Go Back   Steve's Digicams Forums > Digital SLR and Interchangeable Lens Cameras > Sony Alpha dSLR / Konica Minolta dSLR, Sony SLT

Reply
 
Thread Tools Search this Thread
Old Oct 8, 2009, 7:59 PM   #1
Senior Member
 
Join Date: Jun 2006
Location: Rockingham Western Australia
Posts: 592
Default 90mm lens test continued

Thanks TCav for your always helpful input.
As suggested I have just taken (on a rather grey morning) direct comparisons between my 18-250 and the 90mm lens.

The first two are with the 18 -250mm lens on auto, Apature priority, f5.6 no AV compensation. Exposures are 1/80 and 1/160

The second two are with the f2.8 90mm lens, same details but exposures 1/40 and 1/125
Attached Images
    
Bootneck3 is offline   Reply With Quote
Sponsored Links
Old Oct 9, 2009, 9:24 AM   #2
Senior Member
 
TCav's Avatar
 
Join Date: Sep 2005
Location: Washington, DC, Metro Area, Maryland
Posts: 13,572
Default

First, let me say that dSLRs in general may vary the exposure from one shot to the next by up to 1/3 EV. That is, if a proper exposure would occur with a shutter speed of 1/140 second, a dSLR may use a shutter speed of 1/125 for one shot, and 1/160 for the next shot, even if lighting conditions have not noticeably changed. So the first shot may be slightly overexposed, and the second shot might be slightly underexposed. So nitpicking is futile.

That is not to say that you're nitpicking. What I'm saying is that there will be some variance in the exposure settings from one shot to the next, even if the camera and lens settings haven't changed.

Second, the angle of view you obtained when using the 18-250 was different enough from what you got with the 90/2.8 to affect the exposure anyway. Note that the street scene shot with the 90/2.8 has a lot more sky than the similar shot with the 18-250, so it's difficult to be sure what the camera was metering on. In looking at the histograms for the two street scenes, it appears that they were both properly exposed, however, so any dimming of the shadows or brightness of the highlights can be attributable to the slightly different framing as well as the natural variation that might occur.

The histogram of the flower shot from the 18-250, however, clearly shows that it is underexposed. Even when I select the approximate framing of the shot from the 90/2.8 within the shot from the 18-250, that portion of the image is still underexposed.

I think it's interesting that, even considering the minor variations in the angles of view, the exposure settings would be as different as they are, but the images from the 90/2.8 are more correctly exposed than are the images from the 18-250. Your preference is obviously for the less bright images that you get from the 18-250, and if you want the same kinds of images from the 90/2.8, then you may have to dial down the exposure, but I don't think there's anything wrong with the 90/2.8. If I owned the 18-250, I'd do some more tests on it. Superzoom lenses are known to vary their focal length as the focusing distance varies, and since aperture is a function of focal length, perhaps the 18-250 reports an aperture to the camera that isn't corrected for that variation in the focal length, and that might account for the less bright images you get from it.
__________________
  • The lens is the thing.
  • 'Full Frame' is the new 'Medium Format'.
  • "One good test is worth a thousand expert opinions." - Tex Johnston, Boeing 707 test pilot.

Last edited by TCav; Oct 9, 2009 at 9:28 AM.
TCav is offline   Reply With Quote
Old Oct 9, 2009, 7:12 PM   #3
Senior Member
 
Join Date: Jun 2006
Location: Rockingham Western Australia
Posts: 592
Default

Thanks again Tcav - excellent reply.

I have in fact done more comparison tests and my conclusions are that I am an idiot!!!!!

I use a laptop with an LCD screen so made a CD of the same comparison shots I published and looked at them on a friend's computer with an CRT monitor.

You are absolutely correct and the 90mm is correctly exposing whilst the 18 - 250 needs about +.7 EV correction.

Fred
Bootneck3 is offline   Reply With Quote
Old Oct 10, 2009, 8:35 AM   #4
Senior Member
 
TCav's Avatar
 
Join Date: Sep 2005
Location: Washington, DC, Metro Area, Maryland
Posts: 13,572
Default

Quote:
Originally Posted by Bootneck3 View Post
... the 90mm is correctly exposing whilst the 18 - 250 needs about +.7 EV correction.
That sounds like a lot.

I would try a simple test to see where the problem lies. Place the camera on a tripod or other stable surface, switch to A (Aperture Priority) Mode and manual focus, and take shots at different focal lengths and different focusing distances, at each of f/22, f/11 and wide open. Then compare the histograms of the images in PSE.


There are three types of errors:
  • Random Errors
  • Systematic Errors
  • Bias Errors
Random errors would be like the selection of 1/125 or 1/160 from one shot to the next. There's nothing you can do about random errors, except try to better control the environment.

Bias errors are when the operator exerts a preference for a particular result. Most of the time, if the operator is aware that he or she is unduly influencing the outcome, he or she will try to curb their biases, but curbing all biases is impossible.

Systematic errors are precise, measureable and repeatable. By performing the same test, varying the focal length, focusing distance, and aperture, you can find out exactly where the systematic error has the greatest effect on the image. For instance, you really might only need the +.7 EV correction at close focusing distances, or at longer focal lengths. You could then set the appropriate amount of correction as it will be needed. You might also find that a +.3 EV correction might be sufficient for most cases, and you could just tweak the extremes in post-processing.
__________________
  • The lens is the thing.
  • 'Full Frame' is the new 'Medium Format'.
  • "One good test is worth a thousand expert opinions." - Tex Johnston, Boeing 707 test pilot.
TCav is offline   Reply With Quote
Old Oct 11, 2009, 1:59 PM   #5
Senior Member
 
Tullio's Avatar
 
Join Date: Jan 2006
Posts: 7,370
Default

Even though the histogram may indicate that the first two shots are under exposed, I personally think the exposure is a lot better than the shots from the 90mm. On the second image from the 18-20mm, you can actually see some blue patches in the sky. The histogram shows a wide range from under exposure to over exposure. Although the ideal graph may appear to be all concentrated in the middle, as long as the majority of it is not leaning one way or another, it is perfectly fine to have it more towards the left (or the right in some cases). My preference is for a slight under exposure as the colors a better saturated, highlights are less clipped (thus higher DR) and it is always possible to recover shadows (sometimes at the expense of noise) but not clipped highlights, which are gone forever. Each lens has its own characteristics and most importantly, works differently on different cameras. The same exact lens that works wonderfully on a Canon may not do so on the Sony (and vice-verse). I've tried a few Minolta lenses on my A200 and was not impressed by them at all. However, they all got good reviews, probably from the old Minolta days. Does it mean they are not good lenses? Well, they were when used with the old film cameras but technology has changed and now the optics are probably not ideal for digital.
__________________

Tullio
Tullio is offline   Reply With Quote
 
Reply


Thread Tools Search this Thread
Search this Thread:

Advanced Search

Posting Rules
You may not post new threads
You may not post replies
You may not post attachments
You may not edit your posts

BB code is On
Smilies are On
[IMG] code is On
HTML code is Off
Trackbacks are Off
Pingbacks are Off
Refbacks are Off



All times are GMT -5. The time now is 10:46 PM.