The "look" of three 50mm

[back alley]

i disagree frank.

if i never shoot with a tripod then wouldn't a test WITH a tripod be out of the norm and have skewered results for me?

 

[raid]

Frank and Joe: Both of you are right here; a tripod based test is more testing the lenses for "everybody" and not just for me, the user. If and when I repeat the test handheld, the results would be applicable to me, the user, since maybe my hand shakes or maybe I can't focus well and so on. I have done the test for RFF and will then repeat the test for myself.

 

[FrankS]

It won't be applicable to anyone, even Raid, because camera shake during a handheld exposure is not a constant from one shot to another. Some shots will have more shake, others less, independent of what lens is tested, so the results will be a useless mixture of lens performance and the amount of camera shake during each exposure. The critical point is that camera shake is not a constant for each exposure, and is therefore a variable. A very influential variable, I might add. An experiment with more than a single controllable variable is basically flawed, and pretty meaningless.

 

[back alley]

i'm not convinced but then what do i know?

 

[FrankS]

I have a science background and can be very opinionated/bull-headed, so I'm very confident in my statements on this issue. 😉 😉 😉

 

[FrankS]

To explain: let's just say that by chance, every exposure make hand-held with lens A was degraded by more camera shake than exposures made testing other lenses. You could erroneously come to the conclusion that lens A is a dud. The amount of camera shake during each exposure is not a constant and can vary and therefore influence the results independently of the optical performance of the lens tested.

 

[back alley]

i'm really not arguing or looking for one.

my logic is usually off from others in many things.

i'm just thinking that FOR ME, my eyes are not as sharp as they once were. my hands DO SHAKE a bit now.
those same eyes and hands will take every shot i take.
so when i shoot a scene with 4 different 35mm lenses one lens may get a sharper result with me at the helm because it has a physical property that assists me in getting a sharper image, like size or an infinity lock.
that same lens might prove itself a less capable performer on a tripod mounted camera though. so those results would not help me at all.

 

[Doug]

Well, Frank is quite right; in any test you need to eliminate all other variables to the greatest degree possible. Then you can be assured that the test is testing what you want to test! 😀 (errr, Mechanical Engineer, here.)

Edit: Joe, I guess I could add that if in actual practical shooting, other factors that degrade lens performance constantly overwhelm the quality of the lens itself, then there's probably no point to the test at all, and it simply doesn't matter which lens you use.

 

[Honus]

Raid,

Thanks for going to the trouble of doing this. I look forward to seeing the results.

I also agree with Frank. Using a tripod eliminates an unkown variable (camera shake).

- robert

 

[FrankS]

Yes, let's not let this difference in opinion overshadow the efforts made by Raid in doing this. As a collector of 50mm lenses, I am very interested in this!

 

[furcafe]

Not to discount Raid's noble experiment & I'm not disagreeing w/you regarding handholding being a significant variable, but, per Doug's post, isn't the photographer's ability to focus also a non-constant variable (particularly if he's focusing on a 3-dimensional object)? I would think that if the shooter uses a high enough shutter speed, camera shake might not be necessarily more variable/inconsistent than his focusing differences between shots.

It won't be applicable to anyone, even Raid, because camera shake during a handheld exposure is not a constant from one shot to another. Some shots will have more shake, others less, independent of what lens is tested, so the results will be a useless mixture of lens performance and the amount of camera shake during each exposure. The critical point is that camera shake is not a constant for each exposure, and is therefore a variable. A very influential variable, I might add. An experiment with more than a single controllable variable is basically flawed, and pretty meaningless.

 

[celluloidprop]

I understand where FrankS is coming from, it certainly makes sense, but I've never attached my Leica to a tripod. Rangefinder-focusing wouldn't lend itself well to tripods, I should think. Tighten down, focus, loosen to compose/recompose, tighten, etc. For that I'll use an SLR or my Crown Graphic.

Which might, ultimately, mean that 'lens quality' in terms of sharpness is ultimately irrelevant in my case, as I'd never be using it under optimal conditions. Contrast, 'bokeh' and fingerprint would still show through, however, and are probably more important considerations for me (cf. all the folks who love their Elmars enough to make me consider finding one, even though it doesn't blow anyone away in testing).

 

[raid]

Well guys, I have the results. Before uploadiing the images, I need from you one more advise to do or not do something; should I use ADOBE to improve the look of the scans, as it is usually insisted on here and PN, or shoud I for test reasons leave things as they are even if the results may not looj "up to par"? After getting your advices, I will upload the images and start a new thread so that more people can see the results.

 

[regit]

i'm really not arguing or looking for one.

my logic is usually off from others in many things.

i'm just thinking that FOR ME, my eyes are not as sharp as they once were. my hands DO SHAKE a bit now.
those same eyes and hands will take every shot i take.
so when i shoot a scene with 4 different 35mm lenses one lens may get a sharper result with me at the helm because it has a physical property that assists me in getting a sharper image, like size or an infinity lock.
that same lens might prove itself a less capable performer on a tripod mounted camera though. so those results would not help me at all.

To increase the rigor of the test, Raid may need to take 3 shots per aperture setting, with and without tripod and choose the sharpest of the lot (6 choose 2) ... but ...

Though not foolproof, results with tripod is more generalisable. That is, if so desire, anyone can replicate (more or less) tripod-mounted quality. Conversely, if non-mounted shots are subjected to individuals' idosyncracies, it doesn't really matter if Raid made them or not since everyone would produce different results.

Mounted shots can cover (more or less) some of the technical and aesthetic aspects of a lens, i.e. things like relative sharpness, fall-off, bokeh, look, etc...

I guess the point of the exercise is not to produce definitive results, but a starting point for further considerations...

 

[back alley]

understood...

 

[raid]

Take it easy guy. Let's accept this test as a first test with several inherent faults, but maybe with some information on the lenses. Back to my question; should I do any standard manipulations for all photos? Maybe use Autoadjust for all photos so it is fair?

 

[FrankS]

Raid, do you have enough room to post them both ways, autoadjusted and plain?

 

[wlewisiii]

I'd say adjust them except for any sharpening. That seems to hash things the most.

William

 

[raid]

Frank: I don't think so. Each scan has size half MB, and I don't belong to any fancy website where I can have huge amounts of memory. I have PN which lets me do some posting there. My test turned out to be a real flare test. I shot against a sun lit window. I got some surprises ...

 

[FrankS]

I honestly don't know which way of psoting them would be the best. Anyone else?