Interesting subjective look.
It will definately be different for everyone. Personally, I found my GTX 285 to be too loud.
I'm definitely not a "If I can tell it's on, I'll shoot it" kind of guy. I was honestly surprised.
I saw those vids on H where Kyle is yelling out "100C!" or something, and thought they sounded pretty loud. So I thought I'd check it out for myself, and found my case fans are making more racket than the 480.
OTOH I put my hand in front of the side intake fan and thought "Holy crap! This thing would cool lava!" so it's not all bad.
I think is sufficient to say that your case is noisy from the start. Turn off your PC and listen. You will likely discover an eerie silence you did not realize existed (assuming you do not live in a place that is noisy all the time like Las Vegas or New York City). Turn it back on and and you will suddenly realize that your computer is noisy. I have done this with my own computer and it has driven me to extreme lengths to silence it, although I must admit that the recent cards from Nvidia are very quiet, however, that likely depends on the card manufacturer.
I had an ATI AIW 9700 Pro that I considered to be a dustbuster. I replaced its fan with a VGA silencer, which fixed the noise issues. After it died, I was so fed up with ATI for various reasons that I tried an eVGA graphics card, but its heatsink/fan turned out to be worse than the stock heatsink/fan on my old ATI card. I tried replacing it with an aftermarket heatsink, but that did not work very well, so I brought an XFX GeForce 7950 GT, which was passively cooled. It was a dream. After it died on me (which I figured was due to the poor airflow in my case as I use slow spinning fans to make my PC quiet), I brought a XFX GeForce GTS 250. This was after the XFX GeForce 9800 GT that I brought for my cousin's computer against my better judgment because he wanted graphics card that had high performance turned out to be extremely quiet and I have been happy with the GTS 250 since.
Since I mentioned the eVGA card I tried 4 years ago, I have an off topic anecdote I would like to share because I can laugh about it now, which is that I actually had two eVGA graphics cards. The first was a component of my brand new Core 2 Duo system back then and it went on fire the moment I turned on my system; apparently, a dud had passed eVGA's quality assurance. The second was its replacement under warranty, and that is the one that caused me to discover the issue with the fan noise.
Anyway, many noise measurements done by review sites are done 1 cm away from the hub of the graphics card and it has been shown (read about it at silentpcreview.com) that in order to do accurate noise measurements, noise measurements must be taken 1 meter away from the graphics card, rather than 1 centimeter away. While I do not know if the benchmarks that people have been citing have this flaw, it is possible that they do. Although I do not know enough physics to explain why, to my knowledge, it is possible to design a fan A that is quieter than a fan B at 1 centimeter but noisier than fan B at 1 meter. This is what I understand from what I have read at silentpcreview.com, which is the site I visited to learn how to make my computer quieter.
Disclaimer: I over-analyze everything, so try not to be offended if I over-analyze something you wrote.