I have to laugh Paul. When I first saw the title of this video the smart Alek in me stood right up. Lol. But I watched the video knowing exactly what you were referring to. It seems as though a few watchers made comments before actually viewing this informative video first. Thank you for producing these informative and entertaining videos Paul and keep up the good work at PS Audio and I hope one day to be able to have a high quality audio system from your excellent company. R.W.
A corollary question is whether volume level affects sound quality, and of course Fletcher-Munson confirm this. But I also have found that for any given recording there is a playback level that sounds most open and dynamic and natural, and my guess is that this is very close to the level at which it was mastered. When I find myself dissatisfied with a recording, my first reaction is to adjust the volume, and it is often improved dramatically.
Yes, in my old diy passive preamp, I tried three different types, top line alps, Noble and stepped resistor based. I could easily hear the difference between the three and stepped resistor, based on sounded best. I am not familiar with this chip-based approach, so I have no opinion, well I think Paul will be using them because they sounded best.
The attenuator (I'll call it that rather than a volume control) is possibly the most overlooked part of an audio setup. You get really good quality capacitors, resistors etc., then pass the signal through a basic carbon track and some sort of metallic wiper. Everything goes via the attenuator, so it makes total sense to use one that has as little effect a possible on the signal. Trouble is, this can get really expensive. It's a tough call for manufacturers, but having heard the difference it can make, it's worth the cost and effort, Practically it's very hard to implement in a existing system.... It needs to be a design criteria / decision... back to the drawing board folks!🤔
Thanks Paul for answering my question,it's really clarify the matters. Although I still very itchy to mods my current VTL preamp with Khozmo ladder stepped attenuator from std alps motorized blue velvet pot and see how far will it improves the overall sound. But I agree on yiur point that any attenuation should be on neutral where it will not colours the sound too much so the circuit design goal can be achieved
There is a common misconception among audiophiles about pots. The quality of the pot has nothing to do with sound quality. Yes a cheap pot can become intermittent and noisy but that's not the audio quality problem. The problem is the circuit design. Any amplification stage has a finite input capacitance. A pot is basically a resistor in series with the next stage. An RC network is a low pass filter. Depending on the resistance of the volume setting and the next stage capacitance, the higher frequencies get further attenuated than the lower frequencies as the gain is reduced. That is why some pots "sound bad". Replacing the pot with an expensive $50 instrumentation grade part is not going to fix that audio quality problem. The problem is the circuit design. Simple stepped attenuators do not avoid this issue either as they are nothing more than a pot made from switched resistors in series, just like a pot. Better switched attenuator designs maintain a consistent impedance and therefore have minimal effect on audio quality. But as Paul said, the best attenuators are the high quality dedicated switched resistor ladder chips from Burr Brown and Cirrus Logic. These are not found in mass produced consumer HiFi gear. How about digital volume control? That too has a different set of problems, IMO, worse than any analog pot but good enough for an Ipod or sound bar.
That is a brilliant broad-brush misinformation statement, which is wrong. Yep stray/inherent capacitance and the pot resistance will create a low pass filter. But you (deliberately?) provide _no_ numbers. So lets start with a good gulp of resistance, say 22K, and a really bad load with 22pf, no lets go nuts, 47of load capacitance. Where does this worst case low pass put us, around 150kHz, and that’s with a deliberately bad design that you won’t see these days. Pots don’t match (Alps blue volume is ~2dB typical, 3dB max mis-match), and thermally all over the place. Better is use Linear pots with a Log-Fake fixed resistor, less prone to crackle and the lin pot is inherently better matched. Better still is a well designed stepped attenuator (0.1%).
@@Roosville1 I have seen the effects, albeit on vintage tube preamps where the pot values are quite high 500k to 1 megohm. I would assume most SS gear has much lower pot resistance but who knows on some of these exotic designs. Yes it is just some audiophile theory for an example. I do not subscribe to it myself.
