Basically, I would look for the following factors in a good recording:

General Clarity
Properly Miked voices
Properly Miked instruments
Proper volume balance between instruments and vocals
Soundstaging and Imaging
Lack of excessive noise and hiss
Good,tight bass
Crystal clear highs
Detailed

General Clarity

The expression "removing Sunglasses while looking at the moon" [from What Hi-Fi?], really applies to comparisons between 'processed pop' CDs like Jann Arden's Living Under June and Julia Fordham's Falling Forward [yes, I still love them inspite of this] and audiophile CDs Nancy Bryan's XRCD. What do we generally blame for this?

  • Multitrack recording, overdubs, remixing, excessive reverb
  • Synthesiser tracks with hundreds and hundreds of layers of sound. [the 'Enya' effect]
  • Inferior microphones.
  • Trying to hard to get rid of hiss. The Arden/Fordham albums I mentioned are noticeable for their lack of hiss. You can cut down hiss by filters/noise reduction, or by having a short recording chain -- i.e. mike-preamp-ADC-tape.

Volume Balance -- The more natural, the better

This really should be a non-issue. Lets say you're sitting in a club, somewhere in the middle ,the band is playing, and the vocalist is playing. If they were not using microphones or amplification (lets say that synths and electric guitars have speaker/amps next to them), what would the relative volume levels be? The same goes for classical music. The question is: What are the volume levels of each instrument relative to each other and the voice?

Unfortunately, we're getting increasing burdened by recordings that attempt to stuff the voice down your throat at super loud levels while basically ignoring the other instruments. Mary Black's 1997 album, Shine, is guilty as charged. The balance is all horribly wrong (actually, a lot of the recording is horribly wrong). When I listen to her 1983 album, Without the Fanfare, despite the metallic timbre of the guitars, a tiny, tiny loss of detail, it sounds fundamentally 'right' and 'natural' compared to Shine. I invite you to listen to 'Ellis Island' and compare it to any track in Shine. A 1997 DDD album of the same artiste sounding worse than a 1983 AAD (or ADD??), album? I didn't think that it was possible, but if you listen to both of them, I believe you'll agree.

Footnote: After listening to 'new' CDs far harsher than Mary Black's Shine, I don't think its that fair. Maybe I'll retract it and replace 'Shine' with Beth Nielsen Chapman's Sand and Water, which is really bright (at least, until I hear something worse)

Noise and Hiss -- not too much please

For analogue masters being transferred to CD, theres bound to be some hiss. If you removed the hiss, you would remove some of the 'sparkle' from the original music. Examples of CDs trying to minimise the hiss from the analogue master include Kate Bush's The Whole Story, which is a unmitigated sonic disaster. Another example would be Julia Fordham's first 2 albums which mercifully are still quite listenable despite this. Though I must admit that Mobile Fidelity's AAD remasters of some classic rock/pop albums are very good in minimising noise without removing too much of the sparkle from the music.

At the other end, we have no attempt to remove this analogue hiss, a prime example would be The Sheffield Labs remaster of Amanda McBroom's early songs :- just listen to the 'Rose'. For bearing with the hiss, you get rewarded with an insight into all the nuances in Amanda's beautiful voice. Further, its actually only the quiet parts that get affected by the hiss -- when the music starts to move, you really can't hear the hiss at all.

But on to DDD recordings. You would expect hiss to be a non-issue in totally digital (DDD) recordings, but hiss still lurks around. The sad fact is that, the better the microphone, the noisier it is :- mainly because its more sensitive. For some reason, audiophile recordings don't have this mike hiss (unless you stick your ear next to the tweeter) and still manage to achieve super-clear highs. I wonder how they do it. But by and large, for DDD live recordings, like Shawn Colvin's 'Cover girl' , and superb studio recordings like Michelle Shocked's "Kind-Hearted Woman", a slight hiss is present on the quiet parts. IMHO, it doesn't detract from the overall performance at all.

Another source of hiss might be due to the long cables from the microphone to the recording equipment/mixing board, (if its real long, there'll be a pre-amp to boost gain). Of course the cables are fully balanced (i.e. note the XLR or 'cannon' jacks), but sometimes thats not enough (Chesky uses/used Cardas 300Bs -- currently I think they've moved to Nirvana?) To solve this problem, some companies (Chesky and Deutsche Gramaphon I think...) have resorted to having an analogue to digital convertor near to the microphone in order to convert the signal to digital before running it back to the recording equipment/mixing board.

