Modern Photo tests database (LONG)
William I-Wei Chang
====================================
I promised Bertrand I would post some Modern Photo test results, to get
him to do all that research :-) Looking over the Chasseur d'Image examples,
I get the impression that aperture-by-aperture ratings are _roughly_
consistent with Modern Photo numbers, but the "star" system used by C.d'I.
also takes into account price, expectation, and homogeneity (variation
across image area and aperture). (For example, the ** Olympus 50/f2 macro
appears to beat the *** 50/f3.5 macro, at every aperture.) Certainly
these factors affect the "use" of a lens, and should be emphasized in
the comments. But I don't think they should overshadow the basic
capability of a lens.
Generally speaking, the distinguishing factors in a lens' performance are:
(1) performance near wide open (when _can_ one use it); (2) peak performance
(when and how _should_ one use it); (3) corner contrast (the contrastier lens
will always _appear_ sharper); and (4) distortions (comparable modern lenses
tend to have similar distortion figures). Item (2) is more important than
(3) in a portrait or long telephoto lens; the reverse is true for a "noct"
lens to be used in low light, hence low contrast, situations. Modern Photo
test reports are very (perhaps overly) detailed; however, there are the usual
caveats such as sample variation, subjective reading of lpm, and even typos.
The Modern Tests serve as a database for all kinds of comparisons. It would
be nice if Pop Photo retested some outstanding old lenses, so we can figure
out what the SQF's really mean. Also, it would be helpful if we could figure
out the correlations among the various American, European, and Australian
test results. (Are you listening, dbm? :-)
Hope this is of interest!
William Chang (wch...@ucbarpa.Berkeley.EDU)
Table of Contents
I. partial index '77-'89 Modern Tests
II. some numbers and comparisons (long)
I. Modern Tests 1977-1989 partial index
Recently I went through old Modern Photo test reports ('77-89 available
in open stack at Oakland Public Library) and compiled the following index.
I only took down things that caught my attention (systems, exotic cameras,
superb lenses, Tamron SP, Tokina ATX, Sigma APO, Olympus, some Nikon, Canon).
I'd appreciate hearing from anyone who has a better listing or knows how
to get one.
William Chang (wch...@ucbarpa.Berkeley.EDU)
77.1 Questar 700/f8 mirror lens
77.3 Widelux F7; Canon 24/f1.4 asph., 55/f1.2 AL
77.11 Inside OM2
78.8 Panorama expo
79.6 Tamron SP 500/f8 close focus mirror lens
79.9 Olympus OM10, 50/f1.8, 35/f2.8 shift, 180/f2.8 [low corner contrast]
79.10 Novoflex bellows
79.11 Olympus XA 35/f2.8
80.1 Olympus 35-70/f3.6
80.2 Plaubel Mackina 67
80.6 Nikon F3
80.10 Nikon E 75-150/f3.5; Canon 75-150/f4.5; Pentax 24-35/f3.5
80.11 Nikon 55/2.8 micro; Durst M605 6x6 color enlarger
81.1 Pentax LX
81.2 Tamron SP 35-80/f2.8-3.8
81.3 Tamron SP 24-48/f3.5-3.8; Tamron 70-150/f3.5 close focus (1:3)
81.4 Hasselblad 140-280/f5.6 Schneider Variogon; Tamron SP 90/f2.5, 17/f3.5
81.5 Olympus lenses
81.6 How to field test your lenses
81.10 Slide projectors, incl. resolution and brightness falloffs chart
81.11 Dissolve units; pocket cameras
82.2 New lens test standards (lower standards for corner contrast)
82.3 Sigma 21-35/f3.5-4
82.4 How to use 35mm SLR as view camera; make PC lens for $5
82.7 How to make a vacuum easel
83.4 Pentax wide angle lenses
83.5 Polachrome; Olympus OM10; Pentax 85/f2, 40-80/f2.8
83.8 Tamron 35-135, 80-210
