Actually, the digital streams can sound pretty good and if you are doing casual listening that can be quite enjoyable. But it is nice to be able to sit down sometimes and just have an out of body experience with your music. That is harder to obtain.
Well, I just feel like going on a subjective rant on sound stage. There are various degrees of this depending on the recording and the playback equipment and environment. So at the most basic level you can hear acoustic cues in the music that make it sound like the music was recorded in a real space. It is not really 3 dimensional sounding but it has a better environment that can be heard. Also, higher frequency low level sounds are used by the ear to locate directions of sounds. These are lost in the compression process. But, there is a plus side to the compression process also. Early digital recordings from 30-35 years ago are so bad sounding that it is often better to hear them compressed because otherwise they are intolerable to listen to. With the advent of higher resolution digital, modern digital recordings are much better. Especially if care was taken in recording.
At the next level of sound stage there is a more distinctive separation between the instruments and more sense of air between the performers in addition to enhance room acoustic cues. At this stage the sound is pretty enjoyable for most people unless you are very sensitive to digital noise in the recording.
So now the next stage is not the highest I have heard but is about where my current system is sounding. It is now sound a little more 3 dimensional and the musical timbres are much more natural and easy, without hard digital edges on the leading edges of sounds. Dynamics are much more realistic and can at times really get you excited. Big contrasts between the loud and soft passages helps to create the illusion of even more sound stage.
next is about the best of I have heard from really well set up and expensive systems. Even more 3 dimensional and visceral sounding. In some cases sounding even more 3 D than real live music is. This is probably caused because the sound system has only two point sources were live sound can have several. Usually you would have to spent 30-50 thousand USD and well set up room to get this. Most people have never heard this. The high end music stores are few in number and only big cities have one or two anymore. Sticker prices will shock you when you go in. Out of most peoples reach unless you can build your own gear. In which case you can often better a store setup.
It might depending on how the AVR handles the sound. If it keeps the output in the analog domain and just used it's internal preamp and amplifier it can sometimes be better. I would say that this was surely true 20 years ago. But now we have these AVR's that actually have really good DAC's in them when implemented properly. But, these DAC's are typically implemented with many compromises in the interest of profits. Add to that, that many of the new AVR's use bad sounding class D amps in them. There is no way to win with a modern AVR for pure music listening.
You're not seriously lumping in "Red Book" with bad audio quality are you??? It's gotten a really bad rap, very wrongly. There's nothing that wrong with 16/44.1, not even those "horribly" ADCs and DACs of the early 80s. Yet there absolutely were terrible sounding Cds. And there were some spectacular sound Cds. It's all the same "Red Book", so how can you have both? Easy. Rec Book is a "garbage in/garbage out" system, just like any other audio storage method. And the really cause of the bad sounding Cds that messed things up for all of them was some very sloppy digital mastering done by novices. Remember, the 1980s was still the era of Vinyl, and those mastering engineers were not the ones who mastered the CD releases. There were many errors made, everything from mastering the CD from a safety copy, or an equalized copy, or playing a Dolby A tape without decoding (yes, there are several known examples). But there were also some really good Cds too.
This crowd seems to like anecdotes, so here are two of mine. In the early 1980s Sony released a low cost PCM recording system. They already had their pro PCM1600 series units that fromatted PCM onto an NTSC video frame recored on 3/4" tape, but they were very expensive. In 1982 Sony released the PCM-F1 selling for $1900, a real breakthrough. During that time I worked for a studio that recorded, among other things, a major symphony orchestra's live concerts which became broadcasts. Those, and recordings in the studio, were recorded on both analog tape and PCM-F1. Recording consoles have a monitor selector that permits selection of recorders for playback. When an analog recorder was selected during record, it was obvious because of the record-play head delay and in this case, Dolby A noise reduction encoding. But the PCM-F1 would play whatever was fed into its video inputs, and video recorders output their input. The system had very little delay. This often resulted in an engineer forgetting to return the monitor selector to the "2-mix" bus, leaving it monitoring the F1 during takes. The only clue was the monitor switch itself, engineers, myself included, couldn't differentiate between the console's stereo bus and the return from ADC/DAC of what is arguable one of the lowliest PCM devices of the decade. It basically ran at "Red Book" specs.
