Which would be better? An amp going from 1 Hz to 300 kHz, or an amp from 10Hz-30kHz, say, with any deleterious phase shifts compensated for on the way in? I'm guessing the latter would be easier to design, and better behaved.
Last edited:
Preference vs. audibility - please keep them separate.
- Thread starter j_j
- Start date
- Status
Tube or Solid State? ...Euphony or Accuracy? ...Some of the best amp designers know best; Curl, Pass, Hansen, Levinson, d'Agostino, ...
A review of a Quad amp:
http://www.theabsolutesound.com/art...ifier-and-909-stereo-power-amplifier-tas-203/
Could a celebrated virtuoso amplifier designer contemplate building a hybrid system based on DSP correction prior to the amp? Maybe, but it's a slippery slope. Why not mop up the speaker's phase shifts and delays while we're at it? And then extend it to one such system per driver, including the crossover filtering. He might suddenly find that his amps don't actually have to be all that amazing any more.
I'm missing something. If noise and distortion at those frequencies...if even the flattening of an EQ curve at those frequencies can create distortion in the audible range, but we have no examples of supersonic content having a positive impact in the audible range, why would we just hope for well-behaved? Why not limit the range of the amplifier or the media to just slightly above the audible range? Why is it important for an amp to have that sort of bandwidth?
Tim
Between other pieces my Quad collection includes a 303, a 405-II and a 606 mk2 power amplifier, but not a 909. Some people say that the 909 is just a re-badging of the 606 mk2, but I can not confirm it. The 405 and the 606 share the same type of sound, although the 606 is more powerful. Both amplifiers use the current dumping approach designed by Peter Walker - a very different topology from the most usually seen in classic designs of this period. Unless we know what are the other brands and models of amplifiers Paul Seydor is addressing we can not comments on his remark on bandwidth - it is too vague.
A good friend of mine once listened to the Quad 606 mk2 in my older SoundLabs - he was so impressed that he immediately went to a Quad distributor to order one! Although the bass was not up to control and detail of other much more expensive amplifiers and the treble was sometimes a little dry, it had an impressive midrange transparency.
Fast rise time. For people who love measurements, one of things they look at in amplifier measurements is the 1kHz square wave response and the 10kHz square wave response even though music doesn't contain any square waves. How perfect those square waves look is dependent upon the bandwidth of the amplifier. When the left side and the right side of the square wave are actually straight instead of being bent, that indicates an amplifier has a fast rise time. When you see the top of the waveform on the left side overshoot and you see a bunch of noise on the top of the waveform, that usually indicates ringing.
Spectral wants their amplifiers to be very fast and thus they give them a bandwidth into the MHz region. Spectral doesn't want that same bandwidth to actually make it to your speaker terminals which is why they have special MIT speaker cables with termination networks that keep that bandwidth out of your speakers.
Mark,
May be we should be more exact : The UL-38 is a dedicated Spectral cable network which assures the precision time alignment, critical termination and tuned roll-off required for stable operation of Spectral’s megahertz amplifier designs. This proprietary tuning of UL-38 gives Spectral amplifiers broad stability margin into the most complex and difficult speaker loads.
Quoted from the Spectral site.
No, a good amp designer build his amp to be the most accurately and scientifically sound as possible.
It is up to the DSP component's designer (chips and all) to play beyond and above accuracy.
And finally, it's up to us (the music lovers, the music listeners, the audiophiles) to make the final purchase's choice at the end; on deciding our own sound's preference. And not to the designer, not to the DSP maker, and not the audio reviewer, but us.
We're only in it further to learn, improve, and be happier. ...We're the ones who give the final seal of approval according to our own wisdom and knowledge and musical sound experience, and financial disposition.
The audio designers and electronic product's manufacturers are the messengers, with their messages encapsulated within the coffins of their boxes, metal box enclosures and speaker's designs. ...They propose, we select between various choices for the best balanced synergy of our own systems.
Last edited:
IMHO, you will never get a definitive answer in this question. We have excellent sounding amplifiers with very different bandwidths. A few designers have presented their reasons in public - recently TheAbsoluteSound had an interesting issue filled with interviews with well known designers about their approaches. There is much more in an amplifier design than bandwidth - design is a compromise between many features.
Personally I have no hope to correlate sound quality with bandwidth, unless the amplifier response is limited to 10KHz ...
Hope you don't mind if I disagree with you there, Northstar. I suggest that a systems-based approach is the next logical step. There are acknowledged weaknesses in passive multiway speakers that price and exotic materials cannot solve, so the standard High End solution of individual ultra-expensive boxes is a Dead End. Unpalatable as it may be, an active system with much better sound (objectively and subjectively) may cost much less. It won't look so pretty, though.
The compensation you refer to really does not exist in such a way as to not contribute its own signature. I'll take the former rather than the latter.
Actually, I thought I explained that. You go for wide bandwidth in the electronics to reduce phase shift across the audio passband. We *already* have bandwidth-limited media, and bandwidth-limited loudspeakers. Dr Herbert Melchur has already shown that if the audio playback lacks speed (which needs bandwidth to occur) the brain will move the processing from the limbic centers to the cerebral cortex (that's bad, BTW). He has solid objective numbers on this very phenomena BTW, although I don't think he has published. Bottom line is you need speed so that the resulting waveform in the room has the same speed as the original, as much as the recording/playback process will allow.
You'd have to compare to the original. That might be a bit of a trick, as scopes have limited analysis capabilities, so you would need a squarewave as a source.... that would be nice anyway.
I'm sure, but my point is that changes at 50 KHz are inaudible. So the better place to look for changes is within the audible band.
--Ethan
- Status
Similar threads
