I'm still in my learning phase about encoding and decided to encode from a purchased CD today.

I used EAC v0.95 prebeta3 and LAME 3.90.3 with the --alt-preset standard option on the command line. I also set the bit rate to VBR 192kBit/s. I used Winamp v2.81 with NO EQ or effects to listen.

I have to say, my .mp3 files sounded great from this encoding!

However, I zoned in on one song off the CD and encoded it in the following ways:

-using --alt-preset standard and using a bit rate of VBR 192kBit/s
-using no preset switch and using a bit rate of VBR 192kBit/s
-using no preset switch and using a bit rate of CBR 192kBits/s

Honestly, I heard no audible differences between each of the 3 .mp3 files.

Therefore I have to ask, why use any of these preset switches? What are we gaining with using them and should I hear whatever is to be gained? Should any cymbals sound more natural, should my drums thump any harder and sound more clear, should any vocals sound like the singer is standing next to me???

Help me to understand this because I'm not getting the hoopla about using presets, or VBR for that matter....

Thanks.

Hooray, my second post on this board

VBR/alt-preset is good because:
- The filesize will be smaller when the music is easy to encode
- when music is hard to encode the bitrate is increased so these part of the music will sound good too
- thus, a constant good sound quality will be reached

CBR (and ABR) is bad because:
- You might be wasting a lot of bits because you set the bitrate to high
- The quality is low because you set the bitrate too low
- Thus, some part of the music will sound crappy while other parts sound ok

You will sooner or later start to hear the artifacts in cbr or low bitrate mp3s, just listen carefully and compare the mp3s to the original cd.

Most likely it's because you don't have EAC setup correctly. When you use --alt-preset standard, you don't need to specify at bitrate like 192kbps. Have you checked out the links in this thread?

http://www.hydrogenaudio.org/forums/index....=ST&f=15&t=203&

Especially the Tutorial For EAC.

Rob

You will learn to tell the difference in time. It also depends on your listening equipment. I used aps for my collection and everything I listened to was transparent to me. When I got my Creative Zen I was able to notice some slight artifacts in certain tracks. At that point I decided to re-encode my collection with ape. Setting a bit rate and using aps, ape or api is redundant as LAME or EAC (don't know which) will ignore the other switches related to bitrate when using aps, ape or api. It is usually considered the best use of disk space to use a VBR method to encode. This is because it makes the most use of the bits. It uses more where it needs to and less where it does not. This opposed to using a fixed number of bits that may or may not be adequate for that frame. It's not a night and day difference in file size, but it will add up. As you read and search more here you will pick up on things like this. Mostly, I just learned to trust the experts here. Their ears are better than mine and they have more expeirence. If it were me, and this is what I did, just use aps like everyone says to and in time you will gain understanding as to why as you learn to hear artifacts. Good luck.

Besides, it's not always possible to distinguish different encodes at that high a bitrate without good equipment, hearing and training.

edit: hm... beat me. ;-)

--alt-preset standard choses the bitrate itself. How did you tell EAC/lame to use "a bit rate of VBR 192kBit/s"?
What were the other commandlines you used an what options did you chose in EAC's Compression options?

You will only notice differences between --alt-preset standard (and higher) and e.g. CBR 192 sometimes if you're trained to hear artifacts. Lossy encoding is no equalizer, so if you can hardly tell what is the CD and what the mp3 with CBR, there's no room to improve for --alt-preset standard. Otherwise it'd sound better than original ...

For both, artifact training (including howto perform ABX tests) and recommendable lame settings the FAQ is a good starting point.

...and a good cd sample. If your sample is low fidility (mono-ish and little high frequencies) you will most likely hear less difference.

Now, if your alt-preset average bitrate is high (a lot more than 200). Then you should have a nice test sample. You can check average bitrate it in Winamp at the mp3 info.

Or even better, you should use a test sample which float around on this board, which will make it very easy to hear differences.

Congratulations on one major point - you know to trust your own ears! You're right - you can encode an mp3 a lot of different ways and they can all sound about the same, especially with "average" equipment. That's perfectly okay, and doesn't mean you're doing anything wrong.

However, I think there is one quirk with EAC that gets a little confusing. When using a "User Defined Encoder" and specifying your options in the "Additional command line options", I don't believe that the bit rate drop-down box settings have any effect. This is good, as it means you have total control of your settings in the "Additional command line options".

Wow everyone, thanks for all the feedback and advice! Where do I start...ok...



Yes, I followed that tut exactly before encoding. I made changes based on what was recommended (and what my CDROM drive could handle.)



I definitely thought about this and assumed that my audio setup is hindering me from any audible differences I should have heard. I'm using an SB Live MP3+ soundcard with cheap (Numark) powered monitors.



Thanks for pointing this out because I wasn't sure if the preset chose the bit rate or not. If you look on the "External Compression" tab of the Compression Options in EAC, you'll see Additional command line options: then below that you'll see bit rate:. To the right of bit rate there's a drop down window. That window shows various bit rates that can be set and it must be set to something (in other words, there's no way to leave it blank). This confused me because again, I wasn't sure which took precedence, the command line settings or this bit rate setting.



