XLR or SE cable?


After reading the current posts relating to the volumne differences between XLR and SE cables and the recommendation of ARC (to use balanced) I find myself in a quandry as to an avenue of selection for a possibile change and could use some help. I have an ARC CD 3 Mk2 using with a Classe CAP 151 integrated amp. Currently I have them tied with SE Ridge Street Poeima ICs (1 mtr), I must add that listening is nicely relaxed and not fatiuging at all but, I was looking to warm up the presentation a little more and get a bit more bottom end extention as well. ARC recommends using balanced for optimum results (which I could do), after contenplating a change I was wondering if this would be the right route to go? Also after seeing the posting about volumne differences between using the two this additionally leaves me questioning it as well becuse I'm used to the settings I currently use for listening. I have settled on a new pair of ICs from Grover Huffman and before I invest I would like to make the right decision to do so and could use some advise as to XLR or SE being the better or best selection and benefits for optimation? Speakers are Apogee Slant 6s and I listen to soft rock, blues, new age, and jazz.
rsjm80
Got the response back from Classe and is quoted as follows:

"When you use the balanced input on your CAP-151 these 3 signals reach a ”differential circuit” that removes the noise using the principle I mentioned to you before. They are combined into a single low noise signal that is sent to the CAP-151 amplifier stage.
There is only one signal path to the amplifier stage, so single ended signals also travel down it. However they will not have benefited from the “differential circuit” so will be of a lower quality with more noise. This ultimately means that the Balanced input sounds better than the single ended inputs."
It sounds to me that the optimum path for best noise rejection then would be to use Balanced. Mordante & Herman this seems to also confirm the principle of an SE circuit in the signal path that you alerted me of. By lowering the noise floor I could possibily realize better definition and maybe resolution?
I'll also try to borrow a pair of Balanced ICs to check this out and again Thanks everyone for helping to further educate me with your answers, its gone a long way for me.
The rear panel of the CAP-151 is nicely laid out. There are four line level inputs and one balanced input. I was happy to learn that the balanced input is a true balanced differential circuit and not just an unbalanced input terminated with an XLR. This allows a simple CD based system to be balanced from input to output which increases resolution while lowering the perceived noise floor in virtually any system.
That quote is from a TNT review.
I would not expect a quick response from Classe. Been there.
I'm a little skeptical of that claim.... "However they will not have benefited from the “differential circuit” so will be of a lower quality with more noise." Sounds like marketing hype to me. What do they mean by "lower quality?" How much noise can you pick up going a meter or so from the CD to the integrated? I run all SE with 107 dB efficient horns with tube amps and you have to get your ear up close to the horns to hear any residual noise. However, I'll fall back to an earlier post and encourage you to try it with used cables. You have nothing to lose and something to learn as well as gaining the peace of mind of knowing which you prefer.
Here is a quote from SoundStage reviewer Marc Mickelson: "Listening to the CD3 Mk II through its single-ended outputs didn't change the player's sound immensely, but did make for less-refined sound. Retrieval of inner detail was less pronounced, and the entire presentation didn't have quite the same level of easy resolution or verisimilitude. Single ended, the CD3 Mk II is still a very good CD player, but, in meaningful ways, it sounds better balanced"
Perhaps, or maybe the preamp he was using sounded better when being fed a balanced signal.

Beware of reviewers who use the word verisimilitude.