Home |
Search |
Today's Posts |
#1
![]()
Posted to rec.audio.high-end
|
|||
|
|||
![]()
People who have been following discussions on this forum for some time, know
that I have little good to say about MP3. I have gone so far as to say that I don't listen to Internet radio except as background because of the compromises of the lossy compression. While all that was, and remains, true, I've run across a couple of reasons to re-examine my opinion of the resulting sound from this source. First of all, to see from where I'm coming, we have to look at current FM practices in the United States. When I was growing up in the semi-rural Washington DC bedroom community of Woodbridge Virginia, I had a very nice Eico HFT-90 tuner connected to my bedroom stereo system. While the Eico was certainly no H, H. Scott or Marantz 10B, it was, for a non-fringe area, a really good performer that sounded excellent. That excellence continued when I added a Knight-Kit stereo multiplex decoder in 1962. In those days, FM stations, even in an urban market like Washington DC, were actually few and far between. Even in the home of the dread FCC, nobody seemed to care if stations were a little lax on their modulation peaks, and no one really cared if a station went over 100% every now and again. With no need for strict bandwidth control (in FM, over-modulating can result in one station's signal bleeding over into the adjacent station's bandwidth, with resultant distortion. It was (and remains), strictly speaking, illegal) nobody bothered to use limiters which would prevent overmodulation by literally cutting the tops off of peaks that exceeded 100% modulation. While this can be done relatively benignly, it nonetheless represents a source for distortion. With no adjacent FM stations, it didn't really matter. Another bugaboo of modern FM practice is compression. Most FM stations compress heavily. This is done for two reasons, one is to catch "dial twirlers" by being the loudest station on the dial, and two, most FM experts will tell you that their largest audience is in the car. The noise level in an automobile makes compression a real advantage, especially with classical, jazz and other forms of music given to a wide dynamic range. Unfortunately, heavy compression is like any other kind of signal processing. While a little can be done benignly, a lot sounds wretched. Most modern FM stations use a lot. The result is that modern FM sounds nothing like the FM of my youth. When I was young, FM was touted as a true high-fidelity broadcast medium. and it was. A frequency response of 20 - 20 KHz coupled with a full quieting S/N ratio of better than 60 dB meant that FM could sound great - especially live FM carried from the concert hall to the studio via expensive "Class-A" phone lines. When stereo came along in the early 1960's, the FCC chose the technically inferior but mono compatible GE-Zenith System rather than the much better (but stereo only) Crosley System. Stereo-casting had a number of disadvantages. First of all, even with full quieting, a stereo broadcast is significantly noisier than an mono broadcast. You can check this yourself. Tune in a good, strong FM stereo signal and listen over headphones. Now go over to your tuner (or amp) and switch from stereo to mono. You will immediately notice a drop in the noise floor. The second problem with stereo FM is multipath. This is caused by the FM signal bouncing off of hard, high obstacles located around your listening location. These might be tall buildings, mountains, even stands of tall trees. The result is that two signals arrive at your antenna from the same transmitter, but at different times and out of phase with one another. IOW, via "multi-paths". This confuses both the tuner and the multiplex circuitry causing terrible distortion. Now, let's look at Internet radio. Sure, it's compressed, and tou can hear the compression artifacts on some feeds. But, on feeds that stream at 128 KBs or higher, this is not as objectionable as it is at lower bit rates. But, what you do get is a signal that comes right-off the studio console (in the better streaming stations). It is quieter than an FM stereo feed, is not compressed (again, on the better feeds - they are variable) not limited, and is free of multipath. So, in the final analysis, streaming radio is, at it's best, better (IMHO) than the same station off-air (if you can get both the streaming feed and the over-the-air feed of the same station). Here in the SF Bay area, I get our local Classical station, KDFC both over the air with my Yamaha T-85 tuner and via the Internet using a Logitech Squeezebox Touch. I have found that the MP3 Internet stream is quieter, and sounds cleaner than the over-the-air signal. I have also found that the richness of choice afforded by Internet radio is simply fascinating. My favorite feed has become WCRB/WGBH in Boston. They do a lot of live broadcasts of symphony orchestras as well as smaller ensembles that broadcast from the station's own studios. Such a wealth of programming! So, yeah, MP3 is not as good as non-compressed, but even though it's problems arfe different from broadcast FM's problems, I find that the former sounds better, compression artifacts and all, than do the indignities forced upon modern FM broadcasting by broadcast rules and the economics of a competitive FM dial. |
Thread Tools | |
Display Modes | |
|
|
![]() |
||||
Thread | Forum | |||
streaming internet audio, professional application | Pro Audio | |||
Question Online Radio Streaming Software and Hardware | Tech | |||
Bypassing sound card -- worth it for streaming radio? | Tech | |||
Streaming internet radio to go | Tech | |||
Streaming internet radio to go | Pro Audio |