[buzzy]

Life is short.  Don't waste it on stuff like this.

 

[spaark]

Quote:

I would suggest that if you don't detect a difference in abx testing but you think you do detect differences in normal listening then this cannot be reliably taken as an indication of bias and may well be exactly what it seems - a difference.

What I would do is carry out a normal listening test first, memorising the bits that you think are different, and then do a blind listening test with the expectation that you'll find those differences again. Listening fatigue aside, if you can't find those differences with the blind test, then surely they weren't there in the first place. That's sounds reasonable to me.

 

[peterBj]

I just wrote a comment on a different blog,but I think it also goes for this discussion:

There is a big difference.
But the main advantage is gained in the recording process,recording and mixing.
The smooth and warm sound that our recordings are famous for,is to a great deal thanks to 24/96.
We have experimented with different formats at our studio and found that 24/96 was the best sounding format for our purposes.
But once we have finished recording,we found that you can down sample to even mp3 format and the sound quality is still pretty good,and that is not the case with a file recorded in 16 /44, in our experience.
I suggest you get hold of an originally recorded file of 24/96,not an up sample.
This is probably where the root of the misconception lies,in all these older recordings that have somehow magically been transformed to 24/96 or 24/192.
I have some horrible remasters in my CD collection.I.e."Ella and Louis,''is a great sounding LP, but the 24 bit CD remaster I have, is horrible.
(It now lives it's life in my 83 year old mothers car stereo,and there the added smile curve has a purpose,the car,and the stereo in it,is from '92)

And if you compare the same recording in different rates/formats be sure to all ways listen from down to up,i.e. 128mp3 first and flac and Wav last.
Your brain will accommodate the missing bits if you do it the other way around.

 

[spaark]

Quote:

And if you compare the same recording in different rates/formats be sure to all ways listen from down to up,i.e. 128mp3 first and flac and Wav last.

You're saying you can hear a difference between FLAC and WAV?

 

[julian67]

What I would do is carry out a normal listening test first, memorising the bits that you think are different, and then do a blind listening test with the expectation that you'll find those differences again. Listening fatigue aside, if you can't find those differences with the blind test, then surely they weren't there in the first place. That's sounds reasonable to me.

Yes if you need to try to reassure yourself that lossy audio sounds just like lossless you can spend lots of time doing stuff like that. You go ahead if you like. I would only go to such trouble if the result would be something better than the source. That's not possible so for me it would be time wasted. These days my music collection is almost entirely lossless and all my portables support lossless.

The preceding page shows my 16/16 abx of lame -V 0 vs lossless. And there is extrabigmehdi, someone absolutely adamant that there could be no difference and even saying he had tried the exact sample before and found no difference. A quick explanation of a simple listening technique and shortly afterwards he abx's it. In a few minutes someone "objectivist" but without listening skills goes from "no difference" to having learning a basic ear training technique and producing an abx log demonstrating audible difference.

So I don't even care that some of the differences I perceived surely arose from my mind because plenty of them equally surely did not (as unambiguously demonstrated). I did the easiest and most logical thing for anyone who wants the best audio quality: I now simply listen to my lossless collection instead and have ceased caring about data rates, transparency, artefacts and so on.

I use lossy codecs for audio books because they can be really good for speech at low bitrates (HE-AAC is brilliant for this) and some books run to 12, 15 or even 20 CDs and sometimes I want all that on a portable such as my Rockboxed Sansa Clip+.

This isn't the year 2000 and I'm not running a PC with 20 GB hard disk or using a portable with 512 MB or 256 MB storage. I don't need to get an album down to 70-100 MB. 300 MB or 500 MB is perfectly fine. And ironically I saved myself some disk space when I dumped the lossy audio because now I only store a single music library, not two.

Save precious disk space, switch to lossless

 

[spaark]

Uhhh I think you took my comment the wrong way... you said "this cannot be reliably taken as an indication of bias" and I'm saying it can if you do it the way I suggested. For some reason you think I'm advocating that people stick to lossy audio.
 
