You are welcome, and, irrespective of any of my own findings, I would always recommend others to try these things out for themselves.
The only real value in listening to what other people have to say in these cases (including me!) IMHO, is when deciding whether it is worth expending the time and money on such a trial, in the first place.
It may or may not have occurred to you, but don't forget in a case like this that you are not judging the *overall* sound, as much as any*differences* caused by the suggested 'reversal' in question.
Accordingly, when setting things up, it can be on a much more temporary basis (we call it a 'jury rig') when, unless you have done something very drastic, you will still most likely hear the changes, even if the overall quality of sound is not so good.
Peter Danel recently showed a very temporary way of doing this, which I thoroughly uphold, as I have done exactly the same thing, myself, many times.
I seem to recall that some resistors which were under test by him were literally lodged between two connectors (without even soldering) which is fine, provided that if the connection fails for any unexpected reason, it does not cause any damage to the chip etc.
If Peter reads this, maybe he will direct you to his post about this, as I have been posting in many threads recently, and I don't recall which one it was in.:nod:
Here's the link: http://www.diyaudio.com/forums/showt...835#post123835 and the pic of a test setup.;)
Problem with these thingies is to investigate the effect that is supposed to be audible while excluding other factors.
Very probable, a 220 k resistor will sound slighty different from a 110 k resistor (besides the change in value). It all depends on the geometry and material change in the resistor for increasing the value from 110 k to 220 k. Also, there could be noise/temperature effects (like mentioned, the 1/4 w and 1/8 w issue).
Thanks again, especially for your kind manners. I will do the 'rig" hopefully this week.
["Maybe a very obvious point, but do you realize that the value of the resistors is halved by putting two resistors parallel ?
And I wonder how you manage to get the exact same value from the single resistors for the parallel resistors..."]
Yes, this is obvious and i have realised this. In the attached drawing, in the component list you will see the details of the resistor values.
Thanks for the link. I had lost track of it.
What i am thinking of now, is, that i have to do listening tests for each resistor position (there are three) at a time. And for each resistor the mode has to be: 1.Single resistor in one direction. 2 Single resistor in reverse direction. 3. Two parallel resistors in one direction. 4. Two parallel resistors in reverse direction. 5. Two antiparallel resistors. This is 15 tests in total (confining myself to one resistor brand)! This may turn to be the subject of my thesis with the tittle: "Trying to build a logical statement within an unlogical sentence"
Even taking into account resistor directivity I wouldn't have thought that it could make such a radical change in an amplifier that gpapag suggests. I think bad solder joints in point to point wiring on IC's are more probable seeing the confined spaces and the fact that sometimes you have to resolder the same joint.
In case some of you didn't see my posting in another thread I made some A/B listening tests over the weekend between an LM3875 and an OPA549.
I much preferred the OPA 549. It sounds more controlled ,neutral and robust. Some differences were more obvious with piano music (always a good test I think).It is interesting because pianos can sound either too woody or too steely depending on the amplifier. I think the OPA was blending the two in a correct balance while the LM was a bit "steely".Also the OPA could be driven much louder without sounding stressed, maybe as much as the ALEPH 5s. I can't corroborate this scientifically because I lack a scope of course.
Perhaps though all the differences were due to different resistor directivity ;)
I made this set up in 5 min. and it's good enough to compare different brand of resisors. If there is any noise pick up, it will be the same for all componenets under test and the difference is detectable anyway. I can clearly hear the difference between Rikens, holcos and Vishays. I wouldn't have the courage to built two different amps just to compare resistors, unless you do production of course.;)
What you think is a difference between resistors may just be a difference between poor contact with the connectors. Radiation is not steady state, so when performing different tests at different times, one may be subject to radiation that the other is not.
I don't understand how you can go to the extreme lengths to control vibration and "resonance" in your system (based on pics you've posted in these forums) that you do, and then dismiss such effects as completely inconsequential in this type of test.
Do these things matter or don't they?
Hmm.. I really think you need to solder those suckers in there, Peter. Make up one female-male RCA adapter with each resistor type, or perhaps one RCA adapter with a cheaper connector for the resistor (such as a standard 0.1" pin header, gold plated), such that this is a constant variable for each test. Then do a proper job soldering the resistor leads - as they would be in a real circuit. Without soldered joints, you have no idea how the differences in the oxidation properties of the lead materials / coatings and variances in contact pressure are affecting your results. The process of soldering will clean off the oxides, and provide a good, trustworthy connection, thus eliminating this really big unknown. And for the record... I've found low-force connections like this usually sound pretty crappy, reliablility issues aside. C'mon Pete! I know you can do a more scientific test than this! Let's see those construction skills shine! ;)
It may surprise some of you, but I'll always choose convenience over reliability.;)
This supposed to be only a temporary and easy introduction to resistors comparison tests. I'm not claiming that this is a best way to do it, but if someone is reluctant to try those tests because of complexity of a setup, this should show a simple way to do it. If you try my method and will find that there is something to the differences presented, by no means go for more elaborate testing and use solder joints and all the rest of "perfect sound forever" accessories. For me this was good enough.;)
The Sun Don't Shine Out Of My Butt Either.
Have you tried reversing the polarity of the input and output connections of your audio isolating transformers, in order to guage any difference ?.
As part of my test bench, I have a pair of audio isolating transformers too, except that I have a DPDT swith wired as a reversing switch in series with the primary.
In previous experiments, I have reversed the polarity of both primary and secondary connections, and detected an audible difference.
I have also made non-directional star-quad interconnects too, and these made a non-subtle audible difference.
As far as sighted VS non-sighted testing goes, your comment is not woth the bandwidth.
You really ought to realize by now that amongst us here, are listeners who are easily able to divorce expectation from any listening testing, and if you are unable to do so, then please understand that other listeners are - Ummmm, fellows like Peter, Bob, Frank, myself and others I have missed mentioning.
The key is an open mind - indeed if you convince yourself beforehand that these such effects are not possible, well then as a matter of course you will never hear them - period.
You know the origin of the Universe is hitherto still unexplained.
If you have a workable theory, then by all means share your wisdom with us.
Eric / - open minded on most things, and intolerant of sarcasm and baseless naysayers.
|All times are GMT. The time now is 07:27 PM.|
vBulletin Optimisation provided by vB Optimise (Pro) - vBulletin Mods & Addons Copyright © 2013 DragonByte Technologies Ltd.
Copyright ©1999-2013 diyAudio