AVS/AIX High-Resolution Audio Test: Next Step - Page 7 - AVS Forum
Forum Jump: 
 5Likes
Reply
 
Thread Tools
post #181 of 232 Old 06-03-2014, 09:58 PM
Senior Member
 
nx211's Avatar
 
Join Date: Aug 2004
Location: New York
Posts: 297
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 1 Post(s)
Liked: 17
Quote:
Originally Posted by amirm View Post


PICT0867-L.jpg

The handsome dude on the left is me. biggrin.gif The pretty girl in the middle is the singer Jewel.

2210_0075.jpg

Yes, that is me holding an Emmy award for the accomplishments in my team. Have been fortunate enough to have received this honor three times. By the way, you have not lived until you walk around the Las Vegas casino floor with an Emmy award in your hand. Everyone, including all the people from the opposite sex which normally would not look at you, shout at you and smile as they think you are some famous actor biggrin.gif.

*NICE*, *VERY NICE* congrats man!

But, you know, after reading what you posted, the first thing that pops into the mind of a flaming heterosexual is what exactly did you do about the quote: "people from the opposite sex which normally would not look at you, shout at you and smile as they think you are some famous actor"?

nx211
nx211 is offline  
Sponsored Links
Advertisement
 
post #182 of 232 Old 06-03-2014, 10:02 PM
Advanced Member
 
RayDunzl's Avatar
 
Join Date: Sep 2013
Location: Tampa, Florida
Posts: 920
Mentioned: 1 Post(s)
Tagged: 0 Thread(s)
Quoted: 299 Post(s)
Liked: 306
Quote:
Originally Posted by nx211 View Post

Wow, I leave reading this thread for a day or so and now were talking about LP's and cassettes? WTH?

Whatever happened to the discussion about potentially *hearing*, (that would be for humans, not dogs, I really couldn't care less what the neighbors dog has to say about the Ultrasonic audio spectrum, btw), the difference between a 24 bit depth music file vs. a 16 bit depth file in the audio spectrum below 20khz, considering the 24 bit depth's ability to resolve finer gradations?

Has the discussion been relegated to the difference being *only* to an increase in dynamic range?


nx211
As I'm understanding it, any more "detail" between any two sample points of 44.1 than what can already be reconstructed would have to be due to higher (than the 22.05 nyquist) frequency, so that leaves dynamic range.

If you are thinking of amplitude error...

Amplitude resolution on a 2vpp dac output::
The value of a 16bit bit is 0.000,030,517,578,125v, take half of that, call it "rounding error" for choosing between the next higher/lower value for the dac output, at 0.000,015,258,789,0625v

For 24bit the error would be 0.000,000,059,604,644,775,390,600v

Either one of those error levels would seem (to me, the unwashed mass) to be overwhelmed by amplifier or other noise levels or imperfections long before it got to my neuron...

I have been wrong before, shields up!

I'll be back later...



To view links or images in signatures your post count must be 0 or greater. You currently have 0 posts.
> digits >
To view links or images in signatures your post count must be 0 or greater. You currently have 0 posts.
>
To view links or images in signatures your post count must be 0 or greater. You currently have 0 posts.
>
To view links or images in signatures your post count must be 0 or greater. You currently have 0 posts.
>
To view links or images in signatures your post count must be 0 or greater. You currently have 0 posts.
and
To view links or images in signatures your post count must be 0 or greater. You currently have 0 posts.
>
To view links or images in signatures your post count must be 0 or greater. You currently have 0 posts.
RayDunzl is offline  
post #183 of 232 Old 06-04-2014, 02:28 AM
Member
 
symphara's Avatar
 
Join Date: Dec 2004
Location: Switzerland
Posts: 77
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 1 Post(s)
Liked: 19
Quote:
Originally Posted by RayDunzl View Post

As I'm understanding it, any more "detail" between any two sample points of 44.1 than what can already be reconstructed would have to be due to higher (than the 22.05 nyquist) frequency, so that leaves dynamic range.

If you are thinking of amplitude error...

Amplitude resolution on a 2vpp dac output::
The value of a 16bit bit is 0.000,030,517,578,125v, take half of that, call it "rounding error" for choosing between the next higher/lower value for the dac output, at 0.000,015,258,789,0625v

For 24bit the error would be 0.000,000,059,604,644,775,390,600v

Either one of those error levels would seem (to me, the unwashed mass) to be overwhelmed by amplifier or other noise levels or imperfections long before it got to my neuron...

I have been wrong before, shields up!

Not sure if I understand you correctly, but are you saying that amplifiers cannot distinguish between more gradations than 64k (2^16)?

Anyway, the whole purpose of this thread is about the listening test, not to fruitlessly debate the theoretical points on which clearly there is no agreement.
symphara is offline  
post #184 of 232 Old 06-04-2014, 04:12 AM
AVS Addicted Member
 
arnyk's Avatar
 
Join Date: Oct 2002
Location: Grosse Pointe Woods, MI
Posts: 14,381
Mentioned: 2 Post(s)
Tagged: 0 Thread(s)
Quoted: 747 Post(s)
Liked: 1162
Quote:
Originally Posted by symphara View Post

Not sure if I understand you correctly, but are you saying that amplifiers cannot distinguish between more gradations than 64k (2^16)?

It's neck and neck between power amplifiers and 16 bit coding, with the amplifiers usually but not always having the upper hand.

