Monday, September 03, 2007

iTunes, Exact Audio Copy or dBpoweramp?

Last Friday my wife invited some friends and colleagues over and a good time was had by all. That is until I put some music on and we sat out on the patio enjoying the late evening sun, and someone moved his chair to sit next to me.

He said he was moving his seat to get the best audio location. My heart sank – an audio bore. I smiled politely and was pleased when he complimented me on the sound quality. We went through my stereo components and he approved them, although he had ‘auditioned’ my speakers and had gone for a more expensive model. Then the question of CD player.

Actually, I explained, this isn’t a CD player – it’s my iPod. There was a pause. One of the new video iPods? Er, no, this one’s about three years old and its done a million miles, looks like it’s about to fall apart, but sounds good.

“You must have used special software to rip your CDs”. Er, no – iTunes. There was one of those pauses where I’d clearly said the wrong thing. I probed his concern and this was his reply.

“Nobody who’s is halfway serious about music quality would use iTunes.”

That was it. Complete certainty despite the evidence of his own ears. What would be acceptable ripping software? I was given two names – Exact Audio Copy (EAC) and dBpoweramp – which could be relied upon due to the thoroughness with which they read CDs. As my guest explained CDs played via a computer can result in bad CD readings and only advanced software such as these two can read with the accuracy a serious music lover needs. As we later said our goodbyes I thought I’d test that view.

On Saturday I downloaded both applications and set them up on one of my computers. When I set up podServe I’d tried other ripping software so this was a good opportunity to try these, in the case of dBpoweramp re-try. I chose at random five of my own CDs to compare. I wanted to primarily test ripping quality to see if these systems produce better sounds than iTunes.

I started by ripping five CDs into iTunes, stopwatch in hand. Timings carefully noted I repeated the process using EAC. Or at least tried to. EAC is the most confusing product and it took a long time to get a working ripping system. I used the most accurate ripping configuration and eventually the five CDs were added to my hard drive.

dBpoweramp had a more forgiving interface, so I was able to start ripping more quickly. Broadly you can opt for speed of ripping or high accuracy. Thanks to a more communicative interface I could see what had happened when I ripped the CDs. Using the highest accuracy settings each CD had been ripped properly (as was the case with EAC). However it had taken considerably longer to do so – slightly over four times longer.

So I had three sets of rips – iTunes, EAC and dBpoweramp. Bit rates and file formats were the same. Time to sit down and listen for a couple of hours. Did the audiophile rips improve on iTunes? Well not to my ear. As far as I could tell they were identical.
Finally I thought I’d test dBpoweramp’s claim to offer fast ripping. So I switched from great accuracy to speed. Five CDs later I had another set of rips and two notable red flags. On the last tracks of two CDs (The Rolling Stones – Stripped & The Beatles – No 1s) dBpoweramp was flagging ripping errors. What about the speed though? dBpoweramp was marginally slower than iTunes but only by a few seconds. However dBpoweramp got its album data more quickly then iTunes (which was taking a leisurely 15 seconds or more to query CDDB) and at the same time it located the album art. As you may know, iTunes gets its album art in a second pass and only then if its on iTunes.

To be fair to dBpoweramp the fast rips of the good tracks sounded every bit as good as the accurate rips. But there are two tracks where the sound is distinctly flawed, albeit flagged nicely for the user. Given no significant speed advantage I can’t see the point of using dBpoweramp above iTunes.

In this test I found that iTunes was both fast and accurate. It ripped all tracks properly. It ripped all tracks a little faster than dBpoweramp in sprint mode and much, much faster than either EAC or dBpoweramp in their secure ripping modes. And crucially my iTunes tracks sound just as good as the others. Roll on the next home social.

2 comments:

Anonymous said...

I know this is an old post but I feel that it needs a comment.

Your "audio bore" friend has imparted a common misconception on you. This misconception is that the ripping / encoding software can affect the quality or "sound" of the resulting audio file. - This is not correct, hence the reason why you could not hear any difference in sound quality when comparing the files created by the 3 programs.

Digital audio like the type on CD's and in MP3's is a stream of binary data (1's and 0's). The ripping software simply reads the stream of 1's and 0's from the CD and writes it to the hard disk. Leaving you with an exact copy of the data on the cd which will sound exactly the same as it does on the cd. In theory all three programs should write out exactly the same stream to the hard disk.

If that's the case then why shouldn't we use (the much faster) iTunes? Well.. CD's can have scratches which makes reading them accurately more difficult, also your CD-ROM will occasionally read the data incorrectly even if there aren't any scratches. These reading errors can lead to glitches or clicks in the audio or they may not lead to any detectable audible artefact at all. The point of EAC or dbpoweramp is that they try to detect when a reading error has occurred and attempt to fix it - iTunes does not. On the whole EAC and dbpoweramp as very good at preventing reading errors but this is at the sacrifice of speed as you found out.

The problem with iTunes is that it doesn't even warn you that a reading error has occurred. Because of this, you probably have hundreds of errors in your mp3 collection not all of them audible but some will be.

The fact that your first rip on the fastest setting with dbpoweramp resulted in errors is a sign of just how unreliable ripping at these speeds can be. This is nothing to do with problems in the software its all down to your CD-ROM drive. The difference is that dbpoweramp is gracious enough to detect errors and warn you about them.

Anonymous said...

No No No...
A CD, if ripped to Mp3, will not be bit for bit an exact copy on your computer of what is on the CD.
An Mp3 file is smaller than the CD source because the ripping software will throw away select parts that they think we cannot hear. This is why you have different mp3 codecs; LAME, FhG, and what ever iTunes uses.
Yes.... older versions of LAME produced mp3s sound okay, but not as good as the mp3s made by the newer versions.