I received my FS4000U a few weeks ago, got it 2nd hand from B+H.
I have been a film shooter with EOS3 for many years. I got my 10D earlier this year, and out of interest having read sites like luminous landscape etc...I decided to do a little film (superia 100) vs 10D 6.3MP sensor resolution tests.
--------------------------------------------------------- METHOD I'm still in the process of doing these series of tests (I'm shooting newspaper spreads stuck to a wall). Both cameras have CF 12 enabled, tripod mounted, cable released. Superia 100 is scanned at 4000dpi FARE disabled; whilst the 10D captured in RAW, decoded via fileviewer utility and then rescaled to the same width as the film image by PS7.0 bicubic scaling.
Since there is a focal length multiplication factor on the 10D of 1.6x, I tested using the 28-70L using "equivalent" 50mm focal lengths (hence, 32 mm on the 10D, and 50mm on the eos3). Lens was shot at f/11 on both bodies. Manual (same) exposure. ----------------------------------------------------------
The results were quite surprising.
100 speed print film scanned at 4000dpi substantially outperforms the 10D in resolution. Small print not readable on the 10D image is much more legible on the scanned film. I find this quite amazing, as I have about 5 double spreads of newsprint forming the width of the test frame, and the words are still readable with the 28-70/superia 100 film!!
I'm not saying that print film is better than digital. I'm only talking about pure resolution, not colour accuracy, grain, noise or convenience. From what I can tell, print film still out-rezs the 10D's 6.3MP sensor by a considerable amount. I think my preliminary findings also show that the L series zooms outperform the resolving capacity of the 10D's sensor.
I'll be doing some tests over the next few days or so with sensia 100 and velvia (I still have a ton of it left in my fridge!!!! Thanks to going digital!!)
Greg
With a given sensor size, the number of pixels in each dimension determine the maximum resolution that it can resolve. Film can resolve much more; even something like Tri-X can do better than the 1Ds can (which only has 32% higher resolution than the 10D or D-Rebel.
The reason that digital looks so good is due to its lack of grain and resultant smoothness of tone.
When we see an 8x10 print from 35, and compare it with an 8x10 print from medium format, we usually say that the MF print looks 'better'. At this size it usually doesn't show higher resolution, but it usually looks 'smoother'. We associate that look with MF. With a print from a 10D, it looks smoother than a print from film, and we rarely look for the finest detail, which is fairly hard to show on an 8x10 print. So we say that a print from a 10D file is equivalent, or possibly even 'better' than a print from film.
Even with a 1Ds, with its 11mp can't compete with respect to resolution with most films, especially the 100 speed ones. With its complete lack of grain the images are much more like MF than 35mm in their tonality, and that is what makes them look good.
If you make a 20x30" print from a fine grain film image with lots of fine detail shot with a top quality lens and not degraded by camera shake, etc, and compare it with a 1Ds image of the same thing shot with equally good glass, the differences will become apparent. The print from film will show significantly more detail, but will lack the smooth tonality of the digital image. Seen from close up, the print from film might well look better. Seen from further away, the gigital will probably look better.
This shows that we often really don't need all the resolution our equipment can produce.
BTW, I own a D60 and have done and seen some tests of the above type done with a 1Ds.
-- * Henning J. Wulff /|\ Wulff Photography & Design /###\ mailto:[EMAIL PROTECTED] |[ ]| http://www.archiphoto.com * **** ******* *********************************************************** * For list instructions, including unsubscribe, see: * http://www.a1.nl/phomepag/markerink/eos_list.htm ***********************************************************
