The best digital volume controls possible/in existence

[zenpmd]

No I am going totally digital I am thinking of using a standalone DAC and some sort of standalone volume control, like the Weiss IN204.

 

Can anyone share the absolute best digital volume control?

 

Thanks

Paul

[gmgraves]

No I am going totally digital I am thinking of using a standalone DAC and some sort of standalone volume control, like the Weiss IN204.

 

Can anyone share the absolute best digital volume control?

 

Thanks

Paul

 

 

Not really Paul, but I can give you this piece of friendly advice: make sure that whatever you get is controlled by a rotary encoder rather than up/down buttons. Up/down buttons change value at a fixed rate, if you find that you need to lower volume quickly (because different sources, and recordings have different levels), the up/down buttons will lower that volume agonizingly slowly, while a rotary encoder will do so with a quick twist of the knob, virtually instantaneously.

[pooger]

I may not understand your question, but to my way of thinking, the best digital volume control is one that (assuming your DAC is based on a chip that has built in volume control, which most are) works purely in software, sending a code to the chip to do this. For example, using the volume control in A+.

[Panelhead]

No I am going totally digital I am thinking of using a standalone DAC and some sort of standalone volume control, like the Weiss IN204.

 

Can anyone share the absolute best digital volume control?

 

Thanks

Paul

.

 

The digital volume controller built in the 9018 Sabre dacs is great. The LH Infinity dac maybe has the best I have tried, analog or digital.

With the high bit processing it does not lose resolution at high attenuation. Plus small steps.

I used to think the PGA2311 and 4311 were good. The control built into the Sabre chips is much cleaner.

[Jay-dub]

The post is a little confusing, as the Weiss INT204 is a USB->AES/EBU converter without built-in volume control.

 

The best UI for volume control, as gmgraves wrote, has to be a rotary encoder that you can turn with your fingers from the listening position. For a typical speaker setup that would mean a remote with a knob on it.

 

The technically best solution for digital volume control has to be one that performs the adjustment at more than 24 bits immediately before delta-sigma modulation, thus avoiding quantization to 24 bits or less at an attenuated level. If done in-chip, that would require a separate attenuation signal (pooger's suggestion). Alternatively, doing the volume control and modulation together in software, as with HQPlayer converting to DSD128/256, would be another ideal solution.

[barrows]

.

 

The digital volume controller built in the 9018 Sabre dacs is great. The LH Infinity dac maybe has the best I have tried, analog or digital.

With the high bit processing it does not lose resolution at high attenuation. Plus small steps.

I used to think the PGA2311 and 4311 were good. The control built into the Sabre chips is much cleaner.

 

Yup, anything using ESS 9018 volume control is as good as it gets. I have made the comparisons, and the ESS VC allows for plenty of attenuation without any loss. As I recall it runs at 48 bits, so it even has plenty of resolution to work with 32 bit files. I also suspect that Mola Mola's forthcoming DAC with Bruno Putzey's integrated digital VC will be entirely transparent as well.

The days of criticizing digital volume controls are long over, properly implemented, digital volume controls offer the best way to attenuate the signal.

Only someone with an insanely mismatched gain structure in their system need be concerned (like someone needing -100 dB attenuation) and these folks should really make the appropriate changes to their system, as the are going to be leaving a lot of the music behind regardless of how they attenuate the signal.

[mourip]

It is not really a "digital" volume control but Tortuga Audio makes a great LDR (light dependent resistor) based unit with remote volume, balance, 3 input control, auto calibration, and variable/adjustable impedance. It also has left and right channel level displays.

 

They have three versions of it. One with really nice case, one with simple case, and a kit. The innards are all the same with the kit less than half the price of the one with the nice case.

 

I am using it with my Shiit Yggy and it is really great.

[The Computer Audiophile]

Here is an interesting slide from an ESS presentation. The slides were made four years ago, so I'm not positive everything is still 100% relative. According to ESS, analog volume control is still better.