Yes, because audio is a complex waveform entering an audio path after the volume control that is not 100% lineair. So the output signal signature (added harmonics due to nonliniarity) will differ at different audio levels.
I think too that it is. I observe that the more volume is up the more uopper band I notice. - for testing purpose I correct loudness repectively with other adjustment. . My guess is that with using potentiometer we change parameters of circuits - one is input impedance of controlled stage, other is apparently we may change load of previous stage - top up makes lowest impedance because input stage is hunted by wire capacities to ground , third I think potentiometers above 100K are affected by capacity to "ground" - they suppose to have metal body for screeening but it provides some capacity to audio track. I believe it was known to designers of vintage tube gear (or they practiced it by experimenting) In tube constructions they were even able to upbeat sopranos in low pot position. For that reason in many solutions they made loudness correction only with low range circuit With digital controll it is surely possibe to avioid variable resistors but if resistors allow unusual gains I prefer old "resistance games"
Would speakers sound any different with volume? Besides volume. 😅 A siigle driver speaker would have a hard time with the upper treble level when turned down really low. Mine seem to.
The volume level naturally affects the sound due to the Fletcher Munson hearing effect of human hearing being significantly optimized towards more narrow band listening at lower volume levels. It’s great for helping us hearing voices at low levels but horrible for enjoying music at low levels without use of loudness compensation (EQ).
Why not design a low, medium and high volume level optimized for true fidelity and eliminate everything in-between? Of course, that's simplistic, but not everything needs to go to "eleven."
Paul McGowan, PS Audio sir i have a question...is a aftermarket equalizer produce better sound than the bose 901 speakers EQ? my buddy is a hifi nerd he says a aftermarket eq is better,i have a 30 band eq
When using a PC as a source, does volume level (in Windows, for example) influence sound reproduction? Even if you use USB to connect to external DAC, volume lvl in Win still affects volume lvl in the system. So I'm not asking about crappy audio out ports as output :)
Windows especially influences the whole sound reproduction. Try to circumvent this completely, for example using Foobar with a proper ASIO driver for your dac.
Couldn't agree more. Resistors in the signal path are just varying decrease of bad. Arthur Salvatore highlighted a linestage based on autoformer volume controls in his list of exceptional equipment. I secured a pair from Dave Slagle at intact audio and wow. Purity.
Had to repair a 1983 NAD 3155 volume/balance control (my first decent amp). The part was pure unobtainium, so I modded an audio taper stepped attenuator by adding the loudness tap, and created a MN taper stepped attenuator for the balance. Recapped the unit and it sounds not bad at all. I think it's better than the carbon element pots.
I recently read where women's favorite frequency for a male voice is 96 hertz. It didn't say how loud is their favorite volume. But I'm guessing it wouldn't be 96 decibels. That would be like Barry White throwing a tantrum; and they would leave.
Maybe the study meant his most common frequency when talking. You wouldn't be able to express much emotion without raising & lowering pitch. The thing that surprised me the most is that a male voice once registered at 0.18 hertz (by a testing machine). That's less than 1 hertz. That sounds (sounds?) Impossible to me. That is the Guiness Book Of World Records record.
Just get a dac which controls volume in the digital domain, that way you can skip the preamp and its potential coloration. Also you can put more money on a better dac since you are not buying a expensive preamp.
if you thinking on doin it that way just buy a minidsp SHD, or the studio version if you want a separate dac. then you have full dsp that makes all the difference.
