[Dillan]

That's why you up the bitrate and use better codecs like MP4, so there is enough bitrate in the data stream to handle the encoding flawlessly. If the encoding of lossy audio is done well, human ears cannot tell the difference. Scientific testing has shown that, whether you want to believe it or not.

The state of compressed audio isn't the same as it was ten years ago. Many of your DVDs and Blu-Rays use MP4 audio. So does high resolution YouTube, satellite radio and streaming music services like Spotify. You listen to lossy audio all the time and probably don't think twice about it. Odds are, you listen to 128 VBR regularly without even knowing it and think it sounds fine.

I'm sorry, but I have to respectfully disagree with you. Scientific evidence leans more towards the other side of things, there is no actually proof that states that every single high quality lossy file is exactly the same to every healthy human ear as its lossless counterpart. You can link whatever you'd like or tell me anyone's opinion, but there is not a well funded research organization populated with well suited audio technical engineers that have ever had enough proof to end this discussion. That is why it is even an issue to this day, why do you think anyone brings this up on a daily basis? We do not debate if green is the same color as red, even though my friend who is color blind does in fact see those two colors as the same exact color. I'm not exactly saying that your opinion is wrong bigshot, I am just saying that there is a chance that what you believe so passionately about could possibly have holes in it. I never claimed and probably will never claim that lossless is always better to my ears than lossy, I just think that the possibility exists. Im a very reasonable person. Instead of taking a chance (no matter if that chance is the size of a fermion), I choose a codec that has literally no room for debate in and of itself, and that does not affect me negatively in anyway.

This debate can go on forever, at the end of the day you have to be reasonable. Like I said, I believe both can be right, but not one, and we know which that is. I am not against anyone who disagrees with me, but completely looking over the possibility doesn't make logical sense. We could get hit by a meteor at any moment that could wipe out our entire race (in fact we are past due for that to happen), would I bet everything I had that it was going to happen in my lifetime? No I wouldn't, but would I pretend to know for a fact that it couldn't happen? Of course not, that is very possible. Astronomers technology and research is only so advanced right now, much like sound scientists and audio research development.. there is no proof right now saying that it will happen or won't happen, but any sane astronomer would never completely dismiss a possibility unless there was firm evidence backing it up, but there isn't. I think that story can closely relate to one another, because we cannot overcome echoic memory or many other factors which are holding us back from knowing this topic to the point of being unable to debate. Just as we cannot overcome telescope technology and object trajectory in being able to view everything around us so far away that we can predict a century's worth of exact collisions, we only have the ability to predict and understand possibilities. (Until of course you have a smaller time frame to work with - as far as astronomy goes).

Maybe that was a bad way to compare, but at least you understand where I am coming from, saying my way or the highway is the only thing that I will never understand, you can't just dismiss everything because of your tests, or any other persons tests for that matter. In the end no one person can extensively test thousands of albums while overcoming echoic memory. Which makes tests not 100% accurate, only confirms how small a difference can be, if there is one at all.

What I am trying to say is, acknowledge chance, this is not something every logical person can agree on, or they would. There isn't enough proof out there to stop us from talking about this right now, I for one will completely stick to my opinion that being one hundred percent "safe" and future proof is better than not. Either way, you can't say over and over that you are correct and just move on, without accepting some sort of possibility and reason. I will never that my way is better, because I could be the one wasting time here. I don't keep my files on 500 read 500 write solid state drives, there is no way. I keep mine on magnetic hard disks. The transfer rate is terrible compared to flash.. Lossless audio, even FLAC, definitely takes up more space. It is compatible on my portable devices, so that isn't an factor, but the others are.

I without a doubt think about it when I listen to internet radio, youtube, etc. I knew way before this discussion that youtube, at its very best possible quality, uses 192 aac. For that very same reason, I avoid youtube as a source for music, because most of the time it sounds terrible. I sometimes listen to internet radio for something refreshing and to introduce me to new music, not because I am seeking the highest possible quality. It isn't like I run from lossy audio, I just avoid it if I can.

Please listen to what I am saying.

Thank You.

 

[Dillan]

.

 

[logicPwn]

I thought the main difference was in the cutoffs of frequencies, for compressed (why it's called lossy I though)? Then the bitrate of different codecs, where VBR gives you better compression and higher bit rate when needed?

 

[Achmedisdead]

Quote:

Someone pointing out the deficiencies in your argument in a debate does not equate to trolling.

 
This is true....and your point about MP4 being simply a container is also.

 

[iamdacow]

Subbed

 

[bigshot]

Quote:

I thought the main difference was in the cutoffs of frequencies, for compressed

 
Above 192, there are no frequencies cut off above 18kHz. Above that, the frequencies from 18-20kHz are selectively activated. That is so high, most people can't hear it. Even if they can, auditory masking from lower frequencies would make it inaudible. The only musical instruments capable of creating harmonics in that range are cymbals and triangles.

