Argh...opening a bit of a can of worms here I think.
However, I've been using an AVR as a processor for years now, after running an older (7.1) processor before that. I don't feel I've lost anything, though measurements on other forums might say otherwise I can't hear the 'noise' my AVR is supposed to suffer from (but then I run with speakers set to small, so possibly the SINAD is better than that particular 'review' implies).
Having said that I'm one of those people who can't hear a difference between different power amps (so long as they are up to the job in the first place). Maybe I'm not the person to ask, but I do a lot of REW measurements and my avatar might hint that I know my way around a guitar too, so it's not like I'm tone deaf.
The biggest issue is that if you bought an 8805 you would have to run a fresh Audyssy calibration: This alone could give you a different result. If there was some way to load that calibration into your 4500H then I wonder whether in a blind test you would know the difference? In theory it should be possible to do this test with an Arcam AVR verses AVP, but I don't know anyone who has done this: I chose to buy the lowest powered AVR version because I knew I wouldn't use the power amps (well apart from two channels driving the surround backs now). I don't believe that the different power supply makes much difference when used as a processor (they have exactly the same processing stages and even firmware in my case).
Not trying to offend, but I do sometimes think that spending more can bring a certain amount of expectation bias and it's obfuscated when a fresh Audyssey calibration is involved: This will
always make it sound different to the last calibration due to variances in mic position, let alone other influences like slight room layout changes and temperature.
EDIT: My comments apply to the more mainstream companies that produce AVRs and AVPs in the same range, using the same processor stages. Whether the higher end companies that don't produce an AVR equivalent sound 'better' is another discussion.