or Connect
AVS › AVS Forum › Audio › Audio theory, Setup and Chat › [B] NEED HELP - Diganosing audio differences between source components [/B]
New Posts  All Forums:Forum Nav:

[B] NEED HELP - Diganosing audio differences between source components [/B] - Page 11

post #301 of 361
Quote:
Originally Posted by audiophilesavant View Post

So spinning is quoting the >85db from the specifications published by Harman? For both the No. 532H and the No. 53.

If all you had done was point a link to that page maybe that would have been OK. But that's not what you did. Let's look at what you said:

Quote:
Originally Posted by audiophilesavant View Post

Are you planning on getting a better amplifier than the ones offered by Harman too? Krell, for example, has some with signal to noise ratios in the 107db (unweighted) range. Why settle for Harman's 85db?

Let's examine the data in front of us. Here is the spec for the Mark Levinson 532h: http://marklevinson.com/Products/Details/24#specs
Signal-to-Noise Ratio >85dB, reference level: 2.83 Vrms [ 1 Watt into 8 ohms]

This is the spec from Krell: http://www.krellonline.com/s275.html
Signal-to-noise ratio >107 dB, wideband, unweighted, referred to full power output [275 watts]

See the section in red? The Mark Levinson rates its noise relative to 1 watt of output power, the Krell at full output of 275 watts. Clearly those two numbers cannot be compared against each and is the reason the Krell's spec was much higher.

The Krell is rated at 275 Watts. Backing out that difference means subtracting 24 dB from its spec: 107 - 24 = 83 dB. So its apples to apples stated spec is actually a bit lower than the Mark Levinson’s 85 dB! Yet you asked: "Why settle for Harman's 85db?" That's why

As both John Dawson and John Atkinson explained if you add this differential to Mark Levinson's spec/measurement, you easily go past the dynamic range of flat dithered CD. So the format is the bottleneck, not the amp.

Quote:


You are quoting the measurement of the No. 532H that Stereophile measured. If Stereophile measured 93db, then maybe they got a better than average sample.

The machine is rated at > 85 dB. The measurement didn't show 86 dB, or 88 dB or even 90 dB. It showed 93 dB. That is an 8 dB difference or a noise voltage ratio of 2.5:1. We should all be lucky and have reviews show our specs to be 2.5 times lower than specified .

Quote:


Perhaps you could explain why Harman published a >85db specification, instead of something like say, >86db, >87db, >88db...

Simple: they want to be conservative as high-end companies typically are. I know, because I had seen that spec and had asked Harman about it and that is what they told me. You could have arrived at the same answer if you had spent 5 seconds searching for actual measurements. But who needs data when you can just "assume?" Right?

Let's step back and play this as you did with me earlier: did you see that the two specs were rated differently yet omitted that important qualification hoping without a reference we would not know any better? I won't act like you and say no. I think you genuinely didn't know how to interpret the numbers and technical words used to describe them.

Why did you make this mistake then? Bias. You wanted to find a piece of high-end gear not doing well and you thought you found it with that 85 dB number. Others motivated the same way piled on without doing their homework and without asking you where you got your numbers. It reinforced their preconceived notions and the cause they are fighting. Had the number been 130 dB, you would have searched heaven and earth to find a reason it is wrong -- just like you are doing with Fielder research. But there, the 85 db was good enough to run with it.

If you want me to take your arguments seriously about objectively analyzing the science, then demonstrate objectivitiy. Don't fight me tooth and nail on including psychoacoustics in analyzing dynamic range. Don't read words into research which clearly dispute your counter. And importantly, if you are not aware of how to read simple specifications of amplifiers with respect to dynamic range, why not be less aggressive with these posts? Let the people who know how to interpret the data and engineering fight the fight.

Finally, I hope you see how hard it is to make an argument in this forum . The level of knowledge you need to survive a discussion here is very high. Don't dismiss what Arny and I do as something anyone could do. So cut us some slack and don't keep lecturing us on how to post data and defend it.
post #302 of 361
Quote:
Originally Posted by amirm View Post

That's right: 108 to 115. Well in excess of flat dithered 16 bit...

In absence of the industry performing noise shaping...

Continuing to cast aspersions on professional recording and mastering engineers without the slightest support other than a few random quotes pulled off Gearslutz. You can find a quote on Gearslutz to support anything. Participants include everyone from professional mastering engineers like Bob Katz and DAC designers like Dan Lavry to adolescents in their basements who can't figure out how to plug in a microphone, assuming they know what one is. Just like AVS.

Adobe Audition is a fine recording program. There are several others as well. However, the industry standard DAW is ProTools. Walk into virtually any professional studio and there it is. It ships with two dither plug-ins. One has as its only option noise-shaped dither. The other, POWr Dither, has as its only options three different types of noise-shaped dither. Professional recording engineers generally don't use them; there are more sophisticated ones available. But even if you are a rookie using the industry standard software right out of the box for the first time without reading the instructions, you can't choose flat dither.

Your whole argument rests on the assumption that professional recording and mastering engineers don't have a clue what they are doing. Yet they are sophisticated enough to record music 1) with the widest possible dynamic range, 2) in venues with noise levels below the threshold of human hearing, 3) with no audience present, 4) using microphones whose acoustic self noise and the electrical noise of their analog preamplifiers are below the threshold of human hearing, 4) using 24 bit analog to digital converters and digital processors with noise below the threshold of human hearing, 5) played back using 24-bit DACs with noise below the threshold of human hearing, 6) with a digital volume control instead of an analog preamplifier, 6) with little attenuation so that few digital bits have to be thrown away at the expense of dynamic range, 7) using amplifiers with noise below the threshold of human hearing, and 8) professional monitors which can play peak levels of in excess of 120db with no distortion, 9) in listening rooms with noise below the threshold of human hearing. You assume perfection in the music, environment and equipment, with professional recording and mastering engineers at the controls who are incompetent. As the industry standard. It's preposterous.

