Asunder (like most free apps) use Cdparanoia, so the ripping is as good as cdparanoia can handle.
Now the question is, how good is the current Cdparanoia?
We see this in the announce ML:
QUOTE
Hello all,
I'm pleased to announce the release of cdparanoia 10.2. This is the
first release in several years to involve substantial new development
as opposed to simply 'keeping up' and vetting/adding contributed
patches.
In addition to regression testing and bugfixes, 10.2 includes major
drive cache handling improvements over previous versions. The short
story is that a sizable fraction of new cdrom drives that have
appeared since the last major read model update (in 2001) exhibit
cache behaviors or weirdnesses of which previous cdparanoia versions
are partly or fully unaware. These drives, when they skip, stand a
good chance of having the skip or glitch make it through verification
undetected because cdparanoia was not successfully defeating the
drive's data cache. 10.2 implements an improved caching model that
eliminates all the new cache behaviors I've observed.
IN SUMMARY: 10.2 ADDRESSES RIP CORRECTNESS ISSUES. YOU WANT TO UPGRADE.
10.2 also includes a nifty cache analysis utility via the -A option.
This runs an offline cache analysis of the drive, printing a summary
to the screen and dumping a detailed log to 'cdparanoia.log' in the
event that the test finds a problem, or cdparanoia is acting up in
some other way.
As always, the release can be downloaded from
http://www.xiph.org/paranoia/Cheers,
Monty
xiph.org
Would be nice if there were more tests. I have a Pioneer DVR-111D (L is same with different firmware

) which, according to EAC, caches audio data.
I use Ubuntu 8.10 which comes with cdparanoia 10.0, but the packages for the upcoming 9.04 work perfectly so i installed libparanoia and cdparanoia 10.2 to test.
cdparanoia -A shows this:
CODE
CDROM model sensed sensed: PIONEER DVD-RW DVR-111L 8.29
Checking for SCSI emulation...
Drive is ATAPI (using SG_IO host adaptor emulation)
Checking for MMC style command set...
Drive is MMC style
DMA scatter/gather table entries: 127
table entry size: 131072 bytes
maximum theoretical transfer: 6985 sectors
Setting default read size to 27 sectors (63504 bytes).
Verifying CDDA command set...
Expected command set reads OK.
Attempting to set cdrom to full speed...
drive returned OK.
=================== Checking drive cache/timing behavior ===================
Seek/read timing:
[16:51.65]: 51ms seek, 0.55ms/sec read [24.4x]
[10:00.00]: 56ms seek, 0.61ms/sec read [21.7x]
[00:00.00]: 67ms seek, 0.80ms/sec read [16.8x]
Analyzing cache behavior...
Approximate random access cache size: 139 sector(s)
Drive cache tests as contiguous
Drive readahead past read cursor: 133 sector(s)
Cache tail rollbehind: 128 sector(s)
Cache tail granularity: 0 sector(s)
Cache size (considering rollbehind) too small to test cache speed.
Drive tests OK with Paranoia.
Apparently if this test fails, cdparanoia is unable to disable/negate the cache. Could some people verify?
Here is another project using libcdio (instead of libparanoia) claiming "compare 2 eac":
https://gna.org/projects/cued