IPB

Welcome Guest ( Log In | Register )

3 Pages V  < 1 2 3  
Reply to this topicStart new topic
Does adding -q0 to -b320 improve quality in LAME 3.99.5?, [was “Lame Q Switch - definitive answer needed”/TOS #6]
greynol
post Dec 21 2013, 15:44
Post #51





Group: Super Moderator
Posts: 10000
Joined: 1-April 04
From: San Francisco
Member No.: 13167



QUOTE (Roscoe @ Dec 21 2013, 05:23) *
My encoder disagrees:

Did you take the time to read the thread and follow the links that were given before posting?

Specifically, you should have seen this:
http://www.hydrogenaudio.org/forums/index....st&p=850682


--------------------
Your eyes cannot hear.
Go to the top of the page
+Quote Post
Roscoe
post Dec 21 2013, 16:25
Post #52





Group: Members
Posts: 17
Joined: 31-October 07
From: Hungary
Member No.: 48369



QUOTE (greynol @ Dec 21 2013, 15:44) *
Did you take the time to read the thread and follow the links that were given before posting?

Yes, I did.

I've read your comment but we are of different opinion. You say it doesn't worth improving quality beyond transparency, I say it actually does. That is why many people use lossless formats, or ogg and aac with average bitrates above 320kbps.

I've also read the -q document linked earlier and it's a little bit unclear, it's inadequate to settle the question of this topic. In my opinion LAME developers should write a lot more detailed documentation than this to bust every myths and prevent topics like this.

This post has been edited by Roscoe: Dec 21 2013, 16:33
Go to the top of the page
+Quote Post
includemeout
post Dec 21 2013, 16:44
Post #53





Group: Members
Posts: 227
Joined: 16-December 09
From: Maringá, Brazil
Member No.: 76067



QUOTE (Roscoe @ Dec 21 2013, 13:25) *
(...)You say it doesn't worth improving quality beyond transparency, I say it actually does. That is why many people use lossless formats, or ogg and aac with average bitrates above 320kbps.


Congratulations! That's by far the most literal definition of placebo effect I've ever seen.

QUOTE
(...)In my opinion LAME developers should write a lot more detailed documentation than this to bust every myths and prevent topics like this.


Shouldn't we, first of all, be thanking Lame developers (and for that matter most other format's) for a great piece of software that's been adequately serving us non golden-eared mortals for more than a decade, before starting to slag them off?

This post has been edited by includemeout: Dec 21 2013, 16:48


--------------------
Listen to the music, not the media.
Go to the top of the page
+Quote Post
greynol
post Dec 21 2013, 16:53
Post #54





Group: Super Moderator
Posts: 10000
Joined: 1-April 04
From: San Francisco
Member No.: 13167



I never said (nor would I ever say) it isn't worth improving quality beyond transparency because that would imply that I believed such a phenomenon exists.

With lossy encoders you cannot improve sound quality beyond the point of transparency by any objective means. To argue that this is opinion misses the point of perceptual encoding. Divining "better than transparent sound quality" as the reason for using lossless or overkill settings doesn't change the landscape. Even if it is the reason many people do what they do, it doesn't mean they are right or even justified in doing so.

Regarding the q switch as it relates to VBR vs. CBR, if you do a little more research you will find they use different algorithms. As such, there is no point in equating a specific q value in one to that in the other.

I apologize for throwing that link out like I did since it doesn't address the point I thought it was going to address. I am the one guilty of not reading thoroughly enough.

This post has been edited by greynol: Jan 14 2014, 23:39


--------------------
Your eyes cannot hear.
Go to the top of the page
+Quote Post
Roscoe
post Dec 21 2013, 17:05
Post #55





Group: Members
Posts: 17
Joined: 31-October 07
From: Hungary
Member No.: 48369



QUOTE (includemeout @ Dec 21 2013, 16:44) *
Congratulations! That's by far the most literal definition of placebo effect I've ever seen.

This is not black or white. If it were, record companies would provide only mp3 discs and lossless audio and placebo formats would become extinct. What about editing and transcoding?

QUOTE (includemeout @ Dec 21 2013, 16:44) *
Shouldn't we, first of all, be thanking Lame developers (and for that matter most other format's) for a great piece of software that's been adequately serving us non golden-eared mortals for more than a decade, before starting to slag them off?

Yes, we should be and actually we are, but objectively more detailed documents are still needed. I didn't intend to slag them off or something like that.
Go to the top of the page
+Quote Post
greynol
post Dec 21 2013, 17:18
Post #56





Group: Super Moderator
Posts: 10000
Joined: 1-April 04
From: San Francisco
Member No.: 13167



QUOTE (Roscoe @ Dec 21 2013, 08:05) *
transcoding

Transcoding mp3 has been fairly well documented as being a very bad idea.

