Well, I decided to run a test in order to give our skeptics some evidence.
It was a little problematic since WMP is incapable of correctly ripping discs that start with an extended pregap not to mention that regardless of the existence of an extended pregap, they omit the last frame of data from the last track. I noticed this while trying to get the output from EAC synchronized with the output from WMP in order to compare the rips by checksum. So much for easy.
The disc that I used has an extended pregap of 32 frames. As a consequence, every track ripped from this disc using WMP began 32 frames early and the final track was cut short by 33 frames. You guys can just forget about creating bitwise identical copies using WMP.
The purpose of this test wasn't to uncover WMP's inability rip tracks at the correct location in certain circumstances or its inability to preserve the proper length of the last track in all circumstances. Instead it was to show that EAC can rip discs correctly that WMP can't by providing evidence, any evidence.
First, I ripped a single-file image of the disc and created the appropriate cue sheet using EAC. My drive was configured without a read offset correction. I then ripped the same disc using WMP as individual tracks. From there I combined all the individual tracks ripped using WMP into a single wave file using foobar2000. I then appended the last 33 frames of the EAC image to the end of the newly-created WMP image in order to compensate for WMP's shortcoming that I mentioned earlier. Finally, using the CUE sheet that EAC created, I mounted each image to a virtual drive that is offset corrected in order to let AccurateRip verify the results.
As it turns out, all the tracks tested from the EAC image were accurate. All but one of the tracks tested from the WMP image were accurate. I will include data for two tracks from each image to illustrate where they are the same and where they are different.
From the EAC image:
EAC extraction logfile from 31. March 2007, 12:41 for CD
Dave Matthews Band / Under the Table and Dreaming
Used drive : AXV CD/DVD-ROM Adapter: 4 ID: 0
Read mode : Burst
Read offset correction : 30
Track 10
Filename F:\Dave Matthews Band\Under the Table and Dreaming\10 - Warehouse.wav
Peak level 99.8 %
Test CRC 056874C0
Copy OK
Track 11
Filename F:\Dave Matthews Band\Under the Table and Dreaming\11 - Pay for What You Get.wav
Peak level 99.8 %
Test CRC 4AB5F428
Copy OK
------------------------------------------------------------
Track Ripping Status [Disc ID: 0072d79b-5e0ed322]
10 Accurately Ripped (confidence 8) [6eba055a]
11 Accurately Ripped (confidence 8) [04f4eeb9]
_______________________
All Tracks Accurately Ripped.
From the WMP image:
EAC extraction logfile from 31. March 2007, 12:41 for CD
Dave Matthews Band / Under the Table and Dreaming
Used drive : AXV CD/DVD-ROM Adapter: 4 ID: 0
Read mode : Burst
Read offset correction : 30
Track 10
Filename F:\Dave Matthews Band\Under the Table and Dreaming\10 - Warehouse.wav
Peak level 99.8 %
Test CRC 056874C0
Copy OK
Track 11
Filename F:\Dave Matthews Band\Under the Table and Dreaming\11 - Pay for What You Get.wav
Peak level 99.8 %
Test CRC 1D2022EC
Copy OK
------------------------------------------------------------
Track Ripping Status [Disc ID: 0072d79b-5e0ed322]
10 Accurately Ripped (confidence 8) [6eba055a]
11 ** Rip not accurate ** (confidence 8) [04f4eeb9] [df2d3834]
_______________________
Track(s) Accurately Ripped: 1
**** Track(s) Not Ripped Accurately: 1 ****
Track(s) Not in Database: 0
So there you have it, evidence of ripping differences between EAC and WMP. The drive I used, BTW, is a PX-716A by Plextor. I think it qualifies as "a good cd writer."