Summary of the Sound Guideline discussion Last update Fri 11-25-1994 Contains: Jim Widner's original suggestion to adopt a "standardized" description or metric of sound quality, including two of his own description sets as well as one suggested by Bob Burnham. plus Additional suggestions and comments as left by subscribers to the OTR Newsletter, including: Ken Piletic Bob Burnham Frank Gilmore Frank Passage To add your $.02 worth, leave a note in the OTR newsletter by sending email addressed to: otr@airwaves.com ====================================================================== Executive Summary (through Fri 11-25-1994 ) 1. A (voluntary) "Standardized" sound quality metric is desired, by at least some collectors. 2. A unidimensional (one figure, such as "very good" or "poor") metric cannot accurately portray all the pertinent qualities of the recording, and is excessively subjective. 3. A multidimensional metric ("SNP"?) could consider Sound quality, background Noise and Program content, but is also subjective. 4. Some collectors/vendors are reticent to adopt a standardized system because of A) Concerns about who sets the standard B) Subjectivity of the standard C) Difficulty in implementing the standard D) Disinterest on the part of vendors E) Complexity of understanding / implementation 5. A "Period Code" (representing the decade in which the original recording was made) might be worthwhile as part of the metric. ====================================================================== ====================================================================== Date: Tue, 8 Nov 1994 21:36:57 -0500 (EST) From: Jim Widner Subject: Sound guidelines Some time back I had several conversations with other collectors of Old Time Radio and the discussion moved into the area of sound quality. We had all seen various descriptions by vendors on how they determined the sound quality of their own tapes. To say the least, it was generally a subjective process. This has gotten me thinking that given the increasing number of collectors who are online that perhaps some kind of standard could begin to be developed that would help all of us when we are either trying to describe our own collections, or making decisions on what is good to buy. I have been burned a number of times by tapes that were described as having reasonably good sound, only to find that they were generally unlistenable. So within the forum of this digest, I would like to hear from others on their interpretations of sound quality. I propose to offer some suggestions base d upon feedback that I think would create this "standard." What I will offer initially below are some of the descriptions that others - vendors usually - have given to levels of sound quality. VERSION 1: EXCELLENT: Near perfect sound, with few or no flaws of any kind. Generally comparable to current broadcast standards. VERY GOOD: Clear, distinct sound, with only a few minor defects, such as surface noise or scratchiness to mar an otherwise fine sound. GOOD: Audible, listenable sound, but with decided flaws, such as surface noise, hum or other impairments. On the whole, still listenable, but less than easy listening. FAIR: Recommended only for devotees of the show in question. Sound is badly muffled or distorted, and listening requires full attention. VERSION 2: VG - no major sound defects. G - means there may be a little wow and flutter, hum or background noise. F - means you'll have to listen closely. VERSION 3 ARTICLE FROM RLL ON THE AIR BY BOB BURNHAM GREAT SOUND - There should be very little tape hiss or disk hiss. The show is very clear. You can hear the characteristics of the performers voices including 's' words, but without distortion when they are spoken. In addition, the tonal spectrum sounds very natural on Great Sound. You can hear bass during music bridges and openings, the performers' voices are very resonant (and deep- when appropriate) yet are not muddy sounding. There is no distortion, no hum or buzz whatsoever on the recording. CLEAN SOUND - This is a variation of the above, but the show sounds very flat and lifeless. The tonal balance is not appropriate for natural sound. This was probably because when the transcription disk was copied, the wrong equalization or type of pre-amp was used. It is also a characteristic of most 1930's shows. Often, there is not much that can be done to improve shows like these. Equalizers can't boost base when there is no bass to boost! Other than the flat sound, the show is very clear and has low hiss, and no other major flaws. VERY GOOD SOUND - This show is probably the most common type of sound found in today's collectors' hands. The show can have a great tonal balance like the Great Sound category, but may have some other flaws like too much hiss, some hum or rumble (from the transcription player). Using sophisticated equipment, it May be possible to upgrade it to Great Sound. FAIR SOUND - is caused by one or more of the following things: 1) Too many generations of trading in analog format. 2) Dirty tape heads 3) Worn tape heads 4) Equipment not operating properly. 5) Poor recording tape. POOR SOUND - This show requires close attention in listening to, to understand. Muffled sound is the usual description given to this flaw. Some shows only exist in this form due to the fact that the original source master/disk became badly deteriorated and/or it was recorded on a home wire recorder off the radio. In Most Cases, however, Poor Sound is the result of the same bad practices mentioned for Fair Sound. In some cases, sound processors can make the show slightly clearer, but usually at the expense of adding distortion, excessive hiss or other problems. ------------- As you can see there is a bit of variation. Bob Burnham is in the sound engineering business and so, his is more detailed. What I would like to see initially is 1) How should the sound be defined - i.e. Good, Very Good, Fair, etc. or should some kind of rating system based on a numeric scale - e.g. 1 Excellent, 2 Good, etc., then 1.5 might be very good and 1.7 (if this is possible) even better. 