Grey, I see this thread has been bought to the top so spent the last hour rereading. I can only agree 100% … ok, 99%, nothings perfect 
, with your findings. Like many I began by looking at specifications and came to the conclusion that they are effectively meaningless in indicating what the amplifier will sound like. Virtually all amplifiers on the market, at least at the level we are referring to, have superb specifications, yet sound very different.
We live in a society that increasingly wants instant answers for instant gratification; how convenient for specification figures. Yet there is clearly more to an amplifier than is revealed in overly simplistic specifications. While I find it difficult to argue with the concept of “wire with gain”, it is clear in my mind that we still have a long way to go before this is realistically achievable.
Having said this, I am fortunate to travel the world listening to different systems, many are in the >US$100,000 region. I was grateful for the opportunity to recently listen to a German MBL system http://www.mbl-hifi.com/index.htm The first thing that struck me was just how different the sound was. Which suggests that either every other system I’d heard was “wrong” or MBL were taking a different approach. I would suggest the latter, hence one could conclude that the MBL system was “distorting” the sound. The second thing that struck me was just how incredible the MUSIC was that I was listening to. I was in the company of a colleague more knowledgeable in these matters than I, and both of us had never heard anything like what we were listening to. There was no doubt that the sound was coloured, yet what we were experiencing was the “soul” of the music. Try finding that factor in the spec sheets.
Cheers,
Pete
PS Grey, if you want good recordings have you tried Linn? As you know Ivor was obsessed with the idea that the source is the most important component (yeah well you would if you made turntables!) but have taken this a step further with the Linn label. I'm not sure how easy they are to get in the US, but can recommend Claire Martin. As a bonus many are HDCD. http://www.linnrecords.com/
, with your findings. Like many I began by looking at specifications and came to the conclusion that they are effectively meaningless in indicating what the amplifier will sound like. Virtually all amplifiers on the market, at least at the level we are referring to, have superb specifications, yet sound very different. We live in a society that increasingly wants instant answers for instant gratification; how convenient for specification figures. Yet there is clearly more to an amplifier than is revealed in overly simplistic specifications. While I find it difficult to argue with the concept of “wire with gain”, it is clear in my mind that we still have a long way to go before this is realistically achievable.
Having said this, I am fortunate to travel the world listening to different systems, many are in the >US$100,000 region. I was grateful for the opportunity to recently listen to a German MBL system http://www.mbl-hifi.com/index.htm The first thing that struck me was just how different the sound was. Which suggests that either every other system I’d heard was “wrong” or MBL were taking a different approach. I would suggest the latter, hence one could conclude that the MBL system was “distorting” the sound. The second thing that struck me was just how incredible the MUSIC was that I was listening to. I was in the company of a colleague more knowledgeable in these matters than I, and both of us had never heard anything like what we were listening to. There was no doubt that the sound was coloured, yet what we were experiencing was the “soul” of the music. Try finding that factor in the spec sheets.
Cheers,
Pete
PS Grey, if you want good recordings have you tried Linn? As you know Ivor was obsessed with the idea that the source is the most important component (yeah well you would if you made turntables!) but have taken this a step further with the Linn label. I'm not sure how easy they are to get in the US, but can recommend Claire Martin. As a bonus many are HDCD. http://www.linnrecords.com/
Hi Grey,
Hmm, SC. I'm up in Spartanburg myself. I would love to see your system sometime. I from the objective side of the fence (I am cursed in that I am an electronic engineer), but I am open to the idea that I don't know even most of what there is to know about anything.
You play bass too. A good friend of mine in Taylors makes absolutely killer basses (Eshenbaugh Guitars). I make active, parametric tone control boards for his basses. Guitar Center in Atlanta (Robbie Giesler) just started selling his basses. Who's axe do you play?
Drop me a line.
Phil Ouellette
phil.ouellette@mt.com
Hmm, SC. I'm up in Spartanburg myself. I would love to see your system sometime. I from the objective side of the fence (I am cursed in that I am an electronic engineer), but I am open to the idea that I don't know even most of what there is to know about anything.
You play bass too. A good friend of mine in Taylors makes absolutely killer basses (Eshenbaugh Guitars). I make active, parametric tone control boards for his basses. Guitar Center in Atlanta (Robbie Giesler) just started selling his basses. Who's axe do you play?
Drop me a line.
