McGill University in Montreal running tests on Digital Resolution Perception

0 Members and 1 Guest are viewing this topic. Read 10867 times.

Robert D

James
Claude from Sons Ideal had sold a couple of pairs of B&W 800D's to McGill University
Not to long ago

Robert

Levi, I'm sure james will find out the Scoop
on The other Equipment

He told me B&W 801's

james
I was close Robert

James Tanner

  • Facilitator
  • Posts: 20472
  • The Demo is Everything!
    • http://www.bryston.com

werd

HI James

Did you have the chance to talk to him about his studies? If so anything interesting involving the application of digital gear in his studies. I would like to hear what he says about high end gear compared to modest box store gear. I wonder if he has any interesting opinions on it.

James Tanner

  • Facilitator
  • Posts: 20472
  • The Demo is Everything!
    • http://www.bryston.com
HI James

Did you have the chance to talk to him about his studies? If so anything interesting involving the application of digital gear in his studies. I would like to hear what he says about high end gear compared to modest box store gear. I wonder if he has any interesting opinions on it.

Hi werd,

The purpose of this particular study is to assertain whether experienced engineers and listened can hear the difference as the sampling and bit rate is increase. Daniel told me that he should have preliminary findings available in the next month or so.  There is a lot of data to sift through apparently.

It should be very interesting as I think many of us fall under what I call the "seduction of expectations" syndrome when it comes to assessing audio.

james

etcarroll

  • Full Member
  • Posts: 3576
  • SE PA. by way of Boston, MA.
Any update to this?

I searched the professor's web page, but nothing stood out.

James Tanner

  • Facilitator
  • Posts: 20472
  • The Demo is Everything!
    • http://www.bryston.com
Any update to this?

I searched the professor's web page, but nothing stood out.

Hi ET,

About a month ago he said he was working on it and would send me some preliminary findings soon.

james

Freo-1

Hi ET,

About a month ago he said he was working on it and would send me some preliminary findings soon.

james

Hi James,

Could not resist chiming in here:
I have a pair of Sony STR-DA7100ES receivers that I run bi-amped.  The inputs are either HDMI, i-link, or coaxil digital.  When using the i-link or HDMI, it's pretty easy to hear the difference when fed with a hi-res digital input.  The amp here IS the D/A converter.  Would you not reckon that this type of setup could reveal more differences than one where a D/A conversion process occurs prior to the output?

JRace

  • Full Member
  • Posts: 610
  • Greetings one and Everyone!
Hi James,

Could not resist chiming in here:
I have a pair of Sony STR-DA7100ES receivers that I run bi-amped.  The inputs are either HDMI, i-link, or coaxil digital.  When using the i-link or HDMI, it's pretty easy to hear the difference when fed with a hi-res digital input.  The amp here IS the D/A converter.  Would you not reckon that this type of setup could reveal more differences than one where a D/A conversion process occurs prior to the output?
In your example the differences you are hearing are due to the DAC. In the test noted by James, it is not about the DAC being audible, rather the sample rate itself. For this you want to remove as many variables as possible.

Freo-1

In your example the differences you are hearing are due to the DAC. In the test noted by James, it is not about the DAC being audible, rather the sample rate itself. For this you want to remove as many variables as possible.

Right.  That is EXACTLY why I referred to a unit like the 7100ES.  All inputs are sent digitally to the output amp, which acts as the D/A.  Cant' remove any more variables than that setup.  :D

I've tried using an SACD and CD version of the same recording with this setup, and it' pretty easy to hear the deltas.

Napalm

Could it be unveiled what he used as signal source(s)?

Nap.

Napalm

Hi ET,

About a month ago he said he was working on it and would send me some preliminary findings soon.

james

Hi James,

Any news on this?

Thanks,
Nap

James Tanner

  • Facilitator
  • Posts: 20472
  • The Demo is Everything!
    • http://www.bryston.com
Hi James,

Any news on this?

Thanks,
Nap

Yes but I am sworn to silence until published. :D

james

werd

Yes but I am sworn to silence until published. :D

james

Nothing that a bit of waterboarding can't get out of ya...  :lol:

James Tanner

  • Facilitator
  • Posts: 20472
  • The Demo is Everything!
    • http://www.bryston.com
Here's something I can share which I received recently from the professor running the tests:

"James,
 
I'm working on a report that will take me some time to complete but I did want to give you a heads up.  Preliminary data show that, at 192K, the Bryston DAC outperforms the (XXXXXXX-ED) and the (XXXXXXXX-ED) DAC’s, We're continuing to test a lot of different configurations and conditions so that we have a really solid story before we go public.  It could, realistically, be a year or so.
 
Running those three DACs in their highest mode, we found that listeners were better able to detect a soft series of signals in the Bryston BDA-1 DAC, meaning that it had a lower noise floor and offered more clarity.  In that particular shoot-out, we only used the DACs in their 192K mode.

More details to follow, but if I were you, I'd be feeling pretty proud of myself right now for having designed something so transparent."

Dan

Speedskater

  • Full Member
  • Posts: 2680
  • Kevin
A very well known authority sometimes writes about how difficult it is to do this type of test. Even when doing null system test (same source, as a setup test) if all the variables are not perfectly matched then it's easy to tell A from B.
If a difference of 0.1dB or a 100 microsecond alignment error then the test setup is incorrect.

James Tanner

  • Facilitator
  • Posts: 20472
  • The Demo is Everything!
    • http://www.bryston.com
A very well known authority sometimes writes about how difficult it is to do this type of test. Even when doing null system test (same source, as a setup test) if all the variables are not perfectly matched then it's easy to tell A from B.
If a difference of 0.1dB or a 100 microsecond alignment error then the test setup is incorrect.

From what I understand this is a very well thoughtout and scientifically executed test. I give them high marks for attempting it and I do not think we can simply discount it.

james

Napalm

Thanks James for sharing. I already knew that the BDA-1 was the best deal in town (my ears told me so....) I was more interested in the adequacy of 16/44.1 signal for general music reproduction. If you have anything that you can share on this subject I'm all ears.

Nap.  :thumb:

James Tanner

  • Facilitator
  • Posts: 20472
  • The Demo is Everything!
    • http://www.bryston.com
Thanks James for sharing. I already knew that the BDA-1 was the best deal in town (my ears told me so....) I was more interested in the adequacy of 16/44.1 signal for general music reproduction. If you have anything that you can share on this subject I'm all ears.

Nap.  :thumb:


Hi Nap,

If you can do a good job at 192K/24bit, where the DAC's and circuitry are stressed to their performance levels, then 44.1/16bit is a walk in the park.

james

Napalm

Hi James,

Based on your knowledge at this moment, would you recommend that we listen to CDs with upsampling engaged or not?

Nap.  :thumb:

James Tanner

  • Facilitator
  • Posts: 20472
  • The Demo is Everything!
    • http://www.bryston.com
Hi James,

Based on your knowledge at this moment, would you recommend that we listen to CDs with upsampling engaged or not?

Nap.  :thumb:

Good question - we use 'synchronous’ up-sampling in the BDA-1 so there is not a lot of ‘math'; involved in the up-sample process. The feedback I get is about 50/50 on this. Some feel the up-sampling adds more ‘air’ and slightly better ‘bass’. Others feel that the ‘native’ file should not be screwed with and prefer no up-sampling.

James