83.9 Tamron 28-80/f3.5-4.2, 70-210/f3.5
83.11 Olympus 100-200/f5, 50/f1.2
84.1 Tamron, Tokina zoom lenses
84.3 Nikon 24/f2, 24/f2.8, 50/f1.4
84.5 Olympus OM4
84.6 Fujica 645 folder 75/f3.4
84.9 Inside OM4; Olympus 18/f3.5
84.10 Tokina 90/f2.5; Nikon 18/f3.5
85.1 Canon L 20-35/f3.5; Tamron SP 35-210/f3.5-4.2 [surprisingly acceptable];
Nikon 35-70/f3.5, 35-105/f3.5-4.5;
Pentax 24/f2.8, 28/f2.8, 135/f2.8, 300/f4; Olympus 24/f2.8 shift
85.2 Pentax 645 + 45/f2.8, 75/f2.8, 150/f3.5, 80-160/f4.5
85.3 Sigma APO 50-200/f3.5-4.5;
Televue Renaissance 550/f5.5 (10lb 12oz, 30in long, 24ft min focus)
85.4 Bronica GS-1 (6x7) + lenses
85.5 Contax T 38/f2.8
85.6 Leica M6 + lenses
85.7 Olympus OM2S, 50/f1.8;
Tokina ATX 60-120/f2.8, 100-300/f4; Nikon 105/f2.8 micro
85.8 Comparison: Nikon 105/f2.8 micro, 105/f2.8, E 75-150/f3.5
85.9 Tokina ATX 35-70, 35-105; Tokina 35-135; Nikon 180/f2.8 (only 1lb 2oz)
85.10 Canon 300/f4, L 300/f4 [main difference: lateral color flare]
85.11 Olympus OM3, 50/f1.4; Tamron 300/f2.8
86.2 Olympus 35-105/f3.5-4.5 [excellent]
86.6 Olympus 35-70/f3.5-4.5, 100/f2
86.9 Mamiya 645 Super + 55 and 80 lenses
87.6 Olympus OM4T, 50/f1.4
87.8 Olympus 90/f2 macro
89.6 Tamron SP 90/f2.5 macro; Tokina ATX 90/f2.5 macro; Sigma 90/f2.8 macro
89.9 Tamron SP LD 180/f2.5
(I have copies of the following, but alas no dates.)
? Nikon 35/f1.4, 35/f2, 35/f2.8;
Olympus 40/f2 (5oz, 1in long), 500/f8 reflex
? Nikon 200/f2 ED [low corner cntr], 200/f4 micro [low corner cntr];
Olympus 35-70/f4 S (cheaper), 85-250/f5, 21/f2
? Nikon ED 50-300/f4.5 (4lb 13oz, $2875) [great]; Tamron SP 70-210/f3.5
II. Some numbers and comparisons
In a few days I'll post a more succinct form of these (and other) comparisons.
[x/y] = x at center, y at corner
cntr = contrast (% at 30 lines/mm); rsln = resolution (lines/mm)
Nikon = Nikkor, Olympus = Zuiko
(Canon and Olympus f1.2 lenses)
77.3 Canon AL 55/f1.2 (this is an old lens)
[38 36 48 52 52 54 52 42/20 22 36 46 48 46 42 38cntr @f1.2 2 2.8 ... 16;
49 55 55 55 69 69 62 55/31 31 44 49 49 55 55 49rsln @f1.2 2 2.8 ... 16]
83.11 Olympus 50/f1.2
[58 60 64 68 76 78 75 67/26 30 44 60 73 74 66 61cntr @f1.2 2 2.8 ... 16;
45 50 56 63 63 63 56 56/35 40 45 50 56 56 50 45rsln @f1.2 2 2.8 ... 16]
(Nikon version (non-Noct) has very low corner contrast.)
(Nikon and Olympus f1.4 lenses)
85.11 Olympus 50/f1.4
[41 45 47 56 59 57 55 48/30 35 40 45 48 42 41 33cntr;
55 62 62 69 69 69 62 62/49 55 55 62 62 62 55 55rsln];
84.3 Nikon 50/f1.4
[64..80..66/40..76..65cntr;
50..70..56/45..63..50rsln]
(Three sets of numbers for Olympus 50/f1.8)
79.9 Olympus 50/f1.8
[52 68 74 72 70 65 60/30 32 44 60 64 60 52cntr @f1.8 2.8 ... 16;
55 69 69 78 78 61 61/44 49 55 69 69 61 49rsln @f1.8 2.8 ... 16];
85.7 Olympus 50/f1.8
[47 54 56 59 60 53 42/31 39 42 44 46 40 36cntr;
55 62 62 69 69 62 55/49 49 55 62 62 55 49rsln]
? [48 60..67..50/24 39..63..56cntr;
50..71..56/40..63..50rsln];
(Large sample variation, or what?)
(Tamron SP telephotos and zooms)
89.9 Tamron SP LD 180/f2.5
[43 50 65 66 61 52 48 46/38 40 53 55 55 49 39 35cntr;
51 57 64 64 57 57 51 51/45 51 57 57 51 51 45 45rsln]
(Olympus 180/f2.8, Nikon 200/f2, Nikon 200/f4 micro all have
similar resolution but much lower corner contrast)
? Tamron SP 70-210/f3.5
[48 59 63 68 62 54 37/24 28 38 40 41 30 24cntr @210 better @70,135;
45 45 45 50 45 45 45/45 45 45 45 45 45 45rsln @210 sl.better shorter]
? Tamron SP 35-210/f3.5-4.2 [2% pincushion distortion @210]
[43 58 63 64 60 45/39 49 55 57 49 40cntr @210 similar at other f.l.;
50 50 56 50 50 45/45 45 50 45 45 40rlsn @210 similar at other f.l.]
