So there I was, ripping this classical/jazz CD with EAC, thinking all I had to do was extract to WAV, and then encode with LAME MP3.
HOWEVER, as I played the MP3 file, I noticed some 'clickings' when the trumpet goes up high (basically, whenever anything goes up to the higher frequencies, you get that annoying fuzz...sort of like what comes out when you've got crap $2 speakers and they can't handle it.)
I went back and found out the problem originated from the WAV file, directly after extraction using EAC.
I compared it to the CD.
The fuzz showed up on the WAV file at the exact same spots, but didn't on the CD.

I went back to the audio console, and brought down the volume sliders for everything, and re-encoded multiple times. The problem was still there.

I'm using a Lite-on 16x DVD Drive (165H) for the extraction. I'm sure it's not the drive because I used my Lite-on 24x-10x-40x CD-RW Drive to verify. My comp is most certainly not underpowered, nor is anything else hogging the CPU cycles.


----------------------------------------------------
Below are the configs for the Drive Options in EAC::

EXTRACTION METHOD
===========
Secure Mode:
Accurate Stream feature Checked.
Cache Audio feature Checked.
C2 Error Information NOT Checked.

Nothing else checked underneath

DRIVE
===========
Drive read command:
Autodetect

Nothing else checked underneath

OFFSET/SPEED
===========
Use combined read/write sample offset correction selected.
Value set at 0.

Allow speed reduction during extraction Checked.

Nother else checked underneath

GAP DETECTION
===========
Detection method B selected
Secure selected
----------------------------------------------------


The Normalize option in 'EAC Options' is unchecked because I'm not quite sure how to use it


----------------------------------------------------


In 'Compression Options', I'm using the Windows MSAudio Encoder, and the sample format is 128kBits/s 44,100 Hz Stereo


----------------------------------------------------


This problem only appears when things like trumpets go up high. Everything else has no problem.
How do I solve this, and why is this problem here? What am I doing wrong?

By the way, does everyone here us the 'Windows MSAudio Encoder', or do you download another encoder for better results?


Thanks for your time!

Your EAC settings seem to be allright. Some possible reasons for the problem:

1. I'm not sure if I understood correctly: You tried both - extracting to uncompressed .wav file AND MS Audio compression? (BTW: About what compression to choose: Read the FAQ - e.g. EAC tutorials, there are better possibilities for sure.)

2. Copy protection: Is it possible that this CD is copy protected / does this happen if you extract old CDs too?

3. Computer problem (Soundcard/Driver/...). Burn the extracted files (uncompressed .wav) to CD-R(W). If this CD sounds fine the problem lies here.

do you get any errors on extraction? Is track quality 99.9 or 100%? or does eac report 'suspicious' positions? The disk could be possibly damaged then, and glitches are unavoidable (though frequently i can't hear any problems with such positions). Also you should set 'error recovery' to high in EAC options.

as tigre said, I can't understand if you are using LAME, WMA or just plain wav, so some clarification could help.

Sorry for the confusion!

The MSAudio encoder is in the Compression Option menu. I extracted to uncompressed wave. Therefore, the MSAudio encoder wasn't used at all, I just added it in so everyone could get a better idea of my setup.

I've changed the EAC correction setting to High, have enabled C2 correction, and will re-extract.

I'm just wondering why the problem appears in the extracted uncompressed wave file, but DOESN'T appear when I play directly from the CD.

Ok, I re-extracted. The track quality was 100%. The peak level was in the 93% range.
And the problem is still there.

I still notice the same fuzz when the trumpet goes up high...played the CD again, and the problem doesn't appear.

What's wrong?!?

Does this happen with other wav files or just from that cd? As tigre suggested, try to burn the extracted wav files to a cd-r and listen to it. If the new cd has the same problem then it's an extraction problem (copy protected cd maybe?) otherwise it's a hardware/software (drivers) problem when playing wav/mp3/etc files.

With those EAC results, I bet it's a problem with the computer playing back audio files. In the EAC options, set "use alternate CD play routines", and play the CD with EAC. The problem should be there, and come from the computer/soundcard.

