CD audio quality

DannyW

DannyW

DannyW

Man of Honour
Joined
31 Dec 2005
Posts
4,869
Location
England
hi there
can someone please back me on that CDs and Records sound a lot better then compressed mp3's. a n00b keeps arguing with me that mp3s are just a good. NO WAY. 128-320kbs VS over 1500kbs.

thanks, feel like a idiot posting this but i know im right
 
a properly encoded vbr MP3 @ 192k or over is extremely difficult to tell from the real deal. sorry to bust your bubble, but that's the truth. There's a lot more to it than just the bitrate.
 
Rednut05 said:
Well any compression is going to deteriate the sound quality.
Click to expand...
what about lossless compression? :p

Bear in mind the speakers used aswell - it's quite possible he can't tell the difference between a CD and a 128kbs mp3 if he's using headphones he got free out of a packet of crisps...

fini
 
Depends....
I have tried apple lossless vs compressed at a variety of lossy versions on my Ipod, and couldn't tell the difference whilst using some Sennheiser MX500s.
However, tried a 320bps MP3 using CDEX and a highly recommended ripper and thought that the CD sounded a good deal better, but that was played back through my home stereo.
 
Rednut05 said:
Well any compression is going to deteriate the sound quality.
Click to expand...

When you zip up your Word documents, do you notice a deterioration in sound quality?

The reality of the matter is that some people can tell lossy MP3 and CD apart up to ~300kbps. Some people think they can tell the difference between higher but tend to fail blind tests.

People can tell the difference between 128k and 192k but it is hard, ~80% of people fail blind tests in my experience both on here and talking to people who study this sort of thing at doctorate level.

I would put money on an average music listener being totally unable to successfully pick the compressed track over a mix of genres more than two times out of 10.

To the OP:

Interesting to note that you lump CDs and LPs together as being the defacto benchmark in sound quality without any caveats whatsoever!

LPs generally tend to have higher bandwidth on their first play through a top-end player. CD only has a bandwidth of 22050Hz at absolute maximum but in reality its a bit below 20kHz.

Neither are perfect by a LONG way - It is my opinion that digitising at massive bandwidths is preferable to high bandwidth analogue mediums because of the reliability and longevity of the recording as well as how mass-marketable it is. An unaffected analogue recording on something like tape is going to be brilliant until the tape stretches or is affected by magnetism or starts to be affected by UV etc. Vinyl doesnt even come close to being accurate once you have factored in how good the transient response of the needle needs to be.

£5000 carts are all well and good, but how many of those are going to sell when you could realistically use up 15Gb per track for full-bandwidth ~100% lossless audio on something like blu-ray for a tenth of that in a couple of years?
 
DRZ said:
£5000 carts are all well and good, but how many of those are going to sell when you could realistically use up 15Gb per track for full-bandwidth ~100% lossless audio on something like blu-ray for a tenth of that in a couple of years?
Click to expand...

It's an interesting question DRZ, and no one seems to have put forward a real suggestion of what to do with 50GB of space on a Blueray disk for music.
The lack of success of SACD and DVD-A seems to have discouraged manufacturers from even suggesting a higher bandwidth format.
I seem to remember reading that making a studio capable of producing an SACD grade recording would require a number of studios to upgrade, so a higher bandwidth/resolution version seems unlikely.
In addition, it wasn't the SACD hardware being unavailable, but the lack of software that doomed it.
The music industry does seem to be going down the road of happily destroying the recording quality of what we're given.

As for turntables. You're dead right about the bandwith of vinyl. I'm told that over a period of plays, the records elasticity will result in a loss of top end, getting down to around 10khz after 200 plays. Maybe that's why older people like LP, as they're hearing will be shot in the upper regions.
Having said that, I still prefer a really good LP based system to listen to, despite not owning one.
 
Absolutely agree with you on the studios not being content with moving to higher bandwidth - I guess while they still believe you cant hear anything over 20kHz they simply wont bother :(

The best microphones in the world can record up to about 100kHz before they become non-linear, which is fine but they cost ridiculous amounts of money and are fragile beyond belief. No studio is going to want to move to those - silly bottom end shure units seem to pass for acceptable in some studios :eek:

If I am honest (and I am sure a few people here might agree) even CD has been wasted in terms of what is put out on it vs what it is capable of. Jazz artists tend to be the most picky and the quality of the CDs that are released put Pop and Rock to shame. Im sure even after 400 plays on an LP, The Bends still sounds as terribly recorded as it did on play 1.

Ive just had a side-thought about CD vs LP that I am going to have to do some maths (:() to work out: If tests show that some listeners can hear 100kHz I wonder how much of the quantisation noise present is below that, which might have a perceptable effect on sound quality. I probably wont get any answers to that thought though and ultimately a probable reply is that it will come down to the order of the reconstruction filter or something similar...
 
willd58 said:
This bit is very rare on songs downloaded from peer 2 peer networks.
Click to expand...


of course it is. i dont download them, i encode my own collection and it knocks the spots off virtually anything you can download

DannyW said:
exactly most the illeagal mp3 downloads are rubish
Click to expand...

which has nothign to do with the question you asked. You asked if cd's sound 'a lot better than comperssed mp3's'. The answer is no, unless you are comparing a low bitrate, or poorly encoded mp3.

a properly encoded vbr MP3 @ 192k or over is extremely difficult to tell from the real deal
Click to expand...
 
DRZ>
I think you're spot on about the fact that even CD is not fully taken advantage of by studios, such a shame.