(All three lenses are 5-6in long, weigh just under 2lb.)
(Three mirror lenses and a telescope)
77.1 Questar 700/f8 mirror lens (T-stop/11 $1000 4lb 1:4mag)
[40/37cntr; 64/60rsln @1:40]
79.6 Tamron SP 500/f8 close-focus mirror lens (T-stop/8 $735 20oz 1:3mag)
[44/28cntr; 48/36rsln @1:40; 40/30rsln @1:4] (small central obstruction)
? Olympus 500/f8 reflex (21oz, 4in long, 13ft close focus)
[58/55cntr; 45/40rsln]
85,3 Televue Renaissance 550/f5.5 (10 lb 12oz, 30in long, 24ft min focus)
[80/66cntr; 82/64rsln @f5.5; 82/70cntr; 86/68rsln @f7.2 special cap]
(Contax T versus Olympus XA, Olympus 40/f2)
85.5 Contax T 38/f2.8
[51 66 68 70 62 57/35 50 59 62 50 47cntr;
58 66 66 73 73 66/52 58 58 66 66 58rsln]
79.11 Olympus XA 35/f2.8
[38 52 56 60 58 56 60/29 34 42 42 38 24 24cntr;
42 47 59 67 67 59 47/30 32 42 47 53 47 37rsln]
? Olympus 40/f2 (ultra compact lens, 1in long)
[46 54 64 63 63 60 53/28 30 34 38 44 48 43cntr;
55 62 69 69 62 62 55/39 44 44 49 49 49 44rsln]
(I need Rollei 35T and 35S numbers!)
(Fujica 645 folder, Pentax 645, Mamiya 645)
84.6 Fujica 645 folder 75/f3.4
[69 80 90 98 90 89 80/34 40 62 66 84 70 70cntr;
50 50 64 71 64 56 56/40 40 45 50 50 50 50rsln]
85.2 Pentax 645 75/f2.8
[48 66 71 77 72 59 47/43 47 54 60 58 53 42cntr;
41 41 45 51 57 57 51/32 32 36 36 45 45 41rsln]
86.9 Mamiya 645 80/f2.8
[48 54 61 63 57 49 41/36 41 49 52 48 42 34cntr;
63 69 69 79 79 69 63/56 63 63 69 69 63 56rsln]
(Very different lenses?)
(Two Nikon 24's)
84.3 Nikon 24/f2.8
[70..76..50/48..56..35cntr; 60..67..47/37..47..37rsln]
Nikon 24/f2
[@f2 48/43cntr 53/30rsln; @f2.8+ better than above]
----------------------------------------------------------------------
Bertrand Sliwa
|> Lens Testing a la Modern Photography
|> ====================================
|>
|> I promised Bertrand I would post some Modern Photo test results, to get
|> him to do all that research :-) Looking over the Chasseur d'Image examples,
|> I get the impression that aperture-by-aperture ratings are _roughly_
|> consistent with Modern Photo numbers, but the "star" system used by C.d'I.
|> also takes into account price, expectation, and homogeneity (variation
|> across image area and aperture). (For example, the ** Olympus 50/f2 macro
|> appears to beat the *** 50/f3.5 macro, at every aperture.) Certainly
|> these factors affect the "use" of a lens, and should be emphasized in
|> the comments. But I don't think they should overshadow the basic
|> capability of a lens.
|>
Just a small correction about this comment:
- Chasseur d'Images "star" rating system, as included in the tests I posted and as they explain it, does not take into account the price of the lens, but only the performances: so if two lenses have the same number of stars, it means that they will roughly have the same performance (and not that the cheapest of the two will be less performant).
But it does take into account the homogeneity of the lens (variation across image area, aperture and focal length), so that a zoom lens rated good at some focal length and bad at another focal length will receive an average global rating. (Remark: but you have to read the comments provided with the rating to be aware of the precise performance of the lens)
I should have mentioned that they also provide a quality/price ratio (also in the form of a star rating) for each lens, but I did not include it because the prices are so different here in the US (almost 30 to 40% less than in Europe!, but the variations are not uniform even in the same brand).
Concernig the 2 Macro Olympus lenses, the rating was *** for the 50/f2 and **** for the 50/f3.5 (and not ** and ***): their comment is that on the f2 the corners are dark at f2, and the resolution is average in the center and bad in the corners at this same aperture. The performances then increase gradually when you close the diaphragm, stay a little below the f3.5 lens till f5.6 and then are the same on both till f16.
If you include the price in the comparison, the quality/price ratio they give is ** for the f2 and **** for the f3.5.
Hope it was interesting,
BS.