Cartridges and phono settings

[f1eng]

-90db to -100dB with no record playing is state of the art, there won't be an analogue phono stage that quiet in all likelihood.
The background noise on an LP will frequently be up around -45dB on the lead in and lead out grooves on typical LPs, with tape hiss adding to this once the audio starts.

When Meridian were doing their work on lossless packing, being pragmatic engineers, they measured the dynamic range of a huge number of LPs.
IIRC the one with the highest dynamic range they found had 11-bit equivalent SNR. So 16-bit is enough to record any LP if levels are carefully set, and 24 bit makes it easy, even with levels set in a slipshod way.
In terms of sampling frequency whilst it is indeed possible to record over 22 kHz onto an LP, very, very few pickup cartridges can retrieve it, and certainly not at high fidelity.
Even in the mid band top quality cartridges have distortion in the 1 to 4% range, and most are up above 15% in the top octave, even if they make it up there, so most, if not all, of what you record at 96kHz sampling which you wouldn't get at 48kHz are cartridge distortion products rather then music.
The recent review of the outstanding £8750 Techdas Ti cartridge shows a plummeting frequency response from 15kHz, around 2% midband distortion rising to 14% at 20kHz. These are SOTA results for a cartridge.
Recording at 96/24 will capture absolutely all of the output of a record player, all going to 24/192 will do is double the file size whilst containing absolutely no more information, which is not a big deal with cheap mass storage but is a bit stupid.

[marcor]

-90db to -100dB with no record playing is state of the art, there won't be an analogue phono stage that quiet in all likelihood.
The background noise on an LP will frequently be up around -45dB on the lead in and lead out grooves on typical LPs, with tape hiss adding to this once the audio starts.

When Meridian were doing their work on lossless packing, being pragmatic engineers, they measured the dynamic range of a huge number of LPs.
IIRC the one with the highest dynamic range they found had 11-bit equivalent SNR. So 16-bit is enough to record any LP if levels are carefully set, and 24 bit makes it easy, even with levels set in a slipshod way.
In terms of sampling frequency whilst it is indeed possible to record over 22 kHz onto an LP, very, very few pickup cartridges can retrieve it, and certainly not at high fidelity.
Even in the mid band top quality cartridges have distortion in the 1 to 4% range, and most are up above 15% in the top octave, even if they make it up there, so most, if not all, of what you record at 96kHz sampling which you wouldn't get at 48kHz are cartridge distortion products rather then music.
The recent review of the outstanding £8750 Techdas Ti cartridge shows a plummeting frequency response from 15kHz, around 2% midband distortion rising to 14% at 20kHz. These are SOTA results for a cartridge.
Recording at 96/24 will capture absolutely all of the output of a record player, all going to 24/192 will do is double the file size whilst containing absolutely no more information, which is not a big deal with cheap mass storage but is a bit stupid.

Thank you for your detailed explanation, it make me re think the 192 vs 96. I will try this weekend to do some comparison again between the 2 settings.
Do you have any idea why the reading of the Devialet meters and the one from Audacity through Usb are so different? My d250 with phono connected not playing show a noise of -100dB -90dB if I check the same thing through the Audacity recording meter it show noise up to -57dB.
Thanks

Marco

[thumb5]

Do you have any idea why the reading of the Devialet meters and the one from Audacity through Usb are so different? My d250 with phono connected not playing show a noise of -100dB -90dB if I check the same thing through the Audacity recording meter it show noise up to -57dB.

Well, dB is a relative measure so my first question would be: relative to what? Maybe the noise as measured by the Devialet is relative to its full output and the noise measured by Audacity is relative to some fixed reference (e.g. 775 mV rms).

Just a thought. Someone will soon correct me if I've got this wrong...

[f1eng]

What Thumb5 wrote may well be true. I know nothing of Audacity, sorry.

[Mikeeo]

Tried to see if I could make my bars go higher or lower tonight. Tested one output level only 10 uV over (ie 570 uV) the Benz spec output and the a doubled one at 1100 uV.
Here comes the funny bit; No change on the meters behavior, still just slightly under -40 dB for both settings, however at the lower setting I def. had a louder sound than the higher output setting. What does this mean? The bars/meters have no practical use in 'optimizing' signal from phono. So what are they actually showing?

