Just a small warning. In my experience, some selector switches will mess up the sound, and make it a PITA to hear the effects of whatever you are trying to test.
Better to simply make multiple test units that are each different in only one variable (they should look, weigh and feel identical), and label them in a manner that has nothing to do with anyone's perception of what was put inside. The "switching" is accomplished by physically swapping the units (thereby removing the selector switch from the equation), but since the listener has zero idea of what is inside, this shouldn't be an issue.
More comments on these previous posts:
http://diyaudio.com/forums/showthread.php?s=&postid=60831#post60831
http://diyaudio.com/forums/showthread.php?s=&postid=62284#post62284
regards, jonathan carr
Better to simply make multiple test units that are each different in only one variable (they should look, weigh and feel identical), and label them in a manner that has nothing to do with anyone's perception of what was put inside. The "switching" is accomplished by physically swapping the units (thereby removing the selector switch from the equation), but since the listener has zero idea of what is inside, this shouldn't be an issue.
More comments on these previous posts:
http://diyaudio.com/forums/showthread.php?s=&postid=60831#post60831
http://diyaudio.com/forums/showthread.php?s=&postid=62284#post62284
regards, jonathan carr
well...
valid point, jonathan, but i think for the purposes of this test a high-quality switch like Elma would be alright. you could argue trace degradation but that's really nitpicking... right now i'm using a cheap little switchbox i made from a plastic box and radio shack toggle switch, to switch between my surround sound and music preamps, and even with this extra bit in the signal path i hear little if any degradation worth talking about... i can still clearly hear differences in cables etc. i think if the difference in sound is so subtle as to be obscured by a high quality switch, then perhaps the difference is not worth making out anyway...
valid point, jonathan, but i think for the purposes of this test a high-quality switch like Elma would be alright. you could argue trace degradation but that's really nitpicking... right now i'm using a cheap little switchbox i made from a plastic box and radio shack toggle switch, to switch between my surround sound and music preamps, and even with this extra bit in the signal path i hear little if any degradation worth talking about... i can still clearly hear differences in cables etc. i think if the difference in sound is so subtle as to be obscured by a high quality switch, then perhaps the difference is not worth making out anyway...
Dorkus and other potential testers
If you guys are serious about doing any kind of sound test in the NY area you should be ready to account for several environmental effects. Some of which are:
ambient noise: this is particularly troublesome. It is not uncommon to have 50dB of background noise to overcome in testing your music and that ambient noise seldom has a nice even white or pink profile. Frequency cancellations that move up and down your response curve are quite probable.
humidity: you will need to measure and control the moisture content in the room because it will effect the sound. It may also effect the sound of the speakers depending on the materials used in their construction.
mains variation: NY power delivery is pretty good in most areas but be prepared to contend with slow surges and ebbs depending on the times you listen.
I know this from the experience in my own tests which were prompted by the ideas that came out of this thread
If you guys are serious about doing any kind of sound test in the NY area you should be ready to account for several environmental effects. Some of which are:
ambient noise: this is particularly troublesome. It is not uncommon to have 50dB of background noise to overcome in testing your music and that ambient noise seldom has a nice even white or pink profile. Frequency cancellations that move up and down your response curve are quite probable.
humidity: you will need to measure and control the moisture content in the room because it will effect the sound. It may also effect the sound of the speakers depending on the materials used in their construction.
mains variation: NY power delivery is pretty good in most areas but be prepared to contend with slow surges and ebbs depending on the times you listen.
I know this from the experience in my own tests which were prompted by the ideas that came out of this thread
that's why i need 2 - 3 months for the test. over that period of time, those environmental effects will average out and i'll get a "feel" for the intrinsic sound. it's also why i don't like A/B/X tests, switching back and forth and back and forth at gunpoint. so to speak. very valid factors however - in particular, power line fluctuations and ambient noise are key factors, my system sounds so much better in the middle of the night when it's quieter, both acoustically and electrically.
Dorkus:
I assure you that on more than one occasion I have been led astray by testing with switches in place, which is why I now prefer to use them as little as possible. The methodology that I have outlined is one that I have used for years and know will work.
Yes, it is a little more work building an separate chassis for each variation, and yes, it is more convenient to throw a switch than physically swap boxes. But when you are trying to conduct a test, my experience is that it is better to remove as many variables and potentially questionable areas as possible.