Also

Welcome to the world of supposition, probability and uncertainty, decorated with a dash of unqualified and uninvited psychological evaluation.

Warning: If you spend years trying to convince yourself your mp3s sound identical to the CD, without ever really being sure, then this can happen to you too.

I hope this wasn't an attack on me.

 

[julian67]

Yes if you take a snippet of my earlier statement, and then append an "if..." composed of some other conditions and circumstances, apparently ignoring or discarding everything that preceded it, then you can have a satisfying discussion, but probably not with me.

I don't need to do as you suggest because it offers me no worthwhile benefit. If performing those tasks helps you feel confident about your lossy files then that's fine.

On the preceding page you can see that both me and someone else, who previously hadn't thought it even possible, both abx'd lame -V 0 vs lossless. It was even easy.

The single advantage lossy compression has is reduced size. But I do not need to save disk space even on my portables. My Sansa Clip+ with 32GB card has eighteen times the storage space of my first PC. It can hold more than 100 lossless albums. If I want 100 different albums I can swap the cards. That's probably about about 80 hours of music per card and another 15 hours on the internal storage. The battery lasts between 10 and 15 hours per charge. I don't need more albums on it.

I already have a lossless collection derived from CD and I have several terabytes of storage. I don't need any size reduction. I know with 100% certainty that occasionally lossy compression fails badly enough that anyone who knows the original sound will easily notice the failure. So why would I want to use it? What benefit would I get? Why would I want to maintain two different music collections (actually three if you include the optical discs)? Alternatively why would I want to spend time transcoding for portable devices when they can play the untouched originals?

And why do other people want me to use and like lossy even after I clearly state why I prefer not to in plainly expressed, rational terms supported by unambiguous data?

Is it a cult or is it an imperfect compression technique?

 

[spaark]

Quote:

Yes if you take a snippet of my earlier statement, and then append an "if..." composed of some other conditions and circumstances, apparently ignoring or discarding everything that preceded it, then you can have a satisfying discussion, but probably not with me.

So I can't comment on one of your points -- I have to argue with everything? What?
 

I don't need to do as you suggest because it offers me no worthwhile benefit. If performing those tasks helps you feel confident about your lossy files then that's fine.

On the preceding page you can see that both me and someone else, who previously hadn't thought it even possible, both abx'd lame -V 0 vs lossless. It was even easy.

The single advantage lossy compression has is reduced size. But I do not need to save disk space even on my portables. My Sansa Clip+ with 32GB card has eighteen times the storage space of my first PC. It can hold more than 100 lossless albums. If I want 100 different albums I can swap the cards. That's probably about about 80 hours of music per card and another 15 hours on the internal storage. The battery lasts between 10 and 15 hours per charge. I don't need more albums on it.

I already have a lossless collection derived from CD and I have several terabytes of storage. I don't need any size reduction. I know with 100% certainty that occasionally lossy compression fails badly enough that anyone who knows the original sound will easily notice the failure. So why would I want to use it? What benefit would I get? Why would I want to maintain two different music collections (actually three if you include the optical discs)? Alternatively why would I want to spend time transcoding for portable devices when they can play the untouched originals?

And why do other people want me to use and like lossy even after I clearly state why I prefer not to in plainly expressed, rational terms supported by unambiguous data?

Is it a cult or is it an imperfect compression technique?

Huh? I'm not suggesting you do it.
 
Again, I'm not advocating lossy audio, so what are you telling me all this? It has nothing to do with the point I was addressing.

 

[peterBj]

No
I am saying that we find,that there is a benefit with higher resolutions when we are recording a musical instrument.
Afterwards,when we are ready with the mix you can compress it to various formats and it will sound great.
We are selling Wav files only,from the 4 albums we have made so far.The reason for that is,that we want our costumers to get the exact same file as the one we have labored so hard at in the studio.Then when a client has that file,he can compress to whatever he wants,but he knows that he has a one to one copy of the original file.
Several of the big download sites are selling so called master files of older recordings with bit and sampling rates going all the way up to 24/192,and that,in some cases, for recordings that was done in 16/44!
I doubt there is any benefit  to be gained by doing so.
I myself have a sony walkman,and most of the files on that is 256mp3,and I'm perfectly happy with that for that listening situation.  
 
