@cello10_#1 - Thanks for all the detail. Lotsa stuff I did not know, especially about bipolar power amps. Interesting that you would mention piezo resonance of solid state devices because I've always heard that the vacuum tubes had mechanical resonances in the audio frequency range. Microphonics it's called and can be perceived to be good or bad.
And your mention of the decoupling caps on the biasing networks just reminded me of a discussion I attended at the LA Amp Show last October. Several tube guitar amp builders talked about how while many EE's think that the more decoupling the better, subject to cost and size constraints, that's not the case for guitar amps. By extension to hi-fi audio amps, too. Too much capacitance makes an amp sound "dead". Too little and it rings or even squeals.
Not that I've had much experience tinkering with audio, but I do know what sounds good. To me, anyway.
Base charge injection and removal also seems to be responsible for the audible problems with Bipolar transistors. This is particularly true with Class A/B power amplifier stages.
Adding larger than normal emitter ballast resistors - say in the 1 ohm range - helps with base charge removal as well as with softening up the abrupt turn-on turn-off characteristics of these devices. Otherwise, Bipolar Class A/B output stages create an edgy, razor sharp sound not encountered with vacuum tube amplifiers. High negative feedback only makes the problem worse, by increasing the number of higher order harmonics produced, even while reducing overall THD.
Adding large value capacitors across totem pole biasing networks also helps with proper charge injection and removal, and creates a subjective difference to the dynamic sound presentation.
Class 'A' solid state amplifiers always seem to sound better than class A/B solid state.
John Curl once said to me that a vacuum tube is an inherently "purer" device than a transistor because of its physics of operation.
Digital recording added another layer of sonic issues on top of the issues created by solid state amplifier design. MP3 compression then created another layer of issues on top of that.
Thank you for the correction on Alan Blumlein,
the original inventor of stereophonic recording.
As for differences between transistors and tubes,
that topic has been argued for the past 50 years.
One issue often overlooked is that solid state devices create a piezo, electro-mechanical resonance coloration to the original electrical signal.
Electrical signal changes create piezo mechanical changes in the crystal structure which then create a piezo-electrical signal coloration added back on top of the the original electrical signal.
This is probably why audio equipment made with the early Germanium transistors always sounded so horribly tinny and metallic compared to their more modern silicon counterparts. Differences in base charge injection and removal and storage time may also be a factor in the differences in sound between Germanium and Silicon.
If you run a linear A/B solid state amplifier at full output with say a 1KHz sinewave, but with no speaker connected, you can actually hear the power output devices mechanically "sing". As the output devices warm up the silicon/germanium crystal gets "softer" and less prone to mechanically resonating.
Another reason for going to CES was to pick up SACD's for $15 instead of the usual $25 in the stores when I could find them. I don't know what they're going for these days. There are definitely audible differences in the discs themselves as well as the players. One year at CES there was a demonstration of various mfrs SACD players, but they all had different artists' music in them so it wasn't really a fair comparison. They had headphones plugged into them, and I can remember that the Marantz player sounded particularly good, while the Phase Linear one was awful! The Sharp player sounded good, but I wished I could have turned up the volume a bit.
Content is indeed key, and I agree with joe.schoelen and Bill.Graham that Neil Young's weak, wavering voice is better left unheard, to put it politely. As well as his weak, wavering lead guitar, although his song writing and acoustic guitar playing are fine. Let somebody else do his songs, fer God's sakes! On a side note (no pun intended), as some of the folks here have mentioned distortion, some years ago I was mentioning that Young's song "Hey, Hey, My, My" (or is it "My, My, Hey, Hey"?) sounded like he had blown a speaker. One of my fellow guitarists said that he had heard that Neil Young couldn't seem to get enough distortion for the song, so he sliced up his speakers! Yes, there is such a thing as too much distortion, although I'm a big fan of distorted guitars.
And @cello10_#1, that's Alan Blumlein, not Richard. Blumlein was the chap that came up with (patented, anyway) stereo recording and reproduction, among many other useful things such as the long-tailed pair (differential pair).
With that, I'll shut up now.
Ah, a subject near and dear to my heart! @another_bill - I believe there was a discussion online recently in which some tube amp builders theorized that the sound of tube vs. solid state had less to do with the active device and more with the passives that surround it. For example, since a vacuum tube is a high impedance output device, it requires a transformer to be used when driving a low impedance load like a speaker, and the magnetics of the xfmr greatly affect the sound. High voltages used in tubes may also required different types of capacitors, which further affect the sound. These guys actually built a transistor amplifier with all the accoutrements of a tube amp, and lo and behold, it sounded just like a tube amp!