Around the mid 1980s a local audiophile society called me to ask for a tour. That evening, a small group of around 15 guys showed up, and I gave them the cooks tour. Afterwards, we congregated in a smaller studio and talked about recording. One member in particular professed a powerful hate for digital recordings. We decided to have a listen to some recordings we'd made. I had access o the Dolby A masters, the PCM masters, and quite a number of Cds and vinyl (we called them "records" or "albums"). After disc jockey in the adjacent control room, I put on a record. The analog-fan declared it wonderful. I put on a CD, and he condemned it. This went on for a while, then I put on a master tape. He loved it. Then, I admit, I lied. I put on a PCM-F1 tape of an orchestra, and told the group it was analog tape. Again, without hesitation, the group, and our analog fan in particular, extolled it's wonderful analog virtues, the clarity, space, image (soundstage wasn't even a term then), depth and purity. Then I dropped the bomb and informed the group they'd been complimenting a digital recording. Everyone was shocked, and the analog-fan declared that if all digital recordings sounded that good he'd be a convert.
So, my point is, there's nothing all that much wrong with even the cheapest early 1980s Red Book PCM recordings. AND there were a lot of ba CDS. The cause was NOT 16/44.1, and still is not today. There has been no actual proof that anyone can reliably differentiate 16/44.1 vs any high rate you like in blind testing. After 40 years of trying, I'm guessing it's not going to happen now.
All speakers require current. All speakers require power to move their diaphrams to produce sound. There are none that don't. You can't supply current into a load without supplying power too, and you can't deliver power without supplying current. They are not separable.
I recognize your opinion. Please recognize mine. There's no improving his soundstage with a better "front end". It will make not a lick of difference.
I'm sorry, I don't see. You have an old amp that was conservatively-rated. You have an exception. Most from the early days were optimistically-rated. Do you recal the spec, "IHF Power"? It was competely wrong. In your example, your mis-rated amp puts out "almost 110" watts, it's rated for 90wpc. First, that's "almost" .9dB. Wow, massive. Second, did you test at clipping or at 3% THD? Did you test with both channels equally driven and loaded? What frequency? Did you warm the amp per the FTC method? See....lots of questions.
But I really don't think your example proves a thing, other than some devices actually may have been conservatively rated.
Even the entry level Denon AVRS have both coax and optical digital inputs.
The above statement makes no sense. You can't hear a digital stream without an DAC. If you send an uncompressed stream to a DAC you get the original audio. A 128kbps stereo mp3 stream has had 90% of the original PCM data removed based on a perceptual coding scheme develipoed in the mid 1990s. If you feed that, post decoding to PCM, to any DAC, you'll hear at some point what the MPEG2, Layer 3 codec has done. If you can get hold of a 320kbps mp3 stream, the codec artifacts are gone. But not at 128, which is the highest bandwidth that radio stations stream.
Opinion, or fact with evidence? And not really so much a "digital amplifier" as much as a switch-mode PSU driving
Well it's hard to find service manuals for current products, but if you just go back a model year to something mid-market, like the Denon AVR-S730H you find a few interesting things. First, no "digital" amps (though what you're referring to are technicaly not real digital amps but the PWM delta-sigma amps that are around, and some are really good). Those are good old analog amps in there. And what's that DAC? It's the AK4458VN, an 8 channel DAC capable of 115dB s/n, sampling rates up to 768 kHz PCM and 11.2 MHz DSD, and all at 32 bits. Shabby? Not at all. And that's not a top end AVR!
