Here's my question, and I realize the answer(s) will be
somewhat "It depends.." ....

I want my engine (insert any engine) to reach peak HP around
6000, with max RPM of 6500. I want peak torque around 4500.

I'd like to now assume that all the rest of the components
are kind of well-matched, i.e., port sizes matched to
displacement, carb size, header size, etc.

So what sort of generalization can be made about cam duration
to shape the power curve in this RPM range ? Duration and
lobe centers.

Or, is this question too simplistic to yield a useful answer ?

The thread about BradH talking about port sizing got me
pondering.

PipeMax is pretty good at that sort of simulation. It will give you the RPM range of your components.

The software from Rick Jones will also give you that information but it is a little more difficult to use than PipeMax.

You can tweak it somewhat, but when the engine develops power depends primarily on the heads vs. displacement. Hot cam + bad heads may be close to mild cam + great heads but only at high speed.

You have my sincere apologies...

Yep to simplistic. But on a near stock 440 around 240/250* duration @ .050 on a 108 lobe center.

Peak torque occurs at optimum VE, and torque is mainly a product of displacement and compression....if a motor is HP peaking early in the RPM curve because of the CAM the VE and power will fall off at a much slower rate than if it is falling off because of port restriction, If you can't get enough air throught the hole the cam can only really be a band-aid.

Look at an LS3 motor, huge ports with huge 2.19" valves by conventional measure (certainly bigger than out 906 size 440 ports) and for only a 6.2 (376") motor, but with Variable valve timing, EFI (and of course gearing) they still Impressively can move 5200+ pound suburbans and Escalades and they rev to 6400 or so stock.

They could easily make as much torque at say 3000 or 3500 with a conventional 2.02" Valve Old school small block, but they could never get near the kind of peak power with small ports that are drawing ever increasing vacuum beyond the torque peak. They could have developed the same VVT for a conventional small block production head, but they wouldn't have anything close to they same broad torque and peak power.