[Lord Chaos]

What should those of us using a Mac with a DAC1 USB do for volume control, Elias? The headphones I use with the DAC1 are efficient, and the control on the DAC1 is only up to the fourth detent most of the time. I tried an external attenuator but that caused sound problems, as the Wiki suggests.

When I use the DAC1 as a pre-amp the volume is fine; I'm up around 10 o'clock or 11. How about putting a separate volume control on the unit for the headphone outputs? This would enable me to turn off the headphones when using speakers, too.

 

[EliasGwinn]

Quote:

Originally Posted by Jon L /img/forum/go_quote.gif Interesting article. A couple of questions: 1. Winamp is the only player where you specifically state it will play "bit-transparently." Does it mean foobar, iTunes, WMP will not play bit-perfectly? (even with ASIO plugins?) TIA

Thank you for pointing this out!! Foobar, iTunes, WMP will play bit-transparently. The settings I recommend are those necessary for achieving bit-transparency with these players.

Quote:

Originally Posted by Jon L /img/forum/go_quote.gif 2. Do all the comments pertain to only when you use Benchmark DAC-1 USB via the USB input or all DAC's/soundcards w/ USB/spdif? TIA

No... This does not only apply to Benchmark's DAC1. However, not all soundcards and/or USB/spdif are capable of bit-transparent playback.

Quote:

Originally Posted by Jon L /img/forum/go_quote.gif 3. The only player you mention as not having "high-distortion" digital volume control is WMP. Could you tell us what numbers are being measured and how, and especially at how much attenuation levels (-3 dB? -30 dB?). Were you able to subjectively correlate listening impressions of distortions caused by volume control by ear? TIA

I can try to put some specific numbers together for you, but here is our testing method:

We play a test-tone (10 kHz pure sine wave, -1 dBFS) via the media player in question. We set all settings necessary to achieve bit-transparency (sample rate, volume control, etc). We then monitor the FFT. When everything is set properly, the FFT looks like one tone at 10 kHz, with nothing but -135 dBFS noise floor around it (for 16-bit audio).

We then move the volume control, and monitor the FFT simultaneously. A well-built volume control will maintain the same FFT except the tone will be lower in amplitude.

See this page of iTunes FFT measurements for FFT's of the effect of volume control, sample-rate conversion, word-length truncation, etc.

http://extra.benchmarkmedia.com/wiki...n_Measurements

Thanks,
Elias

 

[EliasGwinn]

Quote:

Originally Posted by Lord Chaos /img/forum/go_quote.gif What should those of us using a Mac with a DAC1 USB do for volume control, Elias? The headphones I use with the DAC1 are efficient, and the control on the DAC1 is only up to the fourth detent most of the time. I tried an external attenuator but that caused sound problems, as the Wiki suggests. When I use the DAC1 as a pre-amp the volume is fine; I'm up around 10 o'clock or 11. How about putting a separate volume control on the unit for the headphone outputs? This would enable me to turn off the headphones when using speakers, too.

Do you have the 10 dB headphone pads set in the DAC1?

Otherwise, you may want to use a media player that has well designed volume control. VLC is a Mac media player whose volume control causes very little distortion.

Thanks,
Elias

 

[Lord Chaos]

I haven't opened the DAC1 to look at the jumpers. The manual says that the pads are set for attenuation, but they may not be. I'll take a look.

 

[audioengr]

Quote:

Originally Posted by EliasGwinn /img/forum/go_quote.gif The output attenuators do not affect the performance of the output stage whatsoever (according to our testing and our ears!). I run my DAC1 in my recording studio at the -20 dB level because it allows me to run the volume pot into its most optimal range (>10 o'clock). Now, it should be said that certain equipment (of questionable design) being driven by the classic DAC1 may be affected by the increase in source impedance that the attenuators present. A well designed device (high input impedance, low input capacitance) will not have any performance comprise when using the attenuators. The attenuators would only affect the sound quality if the equipment that was being driven by the DAC1 had low input impedance or high input capacitance, or if the DAC1 was driving a long cable. Unfortunately, some "audiophile" gear has poorly designed input stages with much too low impedance or large noise-draining capacitors which can affect frequency response drastically. In those cases, the source impedance should be as low as possible. I don't know what this reviewer was using as a pre and/or amplifier, so I can't comment on the validity of his experience. But if the device being driven is designed properly, the attenuators will have no affect on the audio whatsoever. The DAC1 USB, however, relieves this problem altogether with the new high-current output drivers. These new drivers will drive even the most difficult loads at any attenuator setting with out any performance comprimises. Thanks, Elias

I ALWAYS set the attenuators to minimum attenuation. IMO, any output impedance higher than 100 ohms tends to act as a low-pass R-C filter with most equipment and cables, not to mention the added noise of the series resistors themselves. Also, the earlier in the chain that you apply most of the gain, the less noise gets amplified by the downstream components, resulting in lower overall noise, so I prefer a much higher output voltage than the 2.25VRMS typical of most DAC's.

BTW, regarding jumpers, I changed the three internal power supply voltage jumpers in the DAC-1 to low-resistance, high-current berillium-copper jumpers and it seems to have made an improvement. The stock jumpers are really designed for low-level signals, not power currents. This is a really cheap change.

Steve N.

 

[EliasGwinn]

Quote:

Originally Posted by audioengr /img/forum/go_quote.gif I ALWAYS set the attenuators to minimum attenuation. IMO, any output impedance higher than 100 ohms tends to act as a low-pass R-C filter with most equipment and cables, not to mention the added noise of the series resistors themselves. Also, the earlier in the chain that you apply most of the gain, the less noise gets amplified by the downstream components, resulting in lower overall noise, so I prefer a much higher output voltage than the 2.25VRMS typical of most DAC's. BTW, regarding jumpers, I changed the three internal power supply voltage jumpers in the DAC-1 to low-resistance, high-current berillium-copper jumpers and it seems to have made an improvement. The stock jumpers are really designed for low-level signals, not power currents. This is a really cheap change. Steve N.

Steve,

It is not just your opinion, it is a fact. Any output impedance WILL cause a RC low-pass filter (high-end roll off) with a capacitive load.

However, the exact roll-off quantity can be determined based on the specific numbers. The 100 ohm standard you have chosen will usually be safe, however, even 100 ohms will have audible roll-off effects with enough cable (> 400 ft @ 32 pf per foot) and/or load capacitance.

If the attenuators in the DAC1 (classic) are set to 30 dB, the output impedance is 160 ohms. This means you can drive 255 ft of cable @ 32 pf per foot before you experience 0.1 dB attenuation at 20 kHz. In other words, driving 100 ft of cable into a well-built amplifier will be done just as well with the 30 dB attenuators as with the 0 dB attenuators.

Also, with the DAC1 USB, the 30 dB attenuation setting results in a lower output impedance (43 ohms) then with no attenuation (60 ohm).

The manuals for the DAC1 and DAC1 USB illustrate their respective impedances when using various XLR attenuation settings, and also lists the length of cable that can be driven at these settings. Here are the links to these manuals:

DAC1 (table is on page 8):
http://www.benchmarkmedia.com/dac1/DAC1-Manual.pdf

DAC1 USB (table is on page 14):
http://www.benchmarkmedia.com/manual...USB_Manual.pdf



Thanks,
Elias

 

[Jon L]

Quote:

Originally Posted by EliasGwinn /img/forum/go_quote.gif I can try to put some specific numbers together for you, but here is our testing method: We play a test-tone (10 kHz pure sine wave, -1 dBFS) via the media player in question. We set all settings necessary to achieve bit-transparency (sample rate, volume control, etc). We then monitor the FFT. When everything is set properly, the FFT looks like one tone at 10 kHz, with nothing but -135 dBFS noise floor around it (for 16-bit audio).