Phil Ouellette
phil.ouellette@mt.com
opt-mos subjetive Vs 'objective analysis
grollins
Nominaly identical amps.,(with respect to distortion specs.), tend to sound different precisely because such distortion readings are obtained with resistive dummy loads that bare no relationship with real-life loudspeaker loads, which are invariably reactive, THD can only meaningfully be related to subjective performance for a given set of test amplifiers if such measurements are taken at the terminals of a given loudspeaker.
Given the above, (and efficient ear plugs!!!!), it should become clear from THD readings thus obtained, that the robustness of the power supply, and immunity of protection circuitry to premature activation with loudspeaker loads of all hues are more significant indicators of sound quality than THD figures obtained with purely resistive, and perforce unrepresentative laboratory test loads.
Amplifiers without correctly designed VI protection represent, in my view, a recipe for near heroic unreliabilty in the face of forbiden voltage-current combinations that the amps. output decices may be subjected to with 'difficult' loudspeaker loads.
grollins
Nominaly identical amps.,(with respect to distortion specs.), tend to sound different precisely because such distortion readings are obtained with resistive dummy loads that bare no relationship with real-life loudspeaker loads, which are invariably reactive, THD can only meaningfully be related to subjective performance for a given set of test amplifiers if such measurements are taken at the terminals of a given loudspeaker.
Given the above, (and efficient ear plugs!!!!), it should become clear from THD readings thus obtained, that the robustness of the power supply, and immunity of protection circuitry to premature activation with loudspeaker loads of all hues are more significant indicators of sound quality than THD figures obtained with purely resistive, and perforce unrepresentative laboratory test loads.
Amplifiers without correctly designed VI protection represent, in my view, a recipe for near heroic unreliabilty in the face of forbiden voltage-current combinations that the amps. output decices may be subjected to with 'difficult' loudspeaker loads.
I use a resistive test load on my bench because that's all that I have to work with, aside from subsequent listening tests on a pair of Acoustic Research AR-11 speakers (4 ohms, nominal) through Kimber 4TC cables, which are probably
moderately capacitive. I do know now that my Leach amp will oscillate nicely without a Zobel network across the output terminals when connected to this real-world load. That should be little surprise to anyone who's designed amplifiers.
It would be nice to have a "standard" simulated reactive load, if we could agree on a standard. I'm open to specific suggestions for designs, otherwise the bulk of my testing will continue to be THD readings on a resistive load.
moderately capacitive. I do know now that my Leach amp will oscillate nicely without a Zobel network across the output terminals when connected to this real-world load. That should be little surprise to anyone who's designed amplifiers.
It would be nice to have a "standard" simulated reactive load, if we could agree on a standard. I'm open to specific suggestions for designs, otherwise the bulk of my testing will continue to be THD readings on a resistive load.
I strongly disagree.
I'm not about to use my speakers for a test load, especially not for a 200 watt/channel amplifier. I need
to know how a amplifier behaves at all power levels. I
need the load at the test bench. I don't need the ear-spliting noise.
For measurements to have any broad applicability, a standardized test load is needed.
At most, I can use my speakers only as a spot check for
stability; even then I have to have at least an oscilloscope handy because I couldn't hear the oscillation that resulted without the output Zobel network installed.
Objectively yours,
Damon
I'm not about to use my speakers for a test load, especially not for a 200 watt/channel amplifier. I need
to know how a amplifier behaves at all power levels. I
need the load at the test bench. I don't need the ear-spliting noise.
For measurements to have any broad applicability, a standardized test load is needed.
At most, I can use my speakers only as a spot check for
stability; even then I have to have at least an oscilloscope handy because I couldn't hear the oscillation that resulted without the output Zobel network installed.
Objectively yours,
Damon
Pete,
I couldn't agree more with your observation about 'instant gratification.' I made notes on what all of you had said; next to this I put a little "Yes!"
I'd like to add that a substantial percentage of my customers back when I was selling audio lacked...how shall I say this without hurting feelings?...the courage to buy something purely on their own initiative. They wanted someone to back them in their decision, even if it was only the mental image of some stalwart gentleman in a white lab coat offering benediction in the guise of objective numbers. I eventually grew to regard it as a form of insecurity--they simply didn't trust their own judgement. My recommendation that they go to hear live music and use that as the benchmark was always treated as somehow suspect. It wasn't complicated enough, hence it couldn't be 'scientific.'
