CD and Pre Emphasis

[Gordon]

We [well a couple of us!] had a vigorous discussion over in the Pristine Audio thread earlier. In there I was moved to comment on the use of Pre Emphasis as a means of dealing with SNR issues. I said this:

“As it happens, CD has a simple one pole PE function defined but I don’t think anyone uses it despite the fact that it will reduce the QD.”

Well, if you google the subject of PE in CD you’ll find some interesting information suggesting that my supposition might be wrong. One quote as example [source last edited last August]:

By the late 1980s, pre-emphasis stopped being used because reliable 16-bit DACs with oversampling and other technologies minimized the conversion & filtering noise without the need for pre-processing the recording.

Most major-label CDs with pre-emphasis were manufactured in Japan in the early and mid-1980s. Relatively recent forum posts indicate that pre-emphasis is still used on newly manufactured CDs by some indie labels, mainly for classical titles.

A pre-emphasis flag for each track is normally stored in the subcode along with the audio data. It's also supposed to be stored in the table of contents (TOC), but many CDs have TOCs that say there's no pre-emphasis when in fact the subcode says there is. There are also some CDs which people believe were mastered with pre-emphasis, but which have no pre-emphasis flags set at all.

and read this

The interesting thing is that if you play a disc that is electronically marked [in more than one place which will not help] as having PE then a compliant player sees the flag and does the right thing. If, however, you rip that disc, the flag may be missed and so the data that gets to your hard drive etc may not be De-Emphasised!! Unless the ripping s/w reads the flags and does the job digitally.

The question is: do CDs get the flags set these days or are they only older ones or from a few manufacturers? Although every little helps PE si not as important as it once might have been. To find out if a given CD has the flags one would have to read the TOC as well as other parts of the data stream. How to do that is described in some of the google results. If you use EAC for example there is means to see it. Does all ripping software see the flags?

So what? Well the resulting playback of that ripped .wav file will sound bright to the tune of a 10dB lift after about 3KHz to about 10.6 before flattening again. That should be audible. Does make you wonder about downloads of CD material which may be ripped and then be coded to MP3.

This will not apply to non-CD formats like 48/16, 96/24 etc.

Anyone out there who knows what the practice with PE on CDs is nowadays?

[johnb]

All my CD collection is ripped to my hard drive (for use with Squeezebox) the only instance I had of pre-emphasis was with the Vegh boxed set of the Beethoven String Quartets on the Naive label. The pre-emphasis was flagged up by EAC - though the default position of the pre-emphasis column in EAC is way to the right, normally off the screen. When I listened to the rip it sounded very wrong indeed!

I have no idea whether the current Vegh set also has pre-emphasis. The current box design seems identical to mine so it might well.

After doing some research I discovered that SOX (a very, very useful command line audio tool that I had used before) had an option to correct pre-emphasis in audio files, so that is what I used.

I seem to remember that Dave also hit the pre-emphasis issue with the Horenstein Mahler 1 on a BBC Music Mag disc a year or so ago.

PS A modified version of SOX is also used by the Logitech Media Server (the part of the Squeezebox system that sits on your PC/NAS/Server) for transcoding.

[Dave2002]

I seem to remember that Dave also hit the pre-emphasis issue with the Horenstein Mahler 1 on a BBC Music Mag disc a year or so ago.

PS A modified version of SOX is also used by the Logitech Media Server (the part of the Squeezebox system that sits on your PC/NAS/Server) for transcoding.

If you mean me, I had a problem with iTunes and a Horenstein Mahler CD from CFM magazine. I'm not quite sure whether the problem was directly due to pre-emphasis or not, but basically some versions of iTunes at the time did not handle some discs correctly. It took Apple a couple of further releases to at least make it manageable. I'm not sure whether iTunes does rip CDs with pre-emphasis correctly, but at least it doesn't make a horrible mess as it was doing for a while. There is software which does apparently. I thought that DBPoweramp could do that, though I've not used that for years.

I did have a few bad rips via iTunes, and I'm not sure that they could all have been due to pre-emphasis issues.

I'm not quite sure how the Logitech/SOX could be exploited. Can the LMS be used for ripping?

[johnb]

Dave,

Apologies, for getting the BBC Mag and CFM mixed up.

I've just checked my CDs and I must have also bought the CFM Magazine for the Horenstein Mahler 1, and I can confirm that it has pre-emphasis.

I mentioned that the Squeezebox system uses SOX in passing, rather than implying one could use it for pre-emphasis or ripping. As that version of SOX is modified by Logitech it would probably be better to download the "official" SOX if one wanted to use SOX as an audio tool.

[David-G]

This sounds a very interesting subject ... or would be, if I understood exactly what you are talking about. For the benefit of those technically interested but not so technically knowledgeable, could you explain what exactly pre-emphasis is in this context?

[Stunsworth]

This sounds a very interesting subject ... or would be, if I understood exactly what you are talking about. For the benefit of those technically interested but not so technically knowledgeable, could you explain what exactly pre-emphasis is in this context?

It it means that the high frequencies were boosted when mastering a CD, and the high frequency boost was automatically reduced when replayed in the CD. The idea was to reduce encoding errors - quantisisation errors - when converting low level analogue signals to digital. It was more of an issue with the original 14 bit specification for CD where there were more likely to encoding errors.