On thought on very high bandwidth speakers. I heard the effect of a pair of Townsend supertweeters last year.
The interesting point was that whilst they did have an effect (for me) that it wasn't what I expected. I personally couldn't hear any additional top end information (hardly surprising as my hearing really is shot at high frequencies). What I did get was much better placement of instruments and musicians within the mix, really locking them down. It was a nice effect, though at £800 a shot for the tweeters (RRP), it's not something I'll be doing in a hurry.
 
Think it quite difficult to compare MP3 Vs CD. I guess you need to use a DVD/Universal player that can read native MP3 files off a data disc ??? Other wise the electronics and playback device has an influence, and your not treat each format equally.
Often I think MP3 are compared to a “copied” version of the CD. Which I maintain sounds worse than the original pressed item……
Secondly if you have the original disc, why rip it other than for a portable devise ?... A good CDP is easier to obtain than PC with sound cards, external DAC’s etc to try and get to the same performance. … Much easier just to stick a disc in draw !

I think you can measure vinyl as much as you like and form a debate and argument, BUT at the end of the day analogue still achieves a more natural and enjoyable, musical sound. Perhaps it has better synergy with the listeners ears ?!?! ….and it doesn’t need £5000 carts to do it….. Vinyl still has the ability, and has got a lot better over recent years to whip ANY current digital format !!!

But it’s all irrelevant; ALL formats are seriously compromised with the current dire recording and mastering quality on many albums….. as consumers we are treat like dirt, ripped off and sold ****.
I believe there are no plans to make a Blue Ray audio format etc….I guess the mess they made with SACD has scared them off…. and “everyone” wants compressed convenience music ! :rolleyes:

My understanding of the output over the range of hearing, 20Khz when your young !!is the effect it has on the harmonics of the audio range, apparently it has a “perceptive” effect without actually hearing the direct sound !!!.... Can’t say a super tweeter is top of my shopping list at present !

9d

PS Monica2 has been shipped :)
 
Well i encoded some an album with Lame @ 192 about 6months ago. Played it through my pc the other day and it sounded worse than the original cd that i bought again last week after losing the original.

Mp3's will always sound retarded on a good hi-fi. COmpression by cutting bits of the music out can never be justified.
 
Last edited:
I find that the music quality that is offered by something like iTunes tends to be rather shabby and that is why I still keep buying CDs myself and then using the latest (usually) lame coder with EAC progy I tend to rip them into 192+ vbr mp3s so that I have greater versatility in listening to whatever sort of music that I like during the time that I spend on the pc & putting on the iPod :).

Also after doing some tests myself, I did find that I can notice the difference in those high quality vbr mp3s and a CD (tested on exactly the same system with all the settings being the same etc), although the difference that I noticed is not something most of the music buying public would notice, on your regular (read as non enthusiast) system.

But I do agree with the statements that CDs now are being very poorly mastered, the sound quality that is being pitched to the regular consumer is that of which allows higher profit margins while still being of "acceptable" quality to give to the end users =\
 
On an average system, its certainly hard to tell the difference between MP3 and the original CD. However on a good enough system, the differences become obvious enough to hear in a blind test.

My Tag/Audiolab system is very detailed and analytical, (although I tempered it slightly, by using Mission speakers), I can certainly tell the difference between a 320kbps encoded mp3, and the original CD.

Can also tell the difference between a poorly mastered CD, and a good one. My DVD player can do MP3 decoding and output the datastream on a standard PCM/SP-DIF interface for my Tag to work with so there is no 'sound card effect' to contend with. My PC's SPDIF output is also connected to the Tag, and does a pretty decent job too.

The best CD I ever listened to turned out to be HDCD encoded, It basically gives the DAC an extra 4 bits of encoding, giving you a 20bit CD. Its a pitty there are only a handfull of HDCD's out their, as they really are considerably better, and only need a new DAC to play properly. (But are encoded in such a way that a 16bit normal DAC can still make music from them)

At home I always listen to the original CD's. However MP3 is great in the car, Even at 320kbps encoding you can get a lot more tracks onto an MP3 CD, which my cars autochanger handles :), and with the roadnoise at 80mph you certainly cant hear the differences.
 
Corasik said:
However on a good enough system, the differences become obvious enough to hear in a blind test.
Click to expand...

Were the blind tests you referred to your own tests on yourself or did you participate in a "proper" blind test, either for a research institution or in informal one (like the one I ran here)? If it was a formal study, could you link me to the institution that carried out the tests (so I could get hold of their research paper) as I would be interested to see their results and how they carried it out.

At the moment, having looked into this in some detail, I would have to say that I find it increasingly difficult to see how someone describe the difference as "obvious" - exceptionally few people seem to be able to get it consistently right - and even those people that do consistently say it was incredibly close between them.
 
non-scientific test I did (not controlled, single blind, no balanced ref) showed a strong correlation to ranking MP3 128, MP3 192, AAC 192vbr, WAV. Interesting point was that the listener found classsical recordings very easy to rank, most pop was harder. Also the listener, on several occasions, was able to rank in terms of compression but preferred the more compressed pop.

I, on the other hand could only differentiate MP3128 with any decent level of accuracy on the kit I had at the time.
 
I'd have a hard time telling the difference between an origonal and a top notch 192kbps mp3, 128k usually sounds a mess though. I reckon I could quickly tell them apart in a blind test.
 
You must log in or register to reply here.
Back
Top Bottom