*Sigh*
I haven't heard the MBL stuff, but would like to. Here in the southern tier of states, it's catch as catch can.
I don't own any Linn recordings, but that's not to say they aren't good. I just haven't run across any down here. (Yes, I think they're available mail order--I just haven't pursued them.) Mostly, I've been building my collection of older jazz recordings recently, which has taken more money than I care to think about. I believe that Mark Levinson (back when ML himself was still with the company) used to make recordings, as well. Some of them had a nice reputation. No, I don't own any of those, either. Now, Miller & Kreisel...those fellas made some excellent direct-to-disk recordings. I have some, and wish I could lay hands on one or two more.
Phil,
I used to live in Spartanburg back in the '60s. Drove past on I-26 headed south today as I came back from camping, in fact.
For the most part, I build my own basses (that is, in part, what my woodworking shop is for, although it does come in handy for building the odd subwoofer, etc.). As far as 'regular' basses, I'm down to the fewest number I've had in years, mostly because I've got my basses sounding pretty much the way I want. At present, I've got a mid-'70s Alembic Series I, a '60 Fender, a Ken Smith 6-string...and, uh, I think that's it. If I think of anything else, I'll drop it in later.
Oddly, I used to build my own speaker cabinets, but don't any more. It's all production stuff.
michel,
God? No comment. It'd be like shooting fish in a barrel.
We are in agreement that load resistors are a poor substitute for real world speaker loads. However, given that each and every speaker in the world has a different impedance curve, I'm not sure how we'd go about making up a standardized test load that would be a good stand-in for all speakers. The challenge, of course, is to build an amplifier that reacts as little as possible to changes in the reactive realm. Then--at least in theory--you can toss in most any speaker you like and feel reasonably confident that you're going to get consistent performance.
On a related note, I took some reading material along whilst I was camping (audio by campfire light...how romantic) and came across something wherein someone was saying that you don't need really big amps if you have efficient speakers (I think in context, he was speaking of horns). However, impedance dips can require vast amounts of stable power (I just glanced a another thread here where someone has an electrostat that he'd measured at .4 ohms at 20kHz [I think I've got those numbers right, don't shoot me if I'm wrong.]) so as not to current limit at those frequencies, which can effect perceived dynamics, frequency response, you name it. I don't listen nearly as loud as I used to, but still tend to keep as much wattage on hand as possible, commensurate with quality. I draw the line at 2kW class D amps for my tweeters. Another thing I do is to, at least for mid-range and tweeters, use purely resistive loads. At least that way I don't get any reactive surprises.
Grey
P.S.: Gracious me...I thought we'd seen the last of this thread.
I couldn't agree more with your observation about 'instant gratification.' I made notes on what all of you had said; next to this I put a little "Yes!"
I'd like to add that a substantial percentage of my customers back when I was selling audio lacked...how shall I say this without hurting feelings?...the courage to buy something purely on their own initiative. They wanted someone to back them in their decision, even if it was only the mental image of some stalwart gentleman in a white lab coat offering benediction in the guise of objective numbers. I eventually grew to regard it as a form of insecurity--they simply didn't trust their own judgement. My recommendation that they go to hear live music and use that as the benchmark was always treated as somehow suspect. It wasn't complicated enough, hence it couldn't be 'scientific.'
*Sigh*
I haven't heard the MBL stuff, but would like to. Here in the southern tier of states, it's catch as catch can.
I don't own any Linn recordings, but that's not to say they aren't good. I just haven't run across any down here. (Yes, I think they're available mail order--I just haven't pursued them.) Mostly, I've been building my collection of older jazz recordings recently, which has taken more money than I care to think about. I believe that Mark Levinson (back when ML himself was still with the company) used to make recordings, as well. Some of them had a nice reputation. No, I don't own any of those, either. Now, Miller & Kreisel...those fellas made some excellent direct-to-disk recordings. I have some, and wish I could lay hands on one or two more.
Phil,
I used to live in Spartanburg back in the '60s. Drove past on I-26 headed south today as I came back from camping, in fact.
For the most part, I build my own basses (that is, in part, what my woodworking shop is for, although it does come in handy for building the odd subwoofer, etc.). As far as 'regular' basses, I'm down to the fewest number I've had in years, mostly because I've got my basses sounding pretty much the way I want. At present, I've got a mid-'70s Alembic Series I, a '60 Fender, a Ken Smith 6-string...and, uh, I think that's it. If I think of anything else, I'll drop it in later.