However the race is settled at the weakest point in the audio reproduction chain,not one of the stronger points.

Performance spaces, microphones, speakers and listening rooms are far more likely to be the weakest links, especially the listening room.
arnyk is offline  
post #185 of 232 Old 06-04-2014, 04:12 AM
Advanced Member
 
RayDunzl's Avatar
 
Join Date: Sep 2013
Location: Tampa, Florida
Posts: 920
Mentioned: 1 Post(s)
Tagged: 0 Thread(s)
Quoted: 299 Post(s)
Liked: 306
Sorry, just thinking out loud while the experts are sleeping... Part of the discussion is about equipment performance to support the test, and other practical considerations. I'm just playing with some numbers that make sense to me while thinking about it... If we all have to be accredited degreed engineers to participate I'll say have a nice day and leave.

From spec sheets for my gear:
SNR resolution (all A-weighted)
109db = 18 bits preamp
114db = 20 bits amp
126db = 22 bits dac

http://www.maximintegrated.com/en/design/tools/calculators/product-design/data-conversion.cfm

specs don't equal reality, but its a start

I'll be back later...



To view links or images in signatures your post count must be 0 or greater. You currently have 0 posts.
> digits >
To view links or images in signatures your post count must be 0 or greater. You currently have 0 posts.
>
To view links or images in signatures your post count must be 0 or greater. You currently have 0 posts.
>
To view links or images in signatures your post count must be 0 or greater. You currently have 0 posts.
>
To view links or images in signatures your post count must be 0 or greater. You currently have 0 posts.
and
To view links or images in signatures your post count must be 0 or greater. You currently have 0 posts.
>
To view links or images in signatures your post count must be 0 or greater. You currently have 0 posts.
RayDunzl is offline  
post #186 of 232 Old 06-04-2014, 04:23 AM
AVS Addicted Member
 
arnyk's Avatar
 
Join Date: Oct 2002
Location: Grosse Pointe Woods, MI
Posts: 14,381
Mentioned: 2 Post(s)
Tagged: 0 Thread(s)
Quoted: 747 Post(s)
Liked: 1162
Quote:
Originally Posted by amirm View Post

Quote:
Originally Posted by arnyk View Post

In name.
Nope, in practice. My company does not deal in turntables or LPs.

Never said it did. It is a business with what appears to be a goodly dollop of professional talent to guide it. It is not unusual for people's businesses to run more pragmatically than their own preferences, especially if they are removed from the day-to-day operations.
Quote:
Quote:
Merits of analog based systems in 2014. Hmmm. Are there any?
Yes, they sound hugely enjoyable when done right.

They don't have to sound bad despite their technical failings if they are running at peak performance. However systems based on analog tape require a lot of maintenance to keep at peak performance.

In actual use a Otari 5050 might require the playback azimuth to be optimized for every roll of tape. Who does that for you?

Media to feed that Otari is pretty hard to find, and is subject to wear and accidental loss of high frequency sounds. Where do you get your 2 track 15 ips masters and how many do you have?

It is ironic that a Sansa Clip can outperform all that fancy, expensive, hard to maintain hardware day in, day out and run on solar power panels that I can carry with my camping gear.
Quote:
I can't believe this is news to so many folks....

It is very old news, and represents a technology that most people are glad to be free of for the exact reasons that I just mentioned.
RobertR likes this.
arnyk is offline  
post #187 of 232 Old 06-04-2014, 08:34 AM
AVS Addicted Member
 
amirm's Avatar
 
Join Date: Jan 2002
Location: Washington State
Posts: 18,085
Mentioned: 3 Post(s)
Tagged: 0 Thread(s)
Quoted: 755 Post(s)
Liked: 435
Hi guys. I am responding to the argumentative posts in the dedicated debate thread instead of here: http://www.avsforum.com/t/1532092/debate-thread-scotts-hi-res-audio-test/390#post_24791820

As a courtesy to everyone else, I hope arguments that have nothing to do with Scott's test protocol go there (or other thread of your choosing).

Amir
Founder,
To view links or images in signatures your post count must be 0 or greater. You currently have 0 posts.

"Insist on Quality Engineering"

amirm is offline  
post #188 of 232 Old 06-04-2014, 11:51 AM
AVS Special Member
 
TVOD's Avatar
 
Join Date: Jun 2003
Posts: 4,881
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 74 Post(s)
Liked: 69
Quote:
Originally Posted by amirm View Post

Hi guys. I am responding to the argumentative posts in the dedicated debate thread instead of here: http://www.avsforum.com/t/1532092/debate-thread-scotts-hi-res-audio-test/390#post_24791820

As a courtesy to everyone else, I hope arguments that have nothing to do with Scott's test protocol go there (or other thread of your choosing).
LOL, that thread is already twice the length of this one and half of this one is the same subject.


To view links or images in signatures your post count must be 0 or greater. You currently have 0 posts.

