Very good analysis, George!
Alex
In a message dated 2000-01-27 9:57:28 PM Eastern Standard Time,
geanders@xxxxxxxxx writes:
<< Folks;
OK, I'm pickin up the gauntlet and I'm gonna try to defend our beloved
Zuikos against this numerical assault. I'm going to place the photodo test
results for our 3 lenses along with results for their competitors, under
the *Totally Unbiased* "Anderscope". I'll be looking for factors that
perhaps are overlooked when photodo comes up with it's big overall rating.
We'll see about this numbers game.
********************************************
First, the lens that Dirk left out, the 35-80/2.8:
I ran advanced search, only zooms, only mtf above 3.5 on Nikon D AF mount.
Had 9 hits. Of those, only 2 are focal range competitors:
Overall wide open
ave mtf@40lpmm
Zuiko 35-80/2.8 3.6 (77.5) 71.0
.49
Nikon 'AF D' 35-70/2.8 3.8 (79.3) 63.3
.51
Nikon 28-70/2.8 ED-IF 3.7 (78.4) 69.3
.51
First off, the Zuiko has the widest focal range, so it's the more
challenging lens design, and the more useful lens, eh? Next, the way
photodo comes up with the rating is based *solely*on the average weighted
mtf. And the weighted mtf has the following weights: F/4 = 40%, F/8 = 60% .
They say that they assume that fast lenses 'suck' (paraphrasing) when tested
wide open, that's why they use F/4. And that after F/8 diffraction becomes
a bigger issue than lens quality. They **totally ignore** wide open
performance.
Now, based on this, our baby is 3.6 But wait, the Anderscope has noted that
our ave mtf is 78. **And so is that for the Nikon 28-70. Yet it received a
3.7 !! ** Hmmm. (We note for the record that Nikon advertises on this site
...) What explains the same mtf # giving 2 different test ratings if it's
only based on that number? Hmmm? Well, the Anderscope crunched the numbers
and found that the Nikon was 78.46 while the Zuiko was 77.53 Maybe this
explains it? They round off when posting the mtf #, but not when deciding
overall rating?
As for the NNow consider the Nikon 35-70. A nice lens, but *where's the
extra 10mm of range? It has a 3.8 rating based on ave mtf of 79.3, but we
notice under the scope that photodo was very *right* in their poor wide open
performance assumption for this lens. But it was very *wrong* about the
Zuiko. Under the Anderscope, our baby is *far* better than the Nikon at
F/2.8 But photodo ignores that data altogether. If that was factored in we
would kick this lenses' butt and been virtually identical to the 28-70.
One other spec that's quite important to me is the mtf @ 40 lpmm, because
higher numbers here indicate better ability to make large prints, which is
my forte. Unfortunately, the Nikons both edge out the Zuiko here.
All in all, I feel the Zuiko is under-rated at 3.6 especially since it's a
big winner wide-open. Nice, sharp photos with great wide-open bokeh! ahhhh.
BTW, I also looked at the rating for the much-ballyhooed Nikon 24-120
F/3.4-5.6 IF - it got a 2.3
Other than Nikon:
Canon EF 28-70/2.8 L USM (3.9) Looks a bit better. But one data point is
missing. Did they make it up?
Minolta 28-70/2.8 G (3.7) Just slightly better,even under the scope.
That's it. Of all zooms in this range, only 3 are in the ballpark. Even
Leica and Contax fail to hit this mark.
*********************************************************
>
>OM 100/2 - (3.9)
>Canon 100/2 (4.2) - Anderscope says "Zuiko better wide open, but overall
:>("
>Minolta 100/2 - (4.4) - Anderscope says "Er, sorry fellas, this is better."
>Nikon 105/2.5 (4.2) - Anderscope says "Too slow"
>Leica Summicron-M 90/2 (4.3) Anderscope says "Zuiko better wide open, but
overall :>("
>
OK, so of *all* F/2 lenses in 90-110 range, only 3 are comparable or better.
****************************************************
>OM 90/2M - 4.2
>Leica Summicron-M 90/2 - 4.3
>Tokina AT-X 90/2.5 - 4.6
>Tamron SP 90/2.5 - 4.2
>Contax G Sonnar 90/2.8 - 4.4
Forget the above chart, see text.
OK, last, the venerable Zuiko 90/2 macro. FLASH: If we insist on F/2
aperture, ***there is NO competition***. So, we have to make it easy on the
other guys.
First, the lenses are tested at infinity. The 90 is best in the macro
range, obviously. So Dirks comparisons above are moot. So I did a search
on Nikon 'AF D' lenses, 80-100 mm, macro lens only, rating >= 4.0 I got
*only* one hit. Surprisingly?, it's a Tamron 90/2.8 macro.
So, how does it compare to our favorite under the unbiased Anderscope?
Well, the Tamron's a stop slower - hey, gimme back my bright lens! As
before, I did the math and Zuiko = 83.2, Tamron = 83.6 Virtually identical.
mtf @ 40 lpmm? Here, the Zuiko blows the Tamron away .63 to .57 I'll take
the Zuiko for my big prints.
If I open up the search to macro lenses from 60 to 105, I get 2 more hits. A
Sigma!! and a Nikon. (I have to search specifically on Adaptall to find the
90/2.5 SP) Again, all are slower than the home team. And the Zuiko beats
them all numerically. The overall numbers below show the Nikon up by 0.2,
but don't be fooled! Under the Anderscope, we see the Zuiko has better mtf
at F/2 and F/2.8 than the Nikon at f/2.8 Again, if wide open performance
were included - Z beats the Nikon hands down. I'll take the Zuiko for speed
and wide open performance. And the Nikon really should be compared to the
50/2 anyway.
It's quite astounding to me that the Tamron and Sigma lenses are actually
more competition for the hometown favorite than the Nikons.
Overall
mtf@40lpmm mtf @ wide open
Zuiko 90/2 macro 4.2(83.2) .63
0.76 (f/2) 0.78(f/2.8)
Tamron 90/2.8 macro 4.3(83.6) .57
0.79(f/2.8)
Tamron SP 90/2.5macro 4.2(83.4) .60
0.68(f/2.5)
Sigma 105/2.8 macro 4.1 (82.2) .60
0.77(f/2.8)
Nikon 60/2.8 micro 4.2 (83.4) .61
0.71(f/2.8)
Just for grins:
Nikon 105/2.8 micro 3.9 0.52
0.75
Also, I did 80 to 135mm macro >4.0 mtf searches on:
Contax got 0 hits
Canon EF got 5 hits. Only the Canon 100/2.8(@4.4) was actually
'better' -but only very slightly under the Andersope and again, the Zuiko
has the speed advantage .
Leica M got 0 hits
Leica R got 1 hit. The APO-macro-Elmarit-R 100/2.8 at 4.5. Again, the
Zuiko is faster, but the Leica is much better wide open (0.87) and slightly
better mtf@40lpmm (0.65)
Minolta AF: got 4 hits, only 1 the Minolta 100/2.8, was 'better'. It was
actually darn good at 4.5 and is a 1:1 macro to boot.
Pentax K AF: got 4 hits, Pentax 100/2.8 was a virtual draw at 4.3 after
several Anderscope adjustments.
So, in summary, of all competitive lenses, only the Leica 100/2.8 and
Minolta 100/2.8 were noticably better than our boy. But did I mention, ours
is a stop faster? :>)
george
>>
< This message was delivered via the Olympus Mailing List >
< For questions, mailto:owner-olympus@xxxxxxxxxxxxxxx >
< Web Page: http://Zuiko.sls.bc.ca/swright/olympuslist.html >
|