Oddly, I used to build my own speaker cabinets, but don't any more. It's all production stuff.
michel,
God? No comment. It'd be like shooting fish in a barrel.
We are in agreement that load resistors are a poor substitute for real world speaker loads. However, given that each and every speaker in the world has a different impedance curve, I'm not sure how we'd go about making up a standardized test load that would be a good stand-in for all speakers. The challenge, of course, is to build an amplifier that reacts as little as possible to changes in the reactive realm. Then--at least in theory--you can toss in most any speaker you like and feel reasonably confident that you're going to get consistent performance.
On a related note, I took some reading material along whilst I was camping (audio by campfire light...how romantic) and came across something wherein someone was saying that you don't need really big amps if you have efficient speakers (I think in context, he was speaking of horns). However, impedance dips can require vast amounts of stable power (I just glanced a another thread here where someone has an electrostat that he'd measured at .4 ohms at 20kHz [I think I've got those numbers right, don't shoot me if I'm wrong.]) so as not to current limit at those frequencies, which can effect perceived dynamics, frequency response, you name it. I don't listen nearly as loud as I used to, but still tend to keep as much wattage on hand as possible, commensurate with quality. I draw the line at 2kW class D amps for my tweeters. Another thing I do is to, at least for mid-range and tweeters, use purely resistive loads. At least that way I don't get any reactive surprises.
Grey
P.S.: Gracious me...I thought we'd seen the last of this thread.
Pixie,
What is your opinion about building the OPTI-MOS amp? The larger version of the BJT output devices and the MOSFET version do look impressive, from a class A/AB point of view. I had a good look and am a bit troubled about the use of 2SB640/2SD660 transistors. Can't remember off hand, but one of these is a 25v 50mA device and is used in the driver position. May be you can seek Randy's clarification on this matter.
What is your opinion about building the OPTI-MOS amp? The larger version of the BJT output devices and the MOSFET version do look impressive, from a class A/AB point of view. I had a good look and am a bit troubled about the use of 2SB640/2SD660 transistors. Can't remember off hand, but one of these is a 25v 50mA device and is used in the driver position. May be you can seek Randy's clarification on this matter.
Grey,
Yes I believe you’re quite right regarding people not wanting to trust their feelings. Indeed I think this is one thing that changes as a person gains more experience in the industry. With exposure one is educated about sound, and the differences between “good” and “bad” components. At the same time we learn to trust ourselves and our feelings regarding what we are hearing.
It’s interesting to compare and contrast the differences between purchasing a new car and a new hi-fi system for example. When buying a car we may look at the specifications to see what the power of the engine is and so on, yet are unlikely to base our purchase decision on that factor alone. Instead we sit in the car, take it for a drive, and generally base our decision on our emotions. Yet with a medium that IS emotions (well ok, I’ll except some of the manufactured pop bands from that equation), such as music, many try to discount their emotions from the purchasing decision. Instead many try to distil their decision down to simple specifications. I guess an automobile is a tactile object that one can see, touch, smell, and hear. On the other hand listening to a hi-fi involves just one of our senses, so we tend to seek confirmation from other sources, such as specifications or the cosmetics.
I find it interesting to observe the differences in attitudes between manufacturers and especially regions in my travels. A very good example comes to mind in the cases on Linn and Naim, 2 brands from what I would call a “traditionalist” hi-fi region, who emphasise the emotion behind the music when auditioning components i.e. does playing this piece of music make you want to tap your foot, make you want to get up and dance, or even make the hairs on your arms stand on end? That’s what music, good music should do.
I would add that I believe the above is a big difference between something that sounds good in a store, and one that I personally would want to live with in my home. In the case of the former it is often the “wow” factor that impresses; incredible detail, wonderful imaging, and so on. Yet often I would hate to have a system like that in my home, there is often either no emotion conveyed, or I listen to every error on the recording. At the end of the day I want to listen to music, not my components or crap on the recordings. It was no coincidence that some of the recordings I listened to on the MBL system were not audiophile quality, indeed they were rather poor quality, and yet the same emotion was conveyed; it was this that impressed, quite magical.
