Park Foundation Music App

Home > Releases

Hungarian Dance No. 5 / Hungarian Dance No. 6

Overview

Hungarian Dance No. 5 / Hungarian Dance No. 6 --- 1923-09-00

Recorded October 24, 1917 at the Victor Office Building no 2, Eighteenth Floor Auditorium, Camden, NJ. Originally released separately on single-sided record nos. 64752 and 64753: "[m=1528768]" and "[r=8675034]." Matrix & take numbers: B-20888- 3 / B-20889- 2. These were the first commercial recordings released by Leopold Stokowski and The Philadelphia Orchestra! These Hungarian Dances were only recorded once, in the acoustic era. Instrumentation: 16 first violins, 15 second violins, 12 violas, 10 cellos, 8 basses, 3 flutes, piccolo, 4 oboes, 4 clarinets, 4 bassoons, 5 horns, 4 trumpets, 3 trombones, tuba, tympani, and 2 percussion. (Instrumentation applicable to take 1; instruments not specified in Victor ledgers for remaining takes. Ledgers include "2 precaution [percussion]" and "2 libarian" [librarians]" among the personnel enumerated.) LPs: Stokowski Society LS 3. CDs: POA 100-1; Pristine PASC 192. Notes by Andrew Rose from Pristine PASC 192: Stokowski made his first recordings with the Philadelphia Orchestra in 1917 though, as Edward Johnson has noted, "between 1917 and 1924, they made an estimated 450 acoustic recordings, but the old method of playing into a large horn gave a very poor representation of orchestral sound, and of all their acoustic discs, only 60 or so were actually issued..." Indeed, Stokowski's very first acoustic recordings were never released - his first attempts at the two Brahms Hungarian Dances included here were recorded on 22nd October, 1917, together with two works by Tchaikovsky. All four recordings were remade - in the case of the Brahms, just two days later, to become the first released recordings conducted by Stokowski. It is indeed true, as Edward Johnson states, that in their 'natural' state, acoustic recordings made a pretty poor job of capturing anything much - though HMV's legendary producer Fred Gaisberg always had an affectionate soft spot for the way they captured the human voice. It has become a generalisation to state that the recordings captured little below 200Hz, and perhaps little above 2500Hz, or perhaps 3000Hz at a push. However, as these recordings demonstrate, this was not necessarily the case - though it often takes the kind of advanced restoration processing that Pristine's XR remastering brings to these recordings to uncover the real range and quality of acoustic recordings. In the case, for example, of the first recording on this collection, there's a full bass going down to (and possibly below) 70Hz - a frequency extension of almost two octaves over the accepted lower limits. Meanwhile, throughout the recording one can clearly hear string harmonics in the 3500-4000Hz range, and where the brass plays at full volume there's genuine audio content up at (and beyond) 5kHz, which is not far off the upper end of electrical recordings. In order to hear these extremes it has been necessary first to dramatically re-equalise sound that has been highly coloured and harmonically distorted by the horn it was recorded through. This therefore comprises the first stage of the XR process in this context. Secondly we then have to apply highly selective, targeted noise reduction to those frequencies which have received what has often been a quite enormous level boost, rooting out the noise and leaving as much of the actual recorded audio intact. The results of this are genuinely astonishing - even more so when the records were slowed down to concert pitch (apparently it was common for Victor to deliberately record at below 78rpm so that their records sounded brighter when played back at their stated speed). The majority were out by up to a full semitone (though some were considerably closer to true concert pitch), and the correction of this adds further depth and authenticity to the instrumental sound, as well as giving a much more sensible idea of Stokowski's tempi.