Numbers matter but are they what consumers are most concerned with?

[Zinda]

Why are tests and evaluations done with little mention of how well amps do with difficult loads. If my speakers have big dips in resistance at high frequencies (Martin Logan Prodigy) how well do the class D amps do with that, especially when I see these amps tend to have difficulties with those high frequencies? Could this bring bigger issues with possible drastic outcomes. Do they have filters built in that prevents high frequencies from entering the amp? Amp output and stability are different for each load, maybe a sweet spot test that tells optimal loads and effect on thermal or auto shut down. 1 minute long tests don't tell me anything of real world use. If it puts out gobs of power for 2 minutes, what good is it to have those numbers when it's not practical or useable?

I realize I would never hear anything above 16KHz no matter how loud you played it, but I also know that AB amps will have ratings that claim proper flat output far past 50KHz with many at 150KHz, speaker makers are claiming 30HKz quite often (even though the only ones to enjoy that would be your pets). D amps seem to stop their tests at 20KHz but that limitation is not considered to be a deduction, yet having an undectible noise floor well beyond hearing gives an amp a super high score. The factors that I think are important are never considered while things I expect to be outside my hearing should have no importance after that threshold is achieved and is expected in this day and age.

This makes me wonder exactly what the most important factors are for scoring. Once any device reaches the human hearing limits wouldn't that be a given, just a check box and not the main point for scoring? I'm thinking that there's much more to test that readers need to know before they could ever use these results for anything.

Since factors that are most important are subjectively different between buyers, that means each person has the one thing that immediately strikes all other things out right from the start regardless of test results.

So for me #1 is capability to handle difficult loads, yet there's nothing that sorts through the results based on that critical factol. Watt output along with current ratings is part of this, I would have this in the #2 spot, but include it at #1.

#2 is the actual physical noise it produces, some amps use fans but some fans are not heard, at least not until they turn on after playing loud, then what's their noise during quiet passages, how fast do they cool down? That leaves another question, why don't makers use sound sensors to control fan speeds in conjunction with typical heat as an overiding 2nd backup. The fans can run full blast during loud times (even if they're still cool) and take a bit of chance by quickly dropping off when it's quiet, since air will continue to move after the fan stops. Allowing the user to set that adjustment for sensing volume. There are also amps that actually make audible noise through the boards and make squeaky sounds.

#3 is the inputs, must have XLR, no way to sort that from others tested.

#4 crossover or shelf filtering, and pass through outputs. Replacing any outboard pieces reduces cost of cables and also reduces a mess of cords and saves money. Since an active crossover ends up being set up once but stays in racks because it has to be for cable management.

#5 speaker connections, having an option to order with bananas or speakon or heavy duty spade type only is a huge factor since none that try to be "all in one" has never worked well for any one of them. This goes for RCA or XLRs as well, the boxes are small on many amps but they try to include all options of ins and outs on that small back panel. Removing the unneeded ones allows to use bigger better quality for little extra cost and easier to attach wires.

#6 overall visual impact, but not only how it fits the decore but how it appears. When you connect 4 XLR cables and 2 runs of 10ga wires to one of those tiny boxes, they not only look rediculous, they have a hard time staying where you want them, constantly being pulled by wires and chancing being pulled to the floor by the slightest tug. Having some weight is actually a good thing once they get too light!

To base the main most important thing on something I can't even hear and I expect it to be in that unheard range (no matter who you are) to begin with, starts to make these numbers arbitrary to an overall score. Not to mention how does it actually sound when playing complex frequencies into a moving coil or electrostatic panel. We all know that these factors are way more important to overall sound than the numbers from inert tests using pure tones.

There was a time when computer benchmarks were really popular and then there were parts made that would increase the numbers of test programs but those numbers didn't mean a better faster computer. They also started making tests that would include factors of the tests that represented one technology or the other for a portion of the score. By fudging their importance (whatever they choose is what they consider most important) percentage, they could make one maker better than others that may be the same or better if tested looking for other attributes. By placing the most important factors on things that are expected to be undetectable in the first place, how does that represent anything that a consumer even thinks about?

I'm thinking if it doesn't meet the lowest audible point, then it's out of testing because it missed that first check box of audible noise floor too high to continue, you just need to have that be a given, rather than the one single most important feature. There's no mention of damping factor given, I've found this to be a big factor with AB amps but never see that rating mentioned.

I'm just providing some examples of how these tests aren't practical when shopping for a new amp, im greatful you do this, I used to enjoy the Wiltson car amp dyno tests on you tube because these other factors were covered. With car audio, raw clean watts that cover the full spectrum is king, load handling was next and noise floor was last since your car is noisy. Of course size and current draw were important, but the ratings provided all important factors that buyers wanted or needed to know before buying.