In summary, a lack of audible hiss while maintaining super-clear highs is indeed a mark of a fine recording, but in most cases there is some hiss present but thats perfectly acceptable if the alternative is applying noise reduction and causing the recording to lose its sparkle.

Compression. The No.1 Enemy?

There are many, many aspects of 'sound' that hi-fi equipment must strive to reproduce in order to be realistic. One of this is dynamics. When you play the piano, there are times when you play pp and times when you have to play ff. What is critical is that you are able to contrast the two. Likewise, if the recording reproduces a pp movement as p and ff as f, you are making the louder part closer in volume to the softer part. This is of course what compression is all about. In truth, even in Classical Music (though some labels don't), it is sometimes necessary to apply some compression (so you don't get hit suddenly by a 100dB peak while listening all along to some quiet music at 75dB). But for pop music, compression is applied basically to make the music radio friendly (radio friendliness. responsible for compression and songs lasting not more than 3 minutes [hmm actually thats not such a bad thing since they're just repeating the chorus again and again anyway]).

Compression is a way of life. The question is : how much? With nothing better to do that to listen to music while monitoring the volume with my Radioshack SPL meter, I've come to a few conclusions. (caveat: extremely generalised, based on the music I listen, whether realistic or not actually depends on a lot of things -- eg, if no drums/brass, the dynamic range need not be a lot for it to sound good)

First, for most music, compression results in really unrealistic bass and brass sounds. If you've listened to jazz live, you know how loud drums and trumpets are. The volume level varies on average about 3dB (if you're lucky). Better sounding CDs have greater dynamic range.

I've recently come across an example of a song that is really great. You can blast in on a mini-compo and it sounds great. But putting it on my humble office system, I realised that there was extreme compression going on. Bringing it back home, the almost total lack of dynamic contrast is apparent. There is only 1 volume: loud. I'm not joking, from the introduction, when she seems to be singing softer, to the first time she hits the infectious chorus 'This Kiss This Kiss', the volume doesn't change. It fact, for dyanmic contrast, what they do is that each subsequent time she repeats the chorus, the volume goes up about 1 dB. So at the end, the last chorus is about 3-4 dB louder than the first chorus. Somehow I don't think she's able to vary her singing volume in 1 dB increments over 4+ occasions, so I guess its all done in the mixing room.

This makes me really disappointed because apart from the compression, the recording is great, particularly the warmth and body in the vocals. The offending song is of course, Faith Hill's 'This Kiss' (she's also guilty of the heinous crime of unrealistic lip-synching in the music video, but thats another matter)

Why Older Recordings can sound better than newer ones...

By the 1990s, you would have expected recording engineers to have figured out how to make good recordings, but no. Too many new CDs suck. Whether a CD sounds good depends more significantly on the 'approach' taken by the recording engineers and subsequently, the mixing engineer. At one time I would have said that primarily 'acoustic' albums would automatically qualify as good sounding because its really difficult to mess it up. Unfortunately, that doesn't seem to be true.

1986 appears to be have been a year when truly great sounding CDs appeared. I do have some earlier classical music CDs from before 1986 (mainly Deutsche Gramaphone) which sounded good then, but don't sound so wonderful now. However, in 1986, Amanda McBroom's 'Dreaming' (using Monster Cables no less), and Jennifer Warne's 'Famous Blue Raincoat' appeared. They still sound great in 1997. Listen and you'll be amazed as to what could have been achieved in 1986.

But back to answering the question. It is entirely possible that current 'bad' recordings could simply have been recorded using crappy equipment. Good microphones cost a lot of money and not all recording studios may have top-class microphones and other equipment. But worse, there is a distinct possibility that the 'Digital' master is actually a DAT tape!!!

But Majority are 'thin sounding'
This truth is, a lot of older CDs are extremely thin sounding. Regrettably, I will have to single out albums from one of my favourite singers, Sting. Undoubtedly, Sting is very concerned with recording and sound quality and Dream of Blue Turtles was an very early digital recordings. However, Dream of Blue Turtles (even the Mo-Fi remix) and Nothing Like the Sun are sound very thin on my system. Only at Soul Cages and onwards did the recording gain 'body' and 'warmth'. In fact, I would hazard a guess that except for audiophile recordings, older ADD recordings would sound better than DDD recordings since digital recorders in the early days seemed to have been quite bad (except perhaps for the very expensive ones [remember how expensive digital was when it first came out?] which many recording studios may not have had)

Audiophile CDs -- Personal Opinions

Audiophile Remasters

Mobile Fidelity Sound Labs (MoFi), DCC, JVC are some companies that produce remastered versions of some very popular CDs. Amanda McBroom's 'Amanda' is also a Sheffield Labs remaster - from analogue master tape into 20bits for remastering and then dithered to 16 bits. Note: MoFi does its remastering in the analogue domain. First off, they do sound better, in varying degrees. Very Variable. I don't think I'll every amass a large collection of these twice the price CDs, but I can comment on some I have at the moment.