To view links or images in signatures your post count must be 0 or greater. You currently have 0 posts.
TVOD is offline  
post #189 of 232 Old 06-04-2014, 12:58 PM
AVS Special Member
 
CharlesJ's Avatar
 
Join Date: Jan 2006
Posts: 3,241
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 87 Post(s)
Liked: 94
Quote:
Originally Posted by RayDunzl View Post

.
Amplitude resolution on a 2vpp dac output::
The value of a 16bit bit is 0.000,030,517,578,125v, take half of that, call it "rounding error" for choosing between the next higher/lower value for the dac output, at 0.000,015,258,789,0625v

For 24bit the error would be 0.000,000,059,604,644,775,390,600v

...
Interesting. Just wondering if a speaker can respond to such voltage variations, not to even bring up the brain's reaction.
CharlesJ is offline  
post #190 of 232 Old 06-04-2014, 02:00 PM
AVS Special Member
 
Frank Derks's Avatar
 
Join Date: Jul 2002
Location: Region A,B,C
Posts: 1,891
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 98 Post(s)
Liked: 89
Quote:
Originally Posted by CharlesJ View Post

Interesting. Just wondering if a speaker can respond to such voltage variations, not to even bring up the brain's reaction.

Speakers can.

But the real question is can the speakers ( better yet the entire playback system ) deal with the +120dB peak levels without distortion at the required volume setting that make it possible to actually hear the digital noise floor.
Frank Derks is offline  
post #191 of 232 Old 06-04-2014, 02:11 PM
AVS Addicted Member
 
arnyk's Avatar
 
Join Date: Oct 2002
Location: Grosse Pointe Woods, MI
Posts: 14,381
Mentioned: 2 Post(s)
Tagged: 0 Thread(s)
Quoted: 747 Post(s)
Liked: 1162
Quote:
Originally Posted by CharlesJ View Post

Quote:
Originally Posted by RayDunzl View Post

.
Amplitude resolution on a 2vpp dac output::
The value of a 16bit bit is 0.000,030,517,578,125v, take half of that, call it "rounding error" for choosing between the next higher/lower value for the dac output, at 0.000,015,258,789,0625v

For 24bit the error would be 0.000,000,059,604,644,775,390,600v

...

Interesting. Just wondering if a speaker can respond to such voltage variations, not to even bring up the brain's reaction.

Speakers are essentially frictionless devices. Their motion always depends on deforming some kind of a spring. So they respond, at least just a little bit, to any kind of force no matter how small.

So yes a speaker can respond to such small voltage variations.

However that's not the weakest link in the chain. The weakest link in the part of the chain that you can control is room noise. The tiny voltage is applied, the speaker diaphragm moves, and the sound it makes is masked by room noise unless it is large enough. Room noise is generally larger than 1 LSB in a 16 bit system when the peak SPL levels are on the order of reference level (105 dB).
arnyk is offline  
post #192 of 232 Old 06-04-2014, 02:34 PM
AVS Special Member
 
Frank Derks's Avatar
 
Join Date: Jul 2002
Location: Region A,B,C
Posts: 1,891
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 98 Post(s)
Liked: 89
The static friction is overcome by the fact that due to the ever existing noise floor the speaker is always in motion.

This is in fact essentially the same as the dithering effect that the vibrating airplane had on the cogs in the mechanical instruments in the cockpit.
Frank Derks is offline  
post #193 of 232 Old 06-04-2014, 03:01 PM
Advanced Member
 
RayDunzl's Avatar
 
Join Date: Sep 2013
Location: Tampa, Florida
Posts: 920
Mentioned: 1 Post(s)
Tagged: 0 Thread(s)
Quoted: 299 Post(s)
Liked: 306
Here are some files that may be of interest to anyone listening, 19 selections in multiple formats.

The first one:




http://www.2l.no/hires/

"We invite you to join us in this evaluation of future consumer delivery formats. FLAC is a lossless encoding of WAV-files derived directly from our production original used for the SACD and Pure Audio Blu-ray. All resolutions and encodings are derived from the same original DXD source files."

I'll be back later...



To view links or images in signatures your post count must be 0 or greater. You currently have 0 posts.
> digits >
To view links or images in signatures your post count must be 0 or greater. You currently have 0 posts.
>
To view links or images in signatures your post count must be 0 or greater. You currently have 0 posts.
>
To view links or images in signatures your post count must be 0 or greater. You currently have 0 posts.
>
To view links or images in signatures your post count must be 0 or greater. You currently have 0 posts.
and
To view links or images in signatures your post count must be 0 or greater. You currently have 0 posts.
>
To view links or images in signatures your post count must be 0 or greater. You currently have 0 posts.
RayDunzl is offline  
post #194 of 232 Old 06-04-2014, 04:41 PM
AVS Special Member
 
CharlesJ's Avatar
 
Join Date: Jan 2006
Posts: 3,241
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 87 Post(s)
Liked: 94
Quote:
Originally Posted by Frank Derks View Post

Speakers can.

But the real question is can the speakers ( better yet the entire playback system ) deal with the +120dB peak levels without distortion at the required volume setting that make it possible to actually hear the digital noise floor.

Quote:
Originally Posted by arnyk View Post

Speakers are essentially frictionless devices. Their motion always depends on deforming some kind of a spring. So they respond, at least just a little bit, to any kind of force no matter how small.

So yes a speaker can respond to such small voltage variations.

However that's not the weakest link in the chain. The weakest link in the part of the chain that you can control is room noise. The tiny voltage is applied, the speaker diaphragm moves, and the sound it makes is masked by room noise unless it is large enough. Room noise is generally larger than 1 LSB in a 16 bit system when the peak SPL levels are on the order of reference level (105 dB).