I think it's time to survey users in what they consider to be the most important things when considering new purchases to set some priorities that will bring people here to get started on making their way through the endless maze products with so many different features that constantly are changing. The numbers matter but only after comparing things that remain after getting rid of all those less important things first. So the numbers matter, don't get me wrong, but they only help when compared between very similarly featured items. Price will always be important as well as reliability, those factors change over time and might be considered as a filtering option only, price when new and maybe suggest checking eBay for current used. Or making a deal with eBay to have links on sales offered for comparison.

Reliability could be done with owners - users/members thoughts ending at 5 years time on each tested item simple time of ownership before repair "Check box" totals for first 5 years after release, beyond that maybe an "I'm still using this" regardless of repairs, would give a feel for things that are just worth owning regardless of tests.

I'm bringing this up now since I'm in the market for new amps and the time it would take just to find 1 amp that I can consider for me to use would take me forever if I have to look through numbers that don't supply me with any of my most important factors. Having a way to filter by things that buyers consider the most important factor would give an idea of how to set up filters that remove all those not meeting each factor.

I may not have much to choose from once I set my filters, but that itself would help me by saving time sifting through thousands of pages that don't apply. I guess I could ask AI and get a bunch of fake info or super biased recommendations.

 

[kyuu]

Why are tests and evaluations done with little mention of how well amps do with difficult loads. If my speakers have big dips in resistance at high frequencies (Martin Logan Prodigy) how well do the class D amps do with that, especially when I see these amps tend to have difficulties with those high frequencies?

Where are you seeing class D amps having "difficulties" with high frequencies? Impedance at high frequencies is really only pertinent in terms of load dependency. Get an amp that has no load dependency and it's no longer an issue. Amps, class D or otherwise, having problems driving high frequency loads isn't really a thing AFAIK.

I realize I would never hear anything above 16KHz no matter how loud you played it, but I also know that AB amps will have ratings that claim proper flat output far past 50KHz with many at 150KHz, speaker makers are claiming 30HKz quite often (even though the only ones to enjoy that would be your pets). D amps seem to stop their tests at 20KHz but that limitation is not considered to be a deduction, yet having an undectible noise floor well beyond hearing gives an amp a super high score. The factors that I think are important are never considered while things I expect to be outside my hearing should have no importance after that threshold is achieved and is expected in this day and age.

Not sure what you're trying to say here. Except for possible intermodulation, which has never been shown to be an actual audible issue, you do indeed not need to worry about distortion or noise beyond 20kHz. Amps aren't given "scores", but having a low noise floor and low distortion in the audible bands is a pertinent factor. Where it becomes "inaudible" depends on many factors so it's hard to say where the line should be drawn, unlike knowing where the audible band stops.

This makes me wonder exactly what the most important factors are for scoring. Once any device reaches the human hearing limits wouldn't that be a given, just a check box and not the main point for scoring? I'm thinking that there's much more to test that readers need to know before they could ever use these results for anything.

There aren't hard and fast limits for "human hearing limits" for many things, and it can change depending on the environment and other equipment.

So for me #1 is capability to handle difficult loads, yet there's nothing that sorts through the results based on that critical factol. Watt output along with current ratings is part of this, I would have this in the #2 spot, but include it at #1.

Driving difficult loads at lower frequencies (where the actual energy in actual content is) is generally covered in amp reviews here.

#2 is the actual physical noise it produces, some amps use fans but some fans are not heard, at least not until they turn on after playing loud, then what's their noise during quiet passages, how fast do they cool down?

The only amps that typically have fans are professional ones, and in those reviews that is mentioned. Consumer amps almost never have fans so there's nothing to worry about.

That leaves another question, why don't makers use sound sensors to control fan speeds in conjunction with typical heat as an overiding 2nd backup.

That's a question for the people designing the amps. For professional amps, they're typically somewhere that their noise level is simply irrelevant, plus they're pushing high power levels for long periods of time which makes active cooling a necessity.

#3 is the inputs, must have XLR, no way to sort that from others tested.

For consumer gear, this is not important unless your setup is very atypical. If you prefer it, that's fine. Putting things like this into a table that can be sorted would be web/database development that someone would likely have to be paid to do. If you'd like to volunteer your time to put that together, I'm sure Amir would love to hear from you.

#4 crossover or shelf filtering, and pass through outputs. Replacing any outboard pieces reduces cost of cables and also reduces a mess of cords and saves money. Since an active crossover ends up being set up once but stays in racks because it has to be for cable management.

#5 speaker connections, having an option to order with bananas or speakon or heavy duty spade type only is a huge factor since none that try to be "all in one" has never worked well for any one of them. This goes for RCA or XLRs as well, the boxes are small on many amps but they try to include all options of ins and outs on that small back panel. Removing the unneeded ones allows to use bigger better quality for little extra cost and easier to attach wires.

