Originally posted by tuco ...
** Look at the MTF curve of Velvia 50 vs T-Max 100, for example. At 50% response Velvia drops to about 50 cycles/mm (line pairs/mm) whereas 100TMX is almost 100 cycles/mm but this ignores the boost in appeared sharpness Velvia gets from the nature of its response curve (the rise before it starts to drop) and fine grain.
When you think about how small 1mm is, it's insane to think that all that information can be stored in such a tiny space. Be it a sensor or a piece of film, it's crazy! I think it's all the more impressive seeing it on a negative though, because it's right there in front of you. You look at it and see a blob, even with a 10x loupe. You then stick it under a microscope or have a ridiculous scan done, and there it is. Think about all the things that have to be perfect to achieve over 100 LP/mm. I think we are going to hit a wall with digital sensors very soon. Even a perfect lens, which there are none, has limits. Sensor size will be the only logical way to increase resolution, particularly for occasions you need DOF and will be diffraction limited.
Edit:
When I got my Rolleiflex, I wanted to see roughly how good the lens was. So I did this silly shot below. I have only ever shot Pan F this one time, I guessed at development time, I didn't focus bracket, the camera was sitting on my $129 tripod, on a tile floor over a basement in a 1929 home(i.e. a little bouncy), and my scanner can not optically scan at this resolution... oh, and the exposure was about 8 seconds long. So do not take this as any type of quality test. About the only thing it told me was the lens on this 56 year old camera is good enough and that I had just wasted a roll of film.
Anyway, the scan was done at 7620SPI reduced to 5080SPI(100LP/mm) to see how much detail I could get with my setup. I adjusted the contrast and sharpening with reckless abandon towards quality. The 2 crops are 1mm x 1mm and the little speck next to them should be about their real life size(depending on the resolution of your monitor). My conclusion from looking at the negative under a microscope was that the film was limiting the resolution, but that it would have to be in the neighborhood of 90-100LP/mm.
/Edit
Anyway, we've went way off topic.