11 steps to a fine latte

01. Visit a local roaster and pick a flavorful, balanced espresso blend.
03. Buy whole beans that have been roasted two to three days ago, or buy freshly roasted ones and let them sit a bit.
04. Don't take home more beans than you can use up in up two weeks.
05. Have the machine fully heated up.
06. Steam the milk before pulling the espresso - which is an art in itself.
07. Use a high quality burr grinder(1).
08. Grind the right amount of beans at the right setting fresh into the heated portafilter's basket(2).
09. Tamp with enough pressure - around 15 kg.
10. Pull your shot while counting down the seconds (25 to 30 sec for a 30 ml single or 60 ml double shot)
11. Use the knock box, clean the group head and quick rinse the portafilter.

Enjoy :-)

(1) Invest as much into the grinder as into your (semi automatic) coffee machine.
(2) Watch a good barista as he experiments with the grind setting to optimize his shots.

[Many thanks to Mark Prince, the CoffeeGeek (http://coffeegeek.com/) for his wisdom]

[Expanded on in Making a better latte]

ORTF off axis

Yesterday I tracked a choral concert in the Catholic Church St. Sophien in Hamburg (http://www.sankt-sophien.de/). Here are some initial observations about the situation:

- The church is highly reverberant.
- The rows of benches are placed on a raised wooden pedestal that is sensitive to shuffling and kicking.
- There were four children's choirs ranging from about 30 to 80 singers performing in differing configurations.
- The altar area was populated asymmetrically as there is a big pulpit to the right.

The smaller choirs I saw rehearse were located more to the left, so I placed the ORTF-setup (two cardioid Oktava MK-102's modified by Attila Czirja'k) about two meters high, pointing at the middle of the choir when the pedestals were used, near the left hand row of benches and about five meters from the front line.

Why five meters? I assumed that the larger setups would take up about eight meters from left to right, but I couldn't be sure they wouldn't spread out a bit more. According to Eberhard Sengpiel (http://www.sengpielaudio.com/AufnahmebereichWichtigeWerte.pdf) ORTF covers an angle of 96°, so your distance from the source has to be about 53% of the width of the sound stage. Then there was a grand piano (with closed lid) close to the left hand side of the choir. I wanted to be a bit more distant from that instrument to not let it sound too prominent.

Since I wanted to have the option to increase the natural reverberance and potentially producing a quadraphonic mix I taped two ambient pickups (DPA 4060) to the walls of the doorways left and right of the altar, in back of the choir podiums. At this location audience noise is minimal.

During the rehearsals I noticed that the raised row of singers of the larger ensembles did fill the space behind the pulpit. Some solos were also performed on the right hand side, and in one song  (what I believe to have been) a Conga was also placed there. I decided to turn the ORTF-setup to face a little bit to the right of it's originally central focus.

So what?

Direct an ORTF-config off axis only when considering the consequences ;-) You don't move the center of the sound stage in the direction you are turning towards but increase the effect of sound pressure reaching one side of the system while decreasing it for the other side.

Luckily the problem was solvable for my recording. I was able to reduce the left hand side by 3 dB and added 3 dB more of the natural reverb to the right hand side. The balance was restored without compromising the overall quality of the sound stage. Still, when the first sopranoes are placed on the left hand side it is more prominent to start with.

What could have also been a solution?

- I could have tracked closer to the ensembles with an AB-setup. That way the vertical localization would not have been as clearly defined. OTOH I would have picked up more of the reverberance and audience noise.
- I could have used flanking omnidirectional microphones to the left and right of the sound stage, perhaps at the closest columns, to be able to manipulate the balance.

In the end it is highly advisable to have a say regarding the placement of the ensemble to be recorded. If it is symmetrically arranged things are much easier. And if the vocals are balanced in themselves, with (especially) the sopranoes singing in a controlled way, and an adequate presence of lower voices, one approaches optimal conditions.