Again, these are all questions for the amp designers. If you expect any of this to be common on consumer gear, I wouldn't hold your breath. The reviews here focus on consumer hi-fi oriented gear, not professional. Even the professional amps are reviewed with an eye towards their use in a home hi-fi setup.

#6 overall visual impact, but not only how it fits the decore but how it appears. When you connect 4 XLR cables and 2 runs of 10ga wires to one of those tiny boxes, they not only look rediculous, they have a hard time staying where you want them, constantly being pulled by wires and chancing being pulled to the floor by the slightest tug. Having some weight is actually a good thing once they get too light!

Ok. Amir mentions things like this typically at the start of reviews. You want these to be listed in a table too?

To base the main most important thing on something I can't even hear and I expect it to be in that unheard range (no matter who you are) to begin with, starts to make these numbers arbitrary to an overall score. Not to mention how does it actually sound when playing complex frequencies into a moving coil or electrostatic panel. We all know that these factors are way more important to overall sound than the numbers from inert tests using pure tones.

Again, there is no "score" outside of the panther rating. It's most important to read through the entire review for a product you may be interested in rather than relying on a single indicator such as the panther.

I may not have much to choose from once I set my filters, but that itself would help me by saving time sifting through thousands of pages that don't apply. I guess I could ask AI and get a bunch of fake info or super biased recommendations.

Your main complaint seems to be that some particular parameters that you in particular care about aren't listed in a searchable, filterable table. I mean yeah that would be nice, but who exactly are you expecting to put this together? Also most of these things are you are on about are not things consumers in general care about for their consumer gear, btw.

 

[RexrothPigeon]

Why are tests and evaluations done with little mention of how well amps do with difficult loads. If my speakers have big dips in resistance at high frequencies (Martin Logan Prodigy) how well do the class D amps do with that, especially when I see these amps tend to have difficulties with those high frequencies? Could this bring bigger issues with possible drastic outcomes. Do they have filters built in that prevents high frequencies from entering the amp? Amp output and stability are different for each load, maybe a sweet spot test that tells optimal loads and effect on thermal or auto shut down. 1 minute long tests don't tell me anything of real world use. If it puts out gobs of power for 2 minutes, what good is it to have those numbers when it's not practical or useable?

I realize I would never hear anything above 16KHz no matter how loud you played it, but I also know that AB amps will have ratings that claim proper flat output far past 50KHz with many at 150KHz, speaker makers are claiming 30HKz quite often (even though the only ones to enjoy that would be your pets). D amps seem to stop their tests at 20KHz but that limitation is not considered to be a deduction, yet having an undectible noise floor well beyond hearing gives an amp a super high score. The factors that I think are important are never considered while things I expect to be outside my hearing should have no importance after that threshold is achieved and is expected in this day and age.

This makes me wonder exactly what the most important factors are for scoring. Once any device reaches the human hearing limits wouldn't that be a given, just a check box and not the main point for scoring? I'm thinking that there's much more to test that readers need to know before they could ever use these results for anything.

Since factors that are most important are subjectively different between buyers, that means each person has the one thing that immediately strikes all other things out right from the start regardless of test results.

So for me #1 is capability to handle difficult loads, yet there's nothing that sorts through the results based on that critical factol. Watt output along with current ratings is part of this, I would have this in the #2 spot, but include it at #1.

#2 is the actual physical noise it produces, some amps use fans but some fans are not heard, at least not until they turn on after playing loud, then what's their noise during quiet passages, how fast do they cool down? That leaves another question, why don't makers use sound sensors to control fan speeds in conjunction with typical heat as an overiding 2nd backup. The fans can run full blast during loud times (even if they're still cool) and take a bit of chance by quickly dropping off when it's quiet, since air will continue to move after the fan stops. Allowing the user to set that adjustment for sensing volume. There are also amps that actually make audible noise through the boards and make squeaky sounds.

#3 is the inputs, must have XLR, no way to sort that from others tested.

#4 crossover or shelf filtering, and pass through outputs. Replacing any outboard pieces reduces cost of cables and also reduces a mess of cords and saves money. Since an active crossover ends up being set up once but stays in racks because it has to be for cable management.

#5 speaker connections, having an option to order with bananas or speakon or heavy duty spade type only is a huge factor since none that try to be "all in one" has never worked well for any one of them. This goes for RCA or XLRs as well, the boxes are small on many amps but they try to include all options of ins and outs on that small back panel. Removing the unneeded ones allows to use bigger better quality for little extra cost and easier to attach wires.

#6 overall visual impact, but not only how it fits the decore but how it appears. When you connect 4 XLR cables and 2 runs of 10ga wires to one of those tiny boxes, they not only look rediculous, they have a hard time staying where you want them, constantly being pulled by wires and chancing being pulled to the floor by the slightest tug. Having some weight is actually a good thing once they get too light!

