Print

Please login or register.
1 Hour 1 Day 1 Week 1 Month Forever
Login with username, password and session length

[samtheman57]

I finally have my HRT Music Streamer II set up with MC 17 after swapping some cable, and adjusting buffering and Windows settings.

For those who don't know the streamer does not include a physical volume control.

After removing my old sound card and installing the streamer, I had the physical volume on my M-Audio monitors set to minimum detent , as I wasn't sure what the balance between the device (USB DAC) and the MC digital volume would be, I didn't want to blow my face off. I adjusted the system volume to about 20 per cent.

I started playback in MC, and sure enough, even with the speakers at minimum detent and system volume set around 20 per cent the volume was WAY too loud, I almost had a heart attack, and started fumbling for the controls.

After some more misadventures, I finally settled on using the internal volume setting in MC as a control (Using my mouse wheel to adjust volume) against permanently setting the M-Audio monitors at the default half detent position.

I am perfectly happy to adjust volume this way, but I generally have to keep the settings in the minus DB ranges.

To the point, I understand the J River volume control is superior to the volume control in other players.

Strictly from a sound quality perspective, do I benefit more from putting a passive physical volume control in the chain, (which I really don't want to do) as it just adds more clutter, or just stay with using MC internal?

Thanks for any thoughts.

[hulkss]

Set JRiver to upsample your audio to 24 bits (CD audio is 16). In that way you will have plenty of dynamic range for digital volume.

[Vincent Kars]

As the HRT is a 24 bit DAC, this is already the case.

In general it is recommended to keep the amount of digital volume control limited to avoid loss of resolution.

I assume the volume control on the speakers is analog.
Simply compare analog set to low and digital to high and the reverse.

even with the speakers at minimum detent and system volume set around 20 per cent the volume was WAY too loud

I’m not good in electronics but might it be that there is a mismatch between the output of the DAC and the input of the speakers?

http://www.thewelltemperedcomputer.com/KB/BitPerfect.htm

[Skogkatt]

M-Audio web site doesn't seem to specify sensitivity for their active monitors that is likely higher than a typical power amplifier.

With JRiver MC would be better to limit attenuation to 15-20dB so that you won't loose too much resolution on the DAC.

Probably the best would be to build a passive attenuator (D.I.Y. resistor divider) placed just at the input of the active monitors. A series resistor of 3.9k ohms followed by a parallel resistor of 1k ohms would provide an attenuation of roughly 14dB (depending on the input impedance of the speakers) that might be sufficient. A series resistor of 9.1k ohms followed by 1k ohms to ground would provide approximately 20dB of attenuation in the case 14dB isn't enough. The use of this type of attenuator should allow you to set the internal volume control of your monitors to the max. (thus eliminating part of the negative effects of the volume potentiometer inside the monitors). Also, a passive attenuator is made of fixed value resistors with no wiper contacts and should sound more neutral than a variable resistor (potentiometer). Of course, this is a viable option only if you have the ability to deal with a solder iron…. unless you have a friend that can help you in building this. 

[samtheman57]

Set JRiver to upsample your audio to 24 bits (CD audio is 16). In that way you will have plenty of dynamic range for digital volume.

Thanks for the replies, much appreciated.

J River is set to upsample to 24 bits, as is the playback device in Windows Sound Properties.

As the HRT is a 24 bit DAC, this is already the case.

In general it is recommended to keep the amount of digital volume control limited to avoid loss of resolution.

I assume the volume control on the speakers is analog.
Simply compare analog set to low and digital to high and the reverse.
I’m not good in electronics but might it be that there is a mismatch between the output of the DAC and the input of the speakers?



http://www.thewelltemperedcomputer.com/KB/BitPerfect.htm

Yes the, speaker controls on the back are analog, the problem is that they are not only inconvenient to reach and impractical as physical volume controls, (based on the tiny size) but also there is only one detent or "click stop" at 50 percent. Anything between minimum half and maximum is your best guess.

