Comparing Stereo Recorders

As most people who know me well can attest, I hate letting things go to waste. This also goes for ideas. Sound Engineer & Producerwas a great brittish magazine of the 80s and early 90s, circulated all over Europe. What I specially liked about them was their extensive comparative tests, be it consoles, monitors or... tape recorders. Some of these tests were spread over several issues otherwise all the info wouldn't fit. So I thought I'd share with you some of that amazing source of data for the modern engineer.
I would like to give full credit to Simon Croft, editor, and Zenon Schoepe, deputy editor, for bringing us such good quality articles at the time ; most magazines (at least in France) could use the inspiration.
Now here is the article :

THE OPEN REEL TEST SE&P Jan 1989

The men from Studer, Sony, Otari, Mitsubishi and Dolby watch the meters in the open reel test

We sit nine audio professionals in Studio 1 at CTS studios and ask them if they can tell the difference between Studer A820 with Dolby SR, Sony PCM 3402 DASH, Otari MTR20 with Dolby SR, Studer D820X DASH and Mitsubishi X86 Prodigi two track recorders. A difficult test with very interesting results.

Of late, many end users have made statements regarding the relative virtues of analogue with SR, DASH and ProDigi digital formats. Although the people issuing these definitive proclamations are no doubt sure of themselves, it is a fact that many of the opinions expressed are contradictory.

Now it is one thing to extol the virtues of a system in which you have invested - when cash or personal reputation is at stake. It is quite another thing to evaluate the various formats under identical conditions and with no clues as to which format is which. These are exactly the conditions we arranged because we thought it was high time we found out what the ears tell us when the eyes haven't had a chance to make up their mind.

For the evaluation sessions to be meaningful, it was essential that the signal be of the highest quality from start to finish and that the source material be both recognisable and varied in terms of timbre and dynamics. We wanted our listeners to be able to check out open-reel recorders, not dodgy room acoustics or interesting synth patches. Fortunately CTS Studios was kind enough to provide the orchestrally dimensioned Studio 1 for the day, complete with Steinway piano. We brought in the best piano, double basse, drum and saxophone players we could find. House engineer Jonathan Miller miked the quartet with Bruel & Kjaër omnis throughout, to provide extreme transparency and the least possible chance of phases anomalies. The only exception was the double bass, which also took a feed from a transducer in the bridge.

Acoustic separation and control of room ambiance was amply achieved by the use of moveable screens. In fact a very small amount of AMS reverb was subsequently added to the final mix. The Neve V series console was used to provide the sound balance and simultaneously feed the five different recorders. No artificial dynamics control was used and in fact, the musicians were encouraged to use as much range as possible.

Under these conditions, it was not easy to provide the ideal feed for all the machines. Apart from the slightly different requirements of analogue and digital, there were variations in reference level and an increasingly 'hot' band to juggle. However, these factors were satisfactorily balanced by the final take. Each recorder was set up and operated by a representative of the company concerned, as were the Dolby SR noise reduction units. Therefore, we can confidently say that all recorders and noise reduction units were delivering their optimum performance. In fact, during the first passes the control room looked like a rehearsal for the chariot race in Ben Hur as each operator grimaced and glared at his particular set of meters !

Because we wanted our listeners to have the best chance of comparing the recorders to each other and to the live band, we gave serious consideration to the possibility of monitoring off-tape while the band played live. We eventually decided against this approach for a number of reasons.

First, the record-to-replay delays on the recorders are different and may have provided clues to the listeners. Second, we wanted to provide individual listening tests - and ten conscutive 'perfect' takes onto five different recorders over the course of many hours is a tall order for the engineers and the band !

It was decided that the band would record a ten minute version of How Deep Is The Ocean onto all formats simultaneously. Each of our listeners had the benefit of an individual session in the control room, consisting of a five minute live version of the number, followed by the extended version on tape(s).

While this meant that the listeners did not have a chance to A/B source and recordings, it did give then a reasonable chance to become accustomed to the band and the ATC monitors. (Incidentally, CTS is equipped for mixing the cinema format Dolby Stereo but only the front L/R speakers were used in the evaluation.)

We felt it important that listeners participated individually because
a) we could then ensure that they did not influence each other,
b) they could sit in the optimum listening position,
c) they could change between formats at will during replay, comparing say drum or double bass solos on different machines.

It is worth mentioning that there was no way in which participants could identify the machines visually. All recorders were behind the desk, with the meters facing away from the listener. The machines were started manually in approximate sync and then the audio was fed to the monitors.

Although we could not hope to produce statistically meaningful results with a panel of nine, we nonetheless wanted to provide the listeners with a framework on which to base their evaluations. Each listener was given a form with various categories for their consideration. After leaving the session, each listener was grabbed and talked through his observations.

This is what they said...