AVS › AVS Forum › Audio › Receivers, Amps, and Processors › ** OFFICIAL ** -- Sound Quality (SQ) Discussion
New Posts  All Forums:Forum Nav:

** OFFICIAL ** -- Sound Quality (SQ) Discussion - Page 6

post #151 of 263
3/8/11 at 1:42pm
Quote:
Originally Posted by DenPureSound View Post
Was listening to Starship - No Protection off a bought CD, and was wondering what Bit Rate are Compact Disc Digital Audio CD's recorded at on the CD?

Does sound better than streaming Napster in at 128Kbps.
Now your'e talking. Now you will be spinning CDs like crazy. I buy alot of my CDs used which is a great way to save money and try new music/artists.

Bill
Reply
Reply

AVS Top Picks

post #152 of 263
3/8/11 at 1:51pm
Quote:
Originally Posted by DenPureSound View Post
Was listening to Starship - No Protection off a bought CD, and was wondering what Bit Rate are Compact Disc Digital Audio CD's recorded at on the CD?

Does sound better than streaming Napster in at 128Kbps.
CDs are 16 bits, 44.1 KHZ sampling rate.
Reply
Reply
post #153 of 263
3/8/11 at 2:01pm
I used to be able to tell the difference between a WAV file and 128kbs MP3 file 100% of the time.

Listen for distortion in the higher frequencies such as cymbals. Have a friend switch between the WAV file (lossless) and the same file at different data rates. 128 should be obvious to most people I think, but not dramatically obvious. 160 kbs gets trickier to tell apart from the WAV file as I recall. 64 kbs sounds like AM radio, it's obviously worse.

Note that the rate is a stereo rate. So 128 kbs is 64 kbs / channel.
Reply
Reply
post #154 of 263
3/8/11 at 3:46pm
Quote:
CDs are 16 bits, 44.1 KHZ sampling rate.
This corresponds to a bit rate of

16 bits per channel x 2 channels x 44,100 samples per second = 1,411,200 bps

Much higher than 128,000 bps so there is a lot of data lost in that compression, and a lot of Audiophiles even regard the CD sample rate and word length as too low.

Most CD's would actually be recorded at higher rates (I think at least 96 kHz and 24 bits), but they must conform to the redbook convention (16/44.1) to be released as CD's in the market.
Reply
Reply
post #155 of 263
3/8/11 at 5:23pm
Thread Starter 
Quote:
Originally Posted by Bill Mac View Post

Mike,

I do not think you would starting a flame here as your advice is very sound (pardon the obvious pun). I have put off all thoughts of any component upgrades till I come up with a plan for some acoustical treatments. Going the DIY route will probably be the only way I can go right now as the savings is quite substantial. I have just started using REW and I just have the basics down for now.

Bill

Bill and Mike, this is the next area to research and learn about for me - Room Treatments/Acoustics. Only takes time, and looks like costs are not to large, so that is good to know.

BTW -- actually put in a CD that I bought some years ago and darn it sounded great here. Could I really hear the difference vs. streaming in Napster or Pandora, maybe a little bit, but not a significant difference as it seems to me, just seemed a little more Crisp.

Now after SteveD61's Post, hell I must be deaf then, as he is showing over 1.411Mbps vs. streaming at 128Kbps. Now, that's one hell of a difference for sure, like 11X if my math is right.

Thanks Guys,

DPS
Den
Reply
Reply
post #156 of 263
3/8/11 at 5:26pm
Thread Starter 
Quote:
Originally Posted by SteveD61 View Post

This corresponds to a bit rate of

16 bits per channel x 2 channels x 44,100 samples per second = 1,411,200 bps

Much higher than 128,000 bps so there is a lot of data lost in that compression, and a lot of Audiophiles even regard the CD sample rate and word length as too low.

Most CD's would actually be recorded at higher rates (I think at least 96 kHz and 24 bits), but they must conform to the redbook convention (16/44.1) to be released as CD's in the market.

Hey Steve,

Hey MATE, Very nice writeup from our good friends Down and Under. Thank you for the technical explanation here.

DPS
Den
Reply
Reply
post #157 of 263
3/8/11 at 5:33pm
Quote:
Originally Posted by SteveD61 View Post

This corresponds to a bit rate of

16 bits per channel x 2 channels x 44,100 samples per second = 1,411,200 bps

Much higher than 128,000 bps so there is a lot of data lost in that compression, and a lot of Audiophiles even regard the CD sample rate and word length as too low.

Most CD's would actually be recorded at higher rates (I think at least 96 kHz and 24 bits), but they must conform to the redbook convention (16/44.1) to be released as CD's in the market.

Not 100% fair comparison, as you can compress losslessly and drop the rate quite a bit.