I will try minimum vs. 50 per cent to see if it sounds better or worse.

As for the output of DAC vs. input of speakers being mismatched, possible, as I switched from balanced input (mini to 1/4 inch TLR on old sound card) to unbalanced. (RCA out from DAC to RCA in on speakers) 

M-Audio web site doesn't seem to specify sensitivity for their active monitors that is likely higher than a typical power amplifier.

With JRiver MC would be better to limit attenuation to 15-20dB so that you won't loose too much resolution on the DAC.

Probably the best would be to build a passive attenuator (D.I.Y. resistor divider) placed just at the input of the active monitors. A series resistor of 3.9k ohms followed by a parallel resistor of 1k ohms would provide an attenuation of roughly 14dB (depending on the input impedance of the speakers) that might be sufficient. A series resistor of 9.1k ohms followed by 1k ohms to ground would provide approximately 20dB of attenuation in the case 14dB isn't enough. The use of this type of attenuator should allow you to set the internal volume control of your monitors to the max. (thus eliminating part of the negative effects of the volume potentiometer inside the monitors). Also, a passive attenuator is made of fixed value resistors with no wiper contacts and should sound more neutral than a variable resistor (potentiometer). Of course, this is a viable option only if you have the ability to deal with a solder iron…. unless you have a friend that can help you in building this. 

Yeah attenuation HAS to be limited to just about that range, as any boost beyond becomes harsh and unpleasant.

I think I have seen these on EBay as kits, but again, I was hoping for something less ugly, and pre built, not that I am afraid to solder.

Just for experiment, I again switched back to system volume and that is way too loud, even with all settings on minimum.

Thanks again for the input.

[mojave]

Set JRiver to upsample your audio to 24 bits (CD audio is 16). In that way you will have plenty of dynamic range for digital volume.

The internal volume control is done at 64 bits regardless of the final output so that you don't lose any dynamic range.

[Vincent Kars]

RCA out from DAC to RCA in on speakers

Should work but as far as I know, line levels in RCA are not standardized.
As balanced uses higher voltage levels than RCA (2 times ?), maybe RCA to TRS is an option.
But again, I’m not very good in electronics.

[Vincent Kars]

The internal volume control is done at 64 bits regardless of the final output so that you don't lose any dynamic range.

64 bits is about precision. This keeps quantisation errors deep down.
But in the end you have to convert to integer to feed the DAC.

If the signal is
1111111111111111 and we want to play it half as loud you get
0000000011111111
So you do lose resolution.
As you lose the LSB first (- 96 dBFS with 16 bits) it won’t be too dramatic.
But losing 8 bits (and applying dither) might become audible

[Skogkatt]

The internal volume control is done at 64 bits regardless of the final output so that you don't lose any dynamic range.

True. Unfortunately it isn't MC that reduces dynamic range for the reason you explained, is in the converter where, reducing the volume
"digitally", the most significant bits remain progressively unused: so more you decrease MC's volume and less dynamic range is available from the DAC.
The DAC it self can exploit its maximum dynamics only with MC's volume set to 100%. A trade off would be achieving a normal listening volume with volume in MC
set not too low (attenuation less than -20dB) but this requires to adjust the gain of the analog chain.

[DarkPenguin]

True. Unfortunately it isn't MC that reduces dynamic range for the reason you explained, is in the converter where, reducing the volume
"digitally", the most significant bits remain progressively unused: so more you decrease MC's volume and less dynamic range is available from the DAC.
The DAC it self can exploit its maximum dynamics only with MC's volume set to 100%. A trade off would be achieving a normal listening volume with volume in MC
set not too low (attenuation less than -20dB) but this requires to adjust the gain of the analog chain.

Does any of this matter if your original content is 16 bits?

[Skogkatt]

Does any of this matter if your original content is 16 bits?

It matters about the same as if the content was 24bits.
Any additional 6dB of attenuation is like losing 1bit of resolution on the converter: on a 24bits converter dither helps to
"contain" the loss and luckily almost all the DACs around are 24bits.