Louis Fielder didn't list as a possible limitation to achieving wide dynamic range a lack of skill on the part of recording and mastering engineers. That's just something you gratuitously threw in. And not only assumed the worst, but stated it as fact. In a technical discussion. You owe your friend Bruce Brown an apology for denigrating his profession. And JA as well.
post #303 of 361
Quote:
Originally Posted by amirm View Post


You shouldn't have bothered.

That is true Amir, because of your inability to learn from other than harsh experience, and even then not so much. ;-)

Quote:


You said you performed a certain analysis and I did the same thing which actually confirmed what you said, except that your conclusions from it were incorrect.

Perhaps that happened in some other universe... I wasn't there! ;-)

Quote:


You said and I quote, " the "Samson and Delilah" HDtracks recording that Amir brought to the discussion does have an approximate 3 second lead out (right before the digitial zeroes that Amir tried to include in his analysis) where it appears that just the residual sound of the hall has been recorded.
[/quite]

You want to dispute this?



Its average level is about 45 dB below the peak of the loudest sound on the recording. Spectral analysis shows its largest peak at about 15 Hz (environmental), followed by one at 50 Hz (power line related - it is an European venue) and it does show evidence of noise shaping."

[/quote]

OK.

Quote:


You said that portion has the residual hall noise and certain frequency peaks. I did the same analysis using the same software you had. As I said, i showed the same things you said.

After I corrected you, I guess.


Quote:


Except that your read and interpretation was wrong:

1. There was no noise shaping as the signal was 24 bits. Wasn't that a digital audio 101 mistake to assume such?

No Amir, I have no need to make audio 101 mistakes as long as you are here to supply them in such great quantity! ;-)

The fact that the signal was 24 bits does not preclude quantization and use of noise shaping. You see Amir, despite your apparent obvious false belief otherwise, even 24 bit signals are quantized. They are at the very least quantized when they are converted to digitial from analog.

Now perhaps Amir you live in a universe where acoustic instruments and singer's voices originate in digital fomr, but in this universe I am told by reliable sources that they are analog. ;-)

The evidence of possible noise shaping that I see would be the rise in the noise floor around 30 KHz. IME that is not a common event in the natural world.


Quote:


2. You ignored psychoacoustics that tells us low frequency noise is much less audible.

No I didn't. I was simply observing a true fact, being its relative amplitude in a certain signal.

Quote:


You proceeded to tell us what the peak at 15 Hz was in amplitude and left it at that.

So what?

Quote:


3. I went on to say that your analysis is not correct anyway with the comment, "While one would need to perform the analysis of Fielder/Stuart do to arrive at audibility of noise relative to tones in the threshold graphs..." But since you ran with it, I showed that analyzing the signal at high level points to you drawing incorrect conclusions from it.

I'm sorry Amir but I can't understand where you are going here.

Quote:


Why don't you provide the spectrum so that we can finish this cat and mouse game of "you don't use the same segment?"

Amir, I'm under the impression that after vigoroos correction by me, you eventually got around to analyzing the right data in a proper way.

Quote:


Whatever is wrong here per above, was your idea. You went to do a spectrum analysis and I did the same thing. That aside, you are mistaken to assume that if you keep increasing the number of points the noise floor keeps going down as if to give you a lower value than the signal itself. FFT can add its own noise. It doesn't subtract from system noise.

Amir your problem here is as you have just stated, you are conflating signal with noise. Signals are coherent and have a periodic nature, noise has a random or at least pseudo-random nature. This has a dramatic difference in the results of applying FFT analysis.

Amir, since you would argue with me if I said that water is wet, I will cite one of many authoritative, independent sources:

http://www.ni.com/white-paper/4278/en

"The measurement of noise levels depends on the bandwidth of the measurement. When looking at the noise floor of a power spectrum, you are looking at the narrowband noise level in each FFT bin. Thus, the noise floor of a given power spectrum depends on the f of the spectrum, which is in turn controlled by the sampling rate and number of points. In other words, the noise level at each frequency line reads as if it were measured through a f Hz filter centered at that frequency line. Therefore, for a given sampling rate, doubling the number of points acquired reduces the noise power that appears in each bin by 3 dB. Discrete frequency components theoretically have zero bandwidth and therefore do not scale with the number of points or frequency range of the FFT."
post #304 of 361
Quote:
Originally Posted by stereoeditor View Post

Whether they want it is a matter of personal choice,

Ah, the personal feeling... So it's not about audible choice when music or movie is playing through, got it.
post #305 of 361
Quote:
Originally Posted by diomania View Post

Quote:
Originally Posted by stereoeditor View Post

Whether they want it is a matter of personal choice...

Ah, the personal feeling...

Yes. To draw an analogy, a $1.7m Bugatti Veyron will do 0-60mph in 2.5s and a $14.3k SmartCar will do 0-60mph in 15s. Yet both get you to the mall in about the same time. It's not up to me or you to tell people which they should prefer.

Quote:


So it's not about audible choice when music or movie is playing through, got it.

In your words, not mine.

John Atkinson
Editor, Stereophile
post #306 of 361
Quote:
Originally Posted by stereoeditor View Post

Yes. To draw an analogy, a $1.7m Bugatti Veyron will do 0-60mph in 2.5s and a $14.3k SmartCar will do 0-60mph in 15s. Yet both get you to the mall in about the same time. It's not up to me or you to tell people which they should prefer.