This sounds almost eactly like what happened to me (See here) I ended up sending it to tigre and he fixed it up using Cool Edit Pro. There was no problem on the CD, still isn't. But rips with any program resulted in a pop. It was also a jazz/blues CD, if it matters. I don't think there's was trumpet in my sample though.

The CD is NOT copy-protected.

I did both suggestion - played back on a burned CD, and used the 'Alternate CD Play Routines'.
The CD has the same problem, meaning there was something wrong with the extraction. When I used the 'Alternate CD Play Routines', there was NO ERROR, so definitely, there's something wrong with the extraction.

BadReligionPR and tigre -- When tigre fixed it up using Cool Edit Pro, what was wrong with it? How did you fix it?

Anybody have any suggestions as to why this would occur? lol does the creator of EAC browse these forums, or does anyone talk to him?

what are you using to play your mp3s? if it's winamp, make sure that the equalizer or any DSP plugins are disabled

>>>'lol does the creator of EAC browse these forums, or does anyone talk to him?'<<<

andre occassionally drops by here, but you can find him "full time" on the EAC message board

www.exactaudiocopy.de

Try to lower speed in Options, it's the main difference between 'Alternate play routines' and ripping.

Your description makes me doubt that you have the same problem, but why not give it a try: Use CoolEditPro 2.0 (e.g. full featured time limited trial version)

Noise Reduction -> Click/Pop Eliminator -> Fill Single Click Now

Maybe the automatic "search and destroy" click/pop function works for you (If I understand correctly the problem occurs several times). The single error on BadReligionPR's CD was a "pop", some 100 samples long, in spectral view noticable as full spectrum event, in waveform view characterized by high-amplitude sample-to-sample changes (random-like) that don't occur in the rest of the sample. I deleted the sample, otherwise I'd post a picture.

I have no idea what could be the reason, but from the tests you did so far, you could try as workaround to rip with another drive and/or use some standalone CD player's digital out and capture with some soundcard's digital in as it seems like standalone player's interpolation handles this well.

Sounds like a plan to me Just in case it doesn't work you could use either totalrecorder or NTONYX virtual audio +some recording software cable to capture and record digitally EAC's (flawless) playback output. (And use this as rip or to replace the erroreous positions in the extracted file(s)).

Check that you don't have any normalisation enabled in EAC options, and that you extract "uncompressed".

Then, can you post a small sample on the web ?

argh!! I feel so stupid.
I went to winamp, turned off the EQ, and MOST of the problem dissapeared (the clicks and pops weren't all that audible anymore)

That's good enough for me =)


I'd like to learn as much as possible from this experience, however, so I'm going to go ahead and ask my questions.

Could someone please the statement "Try to lower speed in Options, it's the main difference between 'Alternate play routines' and ripping." to me?
Also, I'm not sure which speed in which options menu this is referring to.

How does normalization work?

What do the 'peak levels' reported by EAC for each extract represent?

Normal playback is through analog cable directly connected from drive to soundcard. Playback using "Alternate play routines" digitally extracted audio data extracted to memory used for playback. It's at 1x speed (on average), because you listen to CDs at 1x speed generally. Ripping is digitally extracted audio data written to disk as fast as possible (or selected).


EAC->Drive Options->Offset/Speed


- Search the sample with maximum sample value (max_value), e.g. 28314
- Calculate normalize_ratio=max_normalize_value/max_value = 32768/28314 = 1,1573... (for normalizing to 100%)
- Multiply every sample value with normalize_ratio.
This will introduce truncation errors = (mostly inaudible)distortion - if it's done at higher resolution (e.g. 24bit) with dither to 16bit afterwards, it will add dither noise. If done several times it will become audible.

But how can the copy on CD have the problem while the original CD haven't ?

Reread the posts thoroughly, please, before you write...
If the original had no problem, then this thread wouldn't exist.

<edit>
It seems that this CD just clipped, and Winamp EQ was overemphasizing this.
</edit>

The way I understand this, the burned CD has problems with normal CD playback which the original hasn't.

If it's a winamp EQ + clipping issue, but the copy is identical, there can't be differences between original CD and copied CD.

For what it's worth, it's generally recommended to disable caching in EAC as well.

You mean to check "Drive Caches Audio Data" - right?