[Matt]

This isn't a simple question.  We were all standing around the white board arguing the merits of each approach, and basically decided "it depends."  Since that's not too helpful, I'll at least post a few notes:

1) Media Center uses 64-bit precision for applying volume, meaning it's essentially perfect from the software side.

2) You lose 1-bit from your 24-bit output each -6dB.  But since music has to be really loud to hear past 14 or 15 bits, this may not really matter.  In other words, you lose more bits at low volume but you also need less bits at low volume.

3) The comparison between low-software-volume/high-hardware-volume and high-software-volume/low-hardware-volume depends on the hardware.  Sometimes it's better to do hardware volume and sometimes it doesn't matter.  One helpful test is to see if the background hiss (noise) at the speakers while audio is paused increases as you increase the hardware volume.  If it does, this is an argument for not pushing the hardware volume more than you have to.

4) We couldn't think of any reason putting a resistor on the DAC would give a better volume than the internal volume.  Resistors just throw bits away in the analog realm.

5) One tip is that I'd recommend the 'Volume Protection' feature in Media Center.  It's designed exactly for cases where accidentally high volume could, as you said, "blow your face off."

[Vincent Kars]

3) The comparison between low-software-volume/high-hardware-volume and high-software-volume/low-hardware-volume depends on the hardware.  Sometimes it's better to do hardware volume and sometimes it doesn't matter.  One helpful test is to see if the background hiss (noise) at the speakers while audio is paused increases as you increase the hardware volume.  If it does, this is an argument for not pushing the hardware volume more than you have to.

Good one.

What do you think of this article (digital vs analog volume control)?
http://www.resonessencelabs.com/invicta/invicta_analog_vs_digital_volume.pdf

[mojave]

Good one.

What do you think of this article (digital vs analog volume control)?
http://www.resonessencelabs.com/invicta/invicta_analog_vs_digital_volume.pdf

I know you already read this reply, Vincent, but I thought I'd post it for others. Roger Dressler responded to the article at What's Best Forum:

S/N is not relevant in this discussion. Consider the usual cases where S/N is of concern: An audio recording is delivered by various formats: CD, FM radio, audio tape, cassette. The signal would be played at the same loudness each time, but the noise levels would vary from inaudible to very audible.

In the case of the analog volume control, assuming a noiseless output stage and power amp, the signal and the noise both reduce together as the volume is turned down. If the source was noiseless at full volume, it remains noiseless as volume is reduced.

So what about the digital volume control? The DAC has a noise floor which presumably reflects the result of a properly dithered signal. If the DAC is worth its salt when playing at full volume, the noise is acceptably low, probably inaudible except in unusually quiet rooms. Furthermore, since it is properly dithered, there is no quantization noise when signals are present, no matter how softly they are recorded.

Now we turn down the volume digitally, as much as you like, 10dB, 40 dB. The noise floor remains fixed, i.e., virtually inaudible. The quantization noise remains non-existent as it is still properly dithered.

S/N has reduced, but this time the noise is constant while the signal is reduced. Where's the problem?

No one (certainly not me) is suggesting that analog gain controls need to be abandoned. But even analog gain stages have no control on the noise floor of the electronics that follow, such as the output gain stage and the power amplifier. If the speaker sensitivity variation is extreme (20 dB) or the amplifier gain variation is extreme (15 dB), those can conspire to expose the analog noise floor to unacceptable levels.

Now add to the system a high quality bank of PEQs, like Harman or QSC. We see this done routinely in high end home theaters. Maybe noise is less of an issue there?  Here, the A/D and D/A sit right in front of the power amplifier. No analog volume control between. In essence, the dynamic range / distortion / noise floor performance is exactly as would exist with a digital volume control. If anyone is complaining about this post-volume EQ topology, it is about the extra analog/digital/analog conversion cycle (that's why I bought the SSP-800, PEQ inside), not the noise floor and not the lower S/N as the volume is reduced resulting from lower digital resolution. Probably because this is a non-issue in reality.

In fact, the performance of the system would potentially improve if those digital EQ boxes could be slaved to act as the system volume control. In that case, the analog signals feeding the A/D would be a) fixed and b) driving the system closer to 0 dBFS more of the time. That would improve the low-level performance issues of arguably the weaker converter in the box, the A/D, by at least 10 dB.

Someday we will see an all-digital chain, right up to the speaker terminals. All of a sudden our attitude may change due to the compelling concept of staying in the purely digital domain from the mastering desk to the voice coil. Even though it will "suffer" the same "low S/N disadvantage" being discussed here, I somehow suspect when that day comes no one will be claiming that analog volume controls are superior in a digital delivery chain.

IMHO, Resonessence Labs totally shot themselves in the foot with their poor explanation.

[pcstockton]

5) One tip is that I'd recommend the 'Volume Protection' feature in Media Center.  It's designed exactly for cases where accidentally high volume could, as you said, "blow your face off."

Ive seen that setting and always wondered how it worked?  How does it know my preamp is turn all the way up 

thanks,
Patrick

[mojave]

Here is another way to view digital volume control.
You have JRiver/HTPC, a 24-bit DAC, and an amp.

Within JRiver you can adjust the volume down and up a much as you want with the DSP.
It is always identical when truncated back to 32 bits.
Think of sliding the volume down and up in a tube that is 1000 dB long (see first picture).
You can push it all the way down and slide it all the way back up and it will still be the same.

Now you send the sound to the DAC.
With a 24-bit DAC you can attenuate by 48 dB before actually losing any audible info.
We are going to play a piece of classic music that has a range of 30-100 dB when the volume is at maximum.
This means our "tube" has been shortened from 1000 dB to 100 dB.
The music has a dynamic range of 70 dB.
You can hear the trumpets at 100 dB at 4:32.
You can hear the chimes at 30 dB at 0:10.


The DAC has a noise floor of -110 dB. It is so low you can't hear it.
When you lower the volume digitally, the noise floor stays the same.
It doesn't decrease or increase.

The noise floor of your room is 30 dB. Music played below 30 dB will be masked by room noises.

You use the digital volume control and lower the volume 20 dB and send it to the DAC.
Your entire signal has been slid down 20 dB and the chimes are now at 10 dB.
You can no longer hear the chimes due to the noise floor of the room.
The audible portion of music is still identical to when you had full volume.

You keep lowering the volume until you have attenuated by 40 dB.
Since you are only listening at 60 dB, and since the music has a dynamic range of 70 dB,
you are cutting off 10 dB at the bottom. Due to the noise floor of the room, you can only
clearly hear about 30 dB of the music.

You continue to lower the volume until you are at a max of 40 dB. You have now lost 30 dB
of dynamic range from the music and 60 dB overall.
Have you thrown away bits? Yes.
Were they audible? No, you have lowered the volume so that the lost portion wasn't audible anyway.

Is the part you can still hear identical to when you had the volume at 100%?
Yes, it is completely identical when dithered to 32 bits or less.

Has the dynamic range of the DAC changed?
No, it has stayed at -110 dB the whole time.

Has the dynamic range of the music changed?
Yes, the dynamic range is never higher than the loudest portion and
decreases when the volume is lowered below the noise floor of the room.



 

[retro]

5) One tip is that I'd recommend the 'Volume Protection' feature in Media Center

Where do I find this..?

[retro]

Where do I find this..?

Anyone...?!?

[DarkPenguin]

http://yabb.jriver.com/interact/index.php?topic=64574.0

[Blaine78]

always found that a good passive volume is better than a digital volume.

[samtheman57]

I want to thank everyone for their input on this.

I, again, asked Kevin Halverson with HRT technologies about this since I am running a HRT Streamer and he seems to agree with Blaine 78:

"As for the question of a passive volume control and avoiding digital attenuation, this is absolutely true.  Digital attenuation reduces resolution along with amplitude so avoiding it is always the best choice.  Here is a link to a passive volume control that should work well for your needs".

So I ordered the NHT PC PVC (with RCA in and Out) and will try that to see if I get a cleaner sound.

http://www.nhthifi.com/passive_volume_control_PVC_PC

[Alex B]

"As for the question of a passive volume control and avoiding digital attenuation, this is absolutely true.  Digital attenuation reduces resolution along with amplitude so avoiding it is always the best choice.  Here is a link to a passive volume control that should work well for your needs".

This is not true. The "resolution" in the audible range is not reduced at all. It is a common misconception that DSP attenuation would affect the perceivable quality of the louder signals that are not in the range which will be truncated or dithered.

As you can see in mojave's nice graphic, the digital attenuation causes the quietest signals to drop below the hearing threshold. In the worst case scenario, assuming that the source is 16-bit and the output device is also limited to 16-bit, in theory the signals that fall in the least significant bit in the outputted signal (or in my experience actually about half of that, i.e. the signals that are not louder than -93 dBFS) become audibly distorted after the whole range has been attenuated, i.e. been moved to a lower volume level, but you will never hear that inaccuracy. I once created manufactured audio sample files that make this very quietest range audible. A download link to the sample package is in this forum post: http://yabb.jriver.com/interact/index.php?topic=59070.msg399356#msg399356 . The manufactured samples are 48 dB louder than the original 16-bit signal to make the differences audible.

In general, my advice would be to use the powered speakers' or preamp's analog volume control as a calibrator that sets the maximum preferred volume level (i.e. set the internal volume to the max, play some various tracks and calibrate the speakers to play at the highest volume you will ever want to hear.) After the speakers have been calibrated the full range of MC's volume slider will be more usable. I use this system with an integrated amp. Its volume knob is set approximately at "the loudest I'll ever want to use" and MC's internal DSP sets the desired attenuation.

Since the above is not easily possible in your case, the NHT PC-PVC device will probably be useful for adjusting the speakers volume calibration and also if you sometimes want to use other signal sources than the PC.

[samtheman57]

I received the NHT PVC PC (RCA ins and outs) yesterday.

I am running internal volume on J River at 100 per cent for the first time, and attenuating with the NHT PVC, which seems to yield a much more pleasing result to my ears.

Without the PVC in the chain, this (digital slider at 100 per cent system and/or internal volume) was not possible, it would appear the PVC is exactly what was needed to balance the JRiver/Streamer II output signal to my M-Audio active monitors.

The M Audio output "Volume" on the back of the speakers are at 50 percent (default position)

Practically, I have made a few clumsy mouse clicks and nearly blew my face off with volume (as I described before) this can't happen now.

IMHO this is the only way to go if your chain is J River> External DAC> Active Monitor.

Thanks again for the great software and technical explanations even though some are over my head.

[Trumpetguy]

IMHO this is the only way to go if your chain is J River> External DAC> Active Monitor.

I would add "the only way in your system", not any system with the same signal chain. I do not have active monitors, but signal chain is the same:
J River>External DAC>power amplifier>speaker

And it works great. JRiver (and windows) volumes at 100%, digital attenuation in the Lynx DAC driver (somewhere between -12 and -25dB, source and situation dependent).

[theDen]

i listen to music using Wasapi - Event style with following setup:
Arcam rDac (via USB + 24bit\96khz) -> Arcam Delta 290p (amplifier without volume control) -> Mission m62i
so i can tune volume only with internal volume control, wich is most of the time on 20%-30%

when music is played this quiet - it's not very involving...so i suppose if i get pre-amp, where i will be setting maximum "accepted" volume in the current music session, and in MC internal volume would be on 90-100% - it will add much details to music, right?

ps. about 40% of material is 24bit\96kHz

[SamuelMaki]

i listen to music using Wasapi - Event style with following setup:
Arcam rDac (via USB + 24bit\96khz) -> Arcam Delta 290p (amplifier without volume control) -> Mission m62i
so i can tune volume only with internal volume control, wich is most of the time on 20%-30%

when music is played this quiet - it's not very involving...so i suppose if i get pre-amp, where i will be setting maximum "accepted" volume in the current music session, and in MC internal volume would be on 90-100% - it will add much details to music, right?

ps. about 40% of material is 24bit\96kHz

I have thought that internal volume and physical volume are equal quality... So it debends your config which one to use (and DSP choices...) If you are not doing any DSP, then you are fine with internal volume And remember to set windows volume at 100% 

[Zenith]

i listen to music using Wasapi - Event style with following setup:
Arcam rDac (via USB + 24bit\96khz) -> Arcam Delta 290p (amplifier without volume control) -> Mission m62i
so i can tune volume only with internal volume control, wich is most of the time on 20%-30%

when music is played this quiet - it's not very involving...so i suppose if i get pre-amp, where i will be setting maximum "accepted" volume in the current music session, and in MC internal volume would be on 90-100% - it will add much details to music, right?

ps. about 40% of material is 24bit\96kHz

It could be also a problem of impedance mismatching between Dac/Power amplifier...even choose the right passive amplifier (potentiometer is not so easy).
But I'm not an expert I'm getting info on the subject ...It would be interesting to test a Buffer like the Pass b1.

By the way this is what mr. Nelson Pass says about passive preamplifier:

Routinely DIYers opt to make themselves a "passive preamp" — just an input selector and a volume control. What could be better? Hardly any noise or distortion added by these simple passive parts. No feedback, no worrying about what type of capacitors - just musical perfection. And yet there are guys out there who don't care for the result. "It sucks the life out of the music", is a commonly heard refrain (really - I'm being serious here!). Maybe they are reacting psychologically to the need to turn the volume control up compared to an active preamp.

Some people says that also digital volume sounds flat:

http://www.computeraudiophile.com/threads/7362-Conecting-JRiver-DAC-directly-to-Audio-Reseach-Power-Amp?highlight=Conecting+River

Other than how digital volume work I think it also could depend on the analog stage of sound card/dac, matching output /input impedance between source and power amplifier, and in what kind of sound this people is used to listen.
Preamps could add "something"

[icstm]

a couple of comments, but they might not be true, just my thoughts.

1) Rather than thinking should I "reduce" the volume inside software or on my amp, if you turn it around and think "would I rather have my software increase the volume to an acceptable level, or use my amp?”, I think it makes sense to use the software to ensure its volume control is on maximum, before you amp is on maximum. Then following the logic above you are decrease the software from 100% as required. If the software is not on 100% ever during that listening session, then your amp is doing unnecessary work

2) as for preamps “adding” something. Denon amps for years ago used to have a “variable loudness” knob and many amps today try to do the same thing automatically, which is to change the balance between treble and bass at lower volumes (as the bass might not come through at lower volumes)

[Matt]

There's some information that might be useful here:
http://wiki.jriver.com/index.php/Volume

[mojave]

1) Rather than thinking should I "reduce" the volume inside software or on my amp, if you turn it around and think "would I rather have my software increase the volume to an acceptable level, or use my amp?”, I think it makes sense to use the software to ensure its volume control is on maximum, before you amp is on maximum. Then following the logic above you are decrease the software from 100% as required. If the software is not on 100% ever during that listening session, then your amp is doing unnecessary work

Music and movies are mastered at a "maximum level" so you are reducing the level regardless of how you think. This is why volume controls are called attenuators - they only reduce volume. Volume control is ultimately a matter of how much voltage is being passed through to the amplifier. An amp only does work based on the voltage it receives regardless of where the attenuation takes place. In other words, if you listen at the same volume level the amp is doing exactly the same thing regardless of where the volume is controlled.

Some also confuse a receiver with a built in amp vs a standalone amp. A standalone amp does not have volume control. It only has gain control which is used to match the output voltage of the preamp or receiver to the input voltage on the amp to prevent clipping. You can have the gain control turned down and the amp can still produce its maximum rated power output.