I fully agree with Nania, and will add that setting up the speakers with obssessive care and attention to detail usually pays hadsome dividends. Another common source of masking is excessive ground current activity. Observing proper power polarities, using no more audio components than you absolutely need at any given time, and using single-point star-grounding techniques to a real ground are all worthwhile.
regards, jonathan carr
I assure you that on more than one occasion I have been led astray by testing with switches in place, which is why I now prefer to use them as little as possible. The methodology that I have outlined is one that I have used for years and know will work.
Yes, it is a little more work building an separate chassis for each variation, and yes, it is more convenient to throw a switch than physically swap boxes. But when you are trying to conduct a test, my experience is that it is better to remove as many variables and potentially questionable areas as possible.
I fully agree with Nania, and will add that setting up the speakers with obssessive care and attention to detail usually pays hadsome dividends. Another common source of masking is excessive ground current activity. Observing proper power polarities, using no more audio components than you absolutely need at any given time, and using single-point star-grounding techniques to a real ground are all worthwhile.
regards, jonathan carr
i agree that a no-compromise approach, with identical black boxes and no switching, is ideal, but it also makes testing very difficult. also the expensive of building several identical modules becomes prohibitive. i would do it if i had the money/time/motivation, but i'm not sure how realistic it is. but yes, that is the correct way.
i agree about the ground current thing. grounding is a real art form.
i agree about the ground current thing. grounding is a real art form.
Good morning all.
Good input on the test. I agree that using simple test modules and swapping them may be the most revealing. I don't think we are that far along yet. At this point I think we need to consistently be able to detect a "difference" and then be able to measure what we think we hear.
I like Sy's idea for a test box because it usable in any system via tape loop or betweem the preamp and power amp. and is simple to use. We need to start with something that should be easy to detect. Maybe cheep electrolitic v good plastic caps. We need to prove, at least to ourselves, we can hear that and then learn to measrue the difference.
I do think we should use multiple "listerner testers". If Dorkus wants to take three months to deside if he hears a difference thats fine. Thats how he desides. Others of us like ABX testing and I'm one of those. I do long term as well, some times it shows up subtle things I miss with ABX. With only one test box this could take forever if we want to do a fair number of samples or if we want to change the parts under test.
Later
Bruce
Good input on the test. I agree that using simple test modules and swapping them may be the most revealing. I don't think we are that far along yet. At this point I think we need to consistently be able to detect a "difference" and then be able to measure what we think we hear.
I like Sy's idea for a test box because it usable in any system via tape loop or betweem the preamp and power amp. and is simple to use. We need to start with something that should be easy to detect. Maybe cheep electrolitic v good plastic caps. We need to prove, at least to ourselves, we can hear that and then learn to measrue the difference.
I do think we should use multiple "listerner testers". If Dorkus wants to take three months to deside if he hears a difference thats fine. Thats how he desides. Others of us like ABX testing and I'm one of those. I do long term as well, some times it shows up subtle things I miss with ABX. With only one test box this could take forever if we want to do a fair number of samples or if we want to change the parts under test.
Later
Bruce
i'm still thinking about the circuit, and how we're gonna wire it up. need to think out the details.
anybody have parts suggestions for the tweak circuit? i'm thinking Hovland feedback cap, Black Gate coupling cap, maybe Caddock MK132 resistors... for the cheapo version, i think we should use "crappy" metal films. those Radio Shack carbon films actually don't sound so bad.
anybody have parts suggestions for the tweak circuit? i'm thinking Hovland feedback cap, Black Gate coupling cap, maybe Caddock MK132 resistors... for the cheapo version, i think we should use "crappy" metal films. those Radio Shack carbon films actually don't sound so bad.
silver mica is actually pretty nice stuff. but we only need two caps - .22uF film and 10uF 'lytic. i'm not sure what to use for internal wiring, i like Cardas litz but it's harder to work with. any decent solid core with non-PVC insulation will do.
oh yeah, we need a couple larger 'lytics for the power supply bypass. i like Nichicon Muse or Elna Cerafine/Silmic. i hear OSCON is good too but i don't have any.
oh yeah, we need a couple larger 'lytics for the power supply bypass. i like Nichicon Muse or Elna Cerafine/Silmic. i hear OSCON is good too but i don't have any.
I have to go see the Doc this afternoon so I'll talk to you guys later. I'm thinking the simplest possible active circuit would be the best choice, a CCS follower perhaps. I haven't heard the one you favor Dorkus. It could well be the best choice. I have the parts to build up a follower that I have used for years and as far as I can tell has all most no signature of it's own. Maybe we could each build a one channel active stage (no switching) to sent to the others for evaluation and pick one as the test circuit.
Later
Bruce
Later
Bruce
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.