And for the rest of the discussion,I think you all should trust your ears,if you can hear a difference good for you!
If not,good for your wallet.
 
I am now going in to my 35th year as a full time musician,(recording/producing is not my main income)and have met and played with a great deal of musicians all with a very different level of hearing.Some with perfect pitch ,most with out.Some,like the pianist Atsuko Kohashi,on our label,has an extreme sense of color hearing,as we call it,she will only play a tune in a color that fits the tune!
I can,t hear what she means,to me any key will do,but if it makes hear happy to play i.e."How Deep is the Ocean" in G instead of Eb,and it really does,I am happy to go along.
so in short we are all, thank god, very different.

 

[coryeeeee]

My ears can't tell much of a difference. Maybe I burnt them out with too much high-volume music.

 

[julian67]

.....I myself have a sony walkman,and most of the files on that is 256mp3,and I'm perfectly happy with that for that listening situation.......

And for the rest of the discussion,I think you all should trust your ears,if you can hear a difference good for you!......If not,good for your wallet.

Suitability for one situation or another wasn't the issue. The issue is "can a difference be heard between 320 kbps mp3 and a lossless source?"

It has been demonstrated that a difference can be heard easily with some music. Whether that difference is relevant or important to different people in different situations is entirely another discussion.

Also completely irrelevant is price. An audible difference does not become more or less audible as the purchase price changes.

My ears can't tell much of a difference. Maybe I burnt them out with too much high-volume music.

Are you speaking generally or referring to a particular sample? Even if your ability to hear high frequencies is seriously degraded you should still be able to detect obvious smearing and pre-echo as found in the sample I posted. You may be able to hear far more and with more discrimination than you realise. I'll repost this link from the first page of this thread: http://harmanhowtolisten.blogspot.com/ It's a very good listener training package for Windows PCs (it will also work in Linux with Wine).

 

[peterBj]

you are qouteing me out of context.
this Qoute would more agree with you I believe.
 

The smooth and warm sound that our recordings are famous for,is to a great deal thanks to 24/96.
We have experimented with different formats at our studio and found that 24/96 was the best sounding format for our purposes.

 

[Logistics]

Sigh, everyone hears differently so it doesn't matter what another person concludes with an ABX test. If you do or don't hear a difference then that is your conclusion. You cant base it on the conclusions of others, especially when they also have different hardware.

 

[julian67]

Context:

This thread is Trying to hear the difference between 320 mp3 vs flac

It is not "can we please have a series of self promotional messages from a small label?"

Not price, or sound quality of 24/192 vs 16/44 or 24/96 or how much your company loves its customers, or how hi-res is great for recording or mastering.

 

[Mambosenior]

Most of my collection is classical. Even for my old ears, there is a enormous difference between 320 and lossless. This difference manifests itself by a shrunken sound field and unnatural instrumental timbres--most noticeably in the string instruments and high brass. This difference in sound has never been difficult to discern over either speaker or headphone. For my less-sizeable but still plentiful Rock/Pop/Jazz collection that I know well, my conclusions are the same, especially with multi-layered or acoustic-instrument recordings.

I am in agreement that everyone hears differently and that arguments about compression rates are relative only to individuals who can hear the difference, BUT:
considering that prices of storage media are very reasonable and most computer drives are significantly larger than were even a few short years ago, why not rip to a lossless format like FLAC and preserve the integrity of the original recording? What this protects against is a future where your ears may be able to detect more subtle differences in recorded music and/or acquiring a more revealing audio system that magnifies that difference.

(I am from the teen generation where those of us with limited means had portable AM radios with alligator clips for antennae...what most mp3s now sound like to me.)