Varying delays can cause tilts of radiation pattern in speakers, probably a factor, but we're somewhat surprisingly forgiving of just the variations in group delay if made artificially and isolated over headphones.
Thanks,
Chris
Thanks,
Chris
Well, a specific claim about driver time alignment was made several times, but I'd like to know the factual support. Just IME, I haven't found this correlation in the speakers I've listened to, owned, and built; more important was control of polar pattern as a function of frequency. But if someone has studied this in a reasonably rigorous way and shown the opposite conclusions, I'd love to know about it.
Member
Joined 2009
Paid Member
Time alignment -- if only Romy the Cat were around...
I've found that the differences between good amplifiers are relatively tiny, the differences between speakers much larger. The use of GNF vs no-gnf pales against other questions that affect my enjoyment of music i.e. the elephant in the room is the recording quality. Amplifier technology appears to be continuously improving but I'm coming to be of the mind that we can exceed our needs in this area and should turn our attention elsewhere. We're done here, nothing to see, time to move on.
I've found that the differences between good amplifiers are relatively tiny, the differences between speakers much larger. The use of GNF vs no-gnf pales against other questions that affect my enjoyment of music i.e. the elephant in the room is the recording quality. Amplifier technology appears to be continuously improving but I'm coming to be of the mind that we can exceed our needs in this area and should turn our attention elsewhere. We're done here, nothing to see, time to move on.
Last edited:
Even vinyl can have outstanding imaging even though there is a lot of crosstalk that increases with frequency. It crosstalk being far higher than even some of the worst amps. I am for dual mono if we are talking monoblocks, I like having them placed near each speaker with a shirt run speaker cable.
Colin
Thanks for the reply, jcdrisc.
I remembered I have a 1975 copy of Dictionary of Audio & Hi-Fi from Howard W. Sams but neither term has an entry. I did find them at the audioholics.com glossary page. Looks like I wasn't too far off with my own definition.
Siting the amps by the loudspeakers is a good idea. Although I am skeptical of claims for cables beyond characteristics explained by fairly straightforward electrical theory, keeping them short is surely beneficial. Even one of the audiophile spoilsports <jk> who wrote an article about his cable investigations years ago, in Audio magazine, Professor Greiner, said it was better to run line level to power amps (if necessary) than run longer speaker cables.
I believe vinyl imaging is a subtle subject, as the crosstalk is not simply a bit of the other channel's same-polarity signal but can in fact be somewhat akin to a bit of intentional crossfeeding. Or at least so it seems, and I regret I don't have a specific reference to support this. It's not a big effect, but I suspect to simply give a specific crosstalk number can be misleading.
Yet another anecdote: someone that I'd worked with in another company called me to say he had an odd phenomenon when he made a network to attenuate speaker-level signals for a line-level input amplifier. As I looked as his network it turned out that he had inadvertently made a cross-feed network that managed a sort of quasi-surround-sound effect.
Did the professor stipulate balanced drive to the power amps? If not, with normal single ended line level connections, there's the drawback of the much greater loop area for groundloop noise pickup.
Good luck with your arrogant stance. How lucky for you that you already know what other people can and can't hear and you're smarter than the whole "DIY anti opamp community". I'm also still puzzled about your undying support for Bonsai's designs when apparently you've never built any? Even if you did, based on what you're saying you wouldn't believe they would sound any different to other amplifiers so what's the big deal?
In my opinion there are two major flaws in Self's argument (whose works I assume you and Bonsai are taking your gospel from) that an opamp can't possibly have "sound signature" of its own as the sound on all recordings has passed through hundreds of them:
1) most recordings sound like crap
2) not all recordings have been made on gear that contains hundreds of opamps.
If you (well not you obviously, some other person with a less closed mind) took the trouble to find some that aren't recorded on standard modern mixing desks you might be surprised how good they sound. Or even easier: listen to some good jazz recordings from the 50's and '60s. And if you do some recording of your own with minimalist (but high quality) equipment like I have done, it is stunning to hear how much closer it sounds to the real thing than any commercial release. This argument also ignores the fact that many respected artists and recording engineers seek out particular gear for its sound to achieve their desired sound on the recording. If mixing desks were so transparent, this wouldn't happen.