Grey regarding an amps power, I recently posted on this subject, and perhaps this is what you were thinking about. Essentially I was asking if an amplifier of high power (and by this I was referring to capable of high RMS VOLTAGE) is really needed if listening to digital sources at low to moderate volume. Given that a digital source cannot exceed a maximum, voltage headroom above that is immaterial; there are no transients above this level. Whether the amp can supply this CURRENT at this level is another matter entirely. However I’m beginning to believe that it is better to have an amp that will deliver 20W and supply any required current into any load, than one that is capable of 100W but is unable to supply the required current. That is, for the same amount of money to build, one should be able to build a smaller amp that will sound better than a larger amp. This business of voltage transients and headroom (WITH A DIGITAL SOURCE) is simply a nonsense.
Cheers,
Pete
PS sorry ‘bout the length of the post
Yes I believe you’re quite right regarding people not wanting to trust their feelings. Indeed I think this is one thing that changes as a person gains more experience in the industry. With exposure one is educated about sound, and the differences between “good” and “bad” components. At the same time we learn to trust ourselves and our feelings regarding what we are hearing.
It’s interesting to compare and contrast the differences between purchasing a new car and a new hi-fi system for example. When buying a car we may look at the specifications to see what the power of the engine is and so on, yet are unlikely to base our purchase decision on that factor alone. Instead we sit in the car, take it for a drive, and generally base our decision on our emotions. Yet with a medium that IS emotions (well ok, I’ll except some of the manufactured pop bands from that equation), such as music, many try to discount their emotions from the purchasing decision. Instead many try to distil their decision down to simple specifications. I guess an automobile is a tactile object that one can see, touch, smell, and hear. On the other hand listening to a hi-fi involves just one of our senses, so we tend to seek confirmation from other sources, such as specifications or the cosmetics.
I find it interesting to observe the differences in attitudes between manufacturers and especially regions in my travels. A very good example comes to mind in the cases on Linn and Naim, 2 brands from what I would call a “traditionalist” hi-fi region, who emphasise the emotion behind the music when auditioning components i.e. does playing this piece of music make you want to tap your foot, make you want to get up and dance, or even make the hairs on your arms stand on end? That’s what music, good music should do.
I would add that I believe the above is a big difference between something that sounds good in a store, and one that I personally would want to live with in my home. In the case of the former it is often the “wow” factor that impresses; incredible detail, wonderful imaging, and so on. Yet often I would hate to have a system like that in my home, there is often either no emotion conveyed, or I listen to every error on the recording. At the end of the day I want to listen to music, not my components or crap on the recordings. It was no coincidence that some of the recordings I listened to on the MBL system were not audiophile quality, indeed they were rather poor quality, and yet the same emotion was conveyed; it was this that impressed, quite magical.
Grey regarding an amps power, I recently posted on this subject, and perhaps this is what you were thinking about. Essentially I was asking if an amplifier of high power (and by this I was referring to capable of high RMS VOLTAGE) is really needed if listening to digital sources at low to moderate volume. Given that a digital source cannot exceed a maximum, voltage headroom above that is immaterial; there are no transients above this level. Whether the amp can supply this CURRENT at this level is another matter entirely. However I’m beginning to believe that it is better to have an amp that will deliver 20W and supply any required current into any load, than one that is capable of 100W but is unable to supply the required current. That is, for the same amount of money to build, one should be able to build a smaller amp that will sound better than a larger amp. This business of voltage transients and headroom (WITH A DIGITAL SOURCE) is simply a nonsense.
Cheers,
Pete
PS sorry ‘bout the length of the post
Pete,
Yes, cars and stereo equipment have a lot in common. Both are technical devices that have numbers associated with them, yet (can/should) elicit an emotional response. I remember test driving a Datsun 240Z (maybe 260?, it was a long time ago). The car had plenty of horsepower, looked sleek, cost a few pennies, etc. Funny thing, though. When I sat in the driver's seat and nailed the accelerator, the nose popped up, giving the impression of tremendous acceleration without really accomplishing same in reality. The suspension had been tuned such that it was exceedingly floppy from front to back, so that acceleration (and braking) gave a huge subjective sense of Things Are Happening. (Incidentally, the suspension was quite tight from side to side [heavy sway bars] so the car would corner well.) I learned a lot from that experience:
1) The suspension tuning trick didn't show up anywhere in the printed specs. It was purely something that the wily engineers came up with to impress the unwary.