 

Here's al ink to the PDF -> LINK

 

 

 

[ericuco]

It is not really a "digital" volume control but Tortuga Audio makes a great LDR (light dependent resistor) based unit with remote volume, balance, 3 input control, auto calibration, and variable/adjustable impedance. It also has left and right channel level displays.

 

They have three versions of it. One with really nice case, one with simple case, and a kit. The innards are all the same with the kit less than half the price of the one with the nice case.

 

I am using it with my Shiit Yggy and it is really great.

 

The one caveat with the Tortuga Audio units is that they need to be located near your amps. I talked with the founder & owner, Morten, at RMAF this year about this and he confirmed it. I don't recall the distances he mentioned but it needs to be within a few feet for the best results.

 

For most people, this should not be an issue but you should be aware of it.

 

In my case, the balanced cables (XLR) between my DAC & amps are about 15 ft. So if I want to insert a Tortuga preamp, I would need to place it at the amp end (i.e. close to the amps) and not the DAC end.

 

My DAC has the ESS chip so I am sticking with that for volume control. I read the information that Chris provided from ESS to say that analog CAN be better but not necessarily.

[esldude]

Here is an interesting slide from an ESS presentation. The slides were made four years ago, so I'm not positive everything is still 100% relative. According to ESS, analog volume control is still better.

 

Here's al ink to the PDF -> LINK

 

 

 

 

I would say the key point is if the noise level of the analog control is better than the DAC. Most instances of analog volume control will be limited by the thermal noise of the resistance value they present.

 

Any digital control using 48 bit calculations should have no issue on being audible from its use. Dithered 24 bit signals can portray signal below the lowest bit which is usually already below the noise level of the other electronics. Not many systems overall have less noise than -120 db. Where in the signal path you do the volume reduction could still make a difference. In other words worrying about expensive analog volume controls of superb quality versus readily available digital control is mostly splitting hairs of hairs.

 

For years I used a switched resistor network made up of the highest quality Vishay resistors designed for low impedance low noise. I found when I had a Wadia DAC with digital volume you could hear no loss of quality using the digital control. That was some years ago and digital controls are if anything better than they were.

 

With the better DAC chips being 32 bit devices you really have pretty much nothing to worry about with regard to digital volume control. Those chips don't achieve analog results with even 24 bits of range, but they make sure no level of attenuation is going to matter. Doing better with analog would be unlikely, expensive, and still make a difference that is infinitesimal. Spend that money for something better anywhere else in the system and get more bang for your buck.

[ecwl]

Here is an interesting slide from an ESS presentation. The slides were made four years ago, so I'm not positive everything is still 100% relative. According to ESS, analog volume control is still better.

 

Here's al ink to the PDF -> LINK

 

 

 

 

I'm a hobbyist so what I think I know comes from what I've read in magazines and forums. So my understanding to getting great digital volume control (possibly superior to analog) are as follows:

 

The first part is the math:

1) You want the digital attenuation (the math) to be done correctly (which most DACs/computer software) can do.

2) You want enough bit-depth and accuracy for the volume reduction. As pointed out by Barrows, ESS probably uses 48-bit computation whereas JRiver uses 64-bit for computation. But the output matters too. Internal volume control digital output with ESS is 32-bit whereas most DACs are fed 24-bit signals if you're using the computer (or Weiss IN204 or my cheap miniDSP gadget) for digital volume control. So if you need a lot of attenuation and you're feeding your DAC 24-bit attenuated signals, you may be limiting your dynamic range. This is why most people prefer the DAC to do the internal digital volume control because most of them would be using at least 32-bit output which computers can't always offer.

3) You want good dithering algorithm. This is not just to dither in a greater dynamic range but also to make sure there is 2LSB TPDF dithering so that you don't get frequency dependent noise modulation of the noise floor. This is controversial because some people feel this is not audible but some feel this is highly audible and affects the naturalness of sound. Believe what you believe or test it out. JRiver recently added 2LSB TPDF dithering. The BitPerfect blog has an interesting article on this.

