Item 1:
"Playing" a CD is not the same as "ripping" the data from it.
In addition to the error-correction that is built-in to the audio data on the disc -- which is much less than the error-correction that is on data CDs -- the playing process uses the firmware in the DRIVE to create interpolated samples to CONCEAL the existence of errors thatr cannot be corrected. Corrected errors produce correct data; concealed errors produce fake data. When the data coming off the disc cannot be corrected, there's a dropout in the stream, and the PLAYING process interpolates some samples to fill the gap, in a way that is relatively bening, especially as it should be short.
Obviously you would not want interpolated samples when reading a data CD: you want the files back exactly as they were written. There's more error-correction overhead on a data CD for that reason. Audio CDs actually hold more "user data" than data CDs because they have less error-correction overhead, and they get away with it most of the time because listening to a stream is a human-brain process that is more tolerant of fleeting errors than a computer would be.
RIPPING pretends the disc holds data, not audio, and does not invoke the error-concealment code in the drive firmware. If the drive can't correct an error, it probably fills the gap with zeros, and you hear a click (if it's short) or a dropout (if it's long enough to hear).
Ripping at 1x is still ripping, not playing. Computer drive mechanisms will often spin the disc faster than 1x even when playing, and they buffer or cache the audio samples so the computer is free to do other stuff. It also helps if the drive is moved, as in a laptop, and has to re-position the laser and re-read part of the disc. Whether the drive is ripping or playing is not just a matter of rotation speed.
Item 2:
If you rip on a computer and get a file with errors in it, it's often (not always -- it depends on the errors) possible to reconstruct a "correct" copy of the audio using a program and online database called (collectively) CUETOOLS. It has a learning curve, and thankfully I have not had to climb it yet. But I know people who have used it successfully to re-create correct rips of damaged discs in their collections.
If you have bags of abused CDs, it's porbably worth reading up on CUETOOLS.
Item 3:
Sometimes, if a CD has a deep scratch that hasn't quite reached the reflective layer (you should be able to see that by holding the disc up to a light), after you clean and polish it you could try applying a tiny bit of car wax -- NOT POLISH, that scratches, just WAX -- to fill the scratch with something that dries clear. The acrylic coating has a different refractive index from air, and the edges of the scratch could be bending the laser beam enough to misread the disc. Filling the scratch with clear wax can reduce this effect. It might be enough to get a good read.
TMI, I know. Sorry. -- m.