2) (Really an extension on 1) Assuming that you were to buy the car, the trick would soon wear thin; you'd no longer be impressed by the car rocking back when you hit the gas pedal once you noticed that, for all the perceived drama, other cars were going as fast or faster without strain or flashy behavior. This would be directly analogous to a piece of audio gear that has an upper mid-range peak that gives the subjective impression of great detail. Yet once you get it home, you find that it's not true to the sound of what you hear live, and indeed is ultimately wearing on the ear.
Yes, I saw the thread on power, and think that it's a good philosophical think piece. But that's not where I read the comment that sparked my thinking, although the timing was fortuitous. Your thinking there is along the lines of the old Mark Levinson ML-2--the best, no holds barred, damn the expense solid state power that the guys could envision. Then back that up with a power supply the size of Alaska. Overkill for as far as the eye could see. In that same vein, I may...just possibly...be able to acquire a transformer next spring capable of 30A. If it works out, I'm going to have a nervous breakdown trying to decide what the best way to use it might be. Clearly a piece of iron of that size deserves the very best--it'd be a pity to do some hare-brained class B circuit just because it was easy and ran cool to the touch. But I'm trying very hard at the moment not to count those chickens before they hatch.
Grey
Yes, cars and stereo equipment have a lot in common. Both are technical devices that have numbers associated with them, yet (can/should) elicit an emotional response. I remember test driving a Datsun 240Z (maybe 260?, it was a long time ago). The car had plenty of horsepower, looked sleek, cost a few pennies, etc. Funny thing, though. When I sat in the driver's seat and nailed the accelerator, the nose popped up, giving the impression of tremendous acceleration without really accomplishing same in reality. The suspension had been tuned such that it was exceedingly floppy from front to back, so that acceleration (and braking) gave a huge subjective sense of Things Are Happening. (Incidentally, the suspension was quite tight from side to side [heavy sway bars] so the car would corner well.) I learned a lot from that experience:
1) The suspension tuning trick didn't show up anywhere in the printed specs. It was purely something that the wily engineers came up with to impress the unwary.
2) (Really an extension on 1) Assuming that you were to buy the car, the trick would soon wear thin; you'd no longer be impressed by the car rocking back when you hit the gas pedal once you noticed that, for all the perceived drama, other cars were going as fast or faster without strain or flashy behavior. This would be directly analogous to a piece of audio gear that has an upper mid-range peak that gives the subjective impression of great detail. Yet once you get it home, you find that it's not true to the sound of what you hear live, and indeed is ultimately wearing on the ear.
Yes, I saw the thread on power, and think that it's a good philosophical think piece. But that's not where I read the comment that sparked my thinking, although the timing was fortuitous. Your thinking there is along the lines of the old Mark Levinson ML-2--the best, no holds barred, damn the expense solid state power that the guys could envision. Then back that up with a power supply the size of Alaska. Overkill for as far as the eye could see. In that same vein, I may...just possibly...be able to acquire a transformer next spring capable of 30A. If it works out, I'm going to have a nervous breakdown trying to decide what the best way to use it might be. Clearly a piece of iron of that size deserves the very best--it'd be a pity to do some hare-brained class B circuit just because it was easy and ran cool to the touch. But I'm trying very hard at the moment not to count those chickens before they hatch.
Grey
dummy load for amp testing
Hi all,
This is my first post to the forum, so don't kill me (yet).
Basically, I would like specs to reflect the quality of an amplifier; therefore, I intuitively feel more comfortable with the designs of D. Self and R. Slone instead of Nelson Pass.
However, most of you seem to opt for the listen first, measure afterwards approach, and it turns out that the amps with the best specs don't yield the best sound (I don't have enough ABX-comparison experience to confirm or deny this).
Engineering-wise, I think that the measurements are therefore missing something very basic. Most obviously, this is because amplifier testing is performed using resistive dummies, while speakers have both a non-ohmic and a variable impedance with frequency.
Since it is very difficult to test amps with real world speakers (1. bad for your speakers, 2. bad for your ears, 3. everybody has different speakers), a suitable dummy load is needed.
It is quite easy, starting from TS-theory, to design a dummy load with a speaker-like impedance behaviour. If we can agree on the parameters for such a dummy (which I doubt), all measurements could be performed using the dummy.
Maybe we can get a more realistic comparison between 'objectivist' and 'subjectivist' amplifiers.
Regards, Dirk
Hi all,
This is my first post to the forum, so don't kill me (yet).