So say CD is 1400 kbs. And FLAC is 50% (forget the average.) That's 700 kbs.

So some of that difference is from good compression techniques, and some is due to throwing away information.
Reply
Reply
post #158 of 263
3/8/11 at 5:35pm
http://www.noiseaddicts.com/2009/03/...-test-128-320/

Hardly a scientific test...but the results of this sample (over 2000 tries) has yielded interesting results. AES ran a blind test where the participants couldn't tell the difference between DSD/DVD-A and standard redbook. I'm not sure if that proves there are no differences or that blind tests are flawed. The AES has acknowledged or asked if the standardized way of testing audio should be treated like other sensory testing like taste....

Opinions are...opinions...nothing more. Those who ask for a test result have spent far more hours typing as a warrior and cutting and pasting and reading then actually getting behind a closed door and participating. The post counts are an indication of that. Placebo exists...nobody can argue that.
Reply
Reply
post #159 of 263
3/8/11 at 5:38pm
Thread Starter 
Quote:
Originally Posted by Bill Mac View Post

Now your'e talking. Now you will be spinning CDs like crazy. I buy alot of my CDs used which is a great way to save money and try new music/artists.

Bill

Bill Mac, do you have a great low-cost CD online store for those used ones? Might want to take a look at a couple of them. Alot of the used CD's have scratches all over them, I might be concerned in their Quality. What has been your experience on used CD's?

Thanks,
DPS
Reply
Reply
post #160 of 263
3/8/11 at 8:49pm
Quote:


Not 100% fair comparison, as you can compress losslessly and drop the rate quite a bit.

So say CD is 1400 kbs. And FLAC is 50% (forget the average.) That's 700 kbs.

So some of that difference is from good compression techniques, and some is due to throwing away information.

That's true, I didn't want to go too far down this rabbit hole in my previous post.

FLAC and MLP are fantastic compression methods - achieving a 50% reduction in file size without loss is a wonder of modern maths. It's allowing large 96/24 files to be downloaded (legally of course using HD tracks) that would have been a couple of hundred MB each.
Reply
Reply
post #161 of 263
3/8/11 at 9:33pm
I love compression algorithms. One of the more fascinating things in computer science. It feels like magic, then when you study it, it makes sense.

Data is kind of like styrofoam. Pretty solid, but you can compress the heck out of it, because it has holes. Files are like that, they have holes. Text for example, being stored in 8 bits is pretty inefficient. 8 bits allows for a value from 0 to 255. Text is mostly a small subset of 256 values so it's inefficient, hence compression routines which can restructure the data to more efficiently use the space...but I digress.
Reply
Reply
post #162 of 263
3/9/11 at 12:40am
Quote:
Originally Posted by DenPureSound View Post

Bill Mac, do you have a great low-cost CD online store for those used ones? Might want to take a look at a couple of them. Alot of the used CD's have scratches all over them, I might be concerned in their Quality. What has been your experience on used CD's?

Thanks,
DPS

DPS,

There is Bullmoose Music which is a great store and I believe they ship as well. Just search for your favorite artist and it will show the used CDs in with the new. I shop at Bullmoose and Newbury Comics for used CDs. Although Newbury Comics does not show their used inventory on their site. I have had maybe one or two bad CDs but both stores guarantee their used CDs. You should check local to you as there might be stores in your area that sell used.

http://www.bullmoose.com/rel/v2_home...r=258&deptnr=1

Bill
Reply
Reply
post #163 of 263
3/9/11 at 6:35am
Rooms - "Don't get no respect"
Quote:


Audiophiles as a collective, have steadfastly refused to acknowledge that the most important component in the playback chain is the listening room itself. It's more appealing, perhaps, to purchase the latest box of gleaming gold and silver, or towering sculpture of wood and lacquer. We must recognize, however, that no component has as much impact on the performance of an audio system as the room in which it is installed. Quite simply, one will never obtain the performance they paid dearly for if this component is not paid due respect.

Jon T. Gale
Audiophilia Online Magazine
Reply
Reply
post #164 of 263
3/9/11 at 10:49am
Thread Starter 
Quote:
Originally Posted by MichaelJHuman View Post

I love compression algorithms. One of the more fascinating things in computer science. It feels like magic, then when you study it, it makes sense.

Data is kind of like styrofoam. Pretty solid, but you can compress the heck out of it, because it has holes. Files are like that, they have holes. Text for example, being stored in 8 bits is pretty inefficient. 8 bits allows for a value from 0 to 255. Text is mostly a small subset of 256 values so it's inefficient, hence compression routines which can restructure the data to more efficiently use the space...but I digress.