Wrong items to compare. Unlike listening experience, car driving experience constantly involves tactile senses with the item in use.
Even though it still doesn't make sense, the closest you can make it comparable would be to change to something like, a $1.7m Bugatti Veyron will do 0-60mph in 2.5s and a $14.3k car*** will also do 0-60mph in 2.5s. Both get you to the mall in about the same time because the route to the mall is so curvy and bumpy that no matter the horsepower/torque available, you can only go 25 mph at tops. However, Bugatti Veyron's top speed is 268 mph and car***'s is 100 mph. Whether they want Veyron is a matter of personal choice on looks and or bragging rights.

Quote:


In your words, not mine.

I didn't imply those were yours.
post #307 of 361
Quote:
Originally Posted by stereoeditor View Post

Yes. To draw an analogy, a $1.7m Bugatti Veyron will do 0-60mph in 2.5s and a $14.3k SmartCar will do 0-60mph in 15s. Yet both get you to the mall in about the same time. It's not up to me or you to tell people which they should prefer.

The difference being that there is actually a reliably perceptible difference among these cars based on their primary mission. The purpose of a car is to move, and there are very non-trivial differences in how fast the Veyron and the SmartCar move, as well as stopping, agility, and NVH.

In contrast, we are told repeatedly by the major high end ragazines that audio equipment such as amplifiers and DACs sound better, when it fact many if not most of them do not even sound different.

I have no problems with people spending their money any way they want to, including for well-known frauds (e.g. Homeopathy) as long as they in general, are receiving or at least capable of receiving reliable relevant information.

But, when media channels are flooded with false claims, then there is a problem.
post #308 of 361
Quote:
Originally Posted by arnyk View Post

The difference being that there is actually a reliably perceptible difference among these cars based on their primary mission. The purpose of a car is to move, and there are very non-trivial differences in how fast the Veyron and the SmartCar move, as well as stopping, agility, and NVH.

In contrast, we are told repeatedly by the major high end ragazines that audio equipment such as amplifiers and DACs sound better, when it fact many if not most of them do not even sound different.

I have no problems with people spending their money any way they want to, including for well-known frauds (e.g. Homeopathy) as long as they in general, are receiving or at least capable of receiving reliable relevant information.

But, when media channels are flooded with false claims, then there is a problem.

Are you claiming Amir's amps won't sound audibly better than professional grade amps that are driven within their limits even though Amir's amps cost 10 times their price?
post #309 of 361
Quote:
Originally Posted by hd_newbie View Post

Are you claiming Amir's amps won't sound audibly better than professional grade amps that are driven within their limits even though Amir's amps cost 10 times their price?

I manufacture no amps. My company does sell many amplifiers but there, we sell 100X more pro amps than consumer. Here is a picture of a rack from our gallery:



Look through the rest of the pictures and you won't see anything to do with Mark Levinson.

If you are going to continue to make off-topic and personal remarks, at least do a bit of homework so that you pick the right target to throw those rocks at....
post #310 of 361
Quote:
Originally Posted by amirm View Post

I manufacture no amps. My company does sell many amplifiers but there, we sell 100X more pro amps than consumer. Here is a picture of a rack from our gallery:



Look through the rest of the pictures and you won't see anything to do with Mark Levinson.

If you are going to continue to make off-topic and personal remarks, at least do a bit of homework so that you pick the right target to throw those rocks at....

It was an honest question. Thanks for the pictures by the way. It has been a while since we have not seen pictures of your equipment line. I was starting to get worried.
post #311 of 361
Quote:
Originally Posted by amirm View Post

Look through the rest of the pictures and you won't see anything to do with Mark Levinson

Do you no longer carry Mark Levinson amplifiers like the No. 532H and the No. 53? If not, why did you drop them?
post #312 of 361
Quote:
Originally Posted by audiophilesavant View Post

Do you no longer carry Mark Levinson amplifiers like the No. 532H and the No. 53? If not, why did you drop them?

What makes you ask this question?
post #313 of 361
Quote:
Originally Posted by hd_newbie View Post

It was an honest question.

I took it as such. An honest personal attack toward me .

Quote:


Thanks for the pictures by the way. It has been a while since we have not seen pictures of your equipment line. I was starting to get worried.

Actually, it wasn't that long back that I answered the same attack on my views in this very thread:

Quote:
Originally Posted by amirm View Post

It makes not one bit of difference to our sales at Madrona if people believe these things or not. 90% of our revenue is from custom home electronics which includes things like automated lighting, shades, audio/video distribution, home theaters, architectural products (i.e. electronics you don't see), etc. To wit, here is the picture on our home page:



The picture is there to show that there speakers that are completely invisible around that TV. That is more important to most of our customers than any of these things we discuss here. They want elegant electronics and we try to do the best we can to deliver a product to them. The above speakers for example are driven by commercial Crown amplifiers (i.e. very high value) with custom DSP processing by us to make them sound good. And they sound surprisingly good for speakers you can't see at all.

You can read the home page and you see it is devoid of high-end audio talk. Everything is based on explanation of science. If I can't convince you that way, I won't be the one talking to you about buying it. Read through these articles on our web site: http://www.madronadigital.com/Library/Library.html. And then compare them to what you read on an audio store web site. While they cut and paste manufacturer marketing brochures, I put effort toward educating our customers. The more they know, they better it is because our company differentiation is based on design and engineering expertise.

You see how I already made the point regarding commercial amps and their much better price/performance ratio? Since you repeated the question, I thought I provide more proof with those images since my words weren't enough. So if folks are worried, they only need to look to you for the reason .

It is not like I didn't ask repeatedly for us to not dive in this direction. It is still not too late to stay focused on OP's question.
post #314 of 361
Quote:
Originally Posted by amirm View Post

What makes you ask this question?