Something similar is used in the analogue domain with FM radio, and of course LPs, to reduce background noise.

[johnb]

It is a process in which the treble is boosted on the CD and a flag is also set on the CD so that a CD player will recognise that pre-emphasis has been applied to the recording. The CD player should then apply a correction so that you hear the correct audio.

The problem comes when such discs are ripped to a hard drive as the resulting file will still have the pre-emphasis.

[Gordon]

Thanks jb, Dave and Steve, I had not seen any previous dicussion about this so apologies for raising it again. Also hope that David-G found Steve's and jb's explanation helpful!! A bit esoteric.

It seems though that doing and flagging PE may not be standard practice nowadays. I checked a random selection of my collection across different labels and found none!! I have some of the Vegh's, but not the box, if I can find them I'll have a look.

[Dave2002]

I have only encountered a few CDs with pre-emphasis. I had a Philips CD player with a light to indicate pre-emphasis. It only came on once, after I'd been using it for rather a lot of years, and that was on a CD of Japanese origin.

I think the CFM Horenstein Mahler did have pre-emphasis, so it could be used as a test disc to see if ripping software detects this, and also whether it corrects appropriately. In the earlier thread I believe I also mentioned another CD format - Microsoft HDCD - which should rip to 20 bit data if the ripping is done properly. Again, I suspect that only a few ripping tools will do this properly. http://en.wikipedia.org/wiki/High_De...atible_Digital

Re CDs with pre-amphasis, it would be interesting to know if any newly fashionable hardware streamers are designed to detect the pre-emphasis flags in ripped data - as doing the de-emphasis in circuitry might perhaps be preferable to using digital methods - though here I write out of speculation and ignorance!

[johnb]

t would be interesting to know if any newly fashionable hardware streamers are designed to detect the pre-emphasis flags in ripped data - as doing the de-emphasis in circuitry might perhaps be preferable to using digital methods - though here I write out of speculation and ignorance!

In order to do that there would have to be some kind of pre-emphasis "flag" written to the header of the wave/flac/mp3 file and the standardisation of such a flag. I am not aware that ripping software does write such a flag and there doesn't seem to be a field for it in the "standard" wave file header.

As far as streaming hardware/software is concerned I suspect that the manufacturers are unlikely to address pre-emphasis in ripped files as it is relatively uncommon.

[Dave2002]

john

Agreed - commercial pressures, not to mention technical ones, may make any special treatment for "odd" media which use different codecs and processing methods unlikely. Some companies might manage to issue their recordings from masters in current formats, but others may mess up and reissue something which does not get the best out of the recordings due to changes in formats.

[Gordon]

Thinking askance about this, if you put the PE on in every CD anyone ripping it would get the "wrong" sound and that is a mild form of DRM!! They'd have to go to the trouble of fixing it.

When you produce specifications that allow options you're asking for trouble!!

[Stunsworth]

I have only encountered a few CDs with pre-emphasis. I had a Philips CD player with a light to indicate pre-emphasis. It only came on once, after I'd been using it for rather a lot of years, and that was on a CD of Japanese origin

I seem to remember that Denon use pre-emphasis, but I could be wrong about that.

iTunes detects pre-emphasis when ripping CDs, and corrects the data stream accordingly.

[Bryn]

Just checked track 1 of Disc 1 of what I take to be the most recent re-issue of the Naïve Végh Beethoven String Quartets (purchased new a couple of years or so ago). Indeed, it does have PE flagged up by EAC.

[OldTechie]

I just bought a CD with PE on it (from my favourite second-hand shop.) It's a 1986 ASV disc of Emma Johnson playing a Crusell clarinet concerto. I ripped it using dbPowerAmp. It sounded horrid. There was a PE flag on it which showed up in dbPowerAmp when I looked again - but dbPowerAmp did nothing about it. My Denon CD player played it correctly including, to my surprise, de-emphasising the SPDIF output.

SOX can correct it, but you have to make sure you have a reasonably current version because it used to get the equalisation wrong, and not the same in left and right which was not too good. See http://www.radonmaster.de/robernd/tAFILTER.html (in my case via Google translate.)

It seems the pre-emphasis is not a single pole affair - it is 50uS (3,183Hz) boost but with a 15uS (10,610Hz) levelling off again.

The fix with foobar2000 is to install mudlord's Effect DSP plugin. You add a "PRE_EMPHASIS" tag with a value of "1" in the id data and foobar2000 will automatically load the plugin with its CD de-emphasis effect active. The documentation says you can also use a value of 0. I tested that to see whether it turned it off again but it just caused foobar2000 to crash, so don't do it!

I'm left with a question about this concept. Presumably it was designed with the idea that the pre-emphasis was applied in the analogue domain, and that the de-emphasis should be in the analogue domain. So it is digitised after pre-emphasis, and dither is added to compensate for the 16bit quantisation problems. The dither is often shaped and mainly in the HF region. If we de-emphasize it in the digital domain, we attenuate the dither before applying it to the DAC. So do we get a different effect to doing the de-emphasis in the analogue domain? Of course older CDs probably don't have clever shaped dithering (or maybe no dithering at all) so I'm sure it is not of any practical significance.