• WANTED: Happy members who like to discuss audio and other topics related to our interest. Desire to learn and share knowledge of science required as is 20 years of participation in forums (not all true). Come here to have fun, be ready to be teased and not take online life too seriously. We now measure and review equipment for free! Click here for details.

Software Player volume vs DAC digital volume vs Analog volume

Joined
Mar 5, 2018
Messages
18
Likes
3
#1
Dear all and Armirm,

Several topics on audio forum explain/confirm/put the question about the best method to manage the volume in our audio system when we use software player + DAC +Preamp and/or Amp.

I am a little lost and would like if it is possible for you that you write an article with factual measurements on several systems configuration to identify the best solution and clarify all advises:
We could imagine to test several configurations, ie:
- Foobar/windows/Jriver
- DAC: very good (Oppo205/Benchmark), middle and first price
- AMP: class A, Hypex

For sure it is a huge work of preparation/measurement for you but I am fully convinced that it will be very, very,..............., very useful for all Audio crazy men that we are ;-)

Thank you for your feedback, advise.

Hervé
 
Joined
Feb 13, 2018
Messages
74
Likes
44
Location
Laguna, Philippines
#2
If you are listening at a normal volume levels ~80 dBc for me, you are not gonna hear all the bits and information of the digital spectrum anyways so it doesn’t not matter if you control the volume digitally or through an analog preamp. No difference in sound unless you induce clipping by having an integrated amp turned up all the way which would definitely increase SINAD significantly while digitally controlling the sound. If you have a separate passive preamp and separate poweramp there would be zero difference in digitally controlling the volume with the passive preamp on max volume or -0db on digital volume and varying the passive preamp volume during normal listening levels and volume matched as well
 
Joined
Mar 5, 2018
Messages
18
Likes
3
#4
thanks for yours answers.
I know the last archimago article. Nevertheless, I would like to see measurement with different type of configuration. For example Jriver uses a 64 bits resolution for the volume vs 32 bits for Foobar. An other point, OPPO105/205 use ESS sabre 32 bits volume, etc....

Fact, measurement, the only judge to answer to this question several times discussed and shared on audio forum.
 

bennetng

Addicted to Fun and Learning
Joined
Nov 15, 2017
Messages
592
Likes
455
#5
Fact, measurement, the only judge to answer to this question several times discussed and shared on audio forum.
It seems that you think that the article above (I wrote it but Archimago also proofread it and added other information) does not provide fact and measurement. The article linked to some other articles, for example...

ESS provided measurements of their volume control in this presentation:
http://www.esstech.com/index.php/download_file/view/70/267/

Pro Tools also provided measurements at the end of their white paper:
http://akmedia.digidesign.com/support/docs/48_Bit_Mixer_26688.pdf

I measured Reaper's 64-bit volume control in the first part of my article, later I also presented measurements to illustrate the differences between fixed-point and floating point format even if the the bit-depth remains the same (e.g. 32). These are factual results and I did not fabricate those data.

Archimago also pointed to another article with some measurements he made using analog output:
http://archimago.blogspot.com/2019/02/musings-why-bother-with-24-bit-dacs.html

Measurements provide factual information, but it is also important to pay attention to test design, procedure and limitations. For example SPDIF or some audio controllers and drivers are confined to 24-bit data, in some worse cases data would be truncated to 16-bit due to compatibility issues. For example Amir often uses ASIO4all or other ASIO drivers, sometimes issues may show up but it would be completely fine with WASAPI. While these issues are not related to the intrinsic quality of individual playback software's volume control, the results of measurements may indicate so and it would be deceiving.

Therefore even if measurements provide "facts", these "facts" can also vary.
 
Joined
Mar 5, 2018
Messages
18
Likes
3
#6
dear Bennetng,

To avoid misunderstanding, I confirm that your article in Archimago is one of the most clear and factual post related to digital volume that I read on others forum. Your work is very great and interesting. Thank you very much for this.

I would like to see simple comparison between several configurations to observe the result and impact good or not about the digital management of volume. My config does not use preamp: FOOBAR-> HDMI-> DAC OPPO105D -> AMP (hypex NC500). For sure I am convince to limit the number of equipment to transform digital data in analog are for me the best choice. Some measurement would allow to definitively confirm or not this approach
 

Leon

Member
Joined
Mar 5, 2018
Messages
18
Likes
3
#7
Dear Bennetgn,

Your advise will be appreciated on the following point:
- Do you think is better to use the volume on foobar player or on the DAC?

Thank you
 

bennetng

Addicted to Fun and Learning
Joined
Nov 15, 2017
Messages
592
Likes
455
#8
Use the one on foobar if the required volume reduction is less than around 20-25dB. It is based on my observation about the dynamic range of good DACs these days, and assuming the audio driver is at least sending 24-bit data to the DAC.

It is not about the precision of volume control (e.g. how many bits or some fancy dithering), it is about the ability of foobar's volume control to handle floating point induced clipping.

If you need a lot of attenuation (e.g. more than 30dB) it may be useful to use the analog volume control on the amp alongside with foobar's volume control, for example foobar -20dB and amp -20dB. In this way you don't need to turn down the analog volume too much which may induce channel imbalance.
 

Leon

Member
Joined
Mar 5, 2018
Messages
18
Likes
3
#9
Thank you for your answer. If I understand, even OPPO 105 ESS SABRE uses a 32 bit volume resolution but without floating point management, it is better to use Foobar with a 24 bits resolution including floating point management, is it right?

I assume that I use a resolution of 24bit 96khz for music on Foobar output.

Thank you to confirm if I understand correctly.
 

bennetng

Addicted to Fun and Learning
Joined
Nov 15, 2017
Messages
592
Likes
455
#10
If I understand, even OPPO 105 ESS SABRE uses a 32 bit volume resolution but without floating point management, it is better to use Foobar with a 24 bits resolution including floating point management, is it right?
Yes.
I assume that I use a resolution of 24bit 96khz for music on Foobar output.
96kHz is sample rate. It is not foobar's job to change this value. If you are using WASAPI exclusive or ASIO then the DAC will automatically change this value to match the audio file's sample rate. The values on Windows' audio property page is only for applications using shared mode like web browsers, games and so on.
 

Leon

Member
Joined
Mar 5, 2018
Messages
18
Likes
3
#11
Thanks a lot. Your article suggests to use replaygain.
Could you confirm the config in the menu playback : replaygain as to set:
- source mode: album
- processing: apply gain and prevent clipping.

how alter file content menu has to be set : target level set to 89db (default) or much more?

Thank you again for your answer
 

bennetng

Addicted to Fun and Learning
Joined
Nov 15, 2017
Messages
592
Likes
455
#12
If you prefer manual volume adjustment by using foobar's volume control yourself then use "prevent clipping according to peak", otherwise use "apply gain and prevent clipping according to peak".

Alter file content is only for file converter and not for playback. Ignore it.
 
Joined
Sep 26, 2019
Messages
39
Likes
44
#13
IMHO that's a terrible article, and I'm surprised Archimago accepted it as a guest article. The guest writer's writing style meanders between irrelevant points, and the conclusion is unsupported.

There can't possibly be any measurable difference between (i) a floating point digital volume control (32 or 64 bit float, doesn't matter) quantized and dithered to 24 bit at the very end, and (ii) a 24 bit integer digital volume control, dithered on output. Even the very best DACs only manage ~21 bit real world analog performance. Any quantization error from a single multiplication at 24 bit depth is going to be beneath the noise floor.

Sure, if you're running a long multistage DSP pipeline, there are some good reasons to use a float pipeline. But even then, the difference is very small; a lot of good tools, e.g., SoX and many of the most popular commercial WAVES plugins, still use an integer pipeline.
 

g29

Active Member
Joined
May 1, 2019
Messages
289
Likes
166
#14
IMHO that's a terrible article, and I'm surprised Archimago accepted it as a guest article. The guest writer's writing style meanders between irrelevant points, and the conclusion is unsupported.

There can't possibly be any measurable difference between (i) a floating point digital volume control (32 or 64 bit float, doesn't matter) quantized and dithered to 24 bit at the very end, and (ii) a 24 bit integer digital volume control, dithered on output. Even the very best DACs only manage ~21 bit real world analog performance. Any quantization error from a single multiplication at 24 bit depth is going to be beneath the noise floor.

Sure, if you're running a long multistage DSP pipeline, there are some good reasons to use a float pipeline. But even then, the difference is very small; a lot of good tools, e.g., SoX and many of the most popular commercial WAVES plugins, still use an integer pipeline.
Your same argument can be used to discredit using DACs with 21-bits of resolution because the rest of your signal chain can't achieve the same 21-bits of resolution. The OP asked which was the better volume control which appears to be what has been answered.
 

bennetng

Addicted to Fun and Learning
Joined
Nov 15, 2017
Messages
592
Likes
455
#15
...and caused some problems.
https://www.audiosciencereview.com/...oesnt-adhere-to-scarlet-book.5901/post-132540

It is because those classic WAVES plugins work with hardware working in fixed point like Pro Tools TDM. The newer generation of plugins work in 64-bit float already.

http://avid.force.com/pkb/articles/faq/en423351
Q: Why did Avid switch from 24-bit processing/48-bit fixed summing to hardware that offers 32-bit floating-point processing/64-bit floating-point summing, and what does it mean for me?
A: There are several reasons for the transition. A 64-bit floating-point mix bus provides more than 1,000 dB of headroom, which is more than enough to handle the huge track counts that Pro Tools|HDX can deliver. Also, by moving the insert paths to a 32-bit floating-point format, Pro Tools|HDX offers much more dynamic range for plug-in processing, making it nearly impossible to clip the plug-ins, while also being able to handle greater than 24-bit audio file formats.

Many of you have told us that even with dual-precision plug-ins processing at 48-bit, the path between inserts on Pro Tools|HD was still limited to 24-bit — thus, limiting any gains. With Pro Tools|HDX, all data streams are maintained at full 32-bit floating point and then summed in the DSP mixer at 64 bits. Our beta customers told us that they could hear and appreciate the difference.
 
Joined
Sep 26, 2019
Messages
39
Likes
44
#16
That's the SOX plugin for Foobar, not the actual tool or the related library (libsoxr). Both have been stable for years if you look at the commit history.

It is because those classic WAVES plugins work with hardware working in fixed point like Pro Tools TDM. The newer generation of plugins work in 64-bit float already.

http://avid.force.com/pkb/articles/faq/en423351
Some of them do, others still don't. The most popular WAVES resampler still operates in 48 bit integer. Which is actually fine -- there's less quantization error with that than a 32-bit float pipeline.
 
Joined
Sep 26, 2019
Messages
39
Likes
44
#17
Your same argument can be used to discredit using DACs with 21-bits of resolution because the rest of your signal chain can't achieve the same 21-bits of resolution. The OP asked which was the better volume control which appears to be what has been answered.
My point is that the article from Archimago's guest poster actually doesn't answer that question. It's just a set of non-sequiturs and contains no analysis of the quantization issue.
 

bennetng

Addicted to Fun and Learning
Joined
Nov 15, 2017
Messages
592
Likes
455
#18
That's the SOX plugin for Foobar, not the actual tool or the related library (libsoxr). Both have been stable for years if you look at the commit history.
Did you read the hydrognaudio thread carefully? The post was talking about the original command line version of SoX has clipping problem, and the foobar plugin version does not clip since it uses float. Exactly one of the advantages of floating point processing.
Some of them do, others still don't. The most popular WAVES resampler still operates in 48 bit integer. Which is actually fine -- there's less quantization error with that than a 32-bit float pipeline.
Is it so surprising that 48-bit integer has better precision than 32-bit float? Why don't talk about 64-bit float resamplers like the iZotope ones? However it is unimportant anyway since it is a thread about volume control and you are derailing it.
 

bennetng

Addicted to Fun and Learning
Joined
Nov 15, 2017
Messages
592
Likes
455
#19
My point is that the article from Archimago's guest poster actually doesn't answer that question. It's just a set of non-sequiturs and contains no analysis of the quantization issue.
Since this quantization issue is not going to affect a DAC with 21 bits of dynamic range anyway, you also mentioned the same, right?
 

Leon

Member
Joined
Mar 5, 2018
Messages
18
Likes
3
#20
If you prefer manual volume adjustment by using foobar's volume control yourself then use "prevent clipping according to peak", otherwise use "apply gain and prevent clipping according to peak".

Alter file content is only for file converter and not for playback. Ignore it.
Thanks a lot Bennetng !

With the replaygain activated following recommendation , I would like manage the volume from OPPO itself. Is it a good compromise to reduce a little bit the Foobar volume (-6db) and manage the remaining volume control with OPPO to maintain a volume level around 60/70% for a "normal" listening ?
 
Top Bottom