I think this AVR-hate stuff should just stop. You have no evidence other than biased opinion, and haven't even bothered to look under the hood. In short, you don't know what you're talking about.
No, that's not what I said. "Good" is ambiguous. You must state the application, and what you're trying to achieve. In some cases, some AVRs are inadequate. In other cases, they beat a stereo amp quite seriously.
In the case of the OP, where he's listening close to the speakers, and listening at "normal" levels (assuming a speaker sensitivty of 85dB/w/m) he's listening at 1W average, 10W peak power into a relatively easy to drive load. The internal ADCs and DACS are far more than adequate. The amps are conventional analog SS design driven from a high efficiency PSU. The AVR isn't even trying yet. There is nothing to be gained by attempting an amp "improvement" other that the psychological kick in the pants. Assuming he likes his speakers, his issues are entirely acoustic.
In a different situation, someone else might require 500W monoblocks. Then there are the tube/valve people.
Remember, I said, specifically, that all amps DO NOT sound the same in all conditions. In this specific instance, even if there is a radical amp change, as in to a completely different technology, and it results in a small audible difference, no amp of any kind will address his original request for improvement.
IMHO and IME neither specs nor output capability has anything to do for high quality signal reproduction.
Interesting, and I agree somewhat. Power output is tied closely to output impedance/DF, which does make an audible difference if too low. But I meerly used one metric to show now non-challenging the Op's amp needs are. There are certainly other factors. It's just that most of the other amps-make-all-the-difference concepts are completely unverified, and of doubious origin.
When an amplifier is measured and rated it is always done with a resistive load, which has nothing to do with the reactiive load of real speakers. So there will be audible differences between some amplifiers, regardles of measured specifications. And you will never see that difference reflected in the usual FCC compliant measurments,
My little class A amp measures far worse than the Class T (D) amp it replaced in both power and distortion. But the improved sound quality. i.e, imaging, depth, space, detail, bass control, etc... of the class A is very real and very, very obvious.
Remember that a stereo system is just that, a system, a chain. Any lelement in that chain, including the room, that is less than good can degrade the 'sound stage'. And no typical performance statistic is gong to clue you in to what will be good or bad for sure except possbly the decay and group delay in the speakers. And even those can have their euphonic effects.
Sometimes getting a good rig is just luck or good referrals. My current dektop rig is driving Markaudio CHR-70 speakers driven by a Pass ACA Mini fed by a Grace SDAC playing MP3s transcoded to maximum bit rate VBR files. I love it to death. No desktop sound system I have ever owned has ever come close sounding as good. And no specification in any of the components I have reflect how good the whole system will sound. It can only give you reasonable expectations of possibilities. I got lucky. Maybe you will, too.
My little class A amp measures far worse than the Class T (D) amp it replaced in both power and distortion. But the improved sound quality. i.e, imaging, depth, space, detail, bass control, etc... of the class A is very real and very, very obvious.
Remember that a stereo system is just that, a system, a chain. Any lelement in that chain, including the room, that is less than good can degrade the 'sound stage'. And no typical performance statistic is gong to clue you in to what will be good or bad for sure except possbly the decay and group delay in the speakers. And even those can have their euphonic effects.
Sometimes getting a good rig is just luck or good referrals. My current dektop rig is driving Markaudio CHR-70 speakers driven by a Pass ACA Mini fed by a Grace SDAC playing MP3s transcoded to maximum bit rate VBR files. I love it to death. No desktop sound system I have ever owned has ever come close sounding as good. And no specification in any of the components I have reflect how good the whole system will sound. It can only give you reasonable expectations of possibilities. I got lucky. Maybe you will, too.
It makes every bit of sense. Because if you send the digital signal to the AVR from the CD players digital outputs. You are using the AVR's DAC to decode an uncompressed digital stream. And if you send a compressed digital signal to the AVR it will decode the compressed signal. And the uncompressed signal should sound better if all is running though the same DAC's. If it does not, there is something else in the signal chain that is dumbing down the sound quality. And that pretty much leaves the analog sections of the AVR. The modern DAC's in even these budget components are quite good. But the analog sections are going to be very compromised in relation to what is required for higher end audio reproduction.
