Despite a small discrepancy from 10k upwards that sure looks like RIAA to me. I certainly did not expect this kind of hard evidence to appear. One thing, however; this is a very early stereo record and I have a fairly small number of these-they are very thick!-but I seem to remember that these do not have the kind of heavy sound that has been mentioned here. Is it possible that there was a change back to CCIR for some pragmatic reason, perhaps due to characteristics of typical domestic playback equiptment? I hardly think that Thorsten's research can simply be dismissed.
Thanks dimitri. Cold, hard facts. All my melodiya recordings sound pretty OK with the RIAA. If there was some "dark" sound, it wasn't due to a non standard EQ curve.
By the way, the PH-77 manual implies (see pic i attached previously) that Philips stereo records were not cut using RIAA. I'm currently listening to a early Philips stereo record, very early (dark plum label, AY series which was the 1st stereo series.) It sounds beautifully with RIAA.
Ok, from the Stereophile review: PH-77. The secret curves.
Red-blue: RIAA
Cyan-Magenta: "Enhanced RIAA" (freq response is relative to RIAA)
Green-Black: "RIAA DMM" (freq response is relative to RIAA)
The "Enhanced RIAA" is within 0.5dB of the RIAA curve at the audible range. Thus it's basically the same thing. Nice.
The "RIAA DMM" is within 3dB of RIAA so basically is a subtle change, a matter of taste.
Red-blue: "Decca FFSS (stereo)" (curve relative to RIAA)
Cyan-Magenta: "Columbia"
Green-Black: "CCIR 56"
"Decca FFSS" is within 2dB of RIAA. A slight tonal change. But certainly no difference that could be caused by using significanty different time constants (relative to RIAA)
"Columbia": Same comments as above.
"CCIR 56": This one is different indeed.
I spent a month in Moscow in 1990 and heard music for hours every day in the Great Hall of the Moscow Conservatory. If this hall was used extensively for Melodiya recordings, it would possibly explain a lot about the timbre of those recordings.
Whilst on the subject of Melodiya, has anyone any thoughts on equalising of the very earliest mono LP's? They seem to me to reproduce best if they are treated like RCA 78's more or less with not too much bass boost.
Whilst on the subject of Melodiya, has anyone any thoughts on equalising of the very earliest mono LP's? They seem to me to reproduce best if they are treated like RCA 78's more or less with not too much bass boost.
When I first read about the AMR PH-77 I thought that we were dealing with a much overlooked area, and mystifying that we weren't seeing more comments about it.
If you go to the product site, you'll find that they are associating correction curves with specific labels:
Abbingdon Music Research - Products - PH - 77 Features
Then I read through the material at hifimuseum.de (which is the same that has been linked to in the thread, from the BBC archivist). And there it became clear that there was no lock-in between labels and curves, as several labels were using the same cutters; and that the introduction of RIAA came much sooner than stated by some.
But an exception must be made for mono-material in the 50s, as that clearly benefits from curves.
I have found no independent confirmation that DGG stayed off-RIAA well into the 60s.
As concerns Melodiya, I've actually performed an experiment. With a number of different issues of Rozhdestvensky's Manfred, with the large RTV orchestra in Moscow. (A brilliant recording).
Soviet issue (reddish cover) - CM 03151-2. Best orchestral sound reproduction through RIAA, without exaggeration.
Eurodisc version (greyish cover with composer in oval) - 87 781 KK - Substantial weight to the woodwinds, but unnaturally so, as if the bass has been boosted.
Version sold in US, manufactured by Capitol Records. Brown/Yellow cover - SR-40267.
Also with weighty bass, but not as pronounced as in the Eurodisc version.
Sufficiently different to have me thinking they're not supposed to be this different.
If you go to the product site, you'll find that they are associating correction curves with specific labels:
Abbingdon Music Research - Products - PH - 77 Features
Then I read through the material at hifimuseum.de (which is the same that has been linked to in the thread, from the BBC archivist). And there it became clear that there was no lock-in between labels and curves, as several labels were using the same cutters; and that the introduction of RIAA came much sooner than stated by some.
