StreamingSoundtracks.com
Search

 

VIP
Subscribe to become a VIP member of SST!

· Request More Often
· Unshared Requests
· Request Countdown Timer
· Request Ready Indicator
· Your Request History
· Access To The VIP Forum
· Add More Favorites

:: Click Here To Upgrade ::

:: Give VIP as a Gift ::

Click Here To Listen

Follow Us


Donation Meter


Make donations with PayPal!
Monthly Goal:
$500.00

Need:
$500.00

0 Donations:
$0.00


 


Last Month's Donors
StreamingSoundtracks.com (Jul-30) molossus $6.63
Death.FM (Jul-29) diginferno $50.00
Death.FM (Jul-24) htmm $13.37
StreamingSoundtracks.com (Jul-24) megadith $87.00
StreamingSoundtracks.com (Jul-22) danico $10.00
StreamingSoundtracks.com (Jul-21) Anonymous $3.00




Windows Phone App

Android App

SSTore



:: SSTore ::


Transatlantyk Festival Poznan
:: Transatlantyk Festival Poznan ::

:: Tour Dates Thread ::


Windows Media

Partners


:: Link To Us ::
:: Add Your Link ::


SST and codecs

 
Post new topic   Reply to topic    StreamingSoundtracks.com Forum Index -> General
View previous topic :: View next topic 
Author Message
Liquid3D
Cadet 1
Cadet 1



Joined: Aug 28, 2008
Member#: 22442
Posts: 1


Liquid3D is offline View user's profile Send private message Liquid3D's Favorites are Private
PostPosted: Thu Aug 28, 2008 3:37 am   Post subject: SST and codecs Reply with quote


Hi my question or subject alone probably shows I am a noob. I am trying to determine if there is anything such as Lossless streaming, or is all streaming "lossless"?

I am reading and trying to write about music files, compression and digital audio. I've read and if I understand correctly lossless applies to codecs or music files which have the least amount of compression or compression-method which doesn't degrade sound quality as much as others. Hence Audiophiles would try to copy or download music in a file format or codec (if that's correct) such as FLAC,

Here's an example which confused me. Lets say I go to SOMA FM and they give me the following choices:

Listen: 48k aacPlus
128k aacPlus
128k, 56k, 24k MP3
128k, 32k Windows Media

I presume these are codecs for compressing copying music, however clicking on any then gives choices for purchasing smaples of music. But when I go to SOMA FM and just click on the channel icon for streaming audio why must I then click on the file types? What happens when I listen to a specific channel in "streaming" mode?

From what I understand about digital audio basics this is how its broken down (I am copying the pertinent parts or that I think are pertinent):

Digital Audio
• Audio Characteristics
– Synthesized / Sampled Waveforms
– Sample Rate, Sample Size, Channels

• Sample Rate – Samples per second (8000, 44100)
• Sample Size – Bits per sample (8, 16)
• Channels – Number (mono, stereo, 5.1)

Sample Rates
5,000 - Highest human voice
8,000 Speech – Telephony – U-LAW
11,025 Quarter CD
16,000 G.722 compression standard.
18900 CD-ROM/XA standard
20,000 - Limit of human hearing (17k)
22,050 Half CD
32,000 DV; Used in digital radio and other TV
37800 CD-ROM/XA
44,056 Prof. audio, integral samples in video frame
44,100 CD Audio
48,000 DV; DVD-Video; DAT (Digital Audio Tape)
96,000 CD Audio, AAC, DVD PCM
(of course we've now one to 192)

• Audio Compression
– Data Size, File Formats

• Platform Formats
– AIFF – Audio Interchange File Format – Macintosh
AIFC – Compressed
– WAVE – Windows – Typically PCM; Lossy ADPCM – 4:1
– AU, SND – Sun – mu-law – 2:1
• Web Formats
– MP3 – MPEG layer 3 - 32-320 kbps - target 64 kbps
– AAC – Advanced Audio Coding – MPEG-2 (iTunes)
– RA, RM – Real Audio
– ASF, WMA – Windows Media
• Other Formats
– Ogg Vorbis – Open source, royalty free; Vorbis music not voice
– ATRAC – Adaptive Transform Acoustic Encoding
• Audio Processing
– Sample editing
– Multi-track mixing and effects
– Soundtrack generation, Looping
– Composing

