Sure. This will be long, because I'm going to explain how tuners work, which is the first step toward understanding why specifications do, in fact, matter to a larger extent with tuners than with other audio products. Tuners are different from most audio devices in that they not only have to pass an audio signal, but they have to produce the audio signal as well, and in so doing, have to pick that signal out from many other competing signals.
A tuner's front end can only handle so much signal, so if one lives in an area with many strong competing signals, its ability to handle them becomes very important. I know the specs on the Perreaux, and it would be unusable on many stations where I live. When a front end overloads, images of the overly strong station can appear all over the dial, or interfere with the reception of other stations. Basically, you're trying to dump a five gallon bucket of water into a milk jug. In my area, 102.9 and 103.5 will overload many tuners, producing a mixed image at 102.3, either blotting it out or making it unlistenable. A comprehensive set of numbers can help to identify whether this will be a problem with a given tuner.
Assuming we manage to receive the station we're after, we mix that frequency with another frequency to arrive at 10.7MHz, which is where it needs to be to be filtered in the IF stage. This is usually done by a ceramic filter. Remembering that FM works by deviating a signal from a center frequency, we obviously want to get as much of that signal as possible. Remember, however, that there may be another station right next to the one we want. So, a tuners selectivity comes into account. Too much and you're losing a lot of the information a station is putting out. Too little and you'll get interference from stations surrounding it. The best compromise is to have more than one filter width, ie, selectable IF bandwidths. All filters have characteristics as well, and will alter the signal to some extent. Obviously, we want as little alteration of the signal as possible, in addition to getting as much of it as possible.
Perhaps you're now starting to see a number of the problems involved. At this point, we're not even dealing with turning this signal into the stereo audio signal you hear. Skipping over the detector, we'll skip ahead to the multiplex stage. The multiplex is responsible for taking the composite signal out of the detector and splitting it into left and right audio channels. What we have, essentially, in a primary mono channel with encoded left and right information. Doing this stage right is not easy either, and many tuners do it very badly. Anyhow, once out of this stage, we're finally ready for the audio stage. Actually, in many tuners that aren't state of the art, there are still a few more filters and various traps in here.. NOW we're ready for the audio stage, where your typical audiophile company finally has a clue.
Incidentally, this is only the tip of the iceberg, and scarecely brushed the surface of the complexity of what actually has to go one to do this as well as possible.
Various specifications, once understood, will tell you how well each of these stages is designed and is doing its job. In my experience, this is almost always audible with enough tuner listening experience. Obviously, we should desire these stages to be as good as possible without sacrificing audio performance.
There is one more thing to keep in mind: Your fidelity isn't going to get any better than that of the broadcasting station, which is often quite bad with large amounts of compression and the like. One of my longstanding hypotheses is that many audiophiles like tuners that don't have great fidelity. They like tuners that mask much of the junk on lesser stations and make them listenable, not understanding that some stations actually should sound terrible. Magnum's tuners such as the MD-90 have -3dB points which are awful next to what is actually possible. However, by rolling off the edges they cover up not only the station's deficiencies, but the numerous problems in the previous stages of the tuner as well.
Unfortunately, on a great classical or jazz station that isn't compressing and actually pays attention to fidelity, the tuner will never sound as good as a tuner that sounds terrible on those overly compressed commercial stations.
There's actually a picture of a Perreaux TU3 on eBay right now with the lid off. It has a couple of audiophile grade capacitors in it, a nice power supply, and that's it. Everything before that stage stinks, and is little more than a car radio, and an old one at that. Perreaux, being an audiophile company, did what they knew, and left the rest of the tuner to car radio chip manufacturers. Magnum is now following a similar approach in all but their very best tuners.
A tuner's front end can only handle so much signal, so if one lives in an area with many strong competing signals, its ability to handle them becomes very important. I know the specs on the Perreaux, and it would be unusable on many stations where I live. When a front end overloads, images of the overly strong station can appear all over the dial, or interfere with the reception of other stations. Basically, you're trying to dump a five gallon bucket of water into a milk jug. In my area, 102.9 and 103.5 will overload many tuners, producing a mixed image at 102.3, either blotting it out or making it unlistenable. A comprehensive set of numbers can help to identify whether this will be a problem with a given tuner.
Assuming we manage to receive the station we're after, we mix that frequency with another frequency to arrive at 10.7MHz, which is where it needs to be to be filtered in the IF stage. This is usually done by a ceramic filter. Remembering that FM works by deviating a signal from a center frequency, we obviously want to get as much of that signal as possible. Remember, however, that there may be another station right next to the one we want. So, a tuners selectivity comes into account. Too much and you're losing a lot of the information a station is putting out. Too little and you'll get interference from stations surrounding it. The best compromise is to have more than one filter width, ie, selectable IF bandwidths. All filters have characteristics as well, and will alter the signal to some extent. Obviously, we want as little alteration of the signal as possible, in addition to getting as much of it as possible.
Perhaps you're now starting to see a number of the problems involved. At this point, we're not even dealing with turning this signal into the stereo audio signal you hear. Skipping over the detector, we'll skip ahead to the multiplex stage. The multiplex is responsible for taking the composite signal out of the detector and splitting it into left and right audio channels. What we have, essentially, in a primary mono channel with encoded left and right information. Doing this stage right is not easy either, and many tuners do it very badly. Anyhow, once out of this stage, we're finally ready for the audio stage. Actually, in many tuners that aren't state of the art, there are still a few more filters and various traps in here.. NOW we're ready for the audio stage, where your typical audiophile company finally has a clue.
Incidentally, this is only the tip of the iceberg, and scarecely brushed the surface of the complexity of what actually has to go one to do this as well as possible.
Various specifications, once understood, will tell you how well each of these stages is designed and is doing its job. In my experience, this is almost always audible with enough tuner listening experience. Obviously, we should desire these stages to be as good as possible without sacrificing audio performance.
There is one more thing to keep in mind: Your fidelity isn't going to get any better than that of the broadcasting station, which is often quite bad with large amounts of compression and the like. One of my longstanding hypotheses is that many audiophiles like tuners that don't have great fidelity. They like tuners that mask much of the junk on lesser stations and make them listenable, not understanding that some stations actually should sound terrible. Magnum's tuners such as the MD-90 have -3dB points which are awful next to what is actually possible. However, by rolling off the edges they cover up not only the station's deficiencies, but the numerous problems in the previous stages of the tuner as well.
Unfortunately, on a great classical or jazz station that isn't compressing and actually pays attention to fidelity, the tuner will never sound as good as a tuner that sounds terrible on those overly compressed commercial stations.
There's actually a picture of a Perreaux TU3 on eBay right now with the lid off. It has a couple of audiophile grade capacitors in it, a nice power supply, and that's it. Everything before that stage stinks, and is little more than a car radio, and an old one at that. Perreaux, being an audiophile company, did what they knew, and left the rest of the tuner to car radio chip manufacturers. Magnum is now following a similar approach in all but their very best tuners.