In my opinion opamps do have a sonic signature. If you get the power supply decoupling right, it seems to be much reduced. If you look at the complexity of a big mixing console I'm certain not much attention will have been paid to this in most of them other than to scatter a few 100n film caps across the rails with some bulk electros near the power input and then check stability and measure THD. This is understandable since the design task will be focused on the circuit functionality and performance and also people like Self are designing them who don't believe there will be any audible difference as long as the distortion measures ok. I bet many of them could be improved with some attention to this area. The distortion measurements might even improve.
I don't recall his saying that, but it is a good point. With the exception of the increase in input voltage noise at the differential receiver (unless the active devices are quadrupled) balanced is another great way to go. I hesitate to say a "no-brainer", as things rarely really are.
You are not supposed to have ground loops in the first place, in fact one of the reasons to build dual mono amps is to prevent various ground (loop) problems. And... there is such a thing as a co-axial cable even for unbalanced connections 😛 as well as twisted pairs, which are designed to prevent these problems... 😛
I'm talking about ground loops via the mains trafos (inter-winding capacitance) and the mains wiring. Even with dual mono amps, they're pretty much unavoidable. At least running balanced keeps the error voltage from being in series with the wanted signal.
Thanks, very good point, can't argue with that. However I still think there will be an element of subjectivity in which will be the best reproduction methods applied to paintings.
Here's another analogy to pick holes in: when I was buying a TV recently, in a group of the same technology type and resolution etc there were some models that looked much more realistic to me, particularly their colour balance. Nothing about their specifications could explain that. Was I deceiving myself? Do they really all look the same? Would anyone else in the shop believe you if you tried to tell them that? Some might care less than others, but would anyone believe they're all equally good?
It's very easy to create 2 nd harmonic in symmetrical amps by simply loading the VAS or TIS to one of the rails.
BTW, many experts say if you load the VAS, distortion will be higher and you should not do it as you are 'throwing away loop gain'.
On my symmetrical e-Amp http://hifisonix.com/ovation-e-amp/ , I have the option to load the VAS to ground with 2 x 33k resistors via a jumper. When I do this, the distortion goes DOWN. Looking on my sound card, I see that it's the 2nd Harmonic that's decreasing. I since the OPS is an EF3, I surmize that the gain between the two halves is different and when loading the VAS, it flattens out the difference between the two halves and the distortion goes down.
To put the levels in context, the distortion is about 20 ppm at 150 W output. And about 28 ppm without the loading.
3rds and 4ths remain unchanged.
Last edited:
Those 'experts' presumably say this because they prioritize THD above any other effect that might occur due to lower VAS output impedance. They're experts on getting the lowest THD number that's for sure.
The assumptions of those experts would be that the Vas unloaded is perfectly matched/balanced. When you load with 33K on each you force them into a better balance. So, yes, I would agree with your educated guess.
Comment re opamp ic and high gnfb...... it isnt the gnfb which causes the problems as much as the AB output stage..... work-arounds can help there (biased to class A etc).
THx-RNMarsh
ref owdeo's post #1089
Why are we calling it a "sonic signature"? No one seems to want to call it a "performance signature"? Isn't that what it is? Is there a correlated "optical signature" term used in the video fields?
One reason I asked the soundstage/imaging question is that I remember the Stereo/Wide switches when boomboxes were popular. Proof to me those characteristics can be altered electronically.
Liquid sound... That was funny. Cheers.
Why are we calling it a "sonic signature"? No one seems to want to call it a "performance signature"? Isn't that what it is? Is there a correlated "optical signature" term used in the video fields?
One reason I asked the soundstage/imaging question is that I remember the Stereo/Wide switches when boomboxes were popular. Proof to me those characteristics can be altered electronically.
Liquid sound... That was funny. Cheers.
Its a lack of performance signature. But perhaps a little unfair to blame the opamp when high impedance loading and a low enough impedance power supply can eliminate it in many cases.
The distortion is in fact I'd doing the opposite of what is predicted in the various books and articles - It goes down with loading. A good example perhaps of why practical testing is important.