2) Is this too subjective a process? Personally, I believe that sound quality is quantifiable based on levels of hiss etc., though some would argue that electronic equipment is necessary to determine that. If some guidelines are layed out for all to use, then the subjectivity goes away even if the guidelines are based on a non-equipment basis. Anyway, there it is as a start. I'd love to see this discussion take off and I would like to offer similar proposals to the various radio clubs. ----------------------------------------------------------------------- Jim Widner | jwidner@infinet.com ----------------------------------------------------------------------- ----------------------------------------------------------------------- From Ken Piletic: Unless something is really wrong, like "hum", or "hiss", or "bassy", it is difficult to assign a rating. Everybody's ear is different, so what sounds "good" to me might sound "poor" to somebody else. I find that tapes I rated "good" 20 years ago no longer sound good to me. Other tapes that I rated "poor" or "fair", don't actually sound that bad. It's subjective, I guess. Still, I agree with Jim that some kind of rating system should be used. We now have "equalizers", which I use when I play tapes to make them easier to hear, but I don't usually "equalize" tapes when I make copies (there are exceptions to this). If a tape is particularly "bad" I make a note of that, but I don't try to rate every show any longer. Happy listening -- Ken Piletic - Streamwood, IL ----------------------------------------------------------------------- ----------------------------------------------------------------------- From: C. Frank Gilmore" I agree with Ken completely. I went one step further and even added gradations of "-" and "+" to the listings in my catalogs. The borders between G, G+, E-, and E often are subjective and no two people would agree. ----------------------------------------------------------------------- ----------------------------------------------------------------------- From: passage@pluto.dss.com (Frank Passage): I'd like to add to the stream of comments on labeling tapes for sound quality. Maybe there should be a SINPO system used by short wave listeners. That system has a separate rating for Signal strength, Interference, etc. I'd suggest for OTR we have Sound Quality, Background Noise, Program Content. Sound Quality would address the sound quality of the program itself, Background Noise would cover hiss, scratch noises, etc. Program Content would indicate included commercials, lead-in or exit material. Rating Sound Background Program Quality Noise Content ------ --------- -------------------------- --------------------------- 5 excellent none includes lead-in, ads, exit 4 very good almost none includes lead-in, exit 3 good some includes ads 2 fair competes w/program program only 1 poor overrides program in spots program is incomplete Some of the suggestions from the other posts could define what Sound Quality of excellent is, etc. Frank [Frank's method would require three numbers, one from each column. Example: 433 Means "Very good sound quality, some background noise, and includes ads" - lou genco] ----------------------------------------------------------------------- ----------------------------------------------------------------------- From bbs!platecap@gatecoms.gatecom.com (Bob Burnham) Wed Nov 23 17:09:01 1994 QUESTION: By whose "standard" would this be determined/judged? AGREED. But often that [rating to describe several parameters of the recording] is not the intent. The VG / G ratings are to my understanding, merely an OVERALL rating. Additional notation could be added (i.e. very good + / slight tape hiss THE PROBLEM with a very complex rating system is only a small minority will want to bother with them at all, and it would again be difficult to control the use of such as system as far as accuracy from one person to another. And yes, very subjective. >it is surprising how accurate/repeatable subjective measures > can be! YES perhaps among a small circle of elite collectors, but not the majority.. IF IT'S POSSIBLE to do, best of luck. Since the beginning of time there have been variations of sound rating schemes, ranging from letter gradings, numerical or simply words like very good, etc. What I found was the average person's "GOOD" was really only a FAIR to POOR as far as I was concerned. If rated to be good then it should be GOOD SOUND! - My method of hndling it is to use NO sound ratings, but only list major defects. Another collector, Ed Carr in PA has only two ratings: "OK" and "NICE." Ed's "OK" is much better than most people's "very good." I think a factor not mentioned is the reputation of the person. If that person is known for great sound, then you can take their ratings more seriously. Others are less consistant in their analysis. Bob Burnham; BRC Productions; P.O. Box 2645; Livonia, MI 48151 fax voicemail (313) 277-3095; internet platecap@gatecom.com ----------------------------------------------------------------------- ----------------------------------------------------------------------- Date: Wed, 23 Nov 1994 17:39:06 -0500 From: KENPILETIC@aol.com As I mentioned in an earlier article, I think that information as to sound quality should be given whenever possible. Whether this information is listed as SINPO (abbreviated to SNP), or E,VG,G,F,P, or some other meter, it should be accepted by the OTR listeners/traders/collectors at large. There are far more otr listeners (collectors, buyers, sellers, traders, etc) than we might imagine. The number of people who subscribe to the otrdigest is minuscule compared to the WORLD WIDE total of people who are "into" OTR. You can bet than any "standard" proposed here will not be accepted by the OTR people-at-large. ... Unfortunately, the rest of the OTR-world consists of people who are mostly uninterested