Curiosity.
post #315 of 361
Quote:
Originally Posted by hd_newbie View Post

Are you claiming Amir's amps won't sound audibly better than professional grade amps that are driven within their limits even though Amir's amps cost 10 times their price?

I'm suggesting that based on many, many past listening experiences and technical performance measurements, the above could be true in many if not most cases.

One of the first things I notice when I stop looking at pro amps and start looking at consumer amps, is that 2 ohm performance is generally not specified.
post #316 of 361
Quote:
Originally Posted by amirm View Post


Let's examine the data in front of us. Here is the spec for the Mark Levinson 532h: http://marklevinson.com/Products/Details/24#specs
Signal-to-Noise Ratio >85dB, reference level: 2.83 Vrms [ 1 Watt into 8 ohms]

This is the spec from Krell: http://www.krellonline.com/s275.html
Signal-to-noise ratio >107 dB, wideband, unweighted, referred to full power output [275 watts]

See the section in red? The Mark Levinson rates its noise relative to 1 watt of output power, the Krell at full output of 275 watts. Clearly those two numbers cannot be compared against each and is the reason the Krell's spec was much higher.

The Krell is rated at 275 Watts. Backing out that difference means subtracting 24 dB from its spec: 107 - 24 = 83 dB. So its apples to apples stated spec is actually a bit lower than the Mark Levinson's 85 dB! Yet you asked: "Why settle for Harman's 85db?" That's why

..and I explained why the above is yet another rookie mistake. To reiterate, noise can easily increase significantly at high power levels. One common reason is that ripple and other noise on the main power supply buses generally increases dramatically when a great deal of power is being drawn.

Power amplifiers have a characteristic known as power supply rejection, which varies with frequency and may be say 80 dB (one part in 10,000) @ 60 Hz for a good power amplifier. This usually gets worse at higher frequencies on a 6 dB/octave curve. At full output on the test bench there may be 6 volts of ripple on the plus and minus power supply buses. Let's say that we have a power amp that has 160 microvolts of noise at low power levels or while idling. At full output there is a minimum of an additional 6/10,000ths or 600 microvolts of noise creeping past the amp's power supply rejection. This rather obviously is different from presuming that there is no increase in basic noise at full output.

Based on the erroneous information that people are posted here on this topic, I'm now sure that I'm among the very few here who has actually probed the innards of power amps while they are being driven hard. But, that's what happens in the real world. ;-)
post #317 of 361
Quote:
Originally Posted by arnyk View Post

..and I explained why the above is yet another rookie mistake. To reiterate, noise can easily increase significantly at high power levels. One common reason is that ripple and other noise on the main power supply buses generally increases dramatically when a great deal of power is being drawn.

That is speculation on your part in this case and beside the point. Two specs *from manufacturers* were compared by AudioSavant. One was relative to 1 watt, and the other full power. Those two numbers cannot be compared as he did. Period. If you don't like the 600 watts that the ML can put out, use 300 watts. Or 200. Even at 100 watts you still have to add 20 dB to the ML number which based on its measured performance brings its dynamic range up to 113 dB, not "85 dB" that AudioSavant repeatedly mentioned.

Quote:


Based on the erroneous information that people are posted here on this topic, I'm now sure that I'm among the very few here who has actually probed the innards of power amps while they are being driven hard. But, that's what happens in the real world. ;-)

Your chest pumping is noted . But as I just described, immaterial. But if you want to prove John doesn't know what he is talking about, by all means, make your case against his resume:

Quote:
Originally Posted by John Dawson View Post

Just a point on the s/n ratio of power amplifiers.

Harman's figure of >85dB referred to above is quoted ref 1W output into 8 ohms (2.83V rms). If it is a 100W amplifier then that translates to >105dB ref 100W and so on.

-85dB ref 1W (2.83V) is about 160 uV at the speaker terminals. Long years of experience have told me that 200uV or so of white noise is audible in a very quiet room with sensitive speakers, so one should aim at 100uV or less. THX if I recall correctly requires 60uV, which is quite tricky to achieve in practice, though they might allow a little hum too as the lower harmonics will not be audible.

Assuming one achieves 60uV (approx the same as 40uV A-weighted for white noise without too much hum) then one is getting about 113dB unweighted ref 100W output, or 116dB A-wtd. For a 250W amplifier this gets up to 116dB unwtd or 120dB A-wtd.

John Dawson (Arcam)
post #318 of 361
Quote:
Originally Posted by arnyk View Post

To reiterate, noise can easily increase significantly at high power levels. One common reason is that ripple and other noise on the main power supply buses generally increases dramatically when a great deal of power is being drawn.

If you refer back to the spectrum I showed of the Mark Levinson No.532H reproducing a 50Hz tone 2.6dB below actual waveform clipping, all the AC supply components are at or below -125dB. So while inexpensive amplifiers often do have supply ripple appearing on their output as they approach their maximum, this Mark Levinson amplifier doesn't to any significant extent. So your earlier statement that this amplifier suffers from "noise modulation" at high powers - "Amp noise tends to increase with power output and may increase more rapidly as we approach the clipping point," in your words - is still incorrect, Mr. Krueger.

John Atkinson
Editor, Stereophile
post #319 of 361
Quote:
Originally Posted by stereoeditor View Post

If you refer back to the spectrum I showed of the Mark Levinson No.532H reproducing a 50Hz tone 2.6dB below actual waveform clipping, all the AC supply components are at or below -125dB.

I don't see that, and I've asked you politely for more explanation or a clearer graphic. John, reiterating the same unsupported claim over and over makes it seem like you really have nothing else to offer but the same old empty claim over and over again.
post #320 of 361
Quote:
Originally Posted by amirm View Post

That is speculation on your part in this case and beside the point.

