Is Hydrogenaudio losing it's direction?, Going from Hi-Fi -> Mid-Fi Options

[rjamorim]

The issue is that if we encourage with it, low quality musics will be floating around soon, exactly same as what we have seen with MP3 128kbps.

OK, and why are you so afraid of low quality musics floating around? Afraid you won't find APS on Kazaa?


Besides, if anything, my 128kbps tests serve to prove no format is identical to the original at 128kbps - at least, there has never been a 5 score at any of the 12 samples tested, much less at the final plot. So, I'm really having a hard time figuring out your point.

BTW, I suggest you take a look at ff123's last post in this thread. He has a very interesting point that you perhaps missed:
http://www.hydrogenaudio.org/forums/index....ndpost&p=162827

[Floydian Slip]

OK, and why are you so afraid of low quality musics floating around? Afraid you won't find APS on Kazaa?

It's more than just spreading on Kazaa. The world is moving towards digital delivery. If we are not aware from the beginning, the industry will push us buying same music over and over. Also you gotta admit that, online piracy was and most likely will be major reason to make any lossy format so popular.

Besides, if anything, my 128kbps tests serve to prove no format is identical to the original at 128kbps - at least, there has never been a 5 score at any of the 12 samples tested, much less at the final plot. So, I'm really having a hard time figuring out your point.

In your last test it's already been proved. Do you really expect to see a big change just in a 6 month period.

Anyways, if majority of the members think that these repeated tests won't be a major issue in terms of degrading average people's concept of hi quality music then I don't have any problem. But I still believe it would be a good idea to put some sort of disclaimer together with the test results in order to avoid any confusion.

[john33]

I should have thought that these tests are all about preserving the direction of HA.

If the only concern here was audiophile quality at best compression, surely none of us would settle for less than lossless compression.

The whole point is that there are requirements to be able to produce respectable quality at all sorts of different bitrates for different reasons. These tests are all about determining the best codecs and settings to achieve the best possible quality at a range of bitrates. Surely, this is entirely in keeping with HA's mission.

[salpro]

i some time use 96 cbr mp3 resampled to 32kh lame
when i have long trips, i also don't hear the diffrence in a noisy atmosphre like train or street
my personal opinion is that the industry is going towards big storage capacities on optical media 18 gigs on one disc is enough to think that the the size will be secondary compared to the benefits of having the best quality

[DigitalMan]

If these tests are for non-audiophile users then we are doing a disservice by not making them aware that they shouldn't use any of these format  at this bitrate, as they are not transparent. Again if it is targeted at the developers, I don't think this is the kind of test that they need.

There are some fundamental assumptions in your post that underly some of the debate:

1) These bitrates are not / can not be transparent
2) People should not use codecs / settings that are not tranparent

You may want to reconsider both assumptions as they apply to the general music consuming public - I am not certain that they are valid.

That said, my read of your post is that you are concerned about HA accepting mediocrity. This is a valid concern, one that I believe most members share. However, testing the state of the midrange bitrate audio codec art can only serve to increase our knowledge of codecs in general and is very applicable to the real world as Roberto points out.

>>Edit: spelling / clarity.<<

This post has been edited by DigitalMan: Dec 12 2003, 21:07

[sven_Bent]

What i always lovede (and do love) about HA is that it has NO DIRECTION.
It tries to infom the WHOLE truth and both ends of the spectre, so the PEOPLE couild make ther OWN choises, instead of only serving info that benefits there own intereset.

HA always lives by the "if it sound good to you keep doing it... but objectivly you might getter better qualtiy form this and that"
HA always curaged people to do ther own listening test spiced with god objectivly advises, instead of forcing there own preferences down som newbies throat.


BTW it always easier to pick on somebodys work...is much harder to do the "improvede" work youself.

[rjamorim]

But I still believe it would be a good idea to put some sort of disclaimer together with the test results in order to avoid any confusion.

What disclaimer? "Hey, 128kbps IS NOT TRANSPARENT, YOU F00!"

People should be able to infer that from the plots. If they can't, they should get educated elsewhere. My tests are not there to teach people how to read statistical plots.

[fairyliquidizer]

Seasons Greetings!

Ehm I welcome all research as it furthers our understanding. I am interested in how good 128kbps "is" rather than "is believed to be" so let us see.

I only use the format for streaming audio but I am interested in seeing how MP3 stacks up against the newer codec, how much progress (if any) LAME has made against FhG at this bit rate, and finally the results are useful for solid state audio player users (like me!). Although I'd be interested in a study at 160kbps (or there about for this).

Basically an album should average 64Mb in size to make best use of solid state players storage space IMHO (opinion not fact).

Love,
Fairy

[Floydian Slip]

What disclaimer? "Hey, 128kbps IS NOT TRANSPARENT, YOU F00!"

Naah, something like "rjamorim preferred, Apple certified CD quality Audio".

[William]

I am very confused in this discussion.

A listening test that provides information to some people who are interested in, and builds up "knowledge". What's wrong with that?

No matter if it is 64kbps, 128kbps, or 320kbps, these are all knowledges of lossy audio codecs. And these knowledges make HA such a valuable place. Isn't this a good thing?

[Big_Berny]

Yeah, I like this tests too. I daily record two Simpsons-Movies and I want lowest size for those twenty minutes.
ATTENTION IT PROBABLY SOUND TERRIBLE FOR YOU:
I user the Lowest-Profile of Nero (AAC+) which has a bitrate of about 40kbits! But for the simpsons it's good enough! I don't hear to the soundquality all the time, but if you lokk my movies you won't hear that it's 40kbits, believe me! And for the video I use 280kbits! This sound terrible too, I know. But I did some tests with different codecs at this bitrate and I found out that RV9 is the best for me.

Back to Topic:
If you record so much movies from you TV you need low bitrates. My Simpsons-Movies take about 60MB and they are really good quality (not like kazaa-ones).
You see, each of us has different uses and each of us needs different bitrates. So let us test each codec for each purpose.

Big_Berny

[Dologan]

Most arguments against Floydian Slip's post have already been raised, but anyway...

WTF do you think HA.org is? A proselytist movement for transparent quality? Yes, most HA members do want to find out about absolute "transparency without compromises", as you put it; but where did you get the idea that that was the only thing HA.org was after? Even Dibrom fails to get your point. I can assure that many of the "non-plus-ultra" members here are using mid-bitrates for certain purposes, and since those bitrates are seeing the most change, why shouldn't there be tests about them? Why shouldn't anything that provides useful, objective information be allowed? As has already been stated, transparent settings have not seen any major change in at least a year, if the only purpose were to talk about those, HA.org would be dead and deserted for quite a while already.
If you want to be a "transparent audio quality crusader", then be it, but just don't think HA.org's goal is to be at the front of that crusade. HA.org is not an active movement, but a discussion forum, so it doesn't have a "goal". If someone wants to find out more about audio encoding, this is the right place to come and find the truth, but HA.org purpose has never been to go out there and force "quality" down people's throats nor campaigning to rescue people from Xing's evil clutches or their own ignorance/deafness/dumbness. If they are interested, let them come and we will provide information, but we aren't world's daddies to go around protecting people from making uninformed assumptions.
Anyway, why are you so worried about what people might imply from Roberto's tests? HA.org impact on the average listener has never been so great to turn the tide, since there are mountains of old WMA files being played every second, despite WMA being one of the most ferociously attacked codecs on HA.org. It's the big companies which really lead the market and only very few of them do actually care about what is said here.

This post has been edited by Dologan: Dec 13 2003, 17:34