Arny was closer to what I was trying to ask there with such small voltage changes. I think I was trying to wrap myself around such a small voltage change and that small change's audibility. Perhaps it needs several such change or perhaps I am just all s.rewd up.wink.gifbiggrin.gif
CharlesJ is offline  
post #195 of 232 Old 06-04-2014, 05:04 PM
AVS Addicted Member
 
amirm's Avatar
 
Join Date: Jan 2002
Location: Washington State
Posts: 18,085
Mentioned: 3 Post(s)
Tagged: 0 Thread(s)
Quoted: 755 Post(s)
Liked: 435
Quote:
Originally Posted by Frank Derks View Post

Speakers can.
For 24 bit that's impossible.

We can't capture or play anything resembling 24 bits.

Amir
Founder,
To view links or images in signatures your post count must be 0 or greater. You currently have 0 posts.

"Insist on Quality Engineering"

amirm is offline  
post #196 of 232 Old 06-04-2014, 06:18 PM
AVS Addicted Member
 
arnyk's Avatar
 
Join Date: Oct 2002
Location: Grosse Pointe Woods, MI
Posts: 14,381
Mentioned: 2 Post(s)
Tagged: 0 Thread(s)
Quoted: 747 Post(s)
Liked: 1162
Quote:
Originally Posted by CharlesJ View Post

Quote:
Originally Posted by Frank Derks View Post

Speakers can.

But the real question is can the speakers ( better yet the entire playback system ) deal with the +120dB peak levels without distortion at the required volume setting that make it possible to actually hear the digital noise floor.

Quote:
Originally Posted by arnyk View Post

Speakers are essentially frictionless devices. Their motion always depends on deforming some kind of a spring. So they respond, at least just a little bit, to any kind of force no matter how small.

So yes a speaker can respond to such small voltage variations.

However that's not the weakest link in the chain. The weakest link in the part of the chain that you can control is room noise. The tiny voltage is applied, the speaker diaphragm moves, and the sound it makes is masked by room noise unless it is large enough. Room noise is generally larger than 1 LSB in a 16 bit system when the peak SPL levels are on the order of reference level (105 dB).

Arny was closer to what I was trying to ask there with such small voltage changes. I think I was trying to wrap myself around such a small voltage change and that small change's audibility. Perhaps it needs several such change or perhaps I am just all s.rewd up.wink.gifbiggrin.gif

The idea of several changes, each one LSB high, being required in a practical application, is pretty close to the truth. Rule of thumb is that 13-14 bit resolution is all that is required for high quality audio. That's similar to saying that it takes 2 to 4 LSBs in a 16 bit system to create an audible change.
arnyk is offline  
post #197 of 232 Old 06-10-2014, 06:24 PM
AVS Special Member
 
TVOD's Avatar
 
Join Date: Jun 2003
Posts: 4,881
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 74 Post(s)
Liked: 69
So how's the grand experiment going?


To view links or images in signatures your post count must be 0 or greater. You currently have 0 posts.

To view links or images in signatures your post count must be 0 or greater. You currently have 0 posts.
TVOD is offline  
post #198 of 232 Old 06-12-2014, 04:52 AM
AVS Addicted Member
 
arnyk's Avatar
 
Join Date: Oct 2002
Location: Grosse Pointe Woods, MI
Posts: 14,381
Mentioned: 2 Post(s)
Tagged: 0 Thread(s)
Quoted: 747 Post(s)
Liked: 1162
Quote:
Originally Posted by nx211 View Post


Whatever happened to the discussion about potentially *hearing*, (that would be for humans, not dogs, I really couldn't care less what the neighbors dog has to say about the Ultrasonic audio spectrum, btw), the difference between a 24 bit depth music file vs. a 16 bit depth file in the audio spectrum below 20khz, considering the 24 bit depth's ability to resolve finer gradations?

Has the discussion been relegated to the difference being *only* to an increase in dynamic range?
Dynamic range and resolution are two different names for what amounts to being the same thing.

A common mistake is to confuse higher resolution or better dynamic range with improved sound quality. Just because the numbers seem more impressive doesn't mean that there is an audible difference, let alone an improvement.

The principle of the weakest link still applies to the audio chain. The weakest links are performance rooms, microphones, loudspeakers, and listening rooms. Not that ADCs, DACs, power amplifiers, and digital music players are not on the list. That is 100% intentional and experience-based.
arnyk is offline  
post #199 of 232 Old 06-13-2014, 12:44 PM
Advanced Member
 
RayDunzl's Avatar
 
Join Date: Sep 2013
Location: Tampa, Florida
Posts: 920
Mentioned: 1 Post(s)
Tagged: 0 Thread(s)
Quoted: 299 Post(s)
Liked: 306
Quote:
Originally Posted by arnyk View Post
Quote:Originally Posted by MSchu18 

you should change this already scott...




The picture is incredibly misleading. For example the 16 bit picture shows 15 steps when reality is more lke 65,000 steps. In fact there aren't pixels on a printed page that could show the actual fine detail. Unfortunately analog bigots and digiphobes circulate this sort of trash wityh a straignt face.

I thought about it again today, and for 16bit resoultion the picture (in terms of output voltage) looks like this. Each additional most significant bit limits the output voltage to the ranges shown, Can't show 10 bits on down they get too small. The 24 bit image would have "23" at the same (halfway) level as the '15" in this sketch, and get way more un-visible at the bottom.


I'll be back later...