Still a lot of text is 7 bits (ASCII) not the extended version. Save a bit here or there, [Grins] 128 vs. 256.
The issue with compression algorithms is whether or not the listener can detect from an audible perspective whether there is a Loss or Not, at the different compression rates for all types of Analog Audio Waveforms after compression.

You can compress the styrofoam, but is there hysteresis loss and/or memory offsets that are Humanly Perceivable is the question?

Heck, I took a test this AM and can not hear 8KHz off the puter here! They state most people over 25yrs. can not hear 8KHz. or greater!!
Reply
Reply
post #165 of 263
3/9/11 at 10:51am
Thread Starter 
Quote:
Originally Posted by Bill Mac View Post

DPS,

There is Bullmoose Music which is a great store and I believe they ship as well. Just search for your favorite artist and it will show the used CDs in with the new. I shop at Bullmoose and Newbury Comics for used CDs. Although Newbury Comics does not show their used inventory on their site. I have had maybe one or two bad CDs but both stores guarantee their used CDs. You should check local to you as there might be stores in your area that sell used.

http://www.bullmoose.com/rel/v2_home...r=258&deptnr=1

Bill

Bill, thank you and I will check those URL's out to Spin some More Plastic.
Reply
Reply
post #166 of 263
3/9/11 at 10:54am
Thread Starter 
Quote:
Originally Posted by mtbdudex View Post

Rooms - "Don't get no respect"

Mike, that is why a band in an Open Air concert can just sound terrible vs. that Engineered Studio Acoustical Recording.

Point is right on FACT, AMEN !!
Reply
Reply
post #167 of 263
3/9/11 at 10:59am
^^^

well, no... it has nothing to do with that...

but if you ever want to hear what your speakers really sound like, drag them outside sometime and listen to them...

i have a pair of cheap rock speakers outside in my shade garden that sound VERY good...

@mtb... i've attempted to make the point about the room before, but somehow, it's gotten lost...
Reply
Reply
post #168 of 263
3/9/11 at 11:02am
Quote:
Originally Posted by DenPureSound View Post

The issue with compression algorithms is whether or not the listener can detect from an audible perspective whether there is a Loss or Not, at the different compression rates for all types of Analog Audio Waveforms after compression.

You can compress the styrofoam, but is there hysteresis loss and/or memory offsets that are Humanly Perceivable is the question?

DenPureSound, you are referring to so called lossy audio compression. But Michael is almost assuredly writing about lossless compression, in which the decompressed output perfectly matches the input. Such compression schemes take advantage of redundancy & predictability (i.e. the holes in the styrofoam) in data streams.

AJ
Reply
Reply
post #169 of 263
3/9/11 at 11:23am
Thread Starter 
Quote:
Originally Posted by SteveD61 View Post

This corresponds to a bit rate of

16 bits per channel x 2 channels x 44,100 samples per second = 1,411,200 bps

Much higher than 128,000 bps so there is a lot of data lost in that compression, and a lot of Audiophiles even regard the CD sample rate and word length as too low.

Most CD's would actually be recorded at higher rates (I think at least 96 kHz and 24 bits), but they must conform to the redbook convention (16/44.1) to be released as CD's in the market.

Steve, let's break this down here --

Let's look at just One Channel:

16 bits = 2 bytes = 1 word
or 2^16 = 64K = exactly 65,536 bits

and sampling at 44.1Ksps in the time domain.

Therefore,

for a line audio output signal of 1.2Vrms, or 1.697Vp the sensitivity is

1.697V/65,536 = 25.89 µV on the Ordinate (Y-Axis)

Nice, and that is at 16bits, imagine what 24bits would yield.

So, at 16bits it's over 11X vs. a 128Kbps stream in.
Reply
Reply
post #170 of 263
3/9/11 at 11:39am
Thread Starter 
Quote:
Originally Posted by WiWavelength View Post

DenPureSound, you are referring to so called lossy audio compression. But Michael is almost assuredly writing about lossless compression, in which the decompressed output perfectly matches the input. Such compression schemes take advantage of redundancy & predictability (i.e. the holes in the styrofoam) in data streams.

AJ

AJ, exactly MJH was discussing Lossless audio compression like here:

http://en.wikipedia.org/wiki/Audio_c...ion_%28data%29

But to an average human set of ears, do you think we can detect a lossless vs. a lossy compression, both professionally done using all the various standard formats?

I don't think I could, but I'm guessing here.
Reply
Reply
post #171 of 263
3/9/11 at 11:46am
Quote:
Originally Posted by mtbdudex View Post

Rooms - "Don't get no respect"

Quote:
Originally Posted by ccotenj View Post

@mtb... i've attempted to make the point about the room before, but somehow, it's gotten lost...