Oh, I only read the PC portion of the article. Anyway:

"A 16-bit 10k sine wave played through iTunes on OSX 10.4.6 with the volume control set near 50% (-19dB). The resulting distortion is well above the -129 dBFS noise floor and, consequently, is detrimental to the quality of the audio."

So can we assume the distortion decreases as volume control attenuates less? I say that b/c I have my Foobar volume usually around -3 dB, and I can't subjectively detect any audible sound quality difference. In fact, I don't think I hear any distortion at even -8 dB or so..

 

[milkpowder]

That's really all that matters, right? I admit I can't tell the difference between bit perfect, ASIO and whatever other technical jargon word there is for how my digital music is turning into analogue sound waves. I have also never heard any distortion from using software volume controls. This is not to say that there isn't measurable distortion.

 

[audioengr]

Quote:

Originally Posted by EliasGwinn /img/forum/go_quote.gif Steve, It is not just your opinion, it is a fact. Any output impedance WILL cause a RC low-pass filter (high-end roll off) with a capacitive load. However, the exact roll-off quantity can be determined based on the specific numbers. The 100 ohm standard you have chosen will usually be safe, however, even 100 ohms will have audible roll-off effects with enough cable (> 400 ft @ 32 pf per foot) and/or load capacitance. If the attenuators in the DAC1 (classic) are set to 30 dB, the output impedance is 160 ohms. This means you can drive 255 ft of cable @ 32 pf per foot before you experience 0.1 dB attenuation at 20 kHz. In other words, driving 100 ft of cable into a well-built amplifier will be done just as well with the 30 dB attenuators as with the 0 dB attenuators. Also, with the DAC1 USB, the 30 dB attenuation setting results in a lower output impedance (43 ohms) then with no attenuation (60 ohm). The manuals for the DAC1 and DAC1 USB illustrate their respective impedances when using various XLR attenuation settings, and also lists the length of cable that can be driven at these settings. Here are the links to these manuals: DAC1 (table is on page 8): http://www.benchmarkmedia.com/dac1/DAC1-Manual.pdf DAC1 USB (table is on page 14): http://www.benchmarkmedia.com/manual...USB_Manual.pdf Thanks, Elias

Sounds like the 0dB setting is a good one, except for the early gain consideration. I'll give it a try.

Steve N.

 

[smeggy]

Quote:

Originally Posted by milkpowder /img/forum/go_quote.gif That's really all that matters, right? I admit I can't tell the difference between bit perfect, ASIO and whatever other technical jargon word there is for how my digital music is turning into analogue sound waves. I have also never heard any distortion from using software volume controls. This is not to say that there isn't measurable distortion.

Cloth ears

I'm the same really, I normally don't care what it's doing as long as it sounds good, but psychologically I feel better if I know it's doing what it should and can. Even if I can't tell

 

[Scrith]

Some positive feedback on the DAC1 USB:

I received it a few days ago and immediately began comparing it to my existing setup (Lavry DA10 + Empirical Audio Off-Ramp Turbo 2 w/ Superclock 4). They sounded about the same to me. So I brought out my secret uber test: my wife's golden ears and her lack of interest for whatever strange setup I'm making her test. She listened to both setups without knowing which was which. She preferred the DAC1 USB by a slight amount, saying it seemed to have more detail and a broader soundstage (my interpretation of her hand gestures...close together for the DA10/Off-Ramp combo vs. far apart for the DAC1 USB).

I've been living with the DAC1 USB for a couple of days now and I've noticed a few other things: Windows sounds (the clicks and pops and beeps in the UI) actually sound right (previously these were usually a bit distorted, or clipped at the beginning), the sound meter in Foobar2000 is not delayed by several seconds anymore, and my sound during games works a lot better even when playing music in the background with Foobar2000. I attribute all of these problems to M-Audio's lousy USB drivers. I understand that Empirical Audio will soon have a firmware update that makes the Off-Ramp work like the DAC1 USB (using Windows' USB driver), so maybe that will help there.