To view links or images in signatures your post count must be 0 or greater. You currently have 0 posts.
> digits >
To view links or images in signatures your post count must be 0 or greater. You currently have 0 posts.
>
To view links or images in signatures your post count must be 0 or greater. You currently have 0 posts.
>
To view links or images in signatures your post count must be 0 or greater. You currently have 0 posts.
>
To view links or images in signatures your post count must be 0 or greater. You currently have 0 posts.
and
To view links or images in signatures your post count must be 0 or greater. You currently have 0 posts.
>
To view links or images in signatures your post count must be 0 or greater. You currently have 0 posts.
RayDunzl is offline  
post #200 of 232 Old 06-15-2014, 01:32 PM
Advanced Member
 
SiGGy's Avatar
 
Join Date: Jul 2004
Location: Lenexa, Kansas
Posts: 734
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 86 Post(s)
Liked: 86
This "test" doesn't require speakers which can go beyond 20k to hear the benefits of 24bit 96khz audio.

I have a recording studio and record 24bit @ 44khz, 96khz and 192khz. Even on a pair of JBL LSR4328's you can hear the difference between 44khz and 96khz @24bit on dynamic instruments like the drums, 96khz and 192khz is splitting hair and would require better speakers.

96khz @ 24bit sounds a lot more open (dynamic range) than 44khz @ 16bit does; even without hearing past 20,000hz. Just going from 16bit to 24bit is audible... The dynamic range is significant between 16bit 44khz and 24bit 96khz. You don't need to focus on frequency response 100%.

I would use a pair of Stax headphones if you really want to hear stuff that far out cleanly. Occasionally I'll listen to a mix on a pair of them...
symphara likes this.

-SiGGy

Last edited by SiGGy; 06-15-2014 at 01:51 PM.
SiGGy is offline  
post #201 of 232 Old 06-15-2014, 02:12 PM
AVS Special Member
 
mogorf's Avatar
 
Join Date: May 2009
Location: Budapest, Hungary
Posts: 4,393
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 186 Post(s)
Liked: 115
Quote:
Originally Posted by SiGGy View Post
This "test" doesn't require speakers which can go beyond 20k to hear the benefits of 24bit 96khz audio.

I have a recording studio and record 24bit @ 44khz, 96khz and 192khz. Even on a pair of JBL LSR4328's you can hear the difference between 44khz and 96khz @24bit on dynamic instruments like the drums, 96khz and 192khz is splitting hair and would require better speakers.

96khz @ 24bit sounds a lot more open (dynamic range) than 44khz @ 16bit does; even without hearing past 20,000hz. Just going from 16bit to 24bit is audible... The dynamic range is significant between 16bit 44khz and 24bit 96khz. You don't need to focus on frequency response 100%.

I would use a pair of Stax headphones if you really want to hear stuff that far out cleanly. Occasionally I'll listen to a mix on a pair of them...
Hi SiGGy,

Do you have any sample materials for download you might be willing to share with us from your studio work with different resolutions? Hope it's not too much to ask.

Cheers, Feri


To view links or images in signatures your post count must be 0 or greater. You currently have 0 posts.



To view links or images in signatures your post count must be 0 or greater. You currently have 0 posts.



To view links or images in signatures your post count must be 0 or greater. You currently have 0 posts.



To view links or images in signatures your post count must be 0 or greater. You currently have 0 posts.

 


To view links or images in signatures your post count must be 0 or greater. You currently have 0 posts.


To view links or images in signatures your post count must be 0 or greater. You currently have 0 posts.
mogorf is online now  
post #202 of 232 Old 06-15-2014, 02:29 PM
Advanced Member
 
SiGGy's Avatar
 
Join Date: Jul 2004
Location: Lenexa, Kansas
Posts: 734
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 86 Post(s)
Liked: 86
Possibly, we had water damage and had to tear studio apart; then son was born. Been slowly rebuilding, remediation and cleanup is all done.

I was actually thinking of offering this; but I don't have much time as of lately (new father/work). Not sure I can deliver what I want for awhile... Takes a long time to setup the room again for drums (sound panels, reflections...) The room is a disaster looks like a tornado went through. I haven't even unwrapped rack equipment which was wrapped with large syran wrap like stuff... JBL monitors are back in the original boxes so sad.

I was trying to think of a way of doing this without letting anyone know which file is which (and least effort on my part, lol). Difficult to do...

Start a poll... give out 2 files...

1) split mic(s) recording drums. One 16bit 44khz other 24bit 96khz. However it would be obvious which is which simply by file size.

2) split mic(s) recording drums, One 16bit 44khz other 24bit 96khz. up-convert the 16bit 44khz file to 96khz 24bit, leave other file untouched. This would make it very difficult to tell which file is which both would be 96khz 24bit.

3) No need to re-record. take an old 96khz drum session... down convert it to 44khz 16bit, then up-convert it back to 96khz 24bit. (this is the least amount of work as I don't need my drums/room/board/mics setup; but I don't think it's the best idea with all of the conversion)

Let me think about it, also poke through some files tonight. Maybe see if I can con someone into helping me set everything back up

p.s

when I mean split mic(s). I mean one set of mics over the drums where I would split the output from the mic(s) to two separate but identical A/D converters (12 channel converters). One converter sampling at 44khz 16bit, other 96khz 24bit. So one set of mics, and one recording session recording at 2 different sampling rates/bit depths.