Regardless (and assuming this is still on-topic), saying that q0 improves sound quality when transcoding requires objective evidence. Do you have any?

This post has been edited by greynol: Dec 21 2013, 17:19


--------------------
Your eyes cannot hear.
Go to the top of the page
+Quote Post
includemeout
post Dec 21 2013, 17:24
Post #57





Group: Members
Posts: 227
Joined: 16-December 09
From: Maringá, Brazil
Member No.: 76067



QUOTE (Roscoe @ Dec 21 2013, 14:05) *
This is not black or white.


I'm sorry you've lost me here.

QUOTE
If it were, record companies would provide only mp3 discs and lossless audio and placebo formats would become extinct.


Since transparency does not apply to lossless encoding whatsoever, why bringing that up? To compare oranges and apples!?


QUOTE
What about editing and transcoding?


Eew! Editing what? Lossy files!? That's definitely a no-no!! And again, if you by any rate, meant lossless encoding, please refer to my reply above.



--------------------
Listen to the music, not the media.
Go to the top of the page
+Quote Post
Roscoe
post Dec 21 2013, 17:33
Post #58





Group: Members
Posts: 17
Joined: 31-October 07
From: Hungary
Member No.: 48369



QUOTE (greynol @ Dec 21 2013, 17:18) *
Transcoding mp3 has been fairly well documented as being a very bad idea.

I know.

QUOTE (greynol @ Dec 21 2013, 17:18) *
Regardless (and assuming this is still on-topic), saying that q0 improves sound quality when transcoding requires objective evidence. Do you have any?

I didn't claim q0 does inevitably improve transcoding. I'm just saying inaudible quality improvement could be useful in general, for example in the case of transcoding. There are situations when you have to transcode mp3, for example you don't have the original source still you are providing low bitrate mp3 stream from lossy source having higher bitrate.
Go to the top of the page
+Quote Post
greynol
post Dec 21 2013, 17:42
Post #59





Group: Super Moderator
Posts: 10000
Joined: 1-April 04
From: San Francisco
Member No.: 13167



QUOTE (Roscoe @ Dec 21 2013, 08:33) *
I'm just saying inaudible quality improvement could be useful in general

"Could be useful in general" doesn't cut it in a discussion that was created for the sole purpose of obtaining a definitive answer. Furthermore (and again!), there is no such thing as inaudible quality improvement. Improvement necessarily implies audible difference, which requires objective evidence in the way of double-blind test results.

QUOTE (Roscoe @ Dec 21 2013, 08:33) *
There are situations when you have to transcode mp3, for example you don't have the original source still you are providing low bitrate mp3 stream from lossy source having higher bitrate.

You have the lossless source to create a bloated mp3 and then you don't? Please explain why this is a worthwhile scenario.

If you know of any professional service that archives to 320kbit mp3 or employs other such stupid practices, please share so that we know what to avoid.

This post has been edited by greynol: Dec 21 2013, 20:56


--------------------
Your eyes cannot hear.
Go to the top of the page
+Quote Post
matthaeus
post Dec 21 2013, 23:56
Post #60





Group: Members
Posts: 3
Joined: 17-December 13
Member No.: 113382



Did some tests recently using 3.99.5 on this sample:
http://www.hydrogenaudio.org/forums/index....mp;#entry853181

File sizes are the same:
1 200 586 bytes / -b320
1 200 586 bytes / -b320 -q0

But after using mp3packer (without the -z):
1 199 024 bytes / -b320
1 198 463 bytes / -b320 -q0

File with -q0 seems to be smaller here, probably because a better psychoacoustic model removes more information. It is completely the opposite of "inaudible quality improvement" since -q option is to increase perceived quality only, and will not reduce mathematical difference.

Go to the top of the page
+Quote Post
greynol
post Dec 22 2013, 00:57
Post #61





Group: Super Moderator
Posts: 10000
Joined: 1-April 04
From: San Francisco
Member No.: 13167



I'm not sure how to make out the opposite of what is an imaginary concept.

Did you perform a double-blind test and determine the difference was audible?


--------------------
Your eyes cannot hear.
Go to the top of the page
+Quote Post
peteharrison86
post Jan 6 2014, 23:59
Post #62





Group: Members
Posts: 9
Joined: 18-November 13
Member No.: 112296



Is anyone able to ask the developers if q0 is still in an experimental stage of development or if it is now stable / 'safe'?
Go to the top of the page
+Quote Post

3 Pages V  < 1 2 3
Reply to this topicStart new topic
1 User(s) are reading this topic (1 Guests and 0 Anonymous Users)
0 Members:

 



RSS Lo-Fi Version Time is now: 18th April 2014 - 07:29