 

The second part is the DAC:

Because ultimately, you're feeding this digitally attenuated signal to the DAC, so the dynamic range of the DAC and the noise floor of the DAC will ultimately determine the performance of the digital volume control. If your DAC can only handle 96dB dynamic range and you need 50dB attenuation, you're now down to 46dB in dynamic range. Even some very high-end DACs advocate for analog volume control, e.g. MSB. But having a low noise floor and high dynamic range are not enough. You probably want low-level signal linearity because using your DAC as digital volume control means you're relying on high accuracy from the DAC for low-level signals. For example, most R2R DACs have low-level signal linearity issues so you'll either have to use lots of R2R DACs in parallel to average things out and improve the accuracy and linearity or you'll need to use the Metrum Pavane method to use one set of R2R DAC for the higher-level signals and another set of R2R DAC for the lower-level signals. That said, the ultimate performance you get would depend on the level of attenuation, the DAC, and the rest of your system so I can see someone using a relatively simple R2R DAC with some digital attenuation getting truly phenomenal sound. There is my backseat driver theory and then there is the reality of setting up the system and enjoying music.

 

The last part then is whether analog volume control is better than digital. I think the reality is that most people do not have a passive analog volume control/preamplifier. So what usually happens is that the preamp gives 8-9dB gain and then we listen at -10dB to -30dB so the analog volume control has to attenuate -18dB to -39dB because DACs can usually give out a reasonably high voltage output to feed the amplifier (or preamplifier) already. If you're using digital volume control, you're only adding -10dB to -30dB of digital attenuation which is anywhere between 2 and 5 bits. The issue here is that if you have a DAC with really low noise floor and high dynamic range, such as ESS quoted with -135dB, it's going to be really hard for a regular preamplifier, passive or active, to preserve that degree of dynamic range and noise floor. I think that's why ESS's conclusion is that it'll be hard for analog volume control to outperform ESS. My understanding is that you can find passive preamplifiers that are better than digital volume control although I've never heard one. However, I think John Swenson believes passive analog volume attenuation is definitely better. And I think a lot of people find it is very, very difficult to find active preamplifiers that outperform well-implemented digital volume control.

 

Going back to OP's question, if zenpmd is already thinking of upgrading to pure digital volume control, it comes down to what price range. I would definitely choose a new DAC with built-in digital volume control. He already uses Benchmark DAC2 which uses the digital volume control. If he is not happy with it, he can crank the DAC2's volume to maximum and then use the computer to change the volume (JRiver/BitPerfect/etc). I doubt he'll notice a difference or the DAC2's digital volume control may even perform better. I suspect in DACs that are up to double the price of the DAC2, you're mostly stuck with ESS DACs, perhaps up to Resonessence Invicta Mirus. An alternative DAC chip based design would include Hegel HD30. Non pre-fab DAC chip designs in this price range would include Chord Hugo, Hugo TT. I've read varying positive and negative reviews about the digital volume control of the PS Audio Directstream DAC so you definitely want to listen first. If zenpmd is willing to pay more than that, obviously, an audition of the specific DAC is needed, particularly using the digital volume attenuation.

 

Of course, there are many who believe that we are in a minority of faith-based audiophiles because our environmental noise and speaker excursion limits would fundamentally limit the noise floor and dynamic range of our speakers anyway so all this talk of >100dB dynamic range is irrelevant because we are only going to get 40dB dynamic range at home. My ears say they are wrong but what do I know.

 

Anyway, that's one man's opinion, possibly ill-informed. Hope it doesn't generate too much controversy...

[mourip]

The one caveat with the Tortuga Audio units is that they need to be located near your amps. I talked with the founder & owner, Morten, at RMAF this year about this and he confirmed it. I don't recall the distances he mentioned but it needs to be within a few feet for the best results.

 

For most people, this should not be an issue but you should be aware of it.

