives a really clear image which is balanced and tonally pleasing :-)

I will be completing one more recording, with a string quartet, next Wednesday before I conduct listening tests with a small group of people to find out which techniques people prefer on playback.

Continuing Experiment and Research

Since my last entry I have been into the studio to conduct some more experiments.
The first using CLIO, an electrical and acoustic measurement system, I played pink noise from a speaker at a number of different stereo arrays and measured the dBSPL from each microphone in the array for two different locations of the speaker- directly in front and 2 meters off to the side. By comparing the difference between levels for the pair of microphones the ability of each technique to localise sound based on level differences could be assessed. 


The next experiment involved setting up the microphone techniques one at a time and playing a short burst of sound from 4 different locations within the room. These signals were recorded into Pro-Tools and then later by zooming in and using the selector tool to measure the difference in samples between the two microphones within a pair the time difference was calculated. This gave some interesting results showing which techniques use time delays for localisation.


The combination of these two experiments have led me to choose a different room for my final recording, although I would like to use the SSL AWS 900 that comes with the room, it just isn't wide enough to get a large enough stereo field. I've spoken to some local string groups and it seems I will be creating my final recordings at some point in mid march.


My research is continuing, now moving forward to investigate the processes involved with reproducing stereo sound. Today I have been focusing on loudspeakers and the problems of acoustic crosstalk as well as reading a lot about transaural stereo- the process of recreating three dimensional virtual sound sources all around the listener using just two sound sources located infront of the listener. It's very clever but the research indicates it sounds quite unnatural to the listener and tends to cause fatigue pretty quickly. 
Tomorrow I will move onto investigating binaural and headphone stereo, I've already found some fascinating albums recreating Rachmaninov piano concertos recorded in both standard stereo and binaural and the differences between them are striking.  

First Experiments

Blumlein, M-S, ORTF, NOS, XY
not shown- Spaced Pair & Binaural

Spent 4 hours today in the university SSL studio beginning my initial experiments into the different stereo recording methods. I was met with a number of set backs including dilapidated stands which couldn't hold a microphone straight above the base, let alone at the extent of the boom making it very difficult to position the microphones close together without them drooping onto each other. Once they were finally cabled up we'd discover that half of the cables were faulty, the re-wiring requiring the mental agility usually associated with solving chinese puzzles to complete.

Finally signal was flowing all the way through to the AWS, but there it stopped, for some reason that neither I, nor Eddie could understand. Eventually we decided to bypass Pro-Tools and record in Logic instead.

By this point it was getting late in the day and patience was wearing thin, the realisation that these experiments would probably had to be repeated had sunk in however I was determined to get some useable results that at least proved the validity of the different methods. With the help of my friends Mike and Holly, I recorded samples of moving sound sources as well as simultaneous sources to demonstrate the improved intelligibility offered by stereo. 

Listening back to the results confirms a few of the basic assumptions made by the research however it also demonstrated how the originally designed experiments were flawed, with the microphones shadowing each other and interfering with the stereo image. The next set of experiments will be conducted using the acoustic testing device CLIO and only one of the techniques setup at once to control the variables although the recording environment will inevitably compromise the results somewhat.

Progress Report

It's been a while since I've last posted, which isn't to say I haven't been doing work, I just had a lot of other deadlines just before Christmas pushing my dissertation onto the back burner. Now that 2011 is here I've spent the last week putting together my progress report detailing to my tutor exactly what progress I have made, where I'm behind and how my time plan is being adjusted to make up the difference.

Most of the research into microphone techniques is now complete, so I can now move onto researching some methods of reproducing stereo images- over headphones, speaker systems etc. and the processing that goes on to create a coherent image. I will also be investigating the artificial creation of stereo recordings through pan pots, delays and assorted plugins.

This Thursday afternoon will be the first of my experiments into the researched techniques. I will be setting up several techniques simultaneously and then recording sound sources at various points throughout the field, varying the angles of the microphones and the distance from them to gauge the effect this has on the image.
The microphones and techniques I will be investigating are:

Technique	Polar Pattern(s)	Microphones
Co-incident Pair: Blumlien	Figure of 8	AKG C414 x2
Co-incident Pair: XY 90° hypercardioid	Hypercardioid	AKG C1000s x2
Co-incident Pair: XY 180° cardioid	Cardioid	AKG C451 x2
Co-incident Pair: Mid-Side	Figure of 8 Cardioid	AKG C414 Neumann TLM103
Near co-incident Pair: ORTF	Cardioid	AKG C414 x2
Near co-incient Pair: NOS	Cardioid	Matched pair Rode NT1A
Spaced Pair	Omni-directional	Audix TR40 x2
Binaural		Soundman in ear mics
Surround Sound		Holophone H3-D

I will post some pictures and perhaps a few samples soon.