Recently I bought an LP (33 rpm) produced by MCA in 1957. When played through a common RIAA preamp, the high frequencies were obviously exagerated. I guess MCA still didn't use RIAA preemphasis at that time.
Do you know which phono deemphasis curve would be adequate for this record?
Do you know which phono deemphasis curve would be adequate for this record?
What's the catalogue #? MCA records didn't exist in 1957, so this must be a reissue. RIAA was well established by 1957, so maybe your record was poorly mastered.
jeff
jeff
Now I found "Pre-1957" written in small letters on the label. Not sure if this has something to do with the used curve.
The "1957" seems to be with the copyright notes. FWIW, pre-1957 recordings are US copyright until 2067 (110 years), so this is rather pointless on a 1985 release. Copyright in the UK is different and I'm not going there.
Last edited:
Columbia LP quite possible, thanks for the hint. Played through RIAA equalization would result in 2.5dB treble boost which corresponds to how it sounds. The best option is probably making a needledrop with my RIAA preamp and correcting it with SoX in digital domain (2.5dB treble attenuation with f=2122Hz). I'll report the result.
So this is what I've got. Comparing the RIAA curve and the Columbia curve I realized it takes a LF shelf filter with f1 = 50Hz and f2 = 100Hz and HF shelf filter with f1 = 1590Hz and f2 = 2122Hz to make a RIAA preamp compatible with Columbia LP standard.
Having this in mind, I digitized couple of tracks with common RIAA preamp and made correction using the SoX tool with this command (ignoring the bass correction):
sox file1.flac file2,flac treble -2.5 1836 0.3s
I can't be sure if the correction is completely accurate, but the result is much more realistic tonal balance IMHO.
Having this in mind, I digitized couple of tracks with common RIAA preamp and made correction using the SoX tool with this command (ignoring the bass correction):
sox file1.flac file2,flac treble -2.5 1836 0.3s
I can't be sure if the correction is completely accurate, but the result is much more realistic tonal balance IMHO.
I had a Glenn Miller compilation record. It was very bright. That was likely due to the recording. "Simulated stereo" was not helping either.
Ed
Ed
A lot of those old 'reissue' LPs-from-78s seemed to have arbitrary EQ errors. Not to mention hiss-reduction and "stereo". I had a Best Of Billie Holiday that needed a major bass-boost and top-cut to be listenable, nevermind accurate. At this point in history I'd think it is OK to "re-master", EQ for best effect not some lost accuracy.
I find this true even today. I Iisten to a lot of supposedly well-restored stuff from the early 1950s and it all, without exception, sounds too bright. I really have to wonder whether they have really sorted out their NABs from their Columbia LPs from their RCA New Orthophonics &c properly, and also whether they have taken into account the fidelity of the target systems 70 years ago as against today. Speakers were a lot more resonant then, for example, which would make anybody at the time dry out the EQ at least a little.
Scott Wurcer wrote an article about digital RIAA correction in Linear Audio volume 10, https://linearaudio.net/article-detail/2240 . It is very thorough. He ended up doing brute force numerical optimization for both RIAA correction and various historical record correction filters. Unfortunately some of the tables at the end seem to be messed up, at least in the paper edition; I saw tables for 48 kHz and 96 kHz sample rates that looked exactly the same while they should be different. Fortunately they can also be downloaded from the Linear Audio website without these errors:
https://linearaudio.net/downloads
Scroll down to heading: "Test waveforms for Scott Wurcer's digital RIAA article in Volume 10"
Tables:
https://linearaudio.net/sites/linearaudio.net/files/v10 sw app1 table a-1.xlsx
for optimized IIR filter coefficients for RIAA equalization with one biquad at 48 kHz, 96 kHz and 192 kHz sample rates. Using only one biquad results in a +/- 0.13 dB error at 48 kHz, much less at higher sample rates.
For more accurate correction (within +/- 0.01 dB) using two separate biquads at 48 kHz and 96 kHz sample rates (unfortunately no 192 kHz):
https://linearaudio.net/sites/linearaudio.net/files/v10 sw Appendix 1b web.docx
This table contains IIR filter coefficients for RIAA correction as well as for various other types of correction for old records.
To change the treble correction while leaving everything else the same, look up the biquad coefficients for RIAA treble correction and swap the a and the b coefficients. That is, ai becomes bi and the other way around. You then have the coefficients for a biquad that removes the RIAA treble correction. Then look up the coefficients for the treble correction that you want to have. Cascade two biquads, one with the swapped RIAA treble correction coefficients and one with the coefficients for the treble correction you want to have.
The coefficients are not normalized to 0 dB gain at 1 kHz, but that's a matter of adding one multiplication with a gain constant. You can also increase all b coefficients of a biquad by the same factor to increase its gain, if you use floating point or something else that ensures no overload can occur.
According to the article, the Audacity Nyquist syntax for a biquad is biquad-m signal b0 b1 b2 a0 a1 a2. The SoX syntax is biquad a0 a1 a2 b0 b1 b2.
Scott's optimizer is also on the Linear Audio website, so you can try to use it to find 192 kHz sample rate coefficients.