What's good and what's bad about Stereo?

I enjoy tracking musical events stereophonically and I enjoy listening to well recorded stereophonic recordings. Still, there are certain issues with this approach, over which I will reflect. I'm trying to keep these thoughts simple yet accurate.

1. How to record in stereo?

Even though Eberhard Sengpiel pulls our nose by stating that two times mono is also stereo (http://www.sengpielaudio.com/Unterlagen09.htm ;-) what I am talking about is the art of using an integral system comprised of two microphones to recreate a soundstage for the listener.

The sound that microphone A transforms, thereby translating acoustical into electrical impulses, is mapped to the left loudspeaker, the sound microphone B transforms is mapped to the right loudspeaker.

There might be additional microphones used to

- help set right inadequacies in room acoustics - or microphone placement - capture nuances from various sound sources that would not be discernible otherwise - capture uncorrelated ambience from disjoint positions

[Uncorrelated ambience: sound transduced by two or more microphones, placed not in the direct but in the reverberant sound field, does not change significantly when mixed together. Even though the spectral (frequency-related) content is probably similar there is enough difference in phase to make the various sources dissimilar]

Mixing some of the information from the spot-microphones with the sound obtained through the primary setup (Hauptmikrofon), compensating for the runtime-difference, because the sound arrives earlier at closer microphones than at those farther distant, can increase the perceived clarity of the recording. The uncorrelated ambience can be used to make the stereophonic recording sound more spacious.

2. The advantages of stereophonic recordings

A straightforward assumption is that the less microphones the recordist uses the clearer the resultant sound will be.

There are so many reflections in any acoustic space that, even when carefully planning the setup and working primarily with symmetry axis' and equal distances for the placement of spot microphones, any microphone captures a slightly disjoint representation of all that is going on sonically. Obviously there are some aspects which should be perceived differently at each microphone position, but there are aspects that should be identical or similar, and these are converted with differences in spectral and phase composition.

Well executed stereophonic recordings are able to recreate a soundstage that lets the listener perceive width and depth when played back on only two loudspeakers.

3. The disadvantages of stereophonic recordings

- The vertical and horizontal placement of sound sources, as well as phantom images that appear to be located in the middle of the two loudspeakers, are only accurately perceivable by a listener who is located at the right position relative to the setup.

- The listener is not enveloped by the ambience of the recording space.

- A loudspeaker built to reproduce the full spectrum of audible frequencies is difficult to construct. There are many design issues involved, including the construction of inadvertently phase- manipulating crossover networks, the projective abilities of the loudspeaker at different frequencies, internal resonance etc.

- The farther the loudspeakers are placed from the listener the more the quality of the sound being played back is dependant on the sonic characteristics of the listening space, the one aspect the loudspeaker- developer can't get close to solving - unless he creates a near-field system.

4. Surround sound

- LCR (Left Center Right) setups create a real sound source in the middle of the loudspeakers placed left and right of the listener's position in place of the finnicky phantom image.

- Quadraphonic setups enable the creation of enveloping ambience.

- Having a dedicated, centrally placed subwoofer literally takes (a lot of) the load off the other speakers that don't have to create large amounts of sound pressure by moving large amounts of air.

In my opinion 5.1 Surround is a good mix of the solutions to create an even better listening experience. Certainly 7.1 is even neater, as you are able to create three stereophonic zones before the listener combined with enveloping ambience from the back, but in my opinion 5.1 not only will do for quite some time but there is also the aspect of the quality of the components used - and the issue of optimizing the speaker's placement. Go for a good 5.1 system and enjoy :-)

Starting up - again

I set up a blog in 2006, diddled a bit in 2007, but just was not up to the challenge. You have to be persistent, draft ideas as you go along, publish regularly etc. I did manage to create two podCast-episodes, but basically they talked about the stuff that would be forthcoming, which then didn't. Let's see if I'm off to a better start in 2008 :-)