Agree 320kbps is much better sounding than 128 but it is still missing some subtle nuances that lossless compression has. But you won't hear it if the signal and transducer chain can't reproduce it.
Last edited:
This very well might be the biggest problem in these lower priced AVR's. Good linear power supplies are expensive to build and heavy. For many amps the power supplies define the sound quality as much as any other part of the circuit. I would not be surprised that this is the primary issue degrading the sound.
That's simply not true. In fact, it is very hard to find a single modern AVR that uses class D amps. I don't know of one. Do you?
The logic above doesn't work. Yes, amplifiers are tested into resistive loads, and speakers are reactive. But look at speaker impedance plots. When you see impedance dropping with rising frequency, that's capacitive reactance. When you see impedance rising with rising frequency, that's inductive reactance. The speaker load is a complex impedance of all three parameters, inductive, capacitive, and resistive. Again, look at speaker impedance plots, and notice the degree of change from 8 ohms nominal, and over what frequency bandwidth that change occurs. Now you'll get a better picture of what the real load looks like to the amplifier.
Next, look at the source impedance of the typical amplifier. Nearly every SS amp is down well below a fraction of an ohm. This, combined with the power rating vs load (resistive) tells you how an amp can drive power into a reactive load. And generally, all SS amps can quite easily drive the typical reactive speaker load. The reason this MUST be in the design is that speakers are voiced with low source Z amps.
Now realize that if a speaker is voiced (tuned) with anything other than an amp a low source Z, it will not perform according to plan with the typical amp, and that's not desirable at all.
So your logic doesn't work. Amps can drive reactive loads, they are designed to do so because that's what's expected. No, there will not be audible difference because of an amps inability to drive a reactive load, unless the amp is very poorly designed indeed.
The FCC doesn't specific amp measurments other than to insure that all electonic devices meet Part 15. The FTC, Federal Trade Commission did publish amp test methods. You won't see audible differences reflected that way because they the tests are very general and don't reflect even the differences in amplifier technology. That does not mean there are no measurements that can reflect amplifier audible differences!
Thanks for your opinion.
If you "grade" the possible signal modifiers in that chain, you very quickly see that the things that modify the signal in many vectors are the speaker and room. Both impress changes in magnatude response and changes in the time domain. The speakers is dominantly a mangatude response modifier, and lesser in the time domain (but FAR from zero), as well as a strong modifier of linearity (distortion). The room is a massive time-domain modifier, as well as a composite response modifier. Nothing in the electrical chain comes close to those two things at all. Amplifiers, all of them, do present excellent magnatude and time response, even into a reactive load. DACs have no capability to modify magnatude response, and these days, very little in the time domain as well. DAC linearity has vanished to the barely measurable, and definitely inaudible.
You're touching on it here, but I'm not sure what you mean by "typical", because once we're characterizing devices using FFT analysis, frequency and time are the same thing, so you get to window the data differently and see frequency (magnatude), group delay, or another version, phase vs frequeny, and then look out past the stimulus and see what happens in time a bit after it's over. Pretty easy to do, and guess what? They're all pretty darn good. Speakers? Nope, horrible offenders. The only thing worse is the room. Yes, you can see all of this. The fact that people don't look and publish is another matter, but publishing group delay figures on an amp will only confuse the consumer, so I see why they don't.
Yes, you can "luck out", get referrals, etc. But the best rig also can be the result of design. Again, the fact that many here choose to ignore the technololgy and opt for the recommendation (or the mythology) is a different matter.
Well, you're talking out of class here. It would behove you to study the actual technology rather than spout unfounded opinion. Linear power supplies are completely unnecessary today. Swichting PSUs are now refined, and every bit as clean and capable. No, the PSU does not define the sound quality more than any other part of the circuit. In fact, when it's done right (and today it's hard to do it wrong) they simply step out of the way and let the rest of the circuit do its job well.