But an exception must be made for mono-material in the 50s, as that clearly benefits from curves.
I have found no independent confirmation that DGG stayed off-RIAA well into the 60s.
As concerns Melodiya, I've actually performed an experiment. With a number of different issues of Rozhdestvensky's Manfred, with the large RTV orchestra in Moscow. (A brilliant recording).
Soviet issue (reddish cover) - CM 03151-2. Best orchestral sound reproduction through RIAA, without exaggeration.
Eurodisc version (greyish cover with composer in oval) - 87 781 KK - Substantial weight to the woodwinds, but unnaturally so, as if the bass has been boosted.
Version sold in US, manufactured by Capitol Records. Brown/Yellow cover - SR-40267.
Also with weighty bass, but not as pronounced as in the Eurodisc version.
An externally hosted image should be here but it was not working when we last tested it.
Sufficiently different to have me thinking they're not supposed to be this different.
Hi,
Actually, if you really read the site, you find that we have named the curves after the Labels we have good reason to consider to be the originators, with the single exception of RCA/RIAA, as not having a "RIAA" curve at all would have confused too many and having the same curve called one RIAA and once RCA would have seemed to simply make up numbers... CCIR of course is CCIR and is hence listed as CCIR.
To quote from the FAQ section of the site (Abbingdon Music Research - Information - Frequently Asked Questions):
Also:
There was originally actually a fairly long appendix in the User Manual dealing with the minutiae and quite clearly stating "The EQ depends upon who cut the Lacquer and not on the actual label on the record", however that whole was cut, it may make a reappearance in a more extensive article at some point in the future.
It certainly is a difficult task to give enough information to be useful but no so much as to confuse. It may be AMR erred towards the "do not consfuse the customer side" more than towards "educate the customer", however, not by a huge margin I feel.
Ciao T
Actually, if you really read the site, you find that we have named the curves after the Labels we have good reason to consider to be the originators, with the single exception of RCA/RIAA, as not having a "RIAA" curve at all would have confused too many and having the same curve called one RIAA and once RCA would have seemed to simply make up numbers... CCIR of course is CCIR and is hence listed as CCIR.
To quote from the FAQ section of the site (Abbingdon Music Research - Information - Frequently Asked Questions):
AMR said:
AMR said:
Also:
AMR said:
There was originally actually a fairly long appendix in the User Manual dealing with the minutiae and quite clearly stating "The EQ depends upon who cut the Lacquer and not on the actual label on the record", however that whole was cut, it may make a reappearance in a more extensive article at some point in the future.
It certainly is a difficult task to give enough information to be useful but no so much as to confuse. It may be AMR erred towards the "do not consfuse the customer side" more than towards "educate the customer", however, not by a huge margin I feel.
Ciao T
Last edited:
Guys, at what point does the sound that a particular labels production team was trying to achieve and “equalization” begin to blur together a little? After all I don’t find a couple of dB of bass boost or cut out of the realm of personal taste or what a particular recording engineer/producer wanted. Most listening environments have plenty of +-2 dB bumps and grinds. In other words why treat any adjustments at recording time separate from at cutting time? How do you tell the difference?
The “secret” folklore stuff gets a little tiresome. It’s not much different than the retired Jaguar mechanic’s secret to cutting the damping fluid in SU carburetors as they age so you can actually keep two tuned for more than a couple of days.
The “secret” folklore stuff gets a little tiresome. It’s not much different than the retired Jaguar mechanic’s secret to cutting the damping fluid in SU carburetors as they age so you can actually keep two tuned for more than a couple of days.
Last edited:
Scott, on one hand we have the objective types claiming the importance of achievable tolerances of ±.25 dB and that it is sinful to tailor the equalization curve to taste, and on the other hand there are people like me who don't presume that all records are cut to any degree of accuracy and an adjustment to the EQ is a nice touch. It's been my experience that Eurodisc re-issues of Melodiya recordings are of better quality at the upper frequencies and I am simply trying to determine why.