Quote:


We must recognize, however, that no component has as much impact on the performance of an audio system as the room in which it is installed.

Jon T. Gale
Audiophilia Online Magazine

Compare the in room responses of the Wilson Audio Sophia Series 3 (red trace) and the AudioNote AN-E/SPe (blue trace) in Stereophile writer Art Dudley's listening room:



http://www.stereophile.com/content/w...r-measurements

Now, compare the in room responses of the Magico Q5 (red trace) and the Wilson Audio MAXX 3 (blue trace) in Stereophile writer Michael Fremer's listening room:



http://www.stereophile.com/content/m...r-measurements

Finally, compare the in room responses of the Focal Maestro Utopia III (red trace), the Aerial Acoustics 20T V2 (blue trace), and the Revel Ultima Salon2 (green trace) in Stereophile editor John Atkinson's listening room:



http://www.stereophile.com/content/f...r-measurements

What do you note about the responses -- particularly below 100 Hz -- of two or three different loudspeakers in the same listening room? Which has a greater effect, the $15,000+ loudspeakers or the room?



AJ
Reply
Reply
post #172 of 263
3/9/11 at 11:51am
Atkinson's room is impressive... look how flat that is. Are the three lines, left/right and an average?
Reply
Reply
post #173 of 263
3/9/11 at 11:51am
@aj...

talk about a picture being worth a thousand words...
Reply
Reply
post #174 of 263
3/9/11 at 12:01pm
Quote:
Originally Posted by DenPureSound View Post


Therefore,

for a line audio output signal of 1.2Vrms, or 1.697Vp the sensitivity is

1.697V/65,536 = 25.89 µV on the Ordinate (Y-Axis)

Nice, and that is at 16bits, imagine what 24bits would yield.

For consumer digital audio, standard analog output level is 2 Vrms, though many digital components deviate significantly from that standard.

24 bit quantization = 2^24 = 16,777,216 discrete levels

However, w/ standard 2 Vrms output, typically only 18-20 bits rise above the noise floor, even in the best performing components.

AJ
Reply
Reply
post #175 of 263
3/9/11 at 12:11pm
Quote:
Originally Posted by MichaelJHuman View Post

Atkinson's room is impressive... look how flat that is. Are the three lines, left/right and an average?

Nope. Read my legend for each graph. Those three traces represent the in room responses of three different loudspeakers in JA's listening room. Note how many of the peaks & troughs align in frequency, even overlap in level. The room response is largely swamping the responses below 100 Hz of the three $20,000+ loudspeakers.

I greatly enjoy Stereophile and particularly appreciate the equipment measurements. But the reviewers, who nearly all (absent Kal) eschew digital room correction, are cutting off their noses, so to speak, in the name of sonic purity.

AJ
Reply
Reply
post #176 of 263
3/9/11 at 12:55pm
^^^

but... but... but... "the dac"... "the rate"... "the jitter"...

maybe those pictures will get the point about the room across... nice post aj...
Reply
Reply
post #177 of 263
3/9/11 at 12:58pm
@dps...

if you want to learn about compression, it's effects and it's audibility (and how to easily test it for yourself), go spend some time at hydrogen audio... that's a good "science based" site...
Reply
Reply
post #178 of 263
3/9/11 at 3:57pm
Thread Starter 
WiWaveLength --

I want John's room, and the equipment to test even where my crappy room is right now. Point by all is very well made about room treatment/acoustics vs. speakers.

Very nice post AJ - waiting for my Stereophile rag here.
Reply
Reply
post #179 of 263
3/9/11 at 3:59pm
Thread Starter 
@CCO -

thanks Chris, but heck I don't have enough time to watch a movie yet after getting hooked on AUDIO

See, off to dinner and movie night with that mic'd popcorn.
Reply
Reply
post #180 of 263
3/9/11 at 4:06pm
Quote:
Originally Posted by WiWavelength View Post

Nope. Read my legend for each graph. Those three traces represent the in room responses of three different loudspeakers in JA's listening room. Note how many of the peaks & troughs align in frequency, even overlap in level. The room response is largely swamping the responses below 100 Hz of the three $20,000+ loudspeakers.

I greatly enjoy Stereophile and particularly appreciate the equipment measurements. But the reviewers, who nearly all (absent Kal) eschew digital room correction, are cutting off their noses, so to speak, in the name of sonic purity.

AJ

Makes sense. I have not measured my in room response with and without YPAO. Yet another project to try.
Reply
Reply
New Posts  All Forums:Forum Nav:
  Return Home
  Back to Forum: Receivers, Amps, and Processors

AVS Top Picks

AVS › AVS Forum › Audio › Receivers, Amps, and Processors › ** OFFICIAL ** -- Sound Quality (SQ) Discussion