Hello everyone!
I want to convert my FLAC files to AAC, getting the best possible quality. In my researches I found that QAAC is considered the best encoder these days, but I also realized that it has a maximum bitrate of 320, whereas Nero AAC with a quality setting of 1 goes above that. Which would be better to use in that case? I must admit, I have no idea what makes one encoder better than another and am overall a beginner in all of it.
Thanks for any help,
Dennis
I have no idea what makes one encoder better than another
https://hydrogenaud.io/index.php/topic,16295.0.html
http://wiki.hydrogenaud.io/index.php?title=ABX
http://wiki.hydrogenaud.io/index.php?title=ABC/HR
I have no idea what makes one encoder better than another
https://hydrogenaud.io/index.php/topic,16295.0.html
http://wiki.hydrogenaud.io/index.php?title=ABX
http://wiki.hydrogenaud.io/index.php?title=ABC/HR
So basically the notion of one encoder being better than the other is all subjective?
Yes, but whether or not an individual (or group of individuals) has reliably identified audible differences between encoders can be assessed through objective means.
Apart from that though, how big of an advantage is a bit rate above 320? In my opinion, it would be a big one, but since I'm not an expert I thought I'd ask here.
In my opinion, it would be a big one
Before attempting to speculate, let's see if you can provide a ballpark guestimate on the bitrate of this decoded mp3...
Apart from that though, how big of an advantage is a bit rate above 320? In my opinion, it would be a big one, but since I'm not an expert I thought I'd ask here.
That is an absurdly high bitrate for AAC.
In my opinion, it would be a big one
Before attempting to speculate, let's see if you can provide a ballpark guestimate on the bitrate of this decoded mp3...
The tvbr 127 encode with QAAC came out at an average bit rate of 239, the Nero one came out as 363.
Apart from that though, how big of an advantage is a bit rate above 320? In my opinion, it would be a big one, but since I'm not an expert I thought I'd ask here.
That is an absurdly high bitrate for AAC.
I know it might be "overkill" to a certain extent, but I figured better too high than too low. :)
Before attempting to speculate, let's see if you can provide a ballpark guestimate on the bitrate of this decoded mp3...
The tvbr 127 encode with QAAC came out at an average bit rate of 239, the Nero one came out as 363.
He means the bitrate of the MP3 that file came from.
I know it might be "overkill" to a certain extent, but I figured better too high than too low. :)
Then you probably want more like 160-200k VBR.
The tvbr 127 encode with QAAC came out at an average bit rate of 239, the Nero one came out as 363.
That didn't answer my question, though I will tell you that you're far better off not doing what you did with either codec. This doesn't mean you couldn't transcode that mp3 to a smaller aac without any audible quality loss. This also isn't to say I couldn't have given you this sample using an even lower bitrate mp3.
Sorry about any confusion there. The original files are FLAC, not MP3. I have problems deciding whether to use QAAC with the tvbr 127 setting, or Nero AAC with a quality setting of 1, to convert those.
The flac you downloaded was derived as follows:
CD -> WAV -> MP3 -> WAV -> FLAC
Please tell us what you think was bitrate of the mp3.
Doing this does not involve the use of an aac encoder!
I had no idea that FLAC files came from MP3s. Do you have a link to an article that explains it more in detail?
Anyways, the FLAC file is 704k. That is all I know.
flac files can essentially come from anything.
Did you actually listen to the sample? I'm asking you to use your ears, here. Pretend I had simply given you an mp3 file.
I figured that turning an MP3 into FLAC would be counterintuitive, turning lossy into lossless...
Are you saying I should basically just listen to the different files and pick the one I think sounds best?
I gave you just one file and am asking you to guess the bitrate of the mp3 from which it was sourced.
This will go a long way in helping you make sense of any reasonable answer someone might give you in response to your initial post.
I had no idea that FLAC files came from MP3s.
Really? I told you that above.
Ok, once again, my apologies. I had totally overlooked the attached file. My bad, and the source of all the confusion here. I would say the attached file wasn't higher than 96k.
I don't find that answer to be very credible and just underscores the reason why I presented you links on double-blind testing.
