I'm thinking both Rex and 1212 are right- to a point -there is a relationship to finess ratio and speed -(Reynolds number?-help A2).
I was told many years ago that the Goodyear blimp is pretty close to a perfect 60 MPH shape- but looking at the WW2 drop tanks -300 mph tops -they are a bit stretched out!
Food for thought.
Yes, there is a relationship between separation potential and Rn- the other way.
At low Rn, the boundary layer is very thin and has little energy in it. This allows it to separate MORE easily than a thicker, more energetic boundary layer that we get at higher Rn.
The lowest drag fineness ratio is actually about 2.5:1. However, this is almost impossible for any practical vehicle because the very abrupt pressure recovery is too sensitive to other effects, especially intersections with wings, tails, and wheels. Wave drag and shock-induced-separation account for the large drag rise approaching Mach 1, but most wheel driven LSR will never get above 500+ mph where this is a factor. We discussed this before on another thread, and based on the practicalities of axle intersections, decided that a body fineness ratio of 4 to 5 would be about optimum for a lakester. A streamliner could be another 20% shorter. Using this short of a design, I would add a tail aft of the main body for better stability.
Low fineness ratios allow less wetted skin area. At subsonic speeds, this is the dominant drag besides separation. The fact that only half of JCB's drag came from skin friction indicates a need to fix the separation that accounted for the other 50%! Low fineness ratios also provide for more laminar flow as a percentage of wetted area; this is the next most dominant effect after separation and total wetted area. I worked this out from a pile of laminar flow research I have copies of and a streamliner with a fineness ratio of about 5 and the maximum cross section back about 60% of its length would have less than half of the drag of the current crop of 400 mph designs.
I hate to disappoint the streamliner crowd, but short and fat is better than long and skinny.
Beyond Mach 1 the rules go the other way. Wave drag is:
D/q
wave = (9pi/2)(A/L)
2(EWD)
Where,
D/q
wave = flat plate equivalent wave drag
A = maximum cross section
L = total length
EWD = variation of the volume distribution vs. ideal
So beyond Mach 1, long and thin is better than short and fat. But
only beyond Mach 1.