Early disc recordings from the 1900-1925 period of recorded sound. What were the playing speeds?

By 1926 it had finally happened. There was a consistent speed to records... almost.

But before this time and from the very start of the disc recording  industry, there was a really great sense of confusion. Much of this dealt with the question..... what speed do I play my records at?


The standard answer was 78 RPM.

I have seen in my many years of collecting records discs (that we call 78's) running at speeds from 64 RPM to about 84.

Starting with Berliners it is thought that 70 RPM  is a good average for them. But, there you will find many that are not too close to that speed.  From the mid 60's to 72 or so RPM

Climax Records (Columbia 1901-02)  are also around the speed of 72 to 80 RPM's. Give or take a bit. Also you will find on many of the early Climax and silver labeled Columbia discs of 1902 a bit of a problem in keeping a steady speed to the recording. I have found this in a number of early Climax and Columbia recordings.

 By 1903-04 it seems to have been rectified.  The problem with many of the recordings for the first Columbia records was this. They were trying to build up a massive catalog of recordings in the year of 1902. They had discontinued the Climax label and many of the Climax's had an announcement saying it was a Climax record.  All of the Climax records without an announcement were just changed over to the Columbia label. However, there was a massive recording program to as quickly as possible, replace all the Climax records.

Therefore, I am thinking that there may have been a bit of "recording engineer" burn out and exhaustion.

Perhaps several recordings made were sub par, but used for a short while till it could be replaced again. I find that many of the Columbia records from 1902 - 1910 are in the ball park of 78-82 RPM's. Not all of course, but a vast majority.

Consolidated, Improved, Monarch, and Victor records from 1900 to 1902 fall in the 73-82 RPM area as far as I have found.

Post 1902 Victor Talking Machine records run from about 71 to 82. This is where it was a problem for Victor. Many recording engineers had a set speed for some singers. Others did not. There seems to not be any agreement. Case in point with Enrico Caruso's Victor records the speeds vary going as low as 74 RPM and some reaching as high as 81.  There are other singers in which the speeds are quite varied as well.  Some of the early recordings were marked as to the key in which they were recorded. But this practice stopped shortly after it started.

The early days of the Gramophone and Typewriter company in Europe made some wide speed changes in recordings.  The speeds vary from 64 to perhaps 80.

Caruso's  1902 G&T recordings play at the speed of 67 RPM. His 1904 G&T recordings are around 73 RPM.  So as you can see there is a massive gap.

 I have found that most of the very early G&T's are slower than 78 RPM. In many cases much, much slower. Also a large number of the early G&T operatic recordings were recorded quite slow and I have found a number of recordings where 65 to 68 RPM's seems to make sense.

The famous operatic tenor Tamagno's  recordings made for G&T in 1903 and 04 are not only recorded quite slow. Seems 66-68 RPM, but were also pitched down!  RCA reissued some of his recordings on LP and recorded them being played at 78! It made him sound ridiculous.
The famous soprano Patti also falls under this situation. Her recordings made in 1905 and 06 are slower, around 72 RPM, and in her case as well the music was pitched down.

But what this shows a total lack of consistency in the industry. There was no set rule, just a general idea.

If one wishes to try to find the right speeds you have to remember several things. Many recordings were made with a pitch change. Older singers pulled from retirement often opted for changes in the pitch of the music in which they would record. If that is not known, many people will try to play the recording to pitch and find that their singer sounds very much like the hinges of Hell.

So often research and thought have to go into this. Also a sonic memory of what the voice sounds like when played at a proper speed. If you do not have that it will make it harder. But let's try to give a few basics of speed.  Most people will probably have recordings made later than this in their old 78 collection.

So I will just give an idea of the speed in which they should be played at. Remember, there will always be exceptions.  I am not listing all labels of that time, just some.  I am just listing a bit of information, not the whole story.

Columbia Records   80 RPM  By the late 1920's it is also 78.

Victor       78 RPM

Edison Diamond Discs  80 RPM
(Last year of production many DD records were recorded at 78 RPM)

Pathe Lateral  75-80 RPM
Pathe Vertical  60-90 RPM

American Record Company  75-80 RPM

Zonophone (American 1903-1912) 75 -80 RPM

Leeds, Cort, Busy Bee, Harvard, Star, Imperial, Oxford, and more. Almost all of these and many other quick lived labels whose records were made from other companies stampers. Therefore 75-80 RPM would be the rule. The only exception was Leeds and Imperial. They did make some of there own recordings. However in the case of the early Leeds records, some were illegal dubbings.



After electrical recording becomes the norm and calibrated turntables are used for recording.  The standard speed for the records by 1926 is 78.26 RPM.


But the first quarter century left many only to only guess. In some cases we still are.