about Genuine Fractals or Blow-Up plug-ins

[vartkes]

Can anyone shed light on why I should prefer to buy a medium-format digital back vs stick to my 25MP full frame DLSR and resort to one of these two (genuine Fractals or Blow-UP) excellent plug-ins?
The quality of output from the plug-ins appear great to me. There again I havea sneaking suspicion that I am missing something....
thanks

[digitaldog]

Can anyone shed light on why I should prefer to buy a medium-format digital back vs stick to my 25MP full frame DLSR and resort to one of these two (genuine Fractals or Blow-UP) excellent plug-ins?
The quality of output from the plug-ins appear great to me. There again I havea sneaking suspicion that I am missing something....
thanks

At least in terms of GF Pro, its what you'd want to try when taking very small originals and blowing them up huge. Example is some work of my friend and associate, Mac Holbert at Nash Editions. He's got tiny images from X-Ray originals (640x480) that need to go wall size. In this example, he found GF Pro to work better than Photoshop. But going up 200-400%, Photoshop and the proper handling make such 3rd party products unnecessary.

[feppe]

Can anyone shed light on why I should prefer to buy a medium-format digital back vs stick to my 25MP full frame DLSR and resort to one of these two (genuine Fractals or Blow-UP) excellent plug-ins?
The quality of output from the plug-ins appear great to me. There again I havea sneaking suspicion that I am missing something....
thanks

What you are missing is that software interpolates data, or uses some fancy algorithms to guess. It is a mere approximation of the actual data available in the scene with a higher-MP count camera even in the best cases. Whether this is visible in prints - and at what size - is something you would have to test for yourself, though.

Also, MFDBs have no anti-aliasing filters which make the images sharper than the equivalent megapixel sensor with AA filter (ie. DSLR). And MFDBs generally have better quality lenses, and better dynamic range - although these two are becoming less of an issue with current and upcoming DSLR tech.

[dalethorn]

I was under the impression that those plugins did more than just interpolation - like a bit of edge-hardening, for lack of a more accurate term.

[feppe]

I was under the impression that those plugins did more than just interpolation - like a bit of edge-hardening, for lack of a more accurate term.

Correct, that's what was included in the term "fancy algorithms" in my post  

[paulbk]

Imagine an image projected on a white board through your favorite lens. Now divide each axis (x,y) of the projected image in half. That’s a theoretical 4 pixel sensor. Useless. Keep dividing until you get to 25 mp.  Any variation in tone within a single pixel is lost information because a pixel can only have one tone (luminosity).

Now divide each of the 25 mp from above in half. This represents the theoretical information of a 50 mp sensor. More detail. More better.

Of course:
1) Do you need that level of detail?
2) Do you have the money for that level of detail?
I don't on both counts.

[dalethorn]

1) Do you need that level of detail?

Turning it around, the question would be "how badly reproduced would you like your scenic detail to be for only xxxx dollars?"  Thank goodness with the economic downturn that things are still on the upcurve for photo equipment.

[Mark Graf]

I have used BlowUp quite a bit, but still have to check which does a better job - BlowUp or using the techniques as described by Jeff Schewe in the Camera to Print tutorial.    It depends on the particular image.  These fractal algorithms can some times turn fine detail into strange looking blobs - whereas PS BiCubic smoother may render a softer enlargement, but can look more natural once sharpening is applied.