From: Alfred Molon on
In article <hnehls$rio$1(a)news.albasani.net>, stephe_k(a)yahoo.com says...

> No problem? First test I pulled up show only REALLY good ones can pull
> those types of numbers at optimal settings. Most don't
>
> http://www.hevanet.com/cperez/MF_testing.html
>
> Stephanie

What MTF do those lp numbers refer to in the test?
--

Alfred Molon
------------------------------
Olympus E-series DSLRs and micro 4/3 forum at
http://tech.groups.yahoo.com/group/MyOlympus/
http://myolympus.org/ photo sharing site
From: David J. Littleboy on

"Michael Benveniste" <mhb(a)murkyether.com> wrote:
> "David J. Littleboy" <davidjl(a)gol.com> wrote:
>
>>> The 45mm lens is a bit too wide for
>>> the role (35mm equivalent of a 36mm lens), and the 75mm a bit long.
>>
>> The 45mm would be fine here as a normal lens. I use 65mm on 6x7 and 40mm
>> on 24x36.
>
>
> For The Love Of Dog, Montresor! I beg you, don't open up that cask of
> worms. While I prefer a slightly wider "normal" lens as well, even today
> the ghost of Oskar Barnack haunts the industry. I can't see Pentax/Hoya
> bucking that "Conventional Wisdom."

I don't think it was Oskar's fault. The "normal" lens is the widest lens you
can make easily with the Triotar, Tessar, and Planar sorts of designs, and
any longer lens would require a larger camera. This conventional wisdom was
foisted on us by lazy camera designers well before 1912. (The Kodak Brownie
was 1900 and the Planar design was early 1890s, for example.)

> While I though about exchanging my 45mm f/2.8 for an autofocus model,
> I never got around to it. Instead, I replaced the standard focus screen
> with a split-prism one. Should I decide to buy a 645D, I'll regret that
> decision.

Doesn't the AF system beep at you when you use MF lenses on the AF versions
of the Pentax 645?

As in another note, I'm using mostly MF lenses on the 5D2. The extremely
accurate distance scale makes testing for and setting acceptable hyperfocal
distances much easier. I took the 40/2.0 out for some snapshot shooting the
other day, and had no trouble whatsoever. Sure, I had to work marginally
harder, but even with AF, I think about where I want the plane of optimal
focus to be, look for something that the AF will work with, and do the
focus/recompose drill. The only difference with MF is that I have to crank
the focus ring at one point in the process until the AF system beeps. (And
while people complain about the 5D2 AF system, the center AF point is as
good as any AF point in any camera, and being a cross sensor, will focus on
things that you'd have to twist the camera to focus with a split image prism
or rangefinder.)

--
David J. Littleboy
Tokyo, Japan


From: David J. Littleboy on

"Alfred Molon" <alfred_molon(a)yahoo.com> wrote:
> In article <hnehls$rio$1(a)news.albasani.net>, stephe_k(a)yahoo.com says...
>
>> No problem? First test I pulled up show only REALLY good ones can pull
>> those types of numbers at optimal settings. Most don't
>>
>> http://www.hevanet.com/cperez/MF_testing.html
>>
>> Stephanie
>
> What MTF do those lp numbers refer to in the test?

That page is one of my favorites, and represents the results of a lot of
hard, precise, carefully done work. It's incredibly useful for comparing MF
film systems. (Bottom line: Mamiya 7 at f/11.)

But it's not clear what his criterion for OK/NG is when stating lp/mm
numbers.

One could, however, argue that they are overly optimistic and correspond to
a uselessly low MTF, since he's shooting a high contrast target and looking
for a barely detectable response. (And the film used has an MTF50 at 100
lp/mm, so he really can see the numbers he reports for the Mamiya 7 lenses.)

The bottom line, of course, is that the Mamiya 645 lenses I tested produce
sharp images on the 5D2 in a very quick and dirty (handheld, f/8, ISO 800)
test, so there really isn't a problem.

FWIW, from looking at diffraction degradation as lenses are stopped down, I
find that as long as your lens projects an image with an MTF of 50% at 2/3
or 3/4 of the Nyquist frequency, you'll be very happy with the images that
pop up on your screen or appear in your prints.

--
David J. Littleboy
Tokyo, Japan


From: stephe_k on
David J. Littleboy wrote:
> "Alfred Molon" <alfred_molon(a)yahoo.com> wrote:
>> In article <hnehls$rio$1(a)news.albasani.net>, stephe_k(a)yahoo.com says...
>>
>>> No problem? First test I pulled up show only REALLY good ones can pull
>>> those types of numbers at optimal settings. Most don't
>>>
>>> http://www.hevanet.com/cperez/MF_testing.html
>>>
>>> Stephanie
>> What MTF do those lp numbers refer to in the test?
>
> That page is one of my favorites, and represents the results of a lot of
> hard, precise, carefully done work. It's incredibly useful for comparing MF
> film systems. (Bottom line: Mamiya 7 at f/11.)
>
> But it's not clear what his criterion for OK/NG is when stating lp/mm
> numbers.
>
> One could, however, argue that they are overly optimistic and correspond to
> a uselessly low MTF, since he's shooting a high contrast target and looking
> for a barely detectable response. (And the film used has an MTF50 at 100
> lp/mm, so he really can see the numbers he reports for the Mamiya 7 lenses.)

So then look at the SLR 55mm lens numbers compared, ~1/2 this resolution.


>
> The bottom line, of course, is that the Mamiya 645 lenses I tested produce
> sharp images on the 5D2 in a very quick and dirty (handheld, f/8, ISO 800)
> test, so there really isn't a problem.
>


So shooting a hand held quick test of the very center of the lens on a
different camera proves this Pentax can resolve enough for this pixel
density to be useful beyond marketing numbers? OK, I'm sure you believe
this :-)

Stephanie
From: David J. Littleboy on

<stephe_k(a)yahoo.com> wrote:

> So shooting a hand held quick test of the very center of the lens on a
> different camera proves this Pentax can resolve enough for this pixel
> density to be useful beyond marketing numbers? OK, I'm sure you believe
> this :-)

It clearly shows that there's no problem producing sharp images with MF
lenses at that pixel density, and that your claims of unsurmountable
problems are silly, unfounded BS. And, yes, showing that resolution over the
center 24x36mm of the frame for non-superwides is fully adequate proof: the
33x44 crop (from 46 x 56) is plenty to avoid any corners problems I've ever
seen in any lens.

--
David J. Littleboy
Tokyo, Japan