• Platform Formats
– AIFF – Audio Interchange File Format – Macintosh
AIFC – Compressed
– WAVE – Windows – Typically PCM; Lossy ADPCM – 4:1
– AU, SND – Sun – mu-law – 2:1
• Web Formats
– MP3 – MPEG layer 3 - 32-320 kbps - target 64 kbps
– AAC – Advanced Audio Coding – MPEG-2 (iTunes)
– RA, RM – Real Audio
– ASF, WMA – Windows Media
(obviously not all listed)

Audio Rates = Kbps
Low rate - voice = 8Kbps
Low rate - music 20Kbps
CD stereo quality, WMA, AAC 48Kbps
High-quality streaming, portables 64Kbps
Downloaded music, MP3 128Kbps
High-qual MP3, WMA, AAC 192Kbps
CD Audio - uncompressed 1411Kbps

My confusion lies with a term such as the following, what if we say CD is 44k? Is it 44,000 samples per second? And is this better then 192k or am I using lower case "k", upper case "K" improperly here which is confusing me? Why would the recording industry record to a standard at 44k when they have 192K would it be the cost and complexity of changing that standard and the millions of CD-players which would then be obsolete?

Also when looking at Lossless I am looking which codec is basically a 1:1 ratio, meaning no compression and do these exist? This test shows a list;
http://uclc.info/lossless_audio_compression_test.htm

On this list can someone clarify where OptimFROG and FLAC stand relative to each other? If the byte size is the smallest then wouldn't these do the most compression? And if that's the case wouldnt FLAC because it's compressing less "SOUND" better then OptimFROG? The key term is "sound better"
USA j2brown
Commodore
Commodore

aw

Joined: Feb 22, 2002
Member#: 9
Posts: 3176
Location: Sterling, VA

j2brown is offline View user's profile Send private message Send e-mail View j2brown's Favorites
AIM Address Yahoo Messenger MSN Messenger
PostPosted: Thu Aug 28, 2008 6:20 am   Post subject: Reply with quote


I'm no expert, so I don't think I can begin to give you the answers you need/want. There are some experts around, and I know that there used to be pretty detailed discussions in the early day of SST. You might try a forum search and see if you get anything.

To clear things up a bit (or confuse them more, perhaps) keep in mind that there are two different processes at work here. The first process is when the audio is taken from the source, in most cases here a CD, and converted to a digital file for storage on the server. At SST JERIC uses a process that is lossless, so the digital file has all the data of the original.

Next is the stream. The lossless file is played over the streams at the bitrate of the stream you choose and the file is downconverted as appropriate (introducing loss) to fit the proper stream.

Did that help any?

jeff
sdg
Brazil Pesadelo
Admiral (Administrator)
Admiral (Administrator)



Joined: Dec 19, 2007
Member#: 19747
Posts: 205
Location: Rio de Janeiro, Brazil

Pesadelo is offline View user's profile Send private message Visit poster's website View Pesadelo's Favorites
Visit MySpace
PostPosted: Thu Aug 28, 2008 11:32 pm   Post subject: Reply with quote


Hi Liq3D.

I read something about this already. I will try to give you some initial answers. Perhaps, they can help you to form a better shape of the matter and then ask others (and equally important) questions.

Quote:
My confusion lies with a term such as the following, what if we say CD is 44k? Is it 44,000 samples per second? And is this better then 192k or am I using lower case "k", upper case "K" improperly here which is confusing me? Why would the recording industry record to a standard at 44k when they have 192K would it be the cost and complexity of changing that standard and the millions of CD-players which would then be obsolete?