Let's try this again...
It sounds better, but I honestly can't say what quality each file is based on the listening. I guess I'm not experienced enough.
Did you perform a double blind test to determine that? If so, I'd like to see a log of the results.
Would someone else be willing to perform a properly controlled comparison, present a log file and comment on any audible differences so that the OP understands the expectations of this community?
I didn't do a proper test with any software. Just listened to the first file you had sent me and then once you had sent me the second, compared them a bit. What software would you want me to use?
foobar2000, using the ABX comparator utility (http://www.foobar2000.org/components/view/foo_abx), as mentioned in one of the links from my first reply (https://hydrogenaud.io/index.php/topic,112788.msg928790.html#msg928790).
Apart from that though, how big of an advantage is a bit rate above 320? In my opinion, it would be a big one, but since I'm not an expert I thought I'd ask here.
I'd do a comparison (an ABX test) between, say, 200kbps and 320kbps AAC. If you can hear a difference let us know.
In my experience 200kbps is already overkill for AAC. Using 320kbps or higher seems ridiculous and at that point you might as well just use lossless.
foobar2000, using the ABX comparator utility (http://www.foobar2000.org/components/view/foo_abx), as mentioned in one of the links from my first reply (https://hydrogenaud.io/index.php/topic,112788.msg928790.html#msg928790).
I did a comparison, I had 7/16 correct, a possibility of guessing of 77%. I didn't save the log file since I wanted to another test, paying more attention to details. In hindsight, I changed my mind. I'll just use my experience from this and the tool(s) for other cases.
Apart from that though, how big of an advantage is a bit rate above 320? In my opinion, it would be a big one, but since I'm not an expert I thought I'd ask here.
I'd do a comparison (an ABX test) between, say, 200kbps and 320kbps AAC. If you can hear a difference let us know.
In my experience 200kbps is already overkill for AAC. Using 320kbps or higher seems ridiculous and at that point you might as well just use lossless.
Thanks so much for your input. I will do those tests in the future when the need arises. At least I now know better how to do such a thing. :)
I applaud you for your effort.
The mp3 that I converted to flac and uploaded was 156kbps (Lame 3.98.4 -V3) and also happens to be what I use for my lossy library*. The second upload was the lossless source that I used. If I had used QuickTime AAC, I would have used an even lower bitrate and would have been every bit as confident that you would not have been able to tell a difference.
So while you may have struggled and perhaps had a difficult time accepting the fallibility of a-priori comparisons, in the end you win by gaining knowledge from the experience and seeing that you don't need to run a lossy codec at its highest settings in order to achieve great results.
People are being pretty conservative by telling you to try 200kbps. For the vast majority of 2-channel content encoded using AAC, no more than 160kbps will be necessary for results that are indistinguishable from the original source (also known as transparent). It is for only for the extreme cases where something significantly higher is necessary. Here is a recent topic inquiring about what constitutes extreme cases:
https://hydrogenaud.io/index.php/topic,112569.msg927898.html
(*) The average bitrate for my library is 171kbps.
I applaud you for your effort.
The mp3 that I converted to flac and uploaded was 156kbps (Lame 3.98.4 -V3) and also happens to be what I use for my lossy library*. The second upload was the lossless source that I used. If I had used QuickTime AAC, I would have used an even lower bitrate and would have been every bit as confident that you would not have been able to tell a difference.
So while you may have struggled and perhaps had a difficult time accepting the fallibility of a-priori comparisons, in the end you win by gaining knowledge from the experience and seeing that you don't need to run a lossy codec at its highest settings in order to achieve great results.
People are being pretty conservative by telling you to try 200kbps. For the vast majority of 2-channel content encoded using AAC, no more than 160kbps will be necessary for results that are indistinguishable from the original source (also known as transparent). It is for only for the extreme cases where something significantly higher is necessary. Here is a recent topic inquiring about what constitutes extreme cases:
https://hydrogenaud.io/index.php/topic,112569.msg927898.html
(*) The average bitrate for my library is 171kbps.
Thank you. I really appreciate it. :)
Thanks as well for the additional link at the end, it will help even further. :)