Not that even the passives are the main thing that affect the subjective sound quality; I went to the Consumer Electronics Show from 2001 to 2005 (before my 2nd kid was born when I could do such things) mostly to attend the High End Audio Pavilion as well as the nearby The Home Entertainment (THE) Show. I found that from year to year, my perception of the quality of a particular manufacturer's system varied greatly. One year I was blown away by mfr. X's stuff, the next it just sounded so-so or vice-versa. Mfr Y's stuff sounded OK one year and fantastic the next! I concluded that rather than the mfrs changing their equipment radically from year to year (unlikely), that it's really hard to "get it right" when you're actually listening to the entire chain from the musical instrument and/or voice through the recording chain, to the reproduction chain, as well as the spaces where it was recorded and played back.
The analysis above merits further comment:
The fatiguing quality of some CD's often has to do with things like low level / high frequency harmonic distortion being added to the signal. This generally sounds harsh to the ear, even though the THD numbers show little difference compared with "non-fatiguing" sources. Slew rate distortion and class AB output stages in Op-amps are often contributors, as are other factors.
Vacuum tubes are far less likely to produce a fatiguing sound quality than solid state, Bipolar devices. Many solid state designers prefer to use JFETS amd MOSFETs instead of Bipolars in the signal chain for this reason.
Vacuum tubes are also interfaced with D/A converters or used in mixing consoles for the same reason.
The effect of phase shift impacting subjective sound quality has also been ignored by too many "scientific experts".
Brick wall cut-off filters at 22 KHz produce serious sonic aberrations in high quality recording. For one thing, the quality of the stereo image suffers.
Of course, if you do not believe stereo produces a subjective sound image in a room, then by scientific analysis, stereo and mono should sound the same and Richard Blumlein's original patent from the 1930's was nothing but a scam.
Phase shift in loudspeaker crossovers has also been an issue since the late Richard Heyser wrote his landmark paper on time domain spectrometry analysis of loudspeakers. Many knowledgeable audiophiles now use only 1st order, 6/dB per octave crossovers with impedance compensating Zobels so that phase shift and time smear in loudspeakers does not subjectively degrade audio quality.
Although phase shift and time smear can be objectively measured with test equipment, many in the "scientific" audio community still refuses to believe that it impacts subjective sound quality.
Here is another comment on MP3s and recording quality that was sent in via email:
"One of the most important aspects is that modern mastering techniques squash the dynamic range so much that we are not even beginning to use the range available on a CD.
There are a bunch of links at the bottom of the page I am linking to below - if you care about audio you will follow them:
I agree completely with "fatigue". This did not happen in the old days, and MP3 lossy technology is NOT the problem.
Basically, even when I buy a CD of a "digitally remastered" piece of music, I am getting garbage that I don't want to listen to.
Secondly, as you point out in your point 1, MP3 lossy technology can do a great job of 192kHz sampled audio without creating ridiculous sizes specifically by being lossy. That could be vastly superior to CDs, especially now that our tolerance for file sizes has gone up considerably. Standard CDs could NOT capture the high sample rates, but MP3s can. MP3s can be lossy but can be very high quality. What about 320kbps (I used to do that) and VBR technology for reasonable file sizes and excellent audio? MP3 is the punching bag here but if someone wanted to master extremely high quality music MP3 is also the solution!
A lot of CDs, especially older ones (or it may have been the players) suffer from problems with frequency range. The sample rate is too close to Nyquist, so the anti-aliasing filters AND the reconstruction filters contribute to problems. A higher sample rate encoded to lossy MP3 at a good bit-rate would solve that problem."
Having worked on top-quality studio equipment, analog and digital, I learned what sounded good and bad, and why. I listened to live music, badly amplified, and also to good concerts.
Good speakers and acoustics are almost the whole story, but not all good news, because truly transparent monitors will reveal the aliasing and huff-puff in bad MP3s, while every scratch in your vinyl gets its full dynamic range - Oww!
So, for real quality go listen to acoustic live sets; learn to play yourself; or make your own recordings. Otherwise, its a jungle, but if they establish a premium standard the unforeseen consequences may be that 'regular' releases will be dumbed-down and if anyone complains, they can go premium, hey?
Health warning: I respect my ears most of the time, but I occasionally indulge in playing rock at frightening levels to get the adrenaline which was part of the experience.
But I will pay the price...
I would like to think that sound quality will improve. Memory continues to decline in price and this will allow higher fidelity to be stored. Of course, as the price of memory drops, the desire to have ever larger personal music libraries grows. Do the two forces null each other out? Are we doomed to have huge libraries of mediocre quality, or will the size of the library and the quality of the recording improve in parallel?
Or, will we get 64 bit sound, with 63 bits devoted to anti-piracy? ;)
Replay available now: A handful of emerging network technologies are competing to be the preferred wide-area connection for the Internet of Things. All claim lower costs and power use than cellular but none have wide deployment yet. Listen in as proponents of leading contenders make their case to be the metro or national IoT network of the future. Rick Merritt, EE Times Silicon Valley Bureau Chief, moderators this discussion. Join in and ask his guests questions.