-SiGGy

Last edited by SiGGy; 06-15-2014 at 02:43 PM.
SiGGy is offline  
post #203 of 232 Old 06-15-2014, 03:34 PM
Member
 
antoniobiz1's Avatar
 
Join Date: Sep 2007
Posts: 135
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 62 Post(s)
Liked: 68
Quote:
Originally Posted by SiGGy View Post
96khz @ 24bit sounds a lot more open (dynamic range) than 44khz @ 16bit does; even without hearing past 20,000hz. Just going from 16bit to 24bit is audible... The dynamic range is significant between 16bit 44khz and 24bit 96khz. You don't need to focus on frequency response 100%.
Maybe it is me that do not understand what dynamic range means. As I understand it, dynamic range is a ratio (loudest/faintest sound). Using logarithms, it is a difference.

If a very quiet room has a 20db SPL noise floor, and a 120db spl is the threshold of pain (and instant hearing damage can occour), how can you benefit from a higher than 96db dynamic range? Do you listen to your drums recordings at 130db SPL?

Besides, if a microphone has a typical SNR of 90 db, again, how can you appreciate an increased dynamic range, if you cannot record it?
antoniobiz1 is online now  
post #204 of 232 Old 06-15-2014, 04:20 PM
Advanced Member
 
SiGGy's Avatar
 
Join Date: Jul 2004
Location: Lenexa, Kansas
Posts: 734
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 86 Post(s)
Liked: 86
Quote:
Originally Posted by antoniobiz1 View Post
Quote:
Originally Posted by SiGGy View Post
96khz @ 24bit sounds a lot more open (dynamic range) than 44khz @ 16bit does; even without hearing past 20,000hz. Just going from 16bit to 24bit is audible... The dynamic range is significant between 16bit 44khz and 24bit 96khz. You don't need to focus on frequency response 100%.
Maybe it is me that do not understand what dynamic range means. As I understand it, dynamic range is a ratio (loudest/faintest sound). Using logarithms, it is a difference.

If a very quiet room has a 20db SPL noise floor, and a 120db spl is the threshold of pain (and instant hearing damage can occour), how can you benefit from a higher than 96db dynamic range? Do you listen to your drums recordings at 130db SPL?

Besides, if a microphone has a typical SNR of 90 db, again, how can you appreciate an increased dynamic range, if you cannot record it?
I think you are forgetting drums are usually mic'd both close and far from the instrument. Generally close mic'd for the proximity effect (bigger sounding drums). But also used for subtle details on things like cymbals or the snare and other percussion pieces. Mic'd far for adding in the room sound. With this setup your SNR is a bit more complicated than using the spec's from one microphone.

Drums by themselves can be extremely dynamic in a room. Subtle things can be 5-30db in nature, and loud sequences over 120+db (rock drumming). A good jazz drummer would be 110db of dynamic range easily.

Then there's a "sensitivity" part of the microphone outside of the SNR. If you drop your bit-depth you'll also drop some of the details some mic's can pickup; especially condensers. Or they will get smeared/lost/dithered. Good engineers use specific mics for certain things; and apply them in their sweet spots (SPL/sensitivity/distance/angle to instrument)
http://www.analog.com/library/analog...nsitivity.html

There's so many different techniques to recording it's really hard to look at it one way... For example rock drums you might crank up the gain on the pre-amps and drive the transformers into saturation. This will add harmonics ("color") into the sound and make it sound thicker but also lower the SNR a bit (this is why some boards/preamps are known for "that sound"). But then in a jazz recording you want the extra clean headroom (dynamic range) so you can catch both the loud and quiet sequences without running out of headroom or going into distortion. You won't want to lower your mic gain just so you don't clip a loud snare hit, when you need the subtle parts too. In this case you want as much dynamic range as possible.

If you have a LOT of subtle high detail audio samples coming in from multiple mics you are going to want as much resolution as possible on both the X (sampling freq) and Y (bit depth) axis. So it's a bit more than just simply "dynamic range" and "frequency response". Add in sensitivity, transients... sub harmonics... psychoacoustics. Then even what volume you are listening to it back at... our ears hear differently depending on the volume.

-SiGGy

Last edited by SiGGy; 06-15-2014 at 04:58 PM.
SiGGy is offline  
post #205 of 232 Old 06-16-2014, 04:04 AM
AVS Special Member
 
hernanu's Avatar
 
Join Date: Apr 2007
Location: Boston Suburbs
Posts: 2,511
Mentioned: 1 Post(s)
Tagged: 0 Thread(s)
Quoted: 60 Post(s)
Liked: 155
Quote:
Originally Posted by SiGGy View Post
Quote:
Originally Posted by antoniobiz1 View Post
Quote:
Originally Posted by SiGGy View Post
96khz @ 24bit sounds a lot more open (dynamic range) than 44khz @ 16bit does; even without hearing past 20,000hz. Just going from 16bit to 24bit is audible... The dynamic range is significant between 16bit 44khz and 24bit 96khz. You don't need to focus on frequency response 100%.
Maybe it is me that do not understand what dynamic range means. As I understand it, dynamic range is a ratio (loudest/faintest sound). Using logarithms, it is a difference.

If a very quiet room has a 20db SPL noise floor, and a 120db spl is the threshold of pain (and instant hearing damage can occour), how can you benefit from a higher than 96db dynamic range? Do you listen to your drums recordings at 130db SPL?

Besides, if a microphone has a typical SNR of 90 db, again, how can you appreciate an increased dynamic range, if you cannot record it?
I think you are forgetting drums are usually mic'd both close and far from the instrument. Generally close mic'd for the proximity effect (bigger sounding drums). But also used for subtle details on things like cymbals or the snare and other percussion pieces. Mic'd far for adding in the room sound. With this setup your SNR is a bit more complicated than using the spec's from one microphone.