Indeed, the the (current) audio CD is such that the sample rate is of 44,100 samples per second. This means that the A/D converter (Analog to Digital) captures the sound 44,100 times in one second. Remember that when your favorite band/orchestra is playing, they are in the analog (real, concrete) world and that our beloved CDs are representants of the digital (virtual, 0s and 1s) world. Thus the need for conversion from one to the other.

Now, you can capture the sound 44,100 times per seconds, but the next step is to decide in how many bits you will put each sample captured. If you choose to do it with only one bit, you will get only 0 or 1 (2 combinations) at a time. However, if you decide to use 16 bits, you will get one of 65536 different variations in sound at a time. Standard CDs samples are stored in 16 bits each.

Good. We have 44,100 samples per second, each sample stored in 16 bits. But (again!) the standard CD stores its musics in a sterophonic way, i.e. in two distinct and independed channels. This is responsible by the spatial effect we gain when we listen to a stereo program.

Now we have all the elements. 44,100 samples per second, stored in 16bits each, and all this for each of two channels. Ready. We have 44,100 x 16 x 2 = 1411200 bits per second, or 1411Kbps, as you can see in your last list, audio rates, for CD Audio.

Now, about the lossless subject. The lossless is the sound recovered from the source without any kind of change. We use to say about the lossless music for that gotten from the CD in its "natural" state, or as a WAV file (44,100 x 16 x 2). This uses to be a large file.

There are two ways to change the sound in order to store it in less space than we would have if we would go to store it in its lossless form: a resample and a compression.

In the ressample, one transforms that 3 basic parameters of the music. For example, a CD music and its 44,100 x 16 x 2 would be resampled to 22,050 x 16 x 2, and then occupy half the space than the original. Or it would be resampled to 44,100 x 8 x 1, occuping 1/4 of the original size.

Yet the compression is different. It is an algorithm based on certain mathematical features of the music that allow us to "rewrite" the digital pattern according to this features. In a gross way, it is something like to check out the digital set of bits, searching for its parts that are not "well audible" by humans and then cutting them, completing the cut curve in order to not produce noise in the music. Is this process, a kind of resample occurs.

MP3, aacPlus, OOG and etc, are all about compression and, then their streams need to be decompressed (it is what the players do). All compressed sound has loss. Every one. What goes on is that some compressions are better done and then produce better sound quality when decompressed, but at the cost of bigger files and bigger times to compress.

I am not sure if the next paragraph is right. I am almost guessing in it. Hope that someone else can help us with it. Smile

Internet radio works with compressed streamings basically. And when you see that an aacPlus streaming is about 64k, this means that its compressed form is to be transmited in a rate of 64kbps in order to be reproduced accordingly, since the decompressor will do its task at this rate. The result of the decompressed sound will depend of the quality of the compression done.

Each process (MP3, aacPlus, etc) has its own algorithm and each one tends to be better to this or that sound characteristic. Personally, I like aacPlus very much.

And, finally, about the standards: they are settled accordingly with the technology available at the time they are to be decided. The CD standard was enough for a good time, but now, with the technology advance and the demand for more complex sound (countless channels for instance), a new standard is possible and even needed. Lets wait what is coming on! Smile

Well, hope have help you somehow. I have to sleep! Night!

And welcome to SST forums! Smile
_________________
I think I think, therefore I think I am. (Ambrose Bierce)
Display posts from previous:
Post new topic   Reply to topic    StreamingSoundtracks.com Forum Index -> General All times are GMT - 5 Hours
Page 1 of 1

 
Jump to:  
You cannot post new topics in this forum
You cannot reply to topics in this forum
You cannot edit your posts in this forum
You cannot delete your posts in this forum
You cannot vote in polls in this forum



Forums ©


Copyright © 2001-2014 24seven.FM, LLC All rights reserved.
Comments, images, and trademarks are property of their respective owners.
You can syndicate our news using the file backend.php or ultramode.txt. Robots may follow the Sitemap.