Basically, I would like specs to reflect the quality of an amplifier; therefore, I intuitively feel more comfortable with the designs of D. Self and R. Slone instead of Nelson Pass.
However, most of you seem to opt for the listen first, measure afterwards approach, and it turns out that the amps with the best specs don't yield the best sound (I don't have enough ABX-comparison experience to confirm or deny this).
Engineering-wise, I think that the measurements are therefore missing something very basic. Most obviously, this is because amplifier testing is performed using resistive dummies, while speakers have both a non-ohmic and a variable impedance with frequency.
Since it is very difficult to test amps with real world speakers (1. bad for your speakers, 2. bad for your ears, 3. everybody has different speakers), a suitable dummy load is needed.
It is quite easy, starting from TS-theory, to design a dummy load with a speaker-like impedance behaviour. If we can agree on the parameters for such a dummy (which I doubt), all measurements could be performed using the dummy.
Maybe we can get a more realistic comparison between 'objectivist' and 'subjectivist' amplifiers.
Regards, Dirk
Dirk,
It would be nice if all speakers could be represented by a dummy load, but if you look at all the different types of speakers that are out there you can see that this is impossible, just compare an electostatic load to a regular moving coil speaker and you will see what I mean.
Years ago someone came up with eight ohms in parallel with 2uf, which was a joke but a lot of manufacturers used this as a standard to measure performance.
Another problem is that all possible forms of distortion that can occur in an amplifier and the amplifier speaker interface have not been discovered yet.In other words we have to know all the criteria to make an objective measurement possible. Till then I submit to you that the subjective approach is the best method.
Nelson Pass's approach is to keep the signal path as simple as possible, similar to tube circuit, if you will. Needless complexity can result in poor sound even if the distortion numbers look good.
Jam
[Edited by jam on 10-08-2001 at 07:59 PM]
It would be nice if all speakers could be represented by a dummy load, but if you look at all the different types of speakers that are out there you can see that this is impossible, just compare an electostatic load to a regular moving coil speaker and you will see what I mean.
Years ago someone came up with eight ohms in parallel with 2uf, which was a joke but a lot of manufacturers used this as a standard to measure performance.
Another problem is that all possible forms of distortion that can occur in an amplifier and the amplifier speaker interface have not been discovered yet.In other words we have to know all the criteria to make an objective measurement possible. Till then I submit to you that the subjective approach is the best method.
Nelson Pass's approach is to keep the signal path as simple as possible, similar to tube circuit, if you will. Needless complexity can result in poor sound even if the distortion numbers look good.
Jam
[Edited by jam on 10-08-2001 at 07:59 PM]
Yes I believe part of the problem with present measurements is the complexity of a musical signal, yet the simplicity of the testing process. Unfortunately testing is done under essentially static conditions which are nothing like what are seen in normal use. While primitive, it’s really the only solution available in order to provide consistent standards. I would expect the dynamic nature of the load to contribute somewhat to variability between designs, but do not believe this is the full story either. Sadly there is no magic solution here from what I see.
Cheers,
Pete
Cheers,
Pete
Pete,
You make a good point about the dynamic non-repetitive nature of music which I did not mention. Steady-state testing produce results which bear little relationship to musicality.
Maybe we could convince Dirk to to compare a Pass design to a Self or Sloan design.
Cheers,
Jam
[Edited by jam on 10-08-2001 at 10:38 PM]
You make a good point about the dynamic non-repetitive nature of music which I did not mention. Steady-state testing produce results which bear little relationship to musicality.
Maybe we could convince Dirk to to compare a Pass design to a Self or Sloan design.
Cheers,
Jam
[Edited by jam on 10-08-2001 at 10:38 PM]
amplifier testing
Hi all,
Of course, this is right. It is, however, possible to use a load which is a reasonable approximation of a typical closed box or bass reflex. This should at least give an idea of the behaviour of the amplifier under realistic loading conditions.
This notion of the shortest possible signal path bothers me. Since the signal has already gone through sooooo many signal conditioning equipment at the recording and the playback stage (at least for CD), the signal path is ALWAYS complex. This is similar to the people paying large amounts of money for exquisite power cords: If a bad power cord spoils the electric power, the evil has already been done during the kilometers (miles for the US) of cable from the power plant to your home!