Amir, thanks for showing that you really have nothing to offer but dismissive false claims. I've explained this point several times and provided an example based on real world experiences. Its up to you to provide counter evidence or admit that your claims that this effect does not exist are themselves baseless speculations on your part. BTW Amir given the many basic errors you have made in this discussion even just my speculations naturally trump yours because I have the hands-on experience and knowledge of audio theory as applied to power amplifiers to back myself up.

Quote:


Two specs *from manufacturers* were compared by AudioSavant. One was relative to 1 watt, and the other full power. Those two numbers cannot be compared as he did. Period.

Since you can't provide quotes to support your claim above, and I can't find it with a reasonable amount of effort, I am forced to dismiss your claim, Amir.

Quote:


If you don't like the 600 watts that the ML can put out, use 300 watts. Or 200. Even at 100 watts you still have to add 20 dB to the ML number which based on its measured performance brings its dynamic range up to 113 dB, not "85 dB" that AudioSavant repeatedly mentioned.

I don't see specifications or measurements to support any claims at all about the dynamic range of this amp at high power levels. JA does not seem to have anything to offer, either.
post #321 of 361
Quote:
Originally Posted by arnyk View Post

Amir, thanks for showing that you really have nothing to offer but dismissive false claims. I've explained this point several times and provided an example based on real world experiences. Its up to you to provide counter evidence or admit that your claims that this effect does not exist are themselves baseless speculations on your part.

There is no "claim." The man used one spec measured relative the amp barely producing sound vs another that was whaling at maximum volume. You have made no point in this regard.

Quote:


BTW Amir given the many basic errors you have made in this discussion even just my speculations naturally trump yours because I have the hands-on experience and knowledge of audio theory as applied to power amplifiers to back myself up.

Please Arny. I have worked on hundreds of amplifiers. I put myself through school repairing them. And I used o build them when I was younger. I know the theory, I know the practice. You want to set someone right, you need to start with Audiosavant who missed obvious designation that *three* of us say he missed.

Quote:


Since you can't provide quotes to support your claim above, and I can't find it with a reasonable amount of effort, I am forced to dismiss your claim, Amir.

I provided quotes. You are blind to them. Here they are again:

Let's examine the data in front of us. Here is the spec for the Mark Levinson 532h: http://marklevinson.com/Products/Details/24#specs
Signal-to-Noise Ratio >85dB, reference level: 2.83 Vrms [ 1 Watt into 8 ohms]

This is the spec from Krell: http://www.krellonline.com/s275.html
Signal-to-noise ratio >107 dB, wideband, unweighted, referred to full power output [275 watts]


Do you think these two specs can be considered equiv. measurements? Please don't post anything else but answering this simple question. If you had built the ML amp and someone used the Krell amp to tell you that your amp is 22 dB noisier, you would buy that?

Quote:


I don't see specifications or measurements to support any claims at all about the dynamic range of this amp at high power levels. JA does not seem to have anything to offer, either.

You claimed that you are the only one who knows about real life of amplifiers and on that basis, we should take your word. I showed you how John is highly qualified to know this field as well and his view was also 100% consistent with the two of us that Audiosavant did not know how to read the ML spec correctly when he thought its dynamic range is less than the CD. Audiosavant goes to thank him for the information yet keeps going on about the 85 dB number. So I showed him where he is wrong. That you can't measurement the ratios of contaminants relative to a nearly empty can in one instance and a full one in the other. You only need to be able to read English and the very basics of electrical engineering to know that you can't do that.
post #322 of 361
Quote:
Originally Posted by arnyk View Post

Quote:
Originally Posted by stereoeditor View Post

If you refer back to the spectrum I showed of the Mark Levinson No.532H reproducing a 50Hz tone 2.6dB below actual waveform clipping, all the AC supply components are at or below -125dB.

I don't see that. . .

In which case, Mr. Krueger, you are perhaps not as cognizant of amplifier measurements as you claim to be. :-)

I have attached the graph to this posting. Look at the levels of the spectral components at 120Hz, 180Hz, 300Hz, 420Hz etc - not random noise components, please note, but correlated - and tell me what you see. 0dB in this graph is equivalent to 33.15V RMS.

John Atkinson
Editor, Stereophile
LL
post #323 of 361
Quote:
Originally Posted by amirm View Post

Please Arny. I have worked on hundreds of amplifiers. I put myself through school repairing them. And I used o build them when I was younger. I know the theory, I know the practice. You want to set someone right, you need to start with Audiosavant who missed obvious designation that *three* of us say he missed.

Fact is Amir I just told you about something that virtually every power amp you ever worked on did, and you dismissed it out of hand. That says it all - you may have fixed them but you never understood how they worked the way I do.

BTW Amir I also worked on a audio equipment repair bench when I was in school, and fixed a ton of equipment as well. But my experience goes way beyond that. I've been designing amps from scratch since I was in middle school. I designed and built power amps and preamps when tubes were all that we had. I also designed and built SS equipment, and I'd do it again in a heartbeat if there was a need.

The big difference between you and I Amir appears to be one of observation and critical thinking. Remember, I invented ABX for audio which was different and improved as compared to how it was being done until that time under the same name, I did the first consumer and pro audio ABX tests, and I reported the first ABX results.

Amir, you appear to many of us to continue to be in denial about what the nearly 40 years of doing proper subjective tests has taught many of us. You appeaer to be still in denial about the need for even the basics like level matching!

Quote:


Let's examine the data in front of us.



Quote:


Here is the spec for the Mark Levinson 532h: http://marklevinson.com/Products/Details/24#specs
Signal-to-Noise Ratio >85dB, reference level: 2.83 Vrms [ 1 Watt into 8 ohms]

This is the spec from Krell: http://www.krellonline.com/s275.html
Signal-to-noise ratio >107 dB, wideband, unweighted, referred to full power output [275 watts]


Do you think these two specs can be considered equiv. measurements?