/Mike

---

I also tried the 192 kHz setting but I couldn't hear any difference or at least I was ambivalent about it.

/Mike

[f1eng]

Tried to see if I could make my bars go higher or lower tonight. Tested one output level only 10 uV over (ie 570 uV) the Benz spec output and the a doubled one at 1100 uV.
Here comes the funny bit; No change on the meters behavior, still just slightly under -40 dB for both settings, however at the lower setting I def. had a louder sound than the higher output setting. What does this mean? The bars/meters have no practical use in 'optimizing' signal from phono. So what are they actually showing?

/Mike

---

I also tried the 192 kHz setting but I couldn't hear any difference or at least I was ambivalent about it.

/Mike

Changing the gain by only 3dB may well not be noticed on the somewhat small display.
There is no sensible reason why the 192 setting would be different. All the extra data only contains any information in the 48kHz to 96 kHz octave...

[Mikeeo]

Tried to see if I could make my bars go higher or lower tonight. Tested one output level only 10 uV over (ie 570 uV) the Benz spec output and the a doubled one at 1100 uV.
Here comes the funny bit; No change on the meters behavior, still just slightly under -40 dB for both settings, however at the lower setting I def. had a louder sound than the higher output setting. What does this mean? The bars/meters have no practical use in 'optimizing' signal from phono. So what are they actually showing?

/Mike

---

I also tried the 192 kHz setting but I couldn't hear any difference or at least I was ambivalent about it.

/Mike

Changing the gain by only 3dB may well not be noticed on the somewhat small display.
There is no sensible reason why the 192 setting would be different. All the extra data only contains any information in the 48kHz to 96 kHz octave...

Yes you are right, 3 dB will not be enough, I will test again with bigger change in gain at some point.
Though the 192 setting is to our ears smoother in the top freq but without losing details. Blind tried it with the wife as I couldn't trust myself, wanting to hear this or that

Well, try for your selfs/Mike

[marcor]

Yes you are right, 3 dB will not be enough, I will test again with bigger change in gain at some point.
Though the 192 setting is to our ears smoother in the top freq but without losing details. Blind tried it with the wife as I couldn't trust myself, wanting to hear this or that

Well, try for your selfs/Mike

So did you hear some differences between 192 and 96?
I am really trying to figure out if it was all a figment of my immagination or if I really heard a difference.
I am trying again this weekend

Marco

[Mikeeo]

Yes you are right, 3 dB will not be enough, I will test again with bigger change in gain at some point.
Though the 192 setting is to our ears smoother in the top freq but without losing details. Blind tried it with the wife as I couldn't trust myself, wanting to hear this or that

Well, try for your selfs/Mike

So did you hear some differences between 192 and 96?
I am really trying to figure out if it was all a figment of my immagination or if I really heard a difference.
I am trying again this weekend

Marco

Marco,

Yes I did or rather felt that it was. So I had to test it on the wife (blindly) and she also heard this (I did not tell anything about what I changed or what I had experienced).

/M

[marcor]

Yes you are right, 3 dB will not be enough, I will test again with bigger change in gain at some point.
Though the 192 setting is to our ears smoother in the top freq but without losing details. Blind tried it with the wife as I couldn't trust myself, wanting to hear this or that

Well, try for your selfs/Mike

So did you hear some differences between 192 and 96?
I am really trying to figure out if it was all a figment of my immagination or if I really heard a difference.
I am trying again this weekend

Marco

Marco,

Yes I did or rather felt that it was. So I had to test it on the wife (blindly) and she also heard this (I did not tell anything about what I changed or what I had experienced).

/M

Thanks Mike.
To be honest with you I did the same thing with my 9 years old daughter who can hear a lot better than me the higher frequency. I had her listen the same record with both the settings several time without telling her what was the difference and she spotted all the time the 192. She also helped me to set the load of the cartridge....