This website provides reviews and ratings on many different bikes.

I've ridden steel, aluminum, composite, and Carbon Fiber. All are very good. I have a slight frame preference for the Trek Madone 5.2SL. Relative to the other frames I've ridden this frame has less (very little) lateral flex. But, I only notice it during a ride after mile 80 when my legs start getting tired - I feel that the power from my legs transfers directly to the back wheel with no loss of power to frame flex.

Trek also provides and stands by a lifetime warranty on their frames.

"
A 1987 Cannondale:
http://velospace.org/files/Cannondale_1.jpg

But, why is stiffness the main criteria?

""Chainwheel"""

Does anyone have any suggestions?

"Last time I bought a bike I did some investigation of the current state of the art. Here is an executive summary:

Metal frames:

Of the metal options (steel, titanium, aluminum) the one with the lowest density will generally produce the stiffest frame for the same weight (=> aluminum). Some of this is due to material properties (elasticity, etc.) and some of it due to the fact that alluminum frames can (must?) be made with larger diameter tubes which are naturally stiffer.

The down side with all of the metals is that one is always faced with a trade off of lateral stiffness (which you want to reduce power loss to frame flexing) and ride comfort (shock and vibration absorbtion). E.g. the less flex, the harder the ride.

Steel and titanium naturally tend to provide better ride comfort than aluminum due to ""material properties"". From what I've been able to find, modern steel frames with the latest tubing don't have nearly the weight penalty vrs. aluminum or titanium as was the case a few years ago, but still suffer from the corrosion issue. (They do tend to be much cheaper).

With composites, it is theoretically possible to engineer with the right weaves, shapes and resins pretty much any combination of lateral stiffness and shock/vibration absorbtion. The most important word in the last sentance is ""theoretically"".

The issue here is that a bicycle frame is a very difficult thing to construct in a highly controlled fashion with composites. To achieve a particular result with composites (such as that targeted good lateral stiffness combined with good shock absorbtion) one needs to have the lay-up (fabric fibers combined with the resin) match a very specific target.

The fibers and/or weave of the fabric has to lay in a very specific direction and needs to be fully impregnated with the resin, but there can't be any excess resin.

This can be achieved with modern manufacturing techniques for simple structures (such as tubes) but is very, very difficult for complex structures (such as a bottom bracket).

To my knowlege there are three construction methods used for composite frames currently:

-Monocoque (frame laid up in one shot). This is what the pros use predominantly now.
-Composite tubes (or other simple substructures, such as forks or chain-stays) joined with some type of metal lugs (steel or titanium).
-Composite tubes and substructures with composite lugs.

Problems with monocoque: Getting that precise lay-up that one wants. In things with big flat surfaces like airplane wings or the hulls of racing boats (sailboats, kyaks, etc.) this is done with a technique called ""vacume bagging"". Material is laid in a mold, resin is added, then excess resin is removed by sealing the mold with a bag, then pumping all the excess air and resin out (the vacume...).

For things like tubes, this can done using a method where fiber is wound around a mold with resin added incrementally. (In some cases, vacume bagging at the end).

As one can imagine, neither of these options is viable for a monocoque bike frame. What is done is to build a mold (expensive and fairly restrictive to the geometry available) lay the fabric in as well as is possible given the complex shapes involved, add resin then put airbladders of various shapes into the spaces that permit it to try and squeeze out excess resin.

This sounds pretty hit and miss to me. (Did some boat building once upon a time). If you are building a frame for a pro-cyclist you can keep trying until you get what you want. For us mortals I'd suspect that it comes down to getting lucky with the frame we end up with. (I haven't heard of any of these guys doing something like x-raying the finished frames for defects).

The problem with composite tubes with metal lugs is that the materials have very different coefficients o"

In the mid-level price range, Aluminum is probably how you want to go. I have ridden both a Specialized and Pinarello within the past year that were very stiff. The Specialized also had Ksyrium wheels and was the stiffest bike I have ever ridden. After 100 miles, I felt lucky to have any teeth left.
If you want specific suggestions, what do you want to pay?

I sold both my Colnago Ct-1 and my Ct-2(both 6-4) and ended up with an all aluminum Cannondale and I love it.Cannondales are good for racing and so are Colnago's,but Colnago's are better for all day riding.Cannondales you can ride all day long too so long as it fits good and you know how to ride over rough road surface.Colnago's are more expensive.Anything aluminum,Ti,and all carbon is good,I would strongly suggest to stay away from frames that have more than one material.If you like Titanium then get a full titanium frame not one with carbon stays.