 

[bigshot]

Quote:

Scientific evidence leans more towards the other side of things, there is no actually proof that states that every single high quality lossy file is exactly the same to every healthy human ear as its lossless counterpart.

 
There are lot of studies that show that. I've done the test for myself and I know for me it's true. If you took the time and effort to do the same, you would know for sure too, and you wouldn't have to quibble with testing procedure.
 
It's not about being right or wrong. It's about knowing the truth. Generally accepted knowledge is that AAC 192 is transparent. I'm really not stating something that should be surprising.
 
Maintaining a backup as a CD disk is better than a lossless rip. It eliminates the possibility of ripping error.

 

[Brooko]

Quote:

The differences between lossless and well encoded lossy can be subtle. You might find small differences in the amount of bass or treble, or in the decay of notes, or in the definition of transients etc. etc. etc. These are not gross artefacts and may pass completely unnoticed in normal listening, and might even be difficult in a side by side comparison. But they can exist and do. If you've only been listening out for gross artefacts then your listening tests are flawed.

Quote:

I did abx tests and went through the process of doing them with low quality settings so as to easily recognise the differences, then using better quality settings and repeating and so on. There are well known killer samples out there which will defeat lossless compression and, more to the point, there are other samples in my particular music collection (and probably yours too) which are equally difficult to compress.

 
 

Anyway if you listen long enough to enough music you do run into stuff that doesn't sound so good. Often this can be verified to be a problem not with the encode but with the CD. But occasionally a problem comes back to lossy compression. I have to emphasise that these kind of problems are very rare but once you know they are there you do tend to wonder what else is lurking in those many 1000s of lossy files.

 

Hi Julian - I'm following this line of thought with interest - and I have also done many, many abx tests - so I know what my personal limits are (they may be a lot more inferior to others due to my age and hearing  ability - I'm fine with that).  Aside from the back and forth with BigShot (the banter between you two is growing old - would be better if you just chose to ignore each other IMHO) - you've raised some interesting points which I've highlighted.
 
I just want to add - that I'm not calling you out - I have no personal stake in either side - my collection is FLAC for archiving /home use and aac for DAP/portable.  What I am interested in is the tracks / killer samples that you've referenced that can be distinguished lossless vs lossy.  You obviously will know this from your own collection + you'll have ref to the 'killer tracks'.  Can you list them please.  I'd love to abx them myself (FLAC vs aac256) + get a couple of members on this board to do the same.  I know a couple who have very HQ gear + very discerning ears and also the technical knowledge of compression (ie what to look for).
 
This is Sound Science - so if you can name the files, and we can run some tests, it should be interesting to stop the debate and prove your point.  If you could list the tracks (+ year of recording so we get the right master) & also the time-stamp that you noticed the artifacting it would be great.  Many thanks.

 

[Achmedisdead]

Quote:

 
There are lot of studies that show that. I've done the test for myself and I know for me it's true. If you took the time and effort to do the same, you would know for sure too, and you wouldn't have to quibble with testing procedure.
 
It's not about being right or wrong. It's about knowing the truth. Generally accepted knowledge is that AAC 192 is transparent. I'm really not stating something that should be surprising.
 
Maintaining a backup as a CD disk is better than a lossless rip. It eliminates the possibility of ripping error.

For you it is true. For me it is true. But there are a few people out there that do actually hear the difference, so you can't say it is an absolute truth. 
 
And CD disks don't last forever either. I dug some of mine out after years of not playing them and they actually had little holes in them, that I could see through! So IMO it's best to keep the disk and a lossless rip as archive, even if you only are listening to your lossy conversions as many of us do. 

 

[Dillan]

How can you know the truth though. How can you as a person, absolutely know the truth. How? Because you tested a few songs? Because you tested a few hundred songs? Did you test 10,000 songs and with each and every one, remember exactly every little detail that you heard - enough to be absolutely sure without any doubt in your mind that it was the exact same as its lossy counterpart? Can you sit here and tell everyone that your opinion and your testing or anyone else testing is so scientifically impossibly undeniable concrete that we can just forget about this and go with your methods? Testing only goes so far, human memory is the reason it can not be the sole basis of making your opinion a fact.

A possible electronic or music instrumental note at around 19khz that lasts for a few miliseconds, hastily climbing to beyond possible hearing.. is it possible for that less than a second occurrence to happen on this guys lossless file, but not on that guys lossy file. No matter the method of obtaining the lossy encoding?