Also, I'm able to listen to music resampled to 96K in Foobar2000 without any problems using the USB connection (something I have never been able to do with my other USB audio devices).

Sound-wise the DAC1 USB really does seem to be improved. I changed to a Lavry DA10 in mid 2006 from a Benchmark DAC1 (built in late 2005) after my wife picked the Lavry in a blind comparison. And that late 2005 DAC1 sounded significantly better than my original (late 2003?) DAC1. Congratulations on the continued improvements to a wonderful DAC.

By the way, I have had one wish for the DAC1 since I first started using them in late 2003: an on/off switch. Well, and I always wished for a USB or Firewire input, but that wish has finally been granted.

 

[Jetlag]

Quote:

Originally Posted by Scrith /img/forum/go_quote.gif (my interpretation of her hand gestures...close together for the DA10/Off-Ramp combo vs. far apart for the DAC1 USB).

It appears that women are now using the same system and scale to rate DACs as they do to rate men.

 

[laxx]

Quote:

Originally Posted by Jetlag /img/forum/go_quote.gif It appears that women are now using the same system and scale to rate DACs as they do to rate men.

Hah! That's funny.

 

[audioengr]

Quote:

Originally Posted by Scrith /img/forum/go_quote.gif Some positive feedback on the DAC1 USB: I received it a few days ago and immediately began comparing it to my existing setup (Lavry DA10 + Empirical Audio Off-Ramp Turbo 2 w/ Superclock 4). They sounded about the same to me. So I brought out my secret uber test: my wife's golden ears and her lack of interest for whatever strange setup I'm making her test. She listened to both setups without knowing which was which. She preferred the DAC1 USB by a slight amount, saying it seemed to have more detail and a broader soundstage (my interpretation of her hand gestures...close together for the DA10/Off-Ramp combo vs. far apart for the DAC1 USB). I've been living with the DAC1 USB for a couple of days now and I've noticed a few other things: Windows sounds (the clicks and pops and beeps in the UI) actually sound right (previously these were usually a bit distorted, or clipped at the beginning), the sound meter in Foobar2000 is not delayed by several seconds anymore, and my sound during games works a lot better even when playing music in the background with Foobar2000. I attribute all of these problems to M-Audio's lousy USB drivers. I understand that Empirical Audio will soon have a firmware update that makes the Off-Ramp work like the DAC1 USB (using Windows' USB driver), so maybe that will help there. Also, I'm able to listen to music resampled to 96K in Foobar2000 without any problems using the USB connection (something I have never been able to do with my other USB audio devices). Sound-wise the DAC1 USB really does seem to be improved. I changed to a Lavry DA10 in mid 2006 from a Benchmark DAC1 (built in late 2005) after my wife picked the Lavry in a blind comparison. And that late 2005 DAC1 sounded significantly better than my original (late 2003?) DAC1. Congratulations on the continued improvements to a wonderful DAC. By the way, I have had one wish for the DAC1 since I first started using them in late 2003: an on/off switch. Well, and I always wished for a USB or Firewire input, but that wish has finally been granted.

I recommend updating your playback setup for the Off-Ramp:
http://www.audiocircle.com/circles/i...?topic=40068.0

This improves Off-Ramp clarity and focus significantly on a PC. I found it to be equivalent to the driverless firmware. I still recommmend Foobar 0.8.3 and SRC at 24/96.

BTW - you should hear a modded DAC-1 USB with a Superclock clocking the upsampler. Really outstanding, but I still prefer the NOS I2S version with SRC upsampling by a small margin. Gives the customer more flexibility too.

Steve N.

 

[EliasGwinn]

Quote:

Originally Posted by milkpowder /img/forum/go_quote.gif That's really all that matters, right? I admit I can't tell the difference between bit perfect, ASIO and whatever other technical jargon word there is for how my digital music is turning into analogue sound waves. I have also never heard any distortion from using software volume controls. This is not to say that there isn't measurable distortion.

The quality of any solution is limited by the perceived severity of the problem.