Drums by themselves can be extremely dynamic in a room. Subtle things can be 5-30db in nature, and loud sequences over 120+db (rock drumming). A good jazz drummer would be 110db of dynamic range easily.

Then there's a "sensitivity" part of the microphone outside of the SNR. If you drop your bit-depth you'll also drop some of the details some mic's can pickup; especially condensers. Or they will get smeared/lost/dithered. Good engineers use specific mics for certain things; and apply them in their sweet spots (SPL/sensitivity/distance/angle to instrument)
http://www.analog.com/library/analog...nsitivity.html

There's so many different techniques to recording it's really hard to look at it one way... For example rock drums you might crank up the gain on the pre-amps and drive the transformers into saturation. This will add harmonics ("color") into the sound and make it sound thicker but also lower the SNR a bit (this is why some boards/preamps are known for "that sound"). But then in a jazz recording you want the extra clean headroom (dynamic range) so you can catch both the loud and quiet sequences without running out of headroom or going into distortion. You won't want to lower your mic gain just so you don't clip a loud snare hit, when you need the subtle parts too. In this case you want as much dynamic range as possible.

If you have a LOT of subtle high detail audio samples coming in from multiple mics you are going to want as much resolution as possible on both the X (sampling freq) and Y (bit depth) axis. So it's a bit more than just simply "dynamic range" and "frequency response". Add in sensitivity, transients... sub harmonics... psychoacoustics. Then even what volume you are listening to it back at... our ears hear differently depending on the volume.
Thanks for the great explanation.

I did some consulting work on capturing extremely high speed transient signals (the signature of the analog conduction as a fuse blows), and I can attest that having a high sample frequency and bit size is crucial for translating analog signals to digital with high fidelity.

I'm not interested in the signals that go far beyond hearing range as much as making sure what is in the hearing range is represented the best way possible. Missing nuances in a great recording would be a crime (at least a misdemeanor) in my opinion.
hernanu is offline  
post #206 of 232 Old 06-16-2014, 07:44 AM
AVS Addicted Member
 
amirm's Avatar
 
Join Date: Jan 2002
Location: Washington State
Posts: 18,085
Mentioned: 3 Post(s)
Tagged: 0 Thread(s)
Quoted: 755 Post(s)
Liked: 435
Quote:
Originally Posted by antoniobiz1 View Post
If a very quiet room has a 20db SPL noise floor, and a 120db spl is the threshold of pain (and instant hearing damage can occour), how can you benefit from a higher than 96db dynamic range? Do you listen to your drums recordings at 130db SPL?
A quick note on room SPL. Those single digit numbers are of no value in determining audibility. For that, you need the spectrum of that noise and then compare it to our hearing thresholds. I wrote an article on true dynamic range of rooms which explains this. Here is a referenced graph showing spectrum of admittedly very quiet rooms:



Notice how the room noise dips down to threshold of hearing which is actually negative SPL (!) while the same room has 40 db of noise at low frequencies. It is a fortunate thing that even though low frequency noise is very hard to block, it is also landing where our hearing sensitivity is the lowest.

Here is another graph referenced from my article on spectrum of home listening spaces:



We see that the best room ("min") has noise that is well below threshold of hearing at 1-3 KHz. Even the "average" of the rooms has a noise level of 5 dbSPL at 6 Khz.

Put more simply, we need the resolution and dynamic range to represent the spectrum of frequencies that we hear the best. It doesn't matter if we don't need it in low frequencies or high where our dynamic range is lower.

There are also other factors that invalidate using room noise as the floor. Listening tests show that if you have constant omnidirectional noise in the room as would be the case in ambient room noise you can detect noise coming out of a specific source, i.e one of your speakers. The cognitive part of hearing determines that the constant noise is not of significant value and attenuates it. But noise coming from a specific direction is detected. Think of evolutionary need to hear the noise from a threat in presence of environmental noise.

Also, we are able to hear "through the noise." So even if the noise floor was this high, it would not cap what real tones/music we could hear below it.

So all in all, this is much more complicated than taking single figure numbers for room noise and using that to determine our dynamic range. We can show that our hearing has 120 db SPL of dynamic range which represents 20 bits of resolution.

Amir
Founder,
To view links or images in signatures your post count must be 0 or greater. You currently have 0 posts.

"Insist on Quality Engineering"

amirm is offline  
post #207 of 232 Old 06-16-2014, 12:26 PM
AVS Addicted Member
 
arnyk's Avatar
 
Join Date: Oct 2002
Location: Grosse Pointe Woods, MI
Posts: 14,381
Mentioned: 2 Post(s)
Tagged: 0 Thread(s)
Quoted: 747 Post(s)
Liked: 1162
Quote:
Originally Posted by SiGGy View Post
This "test" doesn't require speakers which can go beyond 20k to hear the benefits of 24bit 96khz audio.

I have a recording studio and record 24bit @ 44khz, 96khz and 192khz. Even on a pair of JBL LSR4328's you can hear the difference between 44khz and 96khz @24bit on dynamic instruments like the drums, 96khz and 192khz is splitting hair and would require better speakers.
I can hear the "difference" any time I want to - I just have to exclude level-matching, time synching, and bias controls from the listening evaluation.