Digital attenuation has it's own set of problems. If poorly implemented they can reduce dynamic range. The other more difficult problem is bit rounding. Anytime gain is changed, it is a multiplication process. That generates extra bits beyond the original audio bit depth, like 24 bits becoming 28 bits. This must be rounded back down to 24 bits and to do that well is a complicated process. Just chopping off the extra lower bits, which some cheap designs do, has significant negative impact on the audio quality, Digital gain control is mandatory in applications like mixing consoles. But in a home HiFi, you have to convert to analog at some point. There is no such thing as a digital speaker*. So since you need to convert to analog anyway, why not put the gain control there where you have less issues. A third digital pot problem in mixing consoles is "zipper noise". Not a game changing issue in home HiFi or general room volume control. But in a mixer where gain is sometimes changed on the fly, this is a no go. Getting rid of zipper noise is another complex problem to solve. * A powered speaker with a digital input is not a digital speaker. It's a basic speaker and analog amplifier with a built in DAC. They are mostly for commercial applications where analog noise pickup with long cables is an issue. Then there are IP network based systems like Dante which give you internet based central management of an entire PA system.
@@flargosa No, I'm saying if a digital volume control is not carefully designed it can be worse than an analog pot. I assume the PS Direct Stream is a well designed digital volume control?
Yes, when you crank the volume up it is louder, and when you crank it down it is not so loud. You bunch of upset audiophiles. It would be difficult living with some of you.
Before even watching I know this answer... Yes, they make it louder or quieter.
You want a job answering customer service queries?
@@andrewprettyquick2070
😁👍
Gave me a good laugh!
@@scottyo64 😁
The louder the volume the higher the risk of harmonic distortion.
I have to laugh Paul. When I first saw the title of this video the smart Alek in me stood right up. Lol. But I watched the video knowing exactly what you were referring to. It seems as though a few watchers made comments before actually viewing this informative video first. Thank you for producing these informative and entertaining videos Paul and keep up the good work at PS Audio and I hope one day to be able to have a high quality audio system from your excellent company. R.W.
One of the things i respect about allnic audio is they make their own attenuators which are great!
A corollary question is whether volume level affects sound quality, and of course Fletcher-Munson confirm this. But I also have found that for any given recording there is a playback level that sounds most open and dynamic and natural, and my guess is that this is very close to the level at which it was mastered. When I find myself dissatisfied with a recording, my first reaction is to adjust the volume, and it is often improved dramatically.
Yes, in my old diy passive preamp, I tried three different types, top line alps, Noble and stepped resistor based. I could easily hear the difference between the three and stepped resistor, based on sounded best. I am not familiar with this chip-based approach, so I have no opinion, well I think Paul will be using them because they sounded best.
The attenuator (I'll call it that rather than a volume control) is possibly the most overlooked part of an audio setup. You get really good quality capacitors, resistors etc., then pass the signal through a basic carbon track and some sort of metallic wiper. Everything goes via the attenuator, so it makes total sense to use one that has as little effect a possible on the signal. Trouble is, this can get really expensive. It's a tough call for manufacturers, but having heard the difference it can make, it's worth the cost and effort, Practically it's very hard to implement in a existing system.... It needs to be a design criteria / decision... back to the drawing board folks!🤔
My current preamp has a stepped attenuator and in-spite (or because of ) the clickity click noise when turning the volume up, it works very well.
Thanks Paul for answering my question,it's really clarify the matters. Although I still very itchy to mods my current VTL preamp with Khozmo ladder stepped attenuator from std alps motorized blue velvet pot and see how far will it improves the overall sound. But I agree on yiur point that any attenuation should be on neutral where it will not colours the sound too much so the circuit design goal can be achieved
REALLY good sounding is the Khozmo. But I was not happy with the linearity of the attenuation.
There is a common misconception among audiophiles about pots. The quality of the pot has nothing to do with sound quality. Yes a cheap pot can become intermittent and noisy but that's not the audio quality problem. The problem is the circuit design. Any amplification stage has a finite input capacitance. A pot is basically a resistor in series with the next stage. An RC network is a low pass filter. Depending on the resistance of the volume setting and the next stage capacitance, the higher frequencies get further attenuated than the lower frequencies as the gain is reduced. That is why some pots "sound bad". Replacing the pot with an expensive $50 instrumentation grade part is not going to fix that audio quality problem. The problem is the circuit design. Simple stepped attenuators do not avoid this issue either as they are nothing more than a pot made from switched resistors in series, just like a pot. Better switched attenuator designs maintain a consistent impedance and therefore have minimal effect on audio quality. But as Paul said, the best attenuators are the high quality dedicated switched resistor ladder chips from Burr Brown and Cirrus Logic. These are not found in mass produced consumer HiFi gear. How about digital volume control? That too has a different set of problems, IMO, worse than any analog pot but good enough for an Ipod or sound bar.