No other command line settings were used. In the command line field I entered --alt-preset standard, that was all. On this same tab ("External Compression"), I have "Use external program for compression" checked; the "Parameter passing scheme:" shows "LAME MP3 Encoder." My program path for compression under the passing scheme points to my LAME 3.90.3 executable. The only other items checked on this tab are: "Delete WAV after compression", "Add ID3 tag" and "High Quality." I have no Offset enabled on that tab and the Waveform tab has all options greyed out. That's it...


Why thank you Amadablam!

I guess this unnoticeable change in my files is because of my sucky sound setup.

As had been said many times over and over before, audio equipment has a much less significant role in hearing artifacts than most people would think. If you're using headphones, even if they are a cheap pair and you are listening on a noisy soundcard, this is usually enough to hear problems to at least some degree. The rest is mostly training and/or natural ability.

The ability to hear artifacts has surprisingly little to do with your listening equipment. The biggest factor in whether you'll hear the difference between --aps and another high-bitrate encode is the sample of music you're listening to. It's often hard to guess whether a sample will be troublesome to encode or not. Some music will sound audibly different no matter what bitrate you use, while other music will sound perfect at 128kbps. Although music with lots of transients (snares, hi-hats, or synthesized drums) certainly aren't the only pieces of music that --aps improves, music samples with transients are good to listen to if you're looking for a good showing from --aps. Check Hydrogen Audio's test samples for castanets, drone_clip, and 41_30 if you want to hear audible improvements in transient encoding with --aps over its VBR and CBR counterparts. Mind you, even these samples might not jump out at you as being audibly different; you will still have to concentrate to hear the difference.

The second biggest factor in hearing artifacts is your experience with listening for artifacts. The more you listen and practice, the better you'll be at picking out artifacts, since you'll know what to listen for.

Finally, (edit: as Dibrom has indeed said many times,) your listening equipment does play a role. You'll need halfway-decent headphones, a good soundcard, and a not-too-loud computer/listening-environment to hear high-frequency anomalies, problems with the noise floor, or other quiet artifacts.

Dibrom and SW, thanks for chiming in....

I do have a decent set of headphones (Sony MDR-V700), so I'll use them also to listen to my mp3s & CDs. Maybe what I couldn't hear through the monitors will now jump out at me on the headphones.



How does one learn to hear artifacts or even know what artifacts to listen for?

Thanks again.

Have a look at this thread (taken from the FAQ) : http://www.hydrogenaudio.org/forums/index....ST&f=16&t=7783&

Hi Pio and thanks for the link....

I read the posts there and tried accessing some of the sound files that I could find, but a few were in flac format and I didn't have a way to play them. I saw a few .wav files mentioned but no link to actually hear them. I'll search for similar posts.

I tried my headphones and encoded more mp3s from a different CD last night. I believe heard what everyone refers to as "artifacts."

But, I don't know if these were artifacts as a result of encoding or just blips that carried over in the recording the music onto the original CD? It was a vinyl mix CD (if I'm not mistaken). I can tell an audible difference from the original CD and the mp3s I made of course.

Nevertheless, the new encoded files I made all still sound generally the same. I encoded from 96kBits/s all the way up to --alt-preset insane.

Free Flac decoder : http://flac.sourceforge.net/
It's in the FAQ too (top right of this page) !

QUOTE ( @ --)

Thanks for pointing this out because I wasn't sure if the preset chose the bit rate or not. If you look on the "External Compression" tab of the Compression Options in EAC, you'll see Additional command line options: then below that you'll see bit rate:. To the right of bit rate there's a drop down window. That window shows various bit rates that can be set and it must be set to something (in other words, there's no way to leave it blank). This confused me because again, I wasn't sure which took precedence, the command line settings or this bit rate setting. [/quote] Let me try to explain this. The drop-down selectable bitrates is only good for getting the estimated filesize on EAC's main window. When you open EAC, you see the track list which has a column that reads "Comp. Size". EAC has no VBR calculator. So, if you select 192k which is kinda the average in -aps, you will get kinda approximated predicted filesizes on that column. This is good to roughly know how much space will your songs need. IT IS ONLY COSMETIC and has no effect on quality. There's already a thread that settled that changing the bitrate on that drop-down DOES NOTHING to the files, quality-wise. Read this, for the full explanation/testing (by the way, this is in the FAQ) rohangc Jun 5 2003, 07:30 QUOTE However, I think there is one quirk with EAC that gets a little confusing. When using a "User Defined Encoder" and specifying your options in the "Additional command line options", I don't believe that the bit rate drop-down box settings have any effect. This is good, as it means you have total control of your settings in the "Additional command line options". This link might help clear up some doubts. This is a "lo-fi" version of our main content. To view the full version with more information, formatting and images, please click here. Invision Power Board © 2001-2009 Invision Power Services, Inc.