Models for Slicer Comparison

I would like to test several slicers and try to get demonstrations of their various strengths and weaknesses. I'll run a set of models through each one, with settings that are as identical as possible. I'm thinking of comparing Skeinforge 50, Skeinforge 39, Slic3r 0.9.3, Slic3r 0.7.2b, and KISSlicer. I'm looking for suggestions of models to use that represent various challenges that slicers have to deal with, or features that I should include in a model I design.

To start, I would use a cylinder with a hole through the middle to test how smoothly they handle layer transitions and interior hole diameter. Also the slide from the calibration contest for dimensional accuracy. I want to test performance in bridging, stringing, cooling, support for top layers that appear partway through the print ie does it make a solid layer over the whole part because a portion is exposed (Skeinforge) or does it make just enough solid to do the job (Slic3r).

I would also want some more complex organic shapes. Yoda would be a good test, but I also want to keep the print time down. I want just enough to demonstrate a particular issue without taking something like 3 hours for each print. A longer print would be fine if it is a single model that contains all the needed features. I'm open to doing that, but I'd like some suggestions for what that all in one model should contain.

What other aspects of slicing should I test for?

I think this test will be useful, so thank you for taking the time and effort to undergo it. Might I also suggest you test Cura, which may possibly be the best slicer at the moment (based on what I have seen, and Slic3r seemingly getting worse with each new version). The results will certainly be interesting, perhaps Slic3r 0.7.2b might be the winner!

I haven't even finished calibrating on slic3r 0.7.1 and am wondering if I should be looking at later versions. I'll be interested to see your results.

Isn't Cura just an interface to other Slicers ?

---

Most of my technical comments should be correct, but is THIS one ?
Anyway, as a rule of thumb, always double check what people write.

I don't mention Cura since it is a front end for SF50. That's probably what I'll be using since it runs SF faster using PyPy.

Thanks for the clarification, I was asking because of yydoctt suggestion.

---

Most of my technical comments should be correct, but is THIS one ?
Anyway, as a rule of thumb, always double check what people write.

Ah, I thought it was more like SFact, where the maths that drives the modules had been altered. Good to know, thanks.