That is a brilliant broad-brush misinformation statement, which is wrong. Yep stray/inherent capacitance and the pot resistance will create a low pass filter. But you (deliberately?) provide _no_ numbers. So lets start with a good gulp of resistance, say 22K, and a really bad load with 22pf, no lets go nuts, 47of load capacitance. Where does this worst case low pass put us, around 150kHz, and that’s with a deliberately bad design that you won’t see these days. Pots don’t match (Alps blue volume is ~2dB typical, 3dB max mis-match), and thermally all over the place. Better is use Linear pots with a Log-Fake fixed resistor, less prone to crackle and the lin pot is inherently better matched. Better still is a well designed stepped attenuator (0.1%).
@@Roosville1 I have seen the effects, albeit on vintage tube preamps where the pot values are quite high 500k to 1 megohm. I would assume most SS gear has much lower pot resistance but who knows on some of these exotic designs. Yes it is just some audiophile theory for an example. I do not subscribe to it myself.
Yes, because audio is a complex waveform entering an audio path after the volume control that is not 100% lineair. So the output signal signature (added harmonics due to nonliniarity) will differ at different audio levels.
I think too that it is. I observe that the more volume is up the more uopper band I notice. - for testing purpose I correct loudness repectively with other adjustment. . My guess is that with using potentiometer we change parameters of circuits - one is input impedance of controlled stage, other is apparently we may change load of previous stage - top up makes lowest impedance because input stage is hunted by wire capacities to ground , third I think potentiometers above 100K are affected by capacity to "ground" - they suppose to have metal body for screeening but it provides some capacity to audio track. I believe it was known to designers of vintage tube gear (or they practiced it by experimenting) In tube constructions they were even able to upbeat sopranos in low pot position. For that reason in many solutions they made loudness correction only with low range circuit With digital controll it is surely possibe to avioid variable resistors but if resistors allow unusual gains I prefer old "resistance games"
Would speakers sound any different with volume? Besides volume. 😅 A siigle driver speaker would have a hard time with the upper treble level when turned down really low. Mine seem to.
The volume level naturally affects the sound due to the Fletcher Munson hearing effect of human hearing being significantly optimized towards more narrow band listening at lower volume levels. It’s great for helping us hearing voices at low levels but horrible for enjoying music at low levels without use of loudness compensation (EQ).
Fletchmu? Can that be a slang term?
An often overlooked topic.
Why not design a low, medium and high volume level optimized for true fidelity and eliminate everything in-between? Of course, that's simplistic, but not everything needs to go to "eleven."
Penny and Giles conductive plastic potentiometer forever - perfect 35 years ago, still perfect today in my Daner Detal preamp. But not cheap.😅
What causes display volume not to go up or down smoothly? Like it jumps around, fights with itself, does the opposite or doesn't work at all?
I'm pretty sure they're using high quality op-amps for the signal and digitally adjusting the gain outside the signal path.
Paul McGowan, PS Audio sir i have a question...is a aftermarket equalizer produce better sound than the bose 901 speakers EQ? my buddy is a hifi nerd he says a aftermarket eq is better,i have a 30 band eq
Best quality volume controls are in the newer higher end Technics & Yamaha amps.
Why do you have Elac Debut 2.0 speakers in your office?
