Not everybody judges by SINAD: https://www.stereophile.com/content/pass-labs-hpa-1-headphone-amplifier ...is quite favorable as are most other reviews.
Pass Labs' HPA-1 offers superb measured performance that reflects equally superb audio engineering.—John Atkinson
Besides, when designing for one company then that company may have its own preferences.
In another context transparency can certainly be the main goal:
Hi David; Thank you for the excellent review. In particular you identified one of my engineering objectives over the years as I explored the fascinating field of MC transformers. Some are intended to add a little “seasoning” to the listening experience. Others, such as the MC-1, are intended to be virtually transparent so that every nuance of the master is accurately reproduced. I am proud to have worked with Ned and Justin, as well as Quadratic’s chief engineer Jam Somasundram, to help meet their objectives with custom as well as publically available designs. They, and numerous others, have been inspirational in the eternal quest for the perfect transformer. Kindest regards, David Geren President CineMag, Inc.
https://trackingangle.com/equipment/shootout-at-the-sut-corral
Same designer in both cases, different goals.
Pass Labs' HPA-1 offers superb measured performance that reflects equally superb audio engineering.—John Atkinson
Besides, when designing for one company then that company may have its own preferences.
In another context transparency can certainly be the main goal:
Hi David; Thank you for the excellent review. In particular you identified one of my engineering objectives over the years as I explored the fascinating field of MC transformers. Some are intended to add a little “seasoning” to the listening experience. Others, such as the MC-1, are intended to be virtually transparent so that every nuance of the master is accurately reproduced. I am proud to have worked with Ned and Justin, as well as Quadratic’s chief engineer Jam Somasundram, to help meet their objectives with custom as well as publically available designs. They, and numerous others, have been inspirational in the eternal quest for the perfect transformer. Kindest regards, David Geren President CineMag, Inc.
https://trackingangle.com/equipment/shootout-at-the-sut-corral
Same designer in both cases, different goals.
Last edited:
Regardless I'm 110% sure that his ears don't work with my brain so I'll stick with my own ears. I suggest everybody else does the same.
Aha, I did a search but somehow . . . . . . Thanks.
Yup! But that wasn't really what I was after . What I call calibration is understanding how somebody interprets what they hear so that you can have a worthwhile conversation. Language will be misleading if we assume the listener assigns the same meanings as we do to the words we are using.
The way I sum it up - Ears are not all the same shape, the mechanical parts not all identical, and minds with their orientations certainly differ as well. It would be silly to think that when you say something sounds good I'll know exactly what you're talking about without getting calibrated somehow. One way of doing it would be getting to know what you think of other things that I've also heard. Hence my interest in follow-up.
To return to the initial purpose of this thread, the 1990 user manual of the B&W 800 Matrix recommends at minimum bi-wiring.
Now, I regard B&W as being a company doing quit a lot of decent research so I think they didn't write this out of the blue.
I happen to own these speakers but never tried bi-wiring. I might one day.
Note the reason why they recommend it:
Now, I regard B&W as being a company doing quit a lot of decent research so I think they didn't write this out of the blue.
I happen to own these speakers but never tried bi-wiring. I might one day.
Note the reason why they recommend it:
Yes. Ok perhaps it does minimise the effect, perhaps it doesn't. But can I or anyone else hear that difference?
All I know is I couldn't. I ran bi-wired (qed 42 strand) b&w p6 speakers and couldn't detect any change in sq from bi-wiring.
So I went back to 'normal' cable runs and gave the spare cable to a friend who'd been using bell wire. She couldn't tell the difference and complained that the cable was bulky and impossible to hide 😱
Next week I'll bi-wire them and I will report if I can hear any difference.
Of course, these speakers can handle 800W each and I'm only playing at 1 or 2 watts in my apartment.
There might be a difference at 'concert level' but this I can not test.
Of course, these speakers can handle 800W each and I'm only playing at 1 or 2 watts in my apartment.
There might be a difference at 'concert level' but this I can not test.
As a musician the room was very important to me. I always felt claustrophobic in practice rooms. So, maybe I listen to the room more than most while listening to a recording. How do you deal with different recordings where the room is obviously different and the reverb and weight of the recording is dramatically different? Sometimes it's difficult for me to understand where the soundstage is, it varies sooo much! Jazz recordings from the 50's are horrible with sax on the left piano on the right, bass and drums in the center with Miles Davis. Uggh... Most modern recordings are pure crap mostly due to the high compression, how can you hear anything when it's all at the same level? The best thing ever for anyone who likes music is to go to an acoustic performance where there are no microphones or amplifiers involved.
The listening room needs to be well treated, and the listening position in the room needs to be less than the critical distance from the speakers. The critical distance is where the direct sound from the speakers is the same volume level as the listening room reflections. Then at least its possible to better hear what is on the recording and the quality of reproduction.
This is where dac quality can make a substantial difference. We are auditioning a discrete multibit dac here now that is going for around $450. I like it pretty well in some ways, but it can't match the soundstage that a much better and more expensive dac can reproduce. Also, I had to leave the multibit dac running around the clock for about 3-days before it kind of settled into its most stable state. That's the sound I would expect if the dac is normally kept powered on at all times.
In any case, we have no control over what is put on a CD. If its well recorded, then its possible for the sound to be better than most systems are able fully reproduce. IOW, the limitation isn't usually in what can be had from 16/44 CD audio. It usually the reproduction system quality, and or the quality of the original recording which is then encoded onto the CD. That said, high res can be even better in subtle ways. For some people and some recordings the benefit of high res can seem more than just subtle.