It isn't about being right or wrong, I agree, there isn't enough evidence. Refer to my longer post when I say again, I am very realistic, I choose to be safe and ready for perhaps new codecs in the future or maybe for that quality difference possibility. Either way I don't have to worry because I am losing basically nothing having an open mind and a safe method.

 

[Dillan]

For you it is true. For me it is true. But there are a few people out there that do actually hear the difference, so you can't say it is an absolute truth. 

And CD disks don't last forever either. I dug some of mine out after years of not playing them and they actually had little holes in them, that I could see through! So IMO it's best to keep the disk and a lossless rip as archive, even if you only are listening to your lossy conversions as many of us do. 

Wow. Reading that made me let out a sigh of relief for some reason.

 

[Brooko]

Quote:

It isn't about being right or wrong, I agree, there isn't enough evidence. Refer to my longer post when I say again, I am very realistic, I choose to be safe and ready for perhaps new codecs in the future or maybe for that quality difference possibility. Either way I don't have to worry because I am losing basically nothing having an open mind and a safe method.

 
I agree with what you are saying Rem1x - but perhaps we could put the shoe on the other foot though - and get you or Julian to point to any track (all we need is the album, track, year of recording, and publishing company - so we know we have the correct mastering).
 
Then we get the CD ripped (I'll gladly buy a copy just for this exercise) with EAC & accurate rip, transcoded with Nero-aac 256 & also with latest Lame mp3-320, and then we can get the samples up for everyone to have a go at it.
 
I understand what you are saying - and in theory you are right.  But it's dead easy to prove your point.  Rather than us having to abx every track in our library - you/Julian simply name 2-3 tracks in yours where you know/have heard artifacts being introduced.  Then we can all say ahhhhhh - and move on with knowledge gained

 

[streetdragon]

Dear bigshot, please do realize that not all the treble in the world are only caused by cymbals.
There are other things too that produce high frequencies.

 

[Dillan]

I agree with what you are saying Rem1x - but perhaps we could put the shoe on the other foot though - and get you or Julian to point to any track (all we need is the album, track, year of recording, and publishing company - so we know we have the correct mastering).

Then we get the CD ripped (I'll gladly buy a copy just for this exercise) with EAC & accurate rip, transcoded with Nero-aac 256 & also with latest Lame mp3-320, and then we can get the samples up for everyone to have a go at it.

I understand what you are saying - and in theory you are right.  But it's dead easy to prove your point.  Rather than us having to abx every track in our library - you/Julian simply name 2-3 tracks in yours where you know/have heard artifacts being introduced.  Then we can all say ahhhhhh - and move on with knowledge gained

I completely respect how you are coming at this. However, I have never done an extensive ABX test, only a blind test where I couldn't get through it. I got too frustrated. I would think, yea that sounds like this, play it again, then it would sound like that. It was almost as if my head was making things up. I definitely couldn't put what something sounded like in my head, and play it back again comparatively against something else. Maybe it is just me, but testing in such ways just doesn't work for me as a forgetful human. Even if i could stand here today and give you the information needed that worked for me, hoping it would be the same for you, it still wouldn't be the same situations. Not even to bring gear into the equation, we are two different people. I am under-sensitive to bass, and can not hear over or under some peoples personal frequency response ranges. You see what I am saying?

I am not trying to dodge your request at all. I have never said I personally have concrete proof to back anything up. I just have always accepted the possibilities. Which is why I can collect lossless audio and not have to kill myself with tests. Because believe me.. me being as obsessive over things as sometimes I am.. I wouldn't honestly get any joy out of listening endlessly to music with an analytical perspective. I have found much more joy with researching and reading other people's test results. Other people claim to have heard differences.. Of course if you want me to link you some 128 pieces of beauty, I can do that (Because I can in fact feel confident with almost anyones ears in those situations)

Either way, like I said I never claim to have concrete evidence, I only actively respect the possibility, and will not give up on my research, I can promise you that!

 

[Brooko]

No problem - we are cool - and mutual respect on my part.
 
In a lot of these debates - it comes down to a lot of rhetoric without much substance.  This was simply my way of asking those saying there is a difference to prove it.  It's a lot easier for them to produce a track (any track) where properly transcoded a difference is audible.  If someone can do that - then the debate stops.  It's a lot harder for the other side to prove that every track has no difference - there is no way we can do that.
 
Also - hearing ability is unique to each person - so that is a massive factor.  I'll bow out now.  I do hope Julian comes up with a track - as I'd dearly love to know if I can actually tell a difference.  I suspect I can't (minor but permanent tinnitus).
 
Take the test if you ever get the time Rem1x - it actually is an eye-opener.  But in the meantime don't change what you're doing with FLAC.  For archiving - lossless is the the only way to go IMO.
 
Peace all.