Yes, and the thing in the chain that dumbs down the compressed signal is the codec that created it! I'm sorry if you can't understand this, but it's quite simple. The codec applies a process that eliminates data that is determined to be inaudible, or less audible, than the rest of the signal. This is done by applying the principles of psychoacoustic masking, where a louder sound of a particular frequency "masks" a quieter sound nearby in frequency. The masking curve is a modified bell with the peak at the loud signal, and a sloping curve extending above and below it in frequency. Below the curve, other sounds are inaudble. But, it's acompromise. The shape of the curve is selectable, and the bigger it is the less data is preserved. The curve tracks dominant frequencies, and that, and the selection of shape compromise, results in higher bit-rate reduction, and more audible artifact. There is nothing, and I do mean NOTHING that a DAC can do to help that out! The data has already been irretrievably removed, and the codec artifact built in. The damage is done. Conversely, a non-compressed, full bit-rate stream presents data to the DAC, which reconstruts the original by creating a series of "dots" in time, which the reconstruction filter then connects to form a smooth, stepless output waveform. Nothing to do there either. DACS are not signal modifiers.
Really? I don't think so. Every actually done any testing? DBT/ABX or measurements? They don't support your conclusion.
320kbps is extremely difficult to detect. There are no audible codec artifacts. The one thing that gives it away is the 15kHz LPF baked into mp3, and no DAC, no amp, no speaker and no room can replace that missing top half octave. Fortunately, not many can hear it anyway.
Well said Jaddie, spot on about FCC vs FTC. I used the wrong TLA. And your comments on the room and transducers are also spot on. Definitely the biggest effects. Always. You are preaching to the choir on most of your points. Especially that good equipment is created by design. Alss very true. But that is a very different thing than being able to choose good equipment as a consumer. It ls hard to do that just from considering the specificatoons. In any case, I suspect we are much closer in our assessments than you have chose to fight about. And thanks for your opinions as well.
I was looking for an inexpensive AV receiver the other day. Thought I might make a small headphone system in the bedroom. I was looking through the Best Buy info and was left with the impression that several of the new AV receivers had digital amplifiers. This could have been marketing hype or they may have been confusing the amp with the switching power supply that these lower priced models are supposed to have. In the past I have not known most of them to have class D amps in them.
I don't want you to get the impression that all AV receivers are really bad amplifiers. But a receiver could have a pretty good sounding amp in it and yet the entire sound of that amp can be ruined by driving it with a switching power supply. Which a lot of the lower priced units have. I have a 20 year old Onkyo AV receiver that was in the 900 dollar range back then and in pure direct mode it sounds better than many other bad sounding straight stereo amplifiers that I have heard. Yet, it can not really do a 3D sound stage compared to the Rotel preamp and monoblocks.
Actually, when I was building up my sound system years ago, I had a hard time finding a stereo amplifier in my price range that I was happy with. I had a maybe 1000 dollar budget and I could not find anything new for that price that sounded good to me. Finally I found these Rotel AV monoblocks which are single channel high current amps. All of the amps that I was happy with were over 5000 dollars at that time. Of course, I could not audition every brand that was out there. There was one amp from Classe that I liked for about 3500 bucks. I always preferred the dual mono designs which were much more expensive.
in audio, there are many aspects of design and execution that will determined how the amplifier will sound including the PSU,
i agree with you for most, however even same tech amps with different brand and/or different build will have audible differences.
Yeah, but that's not what you said in your earlier post. What you said then was:
... "Add to that, that many of the new AVR's use bad sounding class D amps in them. There is no way to win with a modern AVR for pure music listening."
Now you are changing your story since you have been called out for being completely wrong. Which brings into question how many other things that you confidently state as facts are not really true at all.