Apache-Talk @lexa.ru 

Inet-Admins @info.east.ru 

Filmscanners @halftone.co.uk 

Security-alerts @yandex-team.ru 

nginx-ru @sysoev.ru 




      :: Filmscanners
Filmscanners mailing list archive (filmscanners@halftone.co.uk)

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

RE: filmscanners: Pixels per inch vs DPI

Austin wrote:
>My largest print size is 17x22 from my 3000.  I can see differences from 
>"standard viewing distances" that have
>convinced me that 180+ is the minimum resolution that is acceptable to me 
>for the type of work I do, if not 240+ preferred.  100 is vastly degraded.

'Vastly'? Well, I bow to your excellent eyesight!  It sounds like your work 
includes a lot of B&W which does make a difference..
You obviously do have high standards, but don't forget us lesser mortals.. :)
(please note - all said in good humour and definitely *not* meant unkindly!)

> > It is generally agreed that your average photo lab print is at
> > best 200 dpi,
>Where has this been agreed upon?  I'm not doubting it, but I never heard
>that.  I also would say that probably doesn't hold true for (especially B&W)
>darkroom prints

In my end of the world a 'photo lab' is the downmarket, 
1-hour-color-processing type place.. so perhaps we are talking at crossed 
purposes - if you are talking B&W or pro-lab prints, then I agree 200 is 
too low.

As to where it is agreed (or maybe argued!) upon:
'When we get right down to it, scanning color prints can rarely yield more 
detail when scanned at more than 300 dpi. And in many cases, that number 
may be closer to 200 dpi. I am carefully saying color prints, to exclude 
film and B&W prints. In particular, I'm speaking of typical 4X size 35 mm 
photographic color prints from the photofinisher.'

http://www.scanjet.hp.com/  (this quote used to be there, but I can't find 
it now, I admit)
'The vast majority of scanning projects require resolutions lower than 300 
dpi. For example, scanning a photograph at resolutions higher than 150 to 
200 dpi only produces a larger file, not more detail.'

On the output side, there's some interesting magnified samples here:
Go down to the samples, move right back until the '360' image is clear, and 
then look at the difference from the 120ppi one.

For the *alternate* view, (yes I often argue with myself) and to show that 
I really don't dispute your opinion, try here (*well* worth a visit):
Here, amongst lots of other interesting stuff,  the author shows that 400 
ppi is required to get everything off a SHARP print (his emphasis).  I 
would argue that the average joe has probably never seen *that* sort of 
sharp print :-( , and that a print from a 1-hour lab is about half as good..

>Point is, whether 100ppi looks "good" at all VASTLY depends on print size.

AND viewing distance, and image content.  :)   It's a bit like the debate I 
have had with people who tell me their 2Mp camera gives superb results when 
printed to 11x8.  Some show me a print, and it can indeed look *darn good*, 
even up close.  But then I say, 'OK, give me the camera and let *me* pick 
the subject', and the debate is lost - they know images with fine detail 
will show up the low resolution..

>Certainly 100ppi will look GREAT if the print is the size of a billboard,
>but for a 13x19, it looks poor, IMO.

Up-close-and-personal, and on an image with detail, I agree.  But what 
about a slightly soft-focus, close-up, color portrait?  I know that's 
cheating, but my point is that low resolutions like 100 ppi shouldn't be 
strenuously avoided at all costs.  I think members of the list (esp newer 
ones) should find out for themselves what low-res prints look like, before 
locking their brains into the 240-and-above zone.

Just more of that 'variable mileage' we all get, I guess!  :-)

mark t


Copyright © Lexa Software, 1996-2009.