AAC, AC3 vs MP3 - Want lowest file size

[kinster]

I used to be under the presumption that AAC produced the highest compression but I was just converting and old Xvid DVDRip to x246 video and AAC audio and found out that was not the case.

First I extracted the raw MP3 from the avi and it came out to 86.9 MB. I then used Avidemux to make an x264/aac MP4.

When I extracted the raw AAC file it was 101 MB (14.1 MB bigger).

From some cursory reading, I understand that AAC can give better quality but since I use cheap headphones or small desktop speakers, I doubt I will notice any difference.

So am I correct in my deducting that recoding all my AC3 or AAC encoded videos to MP3 will reduce the file size?

This is an issue for me because I only have two 1 GB hard drives that I backup by using the second as a "mirror" and was just about out of space. I've managed to create 100GB of space by replacing my Xvid's with BDRip's.

EDIT: OK, I just ran a x264/AAC file through AVIDemux using copy for video and MP3 for audio and the MP3 came out to 116 MB but the original AAC was only 30 MB.

What's going on?

[Midzuki]

EDIT: OK, I just ran a YIFY x264/AAC file through AVIDemux

Please read this (http://forum.doom9.org/forum-rules.htm) again.

[microchip8]

I used to be under the presumption that AAC produced the highest compression but I was just converting and old Xvid DVDRip to x246 video and AAC audio and found out that was not the case.

First I extracted the raw MP3 from the avi and it came out to 86.9 MB. I then used Avidemux to make an x264/aac MP4.

When I extracted the raw AAC file it was 101 MB (14.1 MB bigger).

From some cursory reading, I understand that AAC can give better quality but since I use cheap headphones or small desktop speakers, I doubt I will notice any difference.

So am I correct in my deducting that recoding all my AC3 or AAC encoded videos to MP3 will reduce the file size?

This is an issue for me because I only have two 1 GB hard drives that I backup by using the second as a "mirror" and was just about out of space. I've managed to create 100GB of space by replacing my Xvid's with BDRip's.

EDIT: OK, I just ran a YIFY x264/AAC file through AVIDemux using copy for video and MP3 for audio and the MP3 came out to 116 MB but the original AAC was only 30 MB.

What's going on?

It depends a lot on the bitrate. aac will produce better quality at the same bitrate than mp3 or ac-3. For example, mp3 sounds awful at 96kbps and lower while aac sounds good at these bitrates. file size should be almost equal at the same bitrate

About YIFY, please read the rules. Pirated content is not allowed to be discussed here

[foxyshadis]

What's going on is that you can make a lossy compressed file as big or as small as you like. It's not like lossless, where everything's the same; the smaller, the worse it is, but there can be wide differences in how good something sounds. It sounds like you didn't even bother to look at how much you were compressing, you just threw some things at an encoder with defaults and looked at nothing but the size; did you even listen? You should generally keep reducing the size until you can hear distortions, then stay a little above that. Once you've done that a couple of times, you'll know what to choose each time.

The progression from best to worst sounding at medium bitrates (128kbps 2-channel or 256kbps 6-channel) is: Opus->Vorbis & AAC LC->MP3->AC3

If you want to save mucho storage while still sounding ok (48kbps 2-channel or 96kbps 6-channel), AC3 is off the list. Then it goes: Opus->AAC HE->Vorbis->MP3

If you don't care about saving space then just keep whatever lossy format you started with, AC3 or DTS. TrueHD, LPCM, or (if you have a decoder) DTS-HD can be converted to FLAC or TAK to save space, though.

This is in general. The are specific cases where the order shuffles around slightly, but they're rare. In general, to the untrained ear, Opus, Vorbis, and AAC are virtually indistinguishable, and MP3 only slightly worse; you need to pay a lot of attention and care a lot to be able to hear all the differences.

[kinster]

I set them both at 128 bitrate and I did listen. They both sounder the same.

I'm a little confused by your use of "->" with the dash before the >.
Do you mean larger/better > smaller/worse?

Thanks.

[foxyshadis]

It's an arrow. They're listed from best to worst at a given size. As you've seen, the differences are small at 128kbps stereo.

[kinster]

If you want to save mucho storage while still sounding ok (48kbps 2-channel or 96kbps 6-channel), AC3 is off the list. Then it goes: Opus->AAC HE->Vorbis->MP3

Thanks for the explanation but the above has me a little confused. Hoping you can clarify...

If your first rundown you have an arrow pointing from best quality to worst quality.

In the second rundown (arrow pointing from best to worst again if we are going to be consistent) I think you place the best last or is it in fact that Opus gives the best compression?


Thanks!

[tebasuna51]

Opus is the best compression.

[kinster]

Thanks for the clarification.

I just discovered something that I'm sure you are all aware of but I find odd...

If I re-encode 6 channel AC3 to 2 channel AC3 I save 100's of MB's.
When I perform the same function on AAC 6 channel, the 2 channel video is virtually identical in size! It's actually a couple of megs bigger but I guess that's just some overhead from Avidemux.

[detmek]

File size depends on lenght and bitrate. If you used same bitrate for 6-channel and 2-channel audio output will be the same size.

[kinster]

They where the same bit rate. All I change is the number of channels.

[foxyshadis]

If they're the same bitrate, then I don't know why you expect a different filesize. Bitrate and filesize are the same thing (when length is constant).

If the AC3 has a different size, then it's a different bitrate.

[kinster]

Not sure why people wont take my word for it. I'm not lying or making this up. This is what I am seeing and there is zero debate over this. Maybe something is weird with these files and that may give you a satisfactory explanation but I am not seeing it on only 1 or 2 files.

[Groucho2004]

Not sure why people wont take my word for it.
Simple - You're trying to convince us that you have successfully managed to change the laws of physics.

[kinster]

This is getting downright silly now.

Can a mod please close this thread. I believe it has run it's course.

I don't know how much clearer I can make it...

THIS IS WHAT I AM SEEING!!!! I'M NOT CRAZY AND I'M NOT MAKING THIS UP!

unsubscribing...

[Groucho2004]

This is getting downright silly now.

Can a mod please close this thread. I believe it has run it's course.

I don't know how much clearer I can make it...

THIS IS WHAT I AM SEEING!!!! I'M NOT CRAZY AND I'M NOT MAKING THIS UP!

unsubscribing...
Yet you refuse to post any proof like a simple mediainfo log.

[Asmodian]

Bitrate is, by definition, total length / total size.

If they are the same size and the same length they are the same bitrate. If they are different sizes and the same length they are not the same bitrate.

If you compressed a 6-channel AC3 to a 2-channel AC3 and it was smaller you compressed it to a lower bitrate. This is not surprising as usually the bitrate options for AC3 are different for 5.1 and stereo. You did not compress your two AC3 files to the same bitrate.

AAC will let you pick the same bitrates for 6-channel and 2-channel so, of course, they come out to the same size.

Do you simply want us to agree that maybe in one case 2+2=4 but in another 2+2=3 or do you want to understand what happened?

[tebasuna51]

Can a mod please close this thread.

Thread closed by user request.

BTW, you don't need cry, everybody want only help you.