PDA

View Full Version : Testing



Mircula
15-07-2010, 5:20pm
Hey,

I just went through my library and noticed that a lot of my images taken with me nifty fifty are not really that sharp.

I would like to compare my 28-105, nifty, and kit lens in terms of iq, sharpness, vignetting and CA. etc.

I was just thinking to set my camera up on a tripod somewhere outside and do the same shots.

Do you have any tips how i could setup the tripod to see the most differences (what light, inside or outside, what kind of subject etc.)?


Thanks for the help!

R1titan
15-07-2010, 5:29pm
From my experience the 50/1.8II is a sharp optic even with aperture at its widest, its just that the lack of focus accuracy (esp in poor light) that lets it down...
Ive still got one in my kit although its been replaced by my Sigmalux.

ricktas
15-07-2010, 5:31pm
It could be the lens of the lens and camera combined causing the issue. You can get a camera body and lens calibrated to work with each other perfectly.

Now to some testing. It could be that your version of the lens is a little soft, but it could more than likely be that it is not focusing 100% accurately. This is often referred to as front of back focusing, where the lens focuses either slightly closer or further away than the actual object you lock focus onto. Also be aware that at f1.8 the nifty fifty is not that sharp, it is much sharper at around f8.

Now onto how to test your lens/body combination. Give this a try: http://regex.info/blog/photo-tech/focus-chart

ameerat42
15-07-2010, 5:32pm
Not exhaustive, for sure, but for a start: when I'm not pointing at a little printed grid chart (A4 size, in fact) in the room, I use the very uniform back brick wall lit by the sun. It has straight lines, pretty even colour and tone, and is all approximately the same distance away. Am.

Mircula
15-07-2010, 5:59pm
Thanks for the info guys. The testing from the link sounds pretty complicated and it only addresses the focus, as far as a quick read over tells me...

I would prefer a more reality near setup where i then can compare the overall quality of the shot including the things i mentioned in my first post.

Here an example of a shot with the nifty...

The kid looks kind of grainy noisy and not sharp, although at iso 100.
Maybe i just did not focus properly.....
Hmmm...i guess its kinda hard to see with the upload quality...

Although at iso 100.

What do you think.

RaoulIsidro
15-07-2010, 7:09pm
Going back through your library... are there any images below 1/60 sec? If yes, they may be motion blur and not the lens.

For the test, you should choose a common aperture for all lenses. f4 would be a good start, because the other lenses will not be able to match the f1.8 of the 50mm.

Choose a flat brick wall and stick a length of masking tape as a cross hair. Use a marker to write your name on the masking tape strip, then test away...

Mircula
16-07-2010, 9:10am
It is not motion blur i think. I used shorter shutter speeds.

Thanks for the testing advice.

I will try it out.

R1titan
16-07-2010, 3:56pm
Looks a little soft focused to me.
That is exactly what i mean by the iffy focus of this lens.
Here's a simple test...focus on something about 1-1.5 metres away, and then repress the shutter halfway to refocus on the exact same spot even though it has already found focus from the first try.
If you hear the buzz of the focus motor on the reattempt then the lens has nervous/inconsistent focus issues...

maccaroneski
16-07-2010, 4:02pm
This is often referred to as front of back focusing, where the lens focuses either slightly closer or further away than the actual object you lock focus onto.

I don't mean to be pedantic but there is a little typo there which might give people trouble if googling the term - I think Rick means front or back focusing.