Of course not. Both are highly incomplete. For the record, here is a proper set of specifications:


Low level signal-to-Noise Ratio >xxx dB, operational signal level: 60 dB below 2.83 Vrms with an 8 ohm load. SNR calculated based on 2.83 volts (IOW, a dynamic range test).

High level signal-to-Noise Ratio >yyy dB, operational signal level: full rated output into the lowest rated resistive load.

In both cases the amplifier should be operated with any internal level controls set to maximum and the amplifier should be driven by a source with 100 ohms source impdeance.

Test signal: 1 KHz sine wave.


If someone wanted to really frost the cake, there would be a FFT that illustrates the results of each test.

The numeric test result would be based on the quotient of the 1 KHz tone and everything else with a measurement bandwidth of 20-20 KHz @ -3 dB, with 12 dB/octave skirts. Applying an A weighting could also be done.
post #324 of 361
Quote:
Originally Posted by arnyk View Post

Fact is Amir I just told you about something that virtually every power amp you ever worked on did, and you dismissed it out of hand.

Well, putting aside the fact that it is very doubtful that you have worked on the class of amplifiers we are talking about, and have no ideas of the design of either amplifier in question, you were off topic. The topic was simple: Amir's company carries an amplifier which has 22 dB worse spec than another brand and therefore let's not listen to him showing us AES published research on dynamic range of music. That is the topic and nothing more.

I repeatedly said we should not go there. Meanwhile John Dowson, chief designer at a major company and longtime member of this forum post that Aaudiosavant's read of the specification for said amp was incorrect. Aaudiosavant then confirms that he learned something by:
Quote:
Originally Posted by audiophilesavant View Post

Thanks for the spin-free factual information.

That should have been it and him saying he was mistaken. But no, he asks me what was wrong with what he said and that Mark Levinson probably got lucky that the stereophile measurements of the unit were 8 dB better. No acknowledgement that the amp specs he looked at were stated very differently.

So I point out how he omitted key qualifications from the two manufacturers which clearly showed the test conditions were radically different. And that difference can be explained by how these values are stated based on what John Dawson, John Atkinson and I view the specs and data in hand.

You then jump up and down with a theory of power supply ripple getting into the output. You have no data whatsoever about either amp doing this and at what level. You have a hypothesis which is disputed by John Atkinson who has actually measured the device. He has data and you don't but apparently that doesn't mean anything. Bottom line is that you are throwing something at the wall to see if it sticks. That's cool. You are trying to defend brother Audiosavant. The whole camp must be defended or else, there is a crack in the armor. I get it. But it doesn't mean I don't see past it and try to focus us on the main issue which is him putting down the reputation of a major brand of amplifiers. This is the fact. Everything else about you teaching us about amplifiers and such is debating tactic and an insulting one at that.
Quote:


That says it all - you may have fixed them but you never understood how they worked the way I do.

That would be a neat trick . Fixing amplifiers without knowing how they work. How do you do that? Replace part by part until you get lucky and the thing works? Maybe that is how you fixed amp but not me. I grew up with electronics and analog design. That is what my hobby was. We discuss digital audio but that is the product my last two decades of digital design. For the prior 20 years what I loved and played with was analog. My degree is in Electrical Engineering. It is not an accident that I can take a systems view here whether we are talking about digital or analog.

But as I noted, the problem you have with anointing yourself god of audio is not just me but the three of us. I asked you to compare yourself to John Dawson. If you want to post again on who is bigger than the other, please include how you know more about amplifiers than he does in your next post.
Quote:


BTW Amir I also worked on a audio equipment repair bench when I was in school, and fixed a ton of equipment as well. But my experience goes way beyond that. I've been designing amps from scratch since I was in middle school. I designed and built power amps and preamps when tubes were all that we had. I also designed and built SS equipment, and I'd do it again in a heartbeat if there was a need.

Well, clearly there is no need for your services . Come back when your audio decisions make or break a company. Then we know whether you made the right calls or not. You talk about double blind tests all the time. The type of tests that I participated in resulted in the specifications for the blu-ray disc to be different. They have resulted in technology getting adoption in billions of devices. It was not some audio club work where if you got it wrong, no harm would come from it. I am confident you have never built a 600 watt class AB amplifier that had to be manufactured and become successful in the market but maybe you surprise us with showing its schematic so that we can critique it.
Quote:


The big difference between you and I Amir appears to be one of observation and critical thinking. Remember, I invented ABX for audio which was different and improved as compared to how it was being done until that time under the same name, I did the first consumer and pro audio ABX tests, and I reported the first ABX results.

And you deserve credit for that. You deserve none for taking that as a bat to shoot down technical research as you have been doing in this thread in an attempt to dumb down the performance of audio in every thread you step in. I know a lot about double blind testing and objective evaluation of audio yet I don't let that literally blind me to more knowledge. When I had this debate about dynamic range in the other thread, I put forward the Bob Stuart paper. You called that marketing material and said the thing you value is AES publications. I point out that the paper was an AES publication. Did you back off? No. You continue to then put down the AES research just the same. Here is how it went down:
Quote:
Originally Posted by arnyk View Post

Interestingly enough this paper is neither an AES conference paper or a JAES article." At least I can't find it published that way. It appears to be a rewrite of a 1988 (24 year old!) article in the now-long-departed Audio magazine. It's a corporate white paper that has no standing as an industry standard or recommendation.

This paper is arguably part of the support for SACD and DVD-A which are now known to be failed technical initiatives that failed to make it in the mainstream consumer marketplace.

