[Editing to point out that Ironfish653 said it all, in fewer words, in his post above this one. I'll leave this post here, though, if only to corroborate his.]
For the
nth time:
Read some of the vintage articles that SpeedOfLite has so helpfully posted: specifically, those that assess aluminum bikes, including, among others, road tests of the first-year Cannondale ST-400 and---especially---the comparison of the Trek 570 (steel frame and fork) and Trek 1200 (aluminum frame and steel fork). In both cases, the aluminum bike was praised for its superior comfort.
In fact, the reviewer who compared the Trek aluminum and steel bikes lamented what he saw as the inevitable, more or less rapid and more or less complete disappearance of steel road bikes from bike shop sales floors. Which, of course, is exactly what happened.
I've puzzled over when and how the myth of the uncomfortable aluminum frame originated. My guess:
A significant number of U.S. riders who had started with a Fuji S-10-S or some other sports touring bike in the '70's or early '80's decided to step up to a higher-performance bike in the late '80's. A significant proportion of those riders were attracted to the Cannondale line of bikes. Enthusiastic salespeople (that would have included me at the time) steered them to the latest and greatest: the Cannondale Crit series.
I already owned and raced on a steel Italian criterium-geometry bike, so I was used to the abrupt handling of a super-short-wheelbase bike and loved everything about the Crit series bikes---same handling as the Italian bike, same "comfort" (or lack thereof), but lighter weight.
So (guessing) tens of thousands of riders jumped directly from a sport touring bike with a moderately long wheelbase to what was effectively a track bike with gears and brakes. Of course it was going to feel harsh---exactly as harsh as any steel ultra-short-wheelbase bike.
But while it was true that steel crit bikes rode the same as aluminum crit bikes, only a comparatively tiny number of riders were going to spring for the expensive Italian bike that, in most cases, had to be built from the frame up, usually with an extremely expensive Campagnolo gruppo. And those riders would in almost all cases have been experienced racers, for whom the hard ride was compensated for by the great handling at peloton speeds.
Be that as it may, Cannondale fairly quickly dropped the Crit series from the lineup, no doubt because of complaints from dealers who had fielded complaints from the non-racers who had been suckered into buying those aluminum rockets.
Note that the Italian manufacturers (and the Japanese companies that copied the Italian crit geometry bikes for one or two of their racing models) also quickly dropped the crit-geometry bikes and reverted to their previous designs.
But while no one blamed the steel used in the Italian crit frames for their harsh ride, people whose only exposure to crit geometry was a Cannondale Crit Series bike jumped to the apparently obvious but erroneous conclusion that aluminum bikes ride harshly.
And the myth of harsh-riding aluminum eventually hardened into dogma. It's the bike sideshow among the current black-and-white, us-versus-them, rules/sucks soccer field melees: vinyl versus digital, tubes versus transistors, aluminum versus steel.
Anyway.
There are any number of science-based reconsiderations of the "aluminum frames ride harshly" myth available to read out there, including Sheldon Brown's brisk dismissal of the idea in the page on frames on the Sheldon Brown website. Here's another:
From that article's concluding remarks:
"If you swear black and blue that you can still notice extra comfort in your steel or titanium frame, well, I’m afraid all data suggests it’s just a placebo – which actually isn’t a bad thing if it genuinely works for you!
"
When we consider all components that move on a vertical plane, your frame material – be it steel, aluminium, titanium or carbon – will not affect your ride comfort. Not only is frame compliance a small proportion of the overall spring rate, but it also becomes completely insignificant when we calculate out the springs in a series."