For Jazz, how about for some of the content on "The Essential Herbie Hancock?" Some of the later recordings on the two disk set have a better soundstage as compared to earlier recordings. The song, "stars in your eyes," can to some extent be used as a reference recording as it includes vocal sounds (which can be very telling of reproduction quality, at least to human perception).
I might also mention there is also a famous acoustic folk music CD recorded in a living room with one stereo mic. Its called "Redbird."
Last edited:
Might be worthwhile to try low-cost star-quad cable. Some of it sounds better with the rubber jacket stripped off, but leaving everything else inside perfectly intact (conductor geometry needs to remain exactly the same, so cotton-like spacing material needs to be left in place). A braided plastic cable sleeve can then be slipped over the cable to make it look nice and be more durable. https://www.canare.com/speakercable In case of any uncertainty, conductors of the same color should connected together at the ends where the connectors are.
Last edited:
A low cost mod as in post #7
The elimination of binding posts, and maybe a redo of position of the crossover components along the line. I mean, components like an inductor or a capacitor, could be placed at 1/2 or 1/3 of total length...
The elimination of binding posts, and maybe a redo of position of the crossover components along the line. I mean, components like an inductor or a capacitor, could be placed at 1/2 or 1/3 of total length...
Maybe not that exactly. If using zip cord then according to some reported listening results, the use of "lumped impedance compensation" (based on measurements) could possibly be helpful. A search for posts by Hans Polak in the thread at: https://www.diyaudio.com/community/threads/zip-cord-for-speaker-test.371099/ can explain more.
Last edited:
No, although I wonder how this particular speaker will represent audibility thresholds since we are talking about small differences. (Also, I presume you intend to re-tip/re-make your connections while keeping the currently used wire, before making the comparison..)
I have done some work on the simulator. Here setting up a quick generic cross, with negligible source resistance and showing the load impedance (The 200mΩ resistances are shown shorted in this screenshot).
Here I set the response as an overlay, as the baseline, then changed the source resistance to 200mΩ. There are impedance minima at 150Hz where the response differs by 0.25dB, which may be just audible with a low resonance speaker, and near 5kHz where the response differs by 0.15dB, which by comparison should be somewhat audible.
Last edited:
I set that as the new baseline and then changed the model to bi-wire. 200mΩ in each crossover leg, both fed from the same voltage.
The difference is so small as to likely be inaudible.
I repeated these last steps using 1Ω of cable resistance for single and bi-wiring, looking to demonstrate that the legs will interact given enough source resistance, and towards 4kHz there appears to be enough variation to be potentially audible.
The difference is so small as to likely be inaudible.
I repeated these last steps using 1Ω of cable resistance for single and bi-wiring, looking to demonstrate that the legs will interact given enough source resistance, and towards 4kHz there appears to be enough variation to be potentially audible.
If one were to look at how Hans Polak measured characteristic impedance of real speaker cable using a VNA then perhaps there could be a better model than just resistance?
I mean if the sound changes, and B&W seems to think it can, and if a simple model doesn't explain it, then maybe the problem is with the simplified model?
...It sure wouldn't be the first time some error in an analysis occurred as a result of using an oversimplified model. Just sayin'
Also, from looking at the graphs and reading the text, already I feel an expectation bias of no audible difference. Is that what I am supposed to do?
I mean if the sound changes, and B&W seems to think it can, and if a simple model doesn't explain it, then maybe the problem is with the simplified model?
...It sure wouldn't be the first time some error in an analysis occurred as a result of using an oversimplified model. Just sayin'
Also, from looking at the graphs and reading the text, already I feel an expectation bias of no audible difference. Is that what I am supposed to do?
Last edited:
You seem a little reticent, and I would be the same. It has to be stated that for a proper test the amplifier should be immune to the variations in load, such as cable parasitics. I'd also question what characteristic impedance is going to achieve when the wavelength at 1kHz may be in the order of 200-300km.
It's also necessary to re-make all connections which are going to be touched. Anecdotal evidence (diyAudio members excluded) tends to bring doubt.
It's also necessary to re-make all connections which are going to be touched. Anecdotal evidence (diyAudio members excluded) tends to bring doubt.
As stated, I was talking about the audibility of level differences.
Last edited:
Indeed and understood. It was the discussion in the past by @jneutron about applicability of the Telegrapher's Equation and Bessel function-related eddy currents at lower audio frequencies. That and @Hans Polak observations about audible effects of lumped compensation of zip cord measured Z0 at audio frequencies. All this stuff has me concerned that we may not be thinking about how there can be real audible effects in the right way. Using the wrong models is something we see around here from time to time. As I think I mentioned recently in one of these threads, some people have insisted that shielded power cables cannot possibly have any effect because they are such a tiny amount of the total resistance going all the way back to the power plant generator. That's kind of really obvious case, but similar things are not uncommon. If we start with the assumption that the model is good then we may dismiss real audible effects without even starting to delve into deeper thinking about what might be going on. Of course the whole subject is complicated by the fact that sometimes its very easy for people to fool themselves into thinking they hear a real difference when there isn't one; and likewise they can fool themselves into thinking there is no audible difference when some difference really does exist. Therefore, I tend to start the investigation at the listening end rather than the modeling end. I have too many years of experience with high tech medical devices where when problems were reported by users, the problems were dismissed as user error or imagination by the design engineers because, "the system can't do that." In virtually every case it turned out the engineers' model of the system in their heads was not what the system was actually like. Their models were oversimplified, and or there were missed subtle design errors.
Last edited:
Yes, there's room for doubt with some things. I believe there is a point where we can begin to see it from the other direction. If we eliminate speaker related issues, given that their time based problems are three dimensional, and in a typical room those aspects fall within the realm of audibility..
- Status
- Not open for further replies.