The paper in question is full of unsupported assertions.

How knowledgeable are you if you don't know some of the most famous work in this area published at AES including the Journal? I quoted Fielder research to you there and here. As with this thread, you went after him personally saying he is a failure as an engineer because he was responsible for such things as HDCD and SACD. Is this how we debate science? We take someone who worked for Dolby who provided the licensing for a competing format called DVD-A and put SACD's failure at his feet? And what about relevance? Why would his research be wrong even if he did work on SACD? You either know how to shoot holes in that or you don't and clearly you don't. So you go after him personally.
Quote:


Amir, you appear to many of us to continue to be in denial about what the nearly 40 years of doing proper subjective tests has taught many of us. You appeaer to be still in denial about the need for even the basics like level matching!

No. I know everything you know about your camp because I am in it. I look at audio objectively. In this entire thread not a single subjective argument was used. Everything was about reviewing the research in front of us. Research which has been published at AES. You keep spinning this as arguing with my views. But your arguments are with top researchers in this field.

You are up in arms because their research invalidates your 40 year campaign to dumb down audio performance. You can't let that happen. Looking right is more important to you than the truth about audio being presented. You demonstrate this lack of objectivity in thread after thread, post after post by getting personal. You could have done a million ABX tests but if you are biased and spiteful in how you look at the science, you provide little value in these discussions.

Worse yet, you keep complimenting yourself when you run out of answers. You say dynamic range of music is 70 dB based on noise floors you have measured. I show how you ignored psychoacoustics by using a single number instead of looking at spectrum. How bad is it to be caught in denial about how we hear noise? Pretty bad. Your answer? You are the only one who knows this field.

When we test audio codecs, we know how they work so we use test material that show their weakness. Without such tracks, we would be putting our head in the sand thinking they are all perfect. So please don't keep saying how many tests you have run. If they are done with ignoring the science as basic as our auditory threshold, masking and other psychoacoustics, then they are of little value. They might impress casual posters but not people who are in the industry and are not blind to the science, pun intended .

To wit, I am confident you had never seen the research papers I have presented here. Else, you would have had to acknowledge that you are taking a contrarian position to research in this area. You don't even have an AES membership to read these papers. That is not a sign of someone who still wants to learn. You think you have figured it all out. Good for you. But it is not good for me or researchers who still present papers at these conferences and do this for a living.

No doubt you are going to come back with guns blazing. But I am not the first objectivist you have turned sour toward your views and style of arguing. As I noted, there are others:
Quote:
Originally Posted by terry j View Post

well, in that case let me thank you [amirm] for your contributions. I KNOW I could not have kept my patience as you have, let alone maintained a sense of humour!

It's funny how hard *we* can go to maintain our rightness, and how quickly that line is crossed where we no longer wish to learn (despite our objections to the contrary) where we fight tooth and nail...usually because we know our position is so tenuous that the slightest 'loss' means the whole game is over.

FFS, Amir has sat here page after page and SHOWN how, and under what possible conditions jitter may be audible.

Hey, if it were a cable debate, and we showed with maths and sims that there could not possibly be a difference, well that would have proved it no?

So why the **** in an 'argument' where the shoe is on the other foot does it suddenly become irrelevant what the science says??

My take on what the fear might be is the worry of what might happen if we concede a point of argument. The 'other side' will drive a frickin lorry thru the door if we do.

I mean, there only has to be ONE person who hears a power cord (for sake of illustration) in what seems to be a proper test and the whole frickin lot of the rest of them will claim it as proof that they too can hear it.

No they can't, 'one in a million' means just that. But we KNOW every single one of them thinks they can hear it, using that person as proof, and even less urge to test the truth properly. After all it has been shown.

So, we had better clamp down HARD on the one ever coming out, if only to keep the lid on the rest.

So, move on to something far less controversial than PCs, but as long as it falls into audiofool territory we had better clamp down on that too. It is just safer that way, keep each and every genie in the bottle.

So the need to put amir in his place, and keep the lid hammered on tight. Because the ramifications of this little argument go waaaay past it's tiny borders.

""Oh, but amir has not given any evidence of audibilty"" (apart from the science you mean? The science that would be perfectly acceptable in a different argument, that the one we are talking about???).

Be totally honest here. If he told you that he had found, to his satisfaction, that turning the front panel on and off on his thingamabob had an audible difference, would you accept that?

What then his findings of jitter?

We know you would not accept his results, the genie is too terrifying to contemplate.

So don't come back at me with 'amir has yet to show audibility' ok? It is a definitional thing you know. Some things, by definition, are inaudible.

Bit like cancer, it cannot be cured hence any cure of cancer is untrue (why we are always then exhorted to donate to cancer research is beyond me).

All of you could be right, it may be completely inaudible.

But you sure as hell have not shown it by your arguments. Unless 'nanah nanah nah' counts as an argument.

You keep worshiping my ex-architect JJ as agreeing with your views. With apology to John for quoting him, this is what he has to say about you:
Quote:
Originally Posted by stereoeditor View Post

You remind me of a conversation I once had with JJ when your name came up. He shook his head and pondered aloud on the fact that an unreasonable advocate for a reasonable position is the worst of all worlds, or words to that effect.

I apologize for being blunt but I can't think of anyone giving objectivity in audio a worse name than you. Whether it is the substance or style, you take us places that you should not go as this post of yours shows. You take what could be a proper and constructive discussion and turn into an unprofessional, insulting and grating conversation. Nothing in audio is worth this kind of treatment where the people in your camp distance themselves from you. It is a hobby for heaven's sake.
post #325 of 361
How many times will you quote same comments from JA and Chu. Again and again and again.

And then the pictures of your equipment and promotion start showing. "Look at my design, can you see speakers? No speakers".

