2. misc-audio
  3. 1070296948

What bugs you most about sound quality?

I would have to say that excessive compression is the most bothersome quality issue on recordings. Over the years I have periodically purchased recordings that I have either returned to the seller or demoted to coaster duty because of excessive compression. I cannot even begin to get near the music on these recordings as they take the very life out of the music, all the nuance of a human playing an instrument is lost in the machine-like drone of equalized dynamics.
I guess what bugs me most is that some poducers/engineers and many consumers doesn't give a rip. I can go over to the many review forums and there won't be a single negative comment (out of hundreds) about the sound quality on these garbage recordings. The producer/engineers often contend that highly compressed recordings sound better/jump out at you when played over average audio equipment. I believe this is a disingenuous argument at best. Taking the worst case scenario of a highly compressed recording played over a highly compressed radio signal, how can one tell the difference between a highly compressed recording and one with dynamic range? Perhaps you can with classical music with it's inherent greater dynamic range, but with most pop music I just don't hear it. I know I can't, which is why I purchased the drek in the first place! Therefore, if you can't tell the difference why would you do harm to the recording (at least when played over better equipment)? The contention that compressed recordings sound better on lesser equipment doesn't wash either. I have an average quality car stereo and micro system at work and I can hear the dynamics when playing cd's just fine. The compressed recording sounds more lifeless than the dynamic recording.
I don't believe these producers/engineers have any excuses to give those who have anything better than tin ears. While consumers who have only heard mediocre or worse audio rigs may be excused for having tin ears, most of these producers/engineers have to know better. I don't necessarily hear lack of transparency on most of these compressed recordings, ie. I assume the studio equipment if of high enough quality to hear the dynamic compression. In fact I have heard good quality recordings (good dynamics) come out of the same studio these compressed recordings came from.
I guess the point of this post is that we audiophiles owe it to ourselves and the greater public to bring awareness to this issue. Perhaps we can go on music review forums and post negative reviews in regard to sound quality issues. I have done so in the past and it is often surprising how many then come forward with similar responses (many of the so called tin ears are really just not able to pinpoint why they might not like a recording, they just don't know how to describe it from a technical viewpoint so they keep quiet). I assume most audiophiles don't intentionally purchase poor sounding recordings, perhaps we should make our buying decisions (or lack of) public. On the other hand perhaps I'm just spittin into the wind...
sns
eldartford
4,262 posts
 
Onhwy61...Compression/peak limiting can indeed be undone during playback so as to restore, or even enhance, the original dynamic range. The issue is that with analog logic the playback expansion will always be slightly behind the sound, so that the "attack" may be noticably different from the original. This is less of a problem where the compression during recording, and the expansion during playback are both done in accordance with one standard, Dolby or DBX.

Digital processing could eliminate this problem with delayed logic. The music signal, in digital form, would be delayed by the time interval necessary for the logic to examine it and adjust the gain. Then the gain changes could be implemented at exactly the right time in the music. It would not surprise me to learn that such a device is available. It's an obvious solution to what is evidently a widespread problem.
onhwy61's avatar
onhwy61
5,254 posts
 
Eldartford, Dolby, DBX and the RIAA equalization curve all work because they have set standards for threshold levels and compression/expansion rates. Sns and I are talking about something completely different. In modern pop/rock productions there are no set standards for the amount or type of dynamic compression. As a listener how do I know what value the engineer set the threshold level at? What compression ratio was used? What was the attack set to? Was multi-band compression used? If yes, then was it 2 band, 3 band or 5 band, each with individual threshold, ratio and attack settings. Without this info it's impossible to apply dynamic expansion that would reverse the effects of compression.

Whereas compression can theoretically be undone, limiting cannot. Limiting cuts off the peaks of the music waveform above a set level. Once the signal is altered in this way it cannot be accurately reconstructed. No amount of look ahead processing can determine whether the original signal was .1dB, 1dB or 10dB greater than the limited signal.

Modern pop/rock music involves the use of extensive limiting and compression techniques. A recording may be run through a limiter then compressed, run through a limiter again, compressed again and run through a limiter and a compressor a third time. Different settings may be used for each pass through the chain and EQ may also be applied. The idea that a simple dynamic range expander can undo this sonic manipulation is not realistic.
zaikesman
3,017 posts
 
I have to agree with Onhwy61 about the limitations of attempting to restore dynamic fidelity upon playback in lieu of some formal encode/decode protocol.

(Of course, this doesn't matter if you're listening to Raaaaaaaammmmsteiiiiiiiiinnnnnn!! :-)

Stehno: I've often commented here that IMO, a better system will commonly render even lo-qality recordings more interesting and communicative, despite their flaws, and that audiophiles who curtail their listening variety to only well-recorded material are doing themselves an artistic disservice - in fact defeating the very reason they (presumably) ever got into audio in the first place (da music!).
eldartford
4,262 posts
 
Onhwy61...I guess it is a matter of degree. "Moderate" compression/peak limiting, such as that which is necessary because of limited recording dynamic range, can be undone on playback with pretty good results. And more extreme processing like Dolby and DBX can work because both ends of the process are controlled. This is all I meant when I took issue with your blanket statement that it can't be done.

The Phase Linear Autocorrelator and the DBX equipment that I once used had knob adjustments for the compression/expansion ratios. However I agree that if they do it differently for each cut on a recording it would be tough to get them all right, although if what you say about their practices is true you could probably crank it up all the way and leave it.

I have never been to a live rock/pop concert, so I don't know how much dynamic range actually exists in this stuff. My impression is that it is always LOUD, LOUD, LOUD, but perhaps that is because I have only heard recordings.
More to discover