John
John
What makes sense to me is a total EQ from venue to LP that depended on lable or production team. From what I see here there is nothing worth disavowing knowledege of.
As an aside, with a basic analog RIAA as a starting point, these all coiuld be done (as well as any tweak in between) with an IIR or FIR filter with none of the usual dynamic range complaints applying. At this price a nice feature would be a final QC firmware update that balanced and calibrated both channels against all component mis-match (the test sample appeared to have ~2% channel imbalance).
It's obvious to me that you would want to maintain a fairly accurate (±1 dB) filter for lps that are known to be high quality productions. It may seem like using anything other than the RIAA standard is a bit obscure, but if you are at all interested in classical performance, using the correct correction for Warsaw Pact recordings is worthwhile because there are so many great ones out there. I mean, who wants to listen to Fritz Reiner or Artur Rubinstein (OMG) all of the time?
John
John
Vinyl and CDs. I don't know if they are still a going concern, though. It seems like that were most active during the 'eighties.
John
Scott,
Funnily enough, that generally does not quite appears to be the case based on the records in my own collection. It seems mostly a question of "who cut it.
I used to do that with a separate digital EQ. HOWEVER, this presumes that we first A2D convert the LP Signal, and the re-convert D2A and out. This is something most of the potential customers for a PH-77 will not stand for.
Given that it is not particularly difficult to accommodate additional, switched time constants in a split RIAA EQ design we elected to do it that way...
The channel imbalance comes from the tubes. The PH-77 has three stages, one discrete solid state two tube, all open loop, no feedback (not even local degeneration). In the light of this and the fact that very few cartridges have better than +/-1dB channel balance (and lets not talk about speakers channel matching, especially not in room response) I am not unduly worried about 0.35dB channel imbalance. I'd be more worried if one channel has a severe difference in frequency response from the other.
AMR is a 21st century company. We have USB Inputs on our CD-Players (and upcoming DAC), so it is natural to include an A2D.
Plus, designing in the A2D into the preamp allows us to dispose of a lot of potentially sound degrading circuitry that commonly sits in front of any AD converter chip. This way it is possible to generate archival copies at very high quality.
Don't worry about the connection being USB, we use decent hardware with competent implementation, jitter is much lower than what SPDIF can manage.
Ciao T
Funnily enough, that generally does not quite appears to be the case based on the records in my own collection. It seems mostly a question of "who cut it.
I used to do that with a separate digital EQ. HOWEVER, this presumes that we first A2D convert the LP Signal, and the re-convert D2A and out. This is something most of the potential customers for a PH-77 will not stand for.
Given that it is not particularly difficult to accommodate additional, switched time constants in a split RIAA EQ design we elected to do it that way...
The channel imbalance comes from the tubes. The PH-77 has three stages, one discrete solid state two tube, all open loop, no feedback (not even local degeneration). In the light of this and the fact that very few cartridges have better than +/-1dB channel balance (and lets not talk about speakers channel matching, especially not in room response) I am not unduly worried about 0.35dB channel imbalance. I'd be more worried if one channel has a severe difference in frequency response from the other.
AMR is a 21st century company. We have USB Inputs on our CD-Players (and upcoming DAC), so it is natural to include an A2D.
Plus, designing in the A2D into the preamp allows us to dispose of a lot of potentially sound degrading circuitry that commonly sits in front of any AD converter chip. This way it is possible to generate archival copies at very high quality.
Don't worry about the connection being USB, we use decent hardware with competent implementation, jitter is much lower than what SPDIF can manage.
Ciao T
Hi,
Shame, Melodya had a decent archive of unique performances on pretty high grade masters, perfect fodder for transfers to 24/96 to be available as downloads...
Ciao T
Shame, Melodya had a decent archive of unique performances on pretty high grade masters, perfect fodder for transfers to 24/96 to be available as downloads...
Ciao T
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.