Quote:
96khz @ 24bit sounds a lot more open (dynamic range) than 44khz @ 16bit does; even without hearing past 20,000hz. Just going from 16bit to 24bit is audible... The dynamic range is significant between 16bit 44khz and 24bit 96khz. You don't need to focus on frequency response 100%.
I can hear those "differences" any time I want to - I just have to exclude level-matching, time synching, and bias controls from the listening evaluation.

Many people have done these kinds of evaluations the right way, which is to include level-matching, time synching, and bias controls into the listening evaluation. Do that and all sorts of imagainary differences simply go away.

Ever try doing proper listening tests? It's easy, just use the software ABX comparator that is an option for the FOOBAR2000 music player (freeware).
arnyk is offline  
post #208 of 232 Old 06-16-2014, 01:07 PM
Advanced Member
 
SiGGy's Avatar
 
Join Date: Jul 2004
Location: Lenexa, Kansas
Posts: 734
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 86 Post(s)
Liked: 86
Quote:
Originally Posted by arnyk View Post
Quote:
Originally Posted by SiGGy View Post
This "test" doesn't require speakers which can go beyond 20k to hear the benefits of 24bit 96khz audio.

I have a recording studio and record 24bit @ 44khz, 96khz and 192khz. Even on a pair of JBL LSR4328's you can hear the difference between 44khz and 96khz @24bit on dynamic instruments like the drums, 96khz and 192khz is splitting hair and would require better speakers.
I can hear the "difference" any time I want to - I just have to exclude level-matching, time synching, and bias controls from the listening evaluation.

Quote:
96khz @ 24bit sounds a lot more open (dynamic range) than 44khz @ 16bit does; even without hearing past 20,000hz. Just going from 16bit to 24bit is audible... The dynamic range is significant between 16bit 44khz and 24bit 96khz. You don't need to focus on frequency response 100%.
I can hear those "differences" any time I want to - I just have to exclude level-matching, time synching, and bias controls from the listening evaluation.

Many people have done these kinds of evaluations the right way, which is to include level-matching, time synching, and bias controls into the listening evaluation. Do that and all sorts of imagainary differences simply go away.

Ever try doing proper listening tests? It's easy, just use the software ABX comparator that is an option for the FOOBAR2000 music player (freeware).
Yes, I have used ABX comparator and that player before.

I have also done my homework and time matched, level matched the files then put it through this app as well.

http://www.libinst.com/Audio%20DiffMaker.htm

There are many ways to do a comparison. Just because ABX made some software/hardware doesn't mean is the only/proper way to do it. Your tone/attitude is kinda off putting here.

The calibrated level of the monitors being used can also hide or enhance any differences. usually best to have the artists listen for differences IMO. Most casual listeners are not that in tune with how instruments should sound. Especially solo'd.
amirm likes this.

-SiGGy
SiGGy is offline  
post #209 of 232 Old 06-16-2014, 02:57 PM
Member
 
mjd420nova's Avatar
 
Join Date: Aug 2012
Posts: 161
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 0 Post(s)
Liked: 20
Back in the mid 70's, I experimented in a standards lab to select an adequate new device being brought to market and it was an analog to digital converter. Pretty crude back then as only a four bit processor was being used to collect that data and store it. It was an idea to try and use it with strain/stress detectors but had next to nothing for bandwidth. While working with triangle, sine and square waves I noticed a slight shift in the freqencies being recreated by an accompanying digital to analog converter. The square wave and the triangle were most affected because they are ripe in harmonics, one the odd and the other even. The idea was dropped. Ten years later and things are much different. Before closing the lab from economic pressure I ran some more exacting tests using pure sine waves, three at a time and found the harmonics and beat notes shifted. I thought maybe an up and down shift was normal but it was always one or two cycles low per KC and harmonic/beat notes were lower still. The music we listen to is seldom fewer than three audio sources of dozens of notes at once. Any tones or notes missed are gone and the DAC can't replace what's not in the digital content. Equalization can go a long way to fool the ears by filling ranges missed with more present content, still can't replace the missing content. The technology is coming back full stream again with new vinyl pressing plants opening every year and the mfgrs developing new cartridges and needles. Research is also being done to try using a laser to read the grooves on analog vinyl without touching it.
mjd420nova is offline  
post #210 of 232 Old 06-16-2014, 03:05 PM
AVS Special Member
 
mcnarus's Avatar
 
Join Date: Nov 2007
Posts: 6,120
Mentioned: 1 Post(s)
Tagged: 0 Thread(s)
Quoted: 92 Post(s)
Liked: 303
Quote:
The music we listen to is seldom fewer than three audio sources of dozens of notes at once.
Perhaps you ought to think for a moment about how all those notes are captured by a microphone diaphragm and travel down a microphone cable.

Long before it hits the ADC, the "music" is just a single sinusoid.

(By the way, has anyone at AVS noticed that the quote function doesn't really work anymore?)

If you can't explain how it works, you can't say it doesn't.—The High-End Creed

mcnarus is offline  
Reply Audio theory, Setup and Chat

User Tag List

Thread Tools
Show Printable Version Show Printable Version
Email this Page Email this Page


Forum Jump: 

Posting Rules  
You may not post new threads
You may not post replies
You may not post attachments
You may not edit your posts

BB code is On
Smilies are On
[IMG] code is On
HTML code is Off
Trackbacks are Off
Pingbacks are Off
Refbacks are Off