I have MoFi's remaster of U2's Joshua Tree which is basically the best sounding CD I have ever heard. It does not hurt that the album itself is probably one of the best ever made. Level matched listening comparisions reveal:  

  • Apparently, some sort of lowpass filter is used to remove to harshness and sibilance in Bono's voice, thus improving clarity and soundstage, yet there is no loss in treble extension (maybe they boosted the treble then applied the filter?).
  • There is more bass, and it sounds better too.
  • The whole mix sounds richer and fuller and distinctly un-digital
  • Edge's guitar sound is glorious
  • It beats the LP in many places though some might still find the LP more emotionally involving

I also have MoFi's remaster of Unforgettable Fire. Suprisingly, the differences between the US press of Unforgettable Fire and the MoFi remaster are much closer. This is suprising given the Unforgettable Fire is much older than Joshua Tree and therefore there should have been more room for improvement.

Basically, on my favourite track, Pride, there is really nothing much to distinguish the 2. I could point to tiny differences, but I cannot say that one was better than the other. The LP version blows both away. On 'quieter' tracks like Bad, there improvements become noticeable and are basically what was described for Joshua Tree (though suprisingly, the improvements are less here).

Taking a look at DCC, I've acquired Eagle's Greatest Hits compilation. Having on hand a slightly different compilation (with 17! songs) from Elektra, (probably remastered as well), I noticed improvements similar to what I heard on the MoFi CDs. The differences are not breathtaking, but clearly audible. More bass and more significantly, the electric guitars are more visceral.

The sound is slightly warmer but theres no way I would call it Analogish sounding (the term I would use to describe Amanda McBroom's Dreaming [I have the gold CD which is made in HK]), but certainly, its less fatiguing to listen to. And yes, I would say that the DCC remaster does sound better.

XRCDs
I've just got Nancy Bryan's Lay me down and I've had the opportunity to listen to Carmen Lundy as well. I am amazed at the sound quality. First off, I have to comment on its neutrality. It doesn't sound sweet, it doesn't sound warm, it doesn't sound analogish. It sounds... like music.

Soundstaging :- fabulous, and at times, it sounds like it might extend beyond the speakers. One interesting thing, when I move forward to adjust something on my amp, I'm basically at the same plane as my speakers and I can still hear the voice sharply defined and fixed behind me. On other recordings, the voice is a bit diffuse.

Bass:- It rocks too. Just listen to the intro of track 1. On full range speakers (heard it on the Platinum Trio and my BNS soundcolumn IIIs), the opening bass drum is wonderfully tight and defined.

PopeMusic

The only Popemusic recording I have is the Gold Lori Liebermann CD recorded on some stage/theatre using only 2 microphones. The noise floor is incredibly low and the soundstage quite amazing (and more importantly, amazing in the sense of realism as well as size). The sound is a bit laidback but that should be attributable to the performer. Very impressive indeed. As I understand it, Gene Pope should have gotten his modified Nagra-D by now with the SDIF digital interface (as opposed to S/PDIF or AES) which is nearly jitterless and is used in JVCs XRCD process. Lets see what improvements this brings about...

US Pressings, do they Matter?

Actually, its all quite mysterious. Are US pressings superior because they are pressing from the master tapes or are they superior because they have superior equipment? After hearing quite a few A/Bs, I don't think I want to speculate as there are too many variables.

An alternative theory could be the manufacturing process itself I suppose. But theories are theories, no one actually knows why US press CDs sound so good... Which creates a big problems for UK Hi-Fi mags who audition Hi-Fi equipment using European press CDs...

Jitter (Non-Engineer's Understanding)
Theres of course this mysterious thing called jitter, which is also responsible for CDs containing the exact same 0s and 1s to sound different. Jitter, in general (I'm no technical experts), appears to refer to timing errors caused by a corruption of the master clock signal. In CDs, jitter would be caused by data, as written onto the CD to be spaced apart at irregular distances. So 0 ... 1 ... 0 ... 1 will sound different from 0 .. 1 ... 0 .. 1 (the periods indicating timing intervals) Of course, the master clock signal is supposed to remove these errors but apparently it doesn't do its job well and thats why you have Digital Lenses to reclock the whole thing again.