For the moment, I have PCB's for a low power version of the 'optimum bipolar' design of Slone. If I have (loads of) time, I might also build a more 'subjective' amp like the JLH (the ZEN is too hot for me).
Best regards,
Dirk
Hi all,
jam said:
Of course, this is right. It is, however, possible to use a load which is a reasonable approximation of a typical closed box or bass reflex. This should at least give an idea of the behaviour of the amplifier under realistic loading conditions.
jam said:
This notion of the shortest possible signal path bothers me. Since the signal has already gone through sooooo many signal conditioning equipment at the recording and the playback stage (at least for CD), the signal path is ALWAYS complex. This is similar to the people paying large amounts of money for exquisite power cords: If a bad power cord spoils the electric power, the evil has already been done during the kilometers (miles for the US) of cable from the power plant to your home!
For the moment, I have PCB's for a low power version of the 'optimum bipolar' design of Slone. If I have (loads of) time, I might also build a more 'subjective' amp like the JLH (the ZEN is too hot for me).
Best regards,
Dirk
Funny, AudioXpress is basically a tube magazine. No wonder they don't like Slone. They might have a chip amp here and there ....
PS I built design 11.4 from Slone's book and it is awesome. There are many similarities between it and Self's blameless design. Slone simply added two-pole compensation and played with the VI limiter a bit... Slone and Self's designs are quite similar in theory.
PS I built design 11.4 from Slone's book and it is awesome. There are many similarities between it and Self's blameless design. Slone simply added two-pole compensation and played with the VI limiter a bit... Slone and Self's designs are quite similar in theory.
grataku said:
Scientific evidence? what sceintific evidence?
Sceintists came up with this great "evidence" of big-bang theory and evoloution and shove it down our throats and want us to belive it. They dont know how a human brain stores memory and how memory is kept when brain cells are repoduced. Im not going to trust them with anything they want to come up with..... Especially not my eternity.
Sceintists came up with this great "evidence" of big-bang theory and evoloution and shove it down our throats and want us to belive it. They dont know how a human brain stores memory and how memory is kept when brain cells are repoduced. Im not going to trust them with anything they want to come up with..... Especially not my eternity.
I hope to put a rest to this subject.
Start a new thread if you have something original to say. But, please don't start a thread on religion. We all have strong opinions about, politics, religion and audio. Let's stick to one here.
My experience includes, honestly, about 30 other amps. Surely, not quite as many as Grey. But, I believe I've learned to listen.
I've listened to my Opti-MOS now for many months. The Opti-MOS is almost the best amp I've ever heard. I say almost. There is still some hiss at full volume, about 10-15dB, which is nothing compared to my new TiVo box with it's spinning hard drive! Other than that, I love the sound. period. There is no use describing the sound, because we all know sound cannot be described with words.
So, in conclusion, build this amp. It's well worth the money and time.
pixie
Start a new thread if you have something original to say. But, please don't start a thread on religion. We all have strong opinions about, politics, religion and audio. Let's stick to one here.
My experience includes, honestly, about 30 other amps. Surely, not quite as many as Grey. But, I believe I've learned to listen.
I've listened to my Opti-MOS now for many months. The Opti-MOS is almost the best amp I've ever heard. I say almost. There is still some hiss at full volume, about 10-15dB, which is nothing compared to my new TiVo box with it's spinning hard drive! Other than that, I love the sound. period. There is no use describing the sound, because we all know sound cannot be described with words.
So, in conclusion, build this amp. It's well worth the money and time.
pixie
Folks, please...
The whole thing about religion got started because Slone has--or at least had, don't know if it's still there--religion on his website. Now, I've got my views as to whether that's appropriate or not, but it certainly isn't illegal, either.
You want to talk about audio, that's cool...
You want to get to messing with religion, we'll have to lock the thread down. We've already had one tiff this week, and I'm not anxious to have another.
In short...don't go there.
Period.
Grey
P.S.: Pixie, I'm glad you're happy with your amp. That's what it's all about.
The whole thing about religion got started because Slone has--or at least had, don't know if it's still there--religion on his website. Now, I've got my views as to whether that's appropriate or not, but it certainly isn't illegal, either.
You want to talk about audio, that's cool...
You want to get to messing with religion, we'll have to lock the thread down. We've already had one tiff this week, and I'm not anxious to have another.
In short...don't go there.
Period.
Grey
P.S.: Pixie, I'm glad you're happy with your amp. That's what it's all about.
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.