And the best for the last:

"By the way, can I see your resume?". "Better yet, show me your listening system"

This pretty much summarized 15,000 posts.

post #326 of 361
Admitting that the 107db figure I quickly pulled off the Krell site cannot be compared to the >85db figure from Harman's site, I think you are looking at the wrong end of the spectrum when comparing SNRs of amplifiers. Instead of trumpeting the SNR of amplifiers at near clipping, why not examine the SNR of amplifiers when they are reproducing soft sounds and operating at a fraction of reference level? Assuming an average speaker with a sensitivity of 89db, at reference level (2.83v into 8 ohms = 1 watt), the level in the listening room is 89db, which is pretty loud. What is the SNR of those amplifiers when they are producing soft sounds at, say, 60db or 30db or even 4db which, according to Fielder, is the threshold of human hearing for broadband noise. Apply your magic formula and let us know where the noise of the amplifier exceeds the level of the music itself, for example those delicate note decays and reverb tails audiophiles wax so poetically about. You've gotten the 107db SNR of the Krell amplifier at full power down to 83db at reference level. Why stop there? The same goes for the Levinson amplifier. It's certainly not something Fielder addressed in his paper.
post #327 of 361
Quote:
Originally Posted by audiophilesavant View Post

Instead of trumpeting the SNR of amplifiers at near clipping, why not examine the SNR of amplifiers when they are reproducing soft sounds and operating at a fraction of reference level?

You can work this out for yourself by looking at the THD+N vs power graphs I publish for all the amplifiers Stereophile reviews. (See, for example, figs.3, 4 and 5 at http://www.stereophile.com/content/m...r-measurements . Fig.3 is also shown below.) The uniform upward slope of the trace with decreasing power below a few tens of watts indicates that the percentage is noise rather than THD - the measurement bandwidth is 22Hz-80kHz - so you can extrapolate the S/N ratio at any level you wish.

John Atkinson
Editor, Stereophile
LL
post #328 of 361
Quote:
Originally Posted by stereoeditor View Post

You can work this out for yourself by looking at the THD+N vs power graphs I publish for all the amplifiers Stereophile reviews...so you can extrapolate the S/N ratio at any level you wish.

I am not familiar with the formula you and John Dawson are using and, frankly, the graph is a little confusing to me with the y-axis being expressed as a percentage.

Assuming an average speaker with sensitivity of 89db, I believe the appropriate voltage levels for sound levels I was inquiring about are (halving voltage for each 3db reduction in sound):

89db: 2.83 volts
60db: ~ 0.003 volts
30db: ~ 0.000003 volts
4db: ~ 0.000000001 volts

If it's not too much trouble, could you calculate the SNR at those levels for me.

Also, could you tell me if my thought process is correct that when reproducing very soft sounds like note decays and reverb tails, the level of noise in the amplifier (and preamplifier) is an issue, recognizing that you can hear into the noise floor to some extent.

Thanks.
post #329 of 361
Quote:
Originally Posted by stereoeditor View Post

You can work this out for yourself by looking at the THD+N vs power graphs I publish for all the amplifiers Stereophile reviews. (See, for example, figs.3, 4 and 5 at http://www.stereophile.com/content/m...r-measurements . Fig.3 is also shown below.) The uniform upward slope of the trace with decreasing power below a few tens of watts indicates that the percentage is noise rather than THD - the measurement bandwidth is 22Hz-80kHz - so you can extrapolate the S/N ratio at any level you wish.

The uniform upward slope does not uniquely support the idea that the cause is noise. It does support the idea that the source has a constant amplitude that does not change with increasing power. This very characteristic can be shared by equipment with nonlinear distortion that is entirely composed of even order distortion.
post #330 of 361
Quote:
Originally Posted by audiophilesavant View Post

I am not familiar with the formula you and John Dawson are using and, frankly, the graph is a little confusing to me with the y-axis being expressed as a percentage.

Assuming an average speaker with sensitivity of 89db, I believe the appropriate voltage levels for sound levels I was inquiring about are (halving voltage for each 3db reduction in sound):

89db: 2.83 volts
60db: ~ 0.003 volts
30db: ~ 0.000003 volts
4db: ~ 0.000000001 volts

If it's not too much trouble, could you calculate the SNR at those levels for me.

If you model a power amplifier as containing a noise source that does not change with power levels, which isn't a bad model for a < 100 watt power amp running with output levels below 1 watt, then just subtract the dBs.

IOW if the amp has SNR of 85 dB at 1 watt or 89 dB SPL, at 60 dB SPL the SPL is 29 dB lower and so is the SNR - it is now 56 dB.



Quote:


Also, could you tell me if my thought process is correct that when reproducing very soft sounds like note decays and reverb tails, the level of noise in the amplifier (and preamplifier) is an issue, recognizing that you can hear into the noise floor to some extent.

(1) When you start most media players up playing a recording, the increase in background due to noise on the recording is usually apparent, right? If that's true then the recording is its own worst enemy when it comes to reverb tails.

(2) If the background noise in the room is high enough, then you won't hear an increase in noise when you start playing the recording. Then the room noise is your worst enemy when it comes to hearing reverb tails.

If the room is very quiet, then simply turning on the equipment will cause an audible increase in noise. If you stick your head right next to the speaker, you will probably be able to hear an increase in noise when you power up your equipment. Then start playing the recording, and see whether its background noise increases the noise you hear. If starting up the recording causes an increase in audible noise, then you are back at (1).
New Posts  All Forums:Forum Nav:
  Return Home
  Back to Forum: Audio theory, Setup and Chat
AVS › AVS Forum › Audio › Audio theory, Setup and Chat › [B] NEED HELP - Diganosing audio differences between source components [/B]