Great example of prime vs. zoom sharpness

  • Thread starter Thread starter Ron Parr
  • Start date Start date
As some of you have noticed, Dave over at the imaging resource
goofed and posted a resolution chart with taken with a 28-70L
instead of 100mm fixed focal length lens. He now has both posted:

http://www.imaging-resource.com/PRODS/E10D/E10DPICS.HTM

Some people claim that the 28-70L is as sharp (or even sharper)
than a good prime. Still others have claimed that all lenses are
equally sharp when stopped down. I don't agree with either of
these statements and I don't think they're supported by the tests:
mine, photodo's, and now Dave's.

Here's a comparison of my 28-70L with my 50mm f/1.8 at F8. Last
time I posted this, I was told by some that I had a bad copy of the
28-70L. IMO, the softness in my 28-70L seems comparable to others
that I've seen.

http://www.pbase.com/parr/2870lvs50mm18

I really enjoy my 28-70L, but I've never seen an example where the
28-70L rivals a good prime.
Some of the users of this forum has been abusing the H*LL out of the term "prime".

A "prime" lens is not just any non-zoom lens. It's the "normal" lens for a given format. In other words, the lens where the fixed focal length is approximately equal to the diagonal length of the frame. (i.e. 50mm for a 35mm camera.)

I know it's just nit-picking about semantics, but there's no reason to start redefining terminology that's been used in a specific way for many decades. It's just going to cause confusion. People will read one thing in a book about photography and then see something else here.

Mike
 
I'm guessing that on a 4 by 6 or a 5 by 7 print, the difference would not be noticable. Maybe it shows up at bigger sizes. Ron, at what print size do you see a meaningful difference?
--
Wayne
 
I did test my 28-70 and 50 1.8 at 2.8. The 28-70 was just a littlle
sharper.
I tested my 50/1.4 against my 16-35L ( at 24 mm ), both at f/2.8. To my eyes ( and to the people who responded when I posted the crops here ), the 16 was actually a little sharper than the 50! Which is a shock, not just because of primes-and-zooms, but because the zoom was shot wide-open, and the prime was stopped-down two stops. Not to mention that normal lenses should be sharper than ultrawide ones.

Each time I get a new lens, I test all of them against each other. A lot of photos could be taken with any lens, and for those shots, I'd like to have some idea how each of my lenses would do. The 50 lost one test, and won all of the others.

My 50/1.4 seems to have very bad autofocus; it's reliable enough in the real world, but it struggles with 2D newspapers. When I do these tests, I have to shoot several exposures with the 50, and pick the best focus. In the real-world, I'm not sure if it's DOF, more contrasty subjects, or what, but it isn't a problem nearly as often.
 
My 50/1.4 seems to have very bad autofocus; it's reliable enough
in the real world, but it struggles with 2D newspapers.
I also had bad autofocus with my 50/1.4. When I tested it against the 24-70, the 50/1.4 lost by a significant margin at f/2.8 but it was more due to inaccurate focus than sharpness of the lens. At f/8 the 50/1.4 won.

--
Steve
 
I'm guessing that on a 4 by 6 or a 5 by 7 print, the difference
would not be noticable. Maybe it shows up at bigger sizes. Ron,
at what print size do you see a meaningful difference?
Don't take this wrong way, but if all we care about is printing 4x6 or 5x7 shots straight from the camera then a D30 with any old lens will probably be fine.

If I'm paying for a 6MP camera with a great sensor, then at least some of the time, I'd like to get 6MP worth of crisp, beautiful detail that I can work with. If I decide at some point later that I'd like to crop and print a 4x6 of a small portion of a shot, it's nice to know that the detail is there and that I have the flexibility to do this.

Of course, sometimes we make sacrifices in detail to actually get the shot. I think this is what zoom lenses are all about. I'm not saying that this is bad choice. In fact, I uses zooms much more than I use fixed focal length lenses. I'm just underscoring what the tradeoff is because I think a lot of people don't realize that there is a tradeoff.

--
Ron Parr
FAQ: http://www.cs.duke.edu/~parr/photography/faq.html
Gallery: http://www.pbase.com/parr/
 
I know it's just nit-picking about semantics, but there's no reason
to start redefining terminology that's been used in a specific way
for many decades. It's just going to cause confusion. People will
read one thing in a book about photography and then see something
else here.
Hmmm... While I don't know how the term originated, the use of "prime" as a synomym for "fixed focal length" is quite widespread. I've found references to this in photo.net's learning area and even in Canon's product literature:

http://www.canon.com/bctv/products/pdf/hd.pdf

(Note that they talk about offering a range of zoom and prime lenses and they offer several different 'prime' lenses with different focal lengths. They seem to have just two different imager sizes, but all of the lenses are called 'primes'.)

I don't doubt that you're right that prime means something more specific than "fