AES Paper Presentation on DAC Measurements (Video)

[amirm]

Please note that this is NOT a presentation that replaces the paper. It was meant to hit on high points here and there. Some of the things you mention are covered in the paper.

 

[PlasticDoc]

The abstract was at the link posted in the OP: https://www.aes.org/e-lib/browse.cfm?elib=22278

Your video is much better. But I still think that the main ideas on the video should be printed in a simple pdf.
I really appreciate and value your work in bringing objectivity back to consumer audio.

 

[Multicore]

450 DACs in 6 years is a notable accomplishment that the AES paper and presentation confirms. This landmark provides a nice moment to ask: What is your ambition with the project going forwards, @amirm ?

 

[jcarys]

I would hope that your results would push the AVR manufacturers to at least improve their DAC (and pre-amp) performance. It's clear that could be done with very little money. Improving their amps would cost more, but at least we have an easy upgrade path with external amps if needed. That said, I'm not holding my breath for big improvements in that category in the near future.

 

[Fidji]

Please note that this is NOT a presentation that replaces the paper. It was meant to hit on high points here and there. Some of the things you mention are covered in the paper.

Curious, what was the response. Aren’t lot of AES members in the high-end industry?

 

[amirm]

Curious, what was the response. Aren’t lot of AES members in the high-end industry?

They are mostly into Pro audio, not high-end audio. The feedback was universally very positive. First person asking a question started with, "thank you so much for the service you provide to the industry" or something like that before asking his question. Unfortunately my session was on the first day, first thing in the morning. No signs were put on the rooms as to what talk is being presented either. So the number of people who came was not huge. I think over 40 people had registered but probably 20 showed up. Most busy sessions probably had 40 to 50 people in them.

My presentation was also cut off around slide 5 by the moderator who did not communicated with me that he wanted to allow 5 minute for QA so I only had 10 minutes instead of 15 (AES papers chair kindly reached out and said that was a mistake and they would sort it out next year). So what I presented here in the video is much more complete than what I managed to do at the conference.

The people I knew were very nice and complimentary. There was zero pushback or anything remotely negative.

 

[Multicore]

The part of the presentation that I liked best was the annotations on the SINAD comparison bar chart starting 9:00.

In particular it's good to be reminded that well over half the DACs do better than a theoretically perfect CD player with a perfect CD in it. That's a useful perspective.

 

[amirm]

450 DACs in 6 years is a notable accomplishment that the AES paper and presentation confirms. This landmark provides a nice moment to ask: What is your ambition with the project going forwards, @amirm ?

We have already expanded the scope heavily to include every category of audio sans Subwoofers that someone mentioned. In an ideal world, we would have our own audio conference focused on science and engineering instead of high-end, subjectivist audio that is going on now. It would also be nice to increase the number of testers beyond myself. I don't yet know how to accomplish either one of these goals but feel that we need to.

 

[Pe8er]

On that front, I have been remiss in not giving major credit to member @Pe8er. He provided incredible amount of editorial/review help going way beyond anything I could ever ask someone to do. I would have surely failed to produce a good paper without him!

Thank you, it was a lot of fun! Let's do it again next year

 

[K man]

My favorite takeaway here was the price performance correlation of 0.09… I was actually laughing when you showed that figure. It’s pretty insane.

 

[Multicore]

My favorite takeaway here was the price performance correlation of 0.09… I was actually laughing when you showed that figure. It’s pretty insane.

It's about the same with wine.

 

[sgent]

We have already expanded the scope heavily to include every category of audio sans Subwoofers that someone mentioned. In an ideal world, we would have our own audio conference focused on science and engineering instead of high-end, subjectivist audio that is going on now. It would also be nice to increase the number of testers beyond myself. I don't yet know how to accomplish either one of these goals but feel that we need to.

You ought to reach out to any local Universities. It maybe difficult without funding, but 4 publications in AES isn't bad for an aspiring Ph.D. or someone looking to get into grad school. Often they can bring their departmental resources to bear so journal editors, APA / Klippel / anechoic chambers, etc. Now with this internet thing you can span it across the country since audio engineering isn't taught at every school.

 

[JensH]

Thanks for the AES paper and the video. Good to get some additional information on exactly how the measurements are done and how the numbers are calculated.
One minor detail has slipped though your review process: on page 4 you mention "DACs that mute at 0 dBFS". Some DACs mute when the input is all zeroes, but that is slightly different from 0 dBFS.
I think it would be useful if you could show the output level and basic performance (Dashboard) also at 0 dBFS (in addition to the 2/4 V results).

 

[DualTriode]

I pay my dues and read many AES papers.

Amirm, nice to see you participating and presenting papers.

A couple of days after reading your paper I am still scratching my head wondering your intention in using the correlation coefficient between SINAD and price. Sinad is logarithmic, a measure that compresses the quality scale.

I see increasing the quality compared to price as a steep hill to climb. It seems that there is very much a relationship between the two. (exceptions noted)

The private road that goes up the hill to my little subdivision goes up hill at the rate of 0.09 feet per foot in a couple of places. That does not seem like a large number but if you do not have 4-wheel drive your kids aren't going to school if there is only a little bit of snow and ice.

Thanks DT

 

[amirm]

One minor detail has slipped though your review process: on page 4 you mention "DACs that mute at 0 dBFS". Some DACs mute when the input is all zeroes, but that is slightly different from 0 dBFS.

Oh no! Typed what I was thinking instead of what is correct. Oh well.

 

[amirm]

I think it would be useful if you could show the output level and basic performance (Dashboard) also at 0 dBFS (in addition to the 2/4 V results).

The input value is always 0dBFS unless noted otherwise (which occurs very rarely). At times I adjust the volume control on the DAC if the voltage is too high and again, note that in the dashboard. If that is the case, I often show a sweep up to 0 dBFS to see the effect.

 

[JensH]

If the volume control is digital the internal converter part of the DAC will then see a level lower than 0 dBFS. With some DACs this will have an influence on the distortion.

 

[Hipster Doofus]

On that front, I have been remiss in not giving major credit to member @Pe8er. He provided incredible amount of editorial/review help going way beyond anything I could ever ask someone to do. I would have surely failed to produce a good paper without him!

A true labor of love, thanks for clearing the audio landscape as only an Audio Gardener can do…
to quote Pythagoras…”Every triangle is a love triangle …when you love triangles”

 

[Eddy]

If the volume control is digital the internal converter part of the DAC will then see a level lower than 0 dBFS. With some DACs this will have an influence on the distortion.

I'm very interested in that. I have a Loxjie D30 DAC that I use as a preamp for my AUDIOPHONICS AP300-S250NC power amplifier. On speakers with normal efficiency I have to set the volume control to around -50dB. I wonder if I'm creating distortions by doing this?

 

[Head_Unit]

Not everyone can watch a 32 min video? Some people don't have the time.
Not everyone is a visual learner? Some people like text instead.

Amen, and amen. I can read WAY WAY WAY faster than listening to a video.