Sorry, I have no experience with Win 8. I try to stay away from that newfangled software until I am dragged kicking and screaming in. Perhaps some of the other members has some insight. I am running Jim's software on XP and Win 7 successfully.
If you want to use the microscope as a direct optical comparator, you will need to calibrate it with a known good reticle. The camera came with a reticle but I don't know if I would trust it for accuracy. The Cooling Tech software that came with the camera does have a calibration feature which measures a distance in pixels and you enter in the actual distance in mm or inches. After calibration, it is important that the magnification not be change or you will lose your calibration. Jim's software can do the same as it displays the coordinates of the crosshairs in terms of pixels. You would then have to do the math to get your calibration factor. For a calibration standard, you can use an appropriately sized gage block. Failing that, a lathe bit previously measured withe a micrometer.
One issue with using the microscope as a direct optical comparator is the you are limited to measurements contained within your field of view. While you can zoom out to see the entire feature of interest, your resolution is still only going to be as good as you pixel count. If you are trying to measure a .25" part to +/- .0001", you have to be able to resolve to better than .04%. If your image is 1000 pixels across, you can't resolve to better than +/- .1%.
Another issue is that a cheap camera doesn't have a precision lens. Invariably, there will be distortion at the outer edges of the image.
I use mine in a different manner. I am running on a mill with a DRO and I center the crosshairs on one feature, note the DRO reading and move the mill to center the crosshairs on a second feature and note the second DRO reading. I can then calculate the distance between the two features. I use mine on the highest magnification and can resolve .0001" With this method, I can measure parts as large as 14".