To base the main most important thing on something I can't even hear and I expect it to be in that unheard range (no matter who you are) to begin with, starts to make these numbers arbitrary to an overall score. Not to mention how does it actually sound when playing complex frequencies into a moving coil or electrostatic panel. We all know that these factors are way more important to overall sound than the numbers from inert tests using pure tones.

There was a time when computer benchmarks were really popular and then there were parts made that would increase the numbers of test programs but those numbers didn't mean a better faster computer. They also started making tests that would include factors of the tests that represented one technology or the other for a portion of the score. By fudging their importance (whatever they choose is what they consider most important) percentage, they could make one maker better than others that may be the same or better if tested looking for other attributes. By placing the most important factors on things that are expected to be undetectable in the first place, how does that represent anything that a consumer even thinks about?

I'm thinking if it doesn't meet the lowest audible point, then it's out of testing because it missed that first check box of audible noise floor too high to continue, you just need to have that be a given, rather than the one single most important feature. There's no mention of damping factor given, I've found this to be a big factor with AB amps but never see that rating mentioned.

I'm just providing some examples of how these tests aren't practical when shopping for a new amp, im greatful you do this, I used to enjoy the Wiltson car amp dyno tests on you tube because these other factors were covered. With car audio, raw clean watts that cover the full spectrum is king, load handling was next and noise floor was last since your car is noisy. Of course size and current draw were important, but the ratings provided all important factors that buyers wanted or needed to know before buying.

I think it's time to survey users in what they consider to be the most important things when considering new purchases to set some priorities that will bring people here to get started on making their way through the endless maze products with so many different features that constantly are changing. The numbers matter but only after comparing things that remain after getting rid of all those less important things first. So the numbers matter, don't get me wrong, but they only help when compared between very similarly featured items. Price will always be important as well as reliability, those factors change over time and might be considered as a filtering option only, price when new and maybe suggest checking eBay for current used. Or making a deal with eBay to have links on sales offered for comparison.

Reliability could be done with owners - users/members thoughts ending at 5 years time on each tested item simple time of ownership before repair "Check box" totals for first 5 years after release, beyond that maybe an "I'm still using this" regardless of repairs, would give a feel for things that are just worth owning regardless of tests.

I'm bringing this up now since I'm in the market for new amps and the time it would take just to find 1 amp that I can consider for me to use would take me forever if I have to look through numbers that don't supply me with any of my most important factors. Having a way to filter by things that buyers consider the most important factor would give an idea of how to set up filters that remove all those not meeting each factor.

I may not have much to choose from once I set my filters, but that itself would help me by saving time sifting through thousands of pages that don't apply. I guess I could ask AI and get a bunch of fake info or super biased recommendations.

Any questions about what qualities listeners actually prefer should start by carefully reviewing the controlled, blind listening tests performed by toole, olive, et al.

Also agree with the above that you are laboring under many misapprehensions.

 

[DVDdoug]

"The consumer" cares about how it sounds, how it looks, and what features it has, etc. "The numbers" can give you lots if objective information about how it's going to sound and how loud it will go, etc.

It's not practical, and usually impossible, for the consumer to do their own controlled-blind listening tests and "the numbers" are more reliable than casual listening/auditioning, especially with amplifiers where we expect little or no audible difference. And almost no reviewers do proper controlled listening tests, and even if they did we don't all have equal hearing.

Some class-D amps do have frequency response variations near 20kHz with certain loads. But it's NOT common to all of them.

Class-D seems to be becoming "the standard" and it has lots of (potential) advantages. But it's best to judge an amplifier (or anything else) by its actual performance rather than how it's made.

Because of how class-D works and the fact that it has a high-frequency "switched" output that has to be low-pass filtered, they usually don't go into the radio frequency range. It's not hard to make a regular class-A or class-A/B amp that goes from DC to 100MHz or more, but limiting it to the audio range can help to prevent "problems" caused by getting subsonic and ultrasonic signals into the speakers and/or getting any of this "junk" mixed with audio.

If my speakers have big dips in resistance at high frequencies

There was a recent post about a particular class-D chip that sometimes has "serious problems" with some speakers. It appears to be related to "bad circuit design" where the output is not properly filtered and yes, it seems to be with speakers with low impedance at high (or ultrasonic) frequencies.

Most "regular" speakers are inductive (because of the voice coil) which increases impedance at ultrasonic frequencies. But some technologies are not. Piezo tweeters are basically capacitive, and electrostatic speakers may be capacitive too... They are "built like" a capacitor. Capacitors have lower impedance at higher frequencies... the "opposite" of an inductor. Crossovers also have an effect on speaker impedance (but they rarely make it capacitive overall). Speaker specs rarely show any radio frequency characteristics so it's hard to know what's going on beyond the audio range.