Recent Advances in Wax Recording ^ 



By HALSEY A. FREDERICK 



Synopsis: This paper considers chiefly the frequency-response charac- 

 teristics and limitations of the lateral cut "wax" record. It shows that the 

 frequency range from 30 to 8,000 cycles can be recorded and reproduced 

 from the record with practically negligible deviation from a flat frequency- 

 response characteristic. The paper brings out the ease with which the 

 record can immediately be replayed from the "wax" as an aid in assisting the 

 artist to obtain the best results. A brief description is given of commercial 

 processing methods including both plating and pressing. These methods 

 give essentially a perfect copy of the original "wax." The time required 

 for this work has been considerably reduced of late so that a test pressing can 

 be obtained within three hours of the cutting of the original "wax." 



IN the recording and reproducing of sound by the so-called "electric" 

 method with the "wax" disc, the process may be considered as 

 consisting of eleven steps. In order, these are: (1) studio, with 

 its acoustic conditions, (2) microphone, (3) amplifier, (4) electro- 

 mechanical recorder, (5) "wax" record, (6) copying or reproducing 

 apparatus, (7) hard record or "pressing," (8) electric pickup, (9) 

 amplifier, (10) loud speaker, (11) auditorium. 



With this chain of apparatus the chief problem is that of making 

 the reproduced sound in the auditorium a perfect copy of that in the 

 studio. This is a matter of quality or fidelity of reproduction. There 

 are other problems of cost, reliability, time required, etc., which are 

 important but secondary to that of fidelity. While it may be necessary 

 or convenient to introduce distortion in one of these links to com- 

 pensate for such unavoidable distortion as may occur in other links, 

 experience shows that it is desirable for the sake of simplicity, reliability 

 and flexibility to reduce such corrective warping to a minimum and 

 to make each step in the process as nearly perfect as possible. Per- 

 fection of a complete system may be judged by the practical method 

 of listening to the overall result. It is necessary, however, to analyze 

 each element of the complete system. To do this, other more 

 analytical methods of test and standards of performance must be 

 used. One of the most useful of these is the response-frequency curve. 

 In order that all frequencies be reproduced equally and that the 

 ordinary faults of resonance be avoided, this must be flat and free 

 from sharp peaks. Good reproduction requires that frequencies from 

 50 to 5,000 cycles be included without discrimination. If, however, 



1 Presented before Society of Motion Picture Engineers at Lake Placid, New 

 York, September 26, 1928. 



159 



