Originally posted by Pål Jensen 1) Bjørn Rørslett has no credibility in my book. He was one of the first digital zealots who claimed that 2mp was better than 35mm film and approaching medium format. This was and is clearly nonsense. It is noteworthy that Pentax says the 14,6mp is as good as 35mm film which is probably closer to the truth.
Closer in some ways, but it depends on your basis for comparison. At similar ISO from 100 and up, film grain is much more obvious and has more impact on enlargement than resolution. Generally you can enlarge a 6MP digital print as far as a 100 ISO negative print and what you lose in fine detail you gain in edge sharpness and lack of grain. 14.6MP is probably about as good as the best slide film with top end glass, but 6MP was at least as good as an average Kodak Gold 100 film on a consumer SLR with consumer lens.
Originally posted by Pål Jensen 2) Thom Hogans predictions is usually wrong. Even 35mm film cameras show difference between good or bad lenses. It is not overkill how these cameras resolves bacuse it is not resolution digital excels, but in edge definition, and hence apparent sharpness and lack of noise. Film outresolves them. An it is not true that the differences are only visible at large prints. It is only true if your judgement of quality is not seing noise and that the image still appear sharp. If you look at texture, which is where resolution really hit in, then the difference is obvious in even the smallest print.
But if that texture is grainy then its not real texture, its just grain. Fortunately however Thom Hogan is usually wrong in his business predictions, or we would not be posting on a Pentax forum.
Originally posted by Pål Jensen 3) John Sheehy is right as usual. You need 100mp for APS in order to "outresolve" the lenses. That should put Thom Hogans 12mp outresolving lenses and all the resolution you want in the correct perspective!
Actually this is a bit misleading. You said yourself that even on a 6MP sensor you can tell the difference between a good and a bad lens so 100MP is a little OTT. More to the point, relatively few lenses show any practical resolution improvement between 10 and 15MP. The problem is measuring practical resolution rather than theoretical resolution. Most systems use MTF 50 for visible detail at normal viewing distance and MTF 10 for max visible resolution. Even at MTF10 there are few lenses that would match a 100MP APS sensor and even if they did you wouldnt see the difference in most prints.