or unaware of any show-rating system. ... most vendors don't want to sell poor quality. Their reputation is at stake. The trading of programs between/among collectors is something else. Every time a program is copied, there is some loss of quality. Individual traders should check the quality of outgoing material and make a note of anything unusual or exceptional (good or bad). This is where (I think) the rating system is needed. ... It's up to you and your partner to decide for yourselves how (or if) you want to do this. Within a group (such as the otrdigest people) a "standard" can be devised that everybody can agree with and use as a basis to compare programs/tapes in the individual collections. ... The proposed "SNP" would be a good system for this. However, please don't expect the Hundreds of Thousands of people who buy OTR to accept any single format. Yes, Hundreds of Thousands of people are "into" OTR. Some to a large degree, most to a lesser degree. Most people understand "Good" and "Bad", which was the basis for the widely accepted rating system (E, VG, G, P, F). Few listners-at-large want to go beyond that. It is unlikely that commercial vendors will be changing their rating systems, even though SNP may be a "better" system. Of course some vendors might add an SNP rating to their existing systems, but don't count on it. These people are in business and it costs money and time to make any change. If making a change will not significantly increase business, they probably won't change. In general, I don't plan to change, either. I recognize SNP as a good system, and actually superior to E, VG, G, F, P. It's not that I'm opposed to change, it's just too much work. For those of you who have smaller collections, please consider carefully what kind of rating system you want to use, discuss it among your fellow collectors (otrdigest'ers), and decide on something agreeable to all. Again, there should be a rating system for all shows. ----------------------------------------------------------------------- ----------------------------------------------------------------------- Date: Wed, 23 Nov 1994 23:22:38 -0500 (EST) From: Jim Widner The idea for using a coding system similar to the shortwave SINPO code is a good one. I agree that including Program Content might not be a good idea as there are variations of recording sources. Instead add the following to the Sound/Noise concept: Some kind of code that would place the time period for the program. Because technology changed and a good SHADOW might not be the same sound as a good X MINUS ONE, the code would help place the rating in a realistic light. Thus a highly ranked sound, noise combination for a 1930's program might still sound somewhat tinny due to a lack of equalization equipment that might have existed in the 1950's. A highly ranked 1950's program might have more fidelity. Finally, add an overall rating that could either be weighted by the previous three items. Thus if 1920's recordings are a 1, 1930's = 2, 1940's = 3, 1950's = 4, 1960's = 5, etc. then a 1930's recording with a high rank of 4 for sound and 4 for noise would carry a 442 rating that might carry an overall of either 3.3 or 3, depending on if integers are only allowed. Then a 1950's recording with a 4 for sound and 4 for noise would have a code of 444 and would have an overall of 4. Naturally, this would be a higher overall because of the improved techniques available to sound technicians in that period versus the 1930's. People would have an explanation that the Sound/Noise is a determining factor, not necessarily the Overall. Yet at the same time, for sound closer to what we hear today would carry a higher overall because in the great scheme of things it would normally be better. A 1930's recording of 442 with the overall of 3 (or 3.3) versus a poor 1950's with a rating of 224 would carry an overall of 2.6, or 2, or 3 (rounded up). Now if it was possible to determine, it would be statistically nice if a 1950's recording with an overall of 3 would sound almost the same as a 1930's with an overall of 3 too. By sounding the same, that would be in terms of the Sound/Noise aspect. I am not sure this is possible, but would be interesting and scientifically sound. I'd love to hear others ideas on these problems: 1) Is the SINPO concept the best way? 2) Should Program Content really be part of the "Sound" guideline? 3) Would it be worthwhile having a "period" code (i.e. 1920's, 30's, etc)? 4) Would an overall be desireable? 5) Is it possible to have the overall work according to the last paragraph of this article? ----------------------------------------------------------------------- ----------------------------------------------------------------------- Thursday, November 24 - 1994, 1:20 PM cst From: kenpiletic@aol.com In issue 71, Jim Widner asked five specific questions regarding Sound/Noise in his interesting article. I, too, would like to see the thoughts of fellow otr collectors on these questions. Even though I wrote a rather extensive article (also in issue 71) about Sound Quality Ratings I will offer my "answers" to Jim's questions herewith: 1. Q: Is the SINPO concept the best way? A: It is a good way, but you almost need a chart to understand it. It may be too confusing for the occasional user. 2. Q: Should Program Content really be part of the "Sound" guideline? A: No 3. Q: Would it be worthwhile having a "period" code? A: It's not necessary. Programs with dates (years) on cue sheets are automatically "period-ized". Unknown dates (years) become moot. 4. Q: Would an overall be desireable? A: Yes, but this adds to the confusion (back to the chart). 5. Q: It is possible to have the overall work ... ? A: Yes, but even more confusing (two charts needed?) In a nut shell, I believe SNP with all it's features is too complicated for the average OTR collector. ----------------------------------------------------------------------- -----------------------------------------------------------------------