 

In my case, the balanced cables (XLR) between my DAC & amps are about 15 ft. So if I want to insert a Tortuga preamp, I would need to place it at the amp end (i.e. close to the amps) and not the DAC end.

 

My DAC has the ESS chip so I am sticking with that for volume control. I read the information that Chris provided from ESS to say that analog CAN be better but not necessarily.

 

Good point.

 

I have short IC's and long speaker cables. I am set up for DAC>LDR VC>DHT tube buffer with 3db gain>amp. In my case the buffer provides drive to the amp. Also in my case my Yggy DAC has no volume control and my preamp only has a manual volume control so I "needed" remote control also.

 

Morten is probably pointing out that most passive volume controls are not good at driving long IC's. I think that he is looking into adding a buffer option into his LDR which would be great if well implemented.

 

I do know that using the Tortuga for volume control in my system with my Alps Blue pot turned all the way up was easily better than just using the Alps.

[barrows]

Here is an interesting slide from an ESS presentation. The slides were made four years ago, so I'm not positive everything is still 100% relative. According to ESS, analog volume control is still better.

 

Here's al ink to the PDF -> LINK

 

 

 

 

Chris et al. It is very important to not misunderstand what the folks at ESS are saying here. The factor which they take into account in saying that ultimately the analog volume control is better is in how it relates to the noise floor. Their digital volume does not attenuate noise, so whatever the noise floor of the DAC is at rest, is always going to be the noise floor, regardless of how much attenuation is applied. The difference is that a analog volume control, operating after the DAC chip, will attenuate the noise floor as it attenuates the signal.

 

But! What is important to understand is what this means in the real world. Say we have an ESS chip with a noise floor around -135 dB, thence applies typical attenuation used for critical listening, like -30 dB. With -30 dB of attenuation we now have a noise floor at -105 dB in relation to a peak level signal. In an entirely quiet room (which no one has, even a well built audiophile dedicated room is going to have 20 dB of random noise), a human being can hear a dynamic range of, at best, 120 dB, and with 120 dB peaks we are risking hearing damage if we listen for long. So in reality, the fact that the digital volume control does not attenuate the noise as it attenuates the signal is irrelevant, unless the system is really, really poorly matched for gain and requires huge amounts of attenuation. If one's system is that poorly matched for gain, then you have a world of technical problems anyway which should be addressed if one cares about sound quality at all.

 

So then, one might make the argument that at very low listening volumes there are problems. But again, one should take into account the real world situation: at very low levels, one will not hear low level details anyway, as the are too low level to be discernible, so they do not matter.

 

Now an example of a system where a digital volume control offers the best possible performance: Here is how my system works:

 

1. Components are properly gain matched so I do not have to use excessive attenuation-my DAC outputs about 2.1 volts at full scale signal (0 dB attenuation) and my amplifier produces (400 watts into 4 ohm) full output at 2.0 volts input. My speakers are 4 ohm, and capable of handling just under 500 watts peak. So, technically, at full volume on my DAC, my amp produces full output, and my speakers are just capable of handling that.

 

2. I use my digital VC with anywhere from -20 dB attenuation (usually for higher level type recordings, rock, etc) and just a few dB, maybe -2 to -5 dB for audiophile type recordings with a lower average level (like Reference Recordings HRx, etc). These levels produce average in room volume around 80-85 dB average with peaks around 100-110 dB or so, very good reference listening levels.

 

3. With the above set up, the noise floor of the system is never audible, even when listening at the lowest possible volume; in fact, the noise floor of even the best digital recordings is higher in level than the noise floor of the system.

 

In conclusion: Although Mike is technically correct in what he is explaining, the real world truth is that as long as the system is not terribly compromised in gain matching, there is no advantage in using an analog based volume control, indeed, the extra thermal noise and distortion can be a disadvantage as well implemented digital VCs do not have these analog distortions (and yes, distortion of a resistor can indeed be measured).