When using a PC as a source, does volume level (in Windows, for example) influence sound reproduction? Even if you use USB to connect to external DAC, volume lvl in Win still affects volume lvl in the system. So I'm not asking about crappy audio out ports as output :)
Windows especially influences the whole sound reproduction. Try to circumvent this completely, for example using Foobar with a proper ASIO driver for your dac.
What is an example of one of these digital pots? Any specific component recommendations?
Look up a Burr Brown (now TI) PGA2310. That is the top of the line. They also sell different lower cost versions with lessor specs.
Just being three meters more or less away from the playback source(s) results in drastic changes in the sound!
Couldn't agree more. Resistors in the signal path are just varying decrease of bad.
Arthur Salvatore highlighted a linestage based on autoformer volume controls in his list of exceptional equipment. I secured a pair from Dave Slagle at intact audio and wow. Purity.
Digitally anolog control?
Yes, I need the idiot's explanation.
"Yes, but only in everyone else's products, and even our old ones, so buy our new products."
Are those a set of KEF speakers I see?
They look like Kef's Meta 50, too big for LSX (or Paul is really tiny!)
In other words it is simply cheaper to use a $2 or so chip
It does if you turn it up!😃
Had to repair a 1983 NAD 3155 volume/balance control (my first decent amp). The part was pure unobtainium, so I modded an audio taper stepped attenuator by adding the loudness tap, and created a MN taper stepped attenuator for the balance. Recapped the unit and it sounds not bad at all. I think it's better than the carbon element pots.
I recently read where women's favorite frequency for a male voice is 96 hertz. It didn't say how loud is their favorite volume. But I'm guessing it wouldn't be 96 decibels. That would be like Barry White throwing a tantrum; and they would leave.
A voice doesn't have just one frequency.
Maybe the study meant his most common frequency when talking. You wouldn't be able to express much emotion without raising & lowering pitch. The thing that surprised me the most is that a male voice once registered at 0.18 hertz (by a testing machine). That's less than 1 hertz. That sounds (sounds?) Impossible to me. That is the Guiness Book Of World Records record.
Ah the old Carbon trackers….always noisy. Resistor ladders better. Cheers.
Thanks for many good answers!
Nice KEF speakers on your desk!
Just get a dac which controls volume in the digital domain, that way you can skip the preamp and its potential coloration. Also you can put more money on a better dac since you are not buying a expensive preamp.
if you thinking on doin it that way just buy a minidsp SHD, or the studio version if you want a separate dac. then you have full dsp that makes all the difference.
Digital attenuation has it's own set of problems. If poorly implemented they can reduce dynamic range. The other more difficult problem is bit rounding. Anytime gain is changed, it is a multiplication process. That generates extra bits beyond the original audio bit depth, like 24 bits becoming 28 bits. This must be rounded back down to 24 bits and to do that well is a complicated process. Just chopping off the extra lower bits, which some cheap designs do, has significant negative impact on the audio quality, Digital gain control is mandatory in applications like mixing consoles. But in a home HiFi, you have to convert to analog at some point. There is no such thing as a digital speaker*. So since you need to convert to analog anyway, why not put the gain control there where you have less issues.
A third digital pot problem in mixing consoles is "zipper noise". Not a game changing issue in home HiFi or general room volume control. But in a mixer where gain is sometimes changed on the fly, this is a no go. Getting rid of zipper noise is another complex problem to solve.
* A powered speaker with a digital input is not a digital speaker. It's a basic speaker and analog amplifier with a built in DAC. They are mostly for commercial applications where analog noise pickup with long cables is an issue. Then there are IP network based systems like Dante which give you internet based central management of an entire PA system.
@@andydelle4509 so you’re saying the PSaudio direct stream dacs’ volume control is less transparent than an analog pot?
@@flargosa No, I'm saying if a digital volume control is not carefully designed it can be worse than an analog pot. I assume the PS Direct Stream is a well designed digital volume control?
Yes, when you crank the volume up it is louder, and when you crank it down it is not so loud. You bunch of upset audiophiles. It would be difficult living with some of you.