Canon 50mm f/1.8 Sweet Spot Test

Joined
Nov 26, 2010
Messages
670
Reaction score
21
Location
NJ
Can others edit my Photos
Photos OK to edit
Hello,

I performed a test where I taped a page from a magazine onto a stand and took shots from a 50mm on my tripod mounted xsi with a remote shutter at every available aperture incrementing by 1/3 stops in Manual mode. I was about 5 feet from the page for each shot.

It looks like the sweet spot is at 5. This was clearly the sharpest aperture. Anything above or below was progressively softer. I can't believe how soft f1.8 was. It's funny that I've had this lens for about 2 years and I never conducted a test like this. I'm going to do it next with my 17-40. Any suggestions on what focal length to take these shots from? I was planning on setting it 18 and then run the test again at 28 and again at 39.

Here's the link for the pics from the 50mm test if anyone is interested. They are jpegs straight from the camera.

Test Shots for Canon 50mm f/1.8 - a set on Flickr

Danny
 
Nice Maine advert from Pop Photo (?)
 
Yes, the 'Nifty Fifty' [Canon 50mm F1.8] is well known to be fairly soft at F1.8. Perhaps that's a result of it being designed & built to reach it's low price point. AFAIK, it's the least expensive EF lens.
However, when used at F5 (or wherever you find the sweet spot to be) this lens is usually capable of great image quality, which is why it's still a great bargain.

More expensive lenses 'should' tend to be better....or at least, they shouldn't loose so much of their potential image quality when used wide open. I've read that the big telephoto L lenses are designed so that their sweet spot is very close to wide open, because that's where they will be used 95% of the time.
 
Yes, the 'Nifty Fifty' [Canon 50mm F1.8] is well known to be fairly soft at F1.8. Perhaps that's a result of it being designed & built to reach it's low price point. AFAIK, it's the least expensive EF lens.
However, when used at F5 (or wherever you find the sweet spot to be) this lens is usually capable of great image quality, which is why it's still a great bargain.

More expensive lenses 'should' tend to be better....or at least, they shouldn't loose so much of their potential image quality when used wide open. I've read that the big telephoto L lenses are designed so that their sweet spot is very close to wide open, because that's where they will be used 95% of the time.

I'm going to perform the same test with my 70-200 f/4 as well. I'll post those also. I'm not sure how useful it will be to anyone else, but maybe.

Danny
 
If you are going to be super critical...keep in mind that both cameras and lenses have a tolerance when it comes to focus. For arguments sake, lets say that it's plus or minus 5 points. So your camera may be sitting at -4 points, which is within the tolerance, so it technically OK. Now if you use a lens that is say, -3....you are now at -7, which may be enough to show in your images.
But if you use a lens that is +4, it works out perfectly on the camera and you get very accurate focus.

So because of this, if you called up a Canon service centre and told them about your focus problems...they would likely suggest that you send in your camera and your lenses, so that they could be calibrated together. Most of us don't go that far, but I know a few who do.

Also, the higher end Canon bodes now have micro focus adjust. This lets you dial in a specific amount of focus compensation for a specific lens. So every time you use that lens, the camera remembers how much adjustment to use. Sounds like a great feature, but you'd have to do some careful testing to really nail it down.
 
If you are going to be super critical...keep in mind that both cameras and lenses have a tolerance when it comes to focus. For arguments sake, lets say that it's plus or minus 5 points. So your camera may be sitting at -4 points, which is within the tolerance, so it technically OK. Now if you use a lens that is say, -3....you are now at -7, which may be enough to show in your images.
But if you use a lens that is +4, it works out perfectly on the camera and you get very accurate focus.

So because of this, if you called up a Canon service centre and told them about your focus problems...they would likely suggest that you send in your camera and your lenses, so that they could be calibrated together. Most of us don't go that far, but I know a few who do.

Also, the higher end Canon bodes now have micro focus adjust. This lets you dial in a specific amount of focus compensation for a specific lens. So every time you use that lens, the camera remembers how much adjustment to use. Sounds like a great feature, but you'd have to do some careful testing to really nail it down.

I have a buddy that has a 7d with a 24-70 lens and and when he first got them noticed his pictures were coming out a bit soft. He was about to send it in for calibration, but decided in the end it was ok. I don't think I would go that far with my XSI, but I'm close to picking up a 60d and if I have that kind of problem there, I might do it.

I'm not a perfectionist, but when it comes to sharpness, I need it!
 
I'm not a perfectionist, but when it comes to sharpness, I need it!

Maybe you should have thought about that before testing out a crap lens, well in comparison to a quality prime anyways.

Yea, you have to keep your expectations realistic when your shooting with a lens that cost around $100. I have the same lens, and love it for the price. At the same time, I find my tamron 28-75 f/2.8 to be a tad bit sharper but i paid for it.
 
MWG said:
Yea, you have to keep your expectations realistic when your shooting with a lens that cost around $100. I have the same lens, and love it for the price. At the same time, I find my tamron 28-75 f/2.8 to be a tad bit sharper but i paid for it.

Absolutely! Like you said, for 100 bucks it's a great lens.

Sent from my iPad using PhotoForum
 
I would test at around,at least, 100 times the focal length of the lens for any non-macro lens. I'd suggest 16 to 20 feet for a 50mm lens; you'll actually be able to more easily see some variation in resolving power and acuity at a little bit longer test shooting distance. Also, there is nothing in the magazine test setup that shows the edge performance of the lens...this is testing mostly the central area of the lens. Most 50mm lenses have pretty BAD corner imaging performance at f/1.4, and often have an overall, whole-frame veiling softness at f/1.4--even on pretty good samples from many makers. Stopping down to f/2 usually improves things quite a bit. A 50mm lens that performs best at f/5 to f/5.6 sounds pretty normal to me. My favorite old manual focus 50mm f/2 is superb at f/5.6, but not quite as good at f/4.

Have you tried taping or stapling newspaper sections to a wall and photographing those? They make great test targets, and are pretty affordable.
 
I would test at around,at least, 100 times the focal length of the lens for any non-macro lens. I'd suggest 16 to 20 feet for a 50mm lens; you'll actually be able to more easily see some variation in resolving power and acuity at a little bit longer test shooting distance. Also, there is nothing in the magazine test setup that shows the edge performance of the lens...this is testing mostly the central area of the lens. Most 50mm lenses have pretty BAD corner imaging performance at f/1.4, and often have an overall, whole-frame veiling softness at f/1.4--even on pretty good samples from many makers. Stopping down to f/2 usually improves things quite a bit. A 50mm lens that performs best at f/5 to f/5.6 sounds pretty normal to me. My favorite old manual focus 50mm f/2 is superb at f/5.6, but not quite as good at f/4.

Have you tried taping or stapling newspaper sections to a wall and photographing those? They make great test targets, and are pretty affordable.

I didn't try the newspaper thing, which sounds like a great idea with all the small print. I did find this sheet of paper which was infinitely better than the magazine page I used for the 50 mm testing.

Testing Camera Lenses - Sharpness, Chromatic Aberration and Distortion - Bob Atkins Photography

I used it to test my 17-40 last night, which by the way, is best at 7.1 (for me anyway), and it worked much better than the magazine page.
 

Most reactions

New Topics

Back
Top