If you really want to get into this stuff, look for to Robert's Hi-Fi pages for sites that give introductions to CDs and how they are encoded. I'm just trying to give a vague-ish summary.

Shawn Colvin: Increased Detail and Clarity?

As mentioned in my reviews, I've compared Holly Cole's AAD US press I can see clearly now to the Japanese pressing. I noticed woolier bass in the Japanese press (I believed others have called the Japanese pressing 'lush' -- another word for lack of clarity).

After listening to my friend's CD single of "Get out of this House" which he bought from New York, I promptly begged another friend who was going to New York to get me ,among other things :) US pressings of A Few Small Repairs. After listening; both sighted and 'blind' testing (since the CDs look identical and I evolved a way to handle the CDs so that I couldn't tell by the weight :)) I noticed

  • On the "Get out of this House", there is a hi-hat that enters the track in the beginning and carries on. On the US pressing this hi-hat is clearly heard, on the Austrian press [which the rest of the world seems doomed to have], this hi-hat is muffled and chopped off.
  • On "If I were Brave", which is basically vocals + piano I notice tiny differences in the voice. On the Austrian press, the voice seems to have more 'space' the singer appears smaller, but theres an 'aura' of 'air' (or rather, high frequency noise). On the US press, her voice is one cohesive whole. (depending on the system, this might seem 'thick' and 'congested'). Ultimately, on my system, US pressing.
  • Some of my non-audiophile friends can tell the difference too. But because I don't have test equipment, I can't confirm whether the US press is louder (it doesn't sound louder, but we can't detect tiny changes though this might have an effect on perceived sound quality).

Another thing I noticed about current US pressings are that they're Heavier and Thicker than CDs made in Europe. In addition, I'm pretty sure my Shawn Colvin album 'Steady On' which I bought in 1989, is a US pressing, and its pretty light. So this must have been a recent development or something? Could the added weight have something to do with the superior sound??

Note: Not Just the Volume
Where I've had the chance to compare, I prefer the US press in all cases even after taking into account volume differences. I can deliberately play the US pressing softer, but it still sounds better because the detail is better seperated, and volume increases do not really result in better seperation of detail.

Jann Arden, European Pressing Disaster?

Jann Arden is popular amongst my colleagues. Most of them have the European pressing. After I listened to the European pressing, I declared in a review that it deserved a rating of 7.0 because of its confused sound. Lately, someone with a lot of US pressings (because he goes to New York to buy the CDs, not because hes an audiophile)lent me the Canadian press of Living Under June. It sounded much better. Soon after, Tower brought it the US press of Living Under June for the princely price of $26.90 (ulp!) Having heard the Canadian press, I bought the US one. The quality is just as great as the Canadian one and theres an extra track. As mentioned, the differences in quality between the US and European pressings appear to be quite variable.

Idde Schultz - Malaysian Pressing better than EU pressing?

I've also had the opportunity to listen to a Malaysian press of Idde Shultz's album vs a European pressing (recorded in Stockholm anyway). The CD labels are almost identical so its easy to do a blind A/B test. After some listening, I picked 1 of them as sounding better as it was definitely less harsh and less forward sounding than the other. To my suprise, I picked the Malaysian press. Anyway, this suggests that the 'original' pressings have more treble detail than subsequent pressings and amazing, this can be beneficial for overly bright recordings.

Suzanne Vega, Increased Clarity

The only non-US pressing of Suzanne's CDs I have is solitude standing. This is a famous CD beloved by audiophiles though its far from perfect recording wise. Perhaps its the smoky voice that sounds so seductive over the best systems. Anyway, the US press is overall better, with clearer seperation etc.

But not everyone thinks so. Demoing the 2, a friend preferred the German press because the vocals were more spacious vs the US press which seemed more closed in. (this demo was done on a mini-compo, and personally, on a mini-compo, I was not sure which was better, but I picked the US press because I could hear the better instrument seperation)

Conclusions If Any:

In my view, in the case of US artists, the US pressing sounds better. I'll buy US presses here if they're not too pricey or endure until I go the the US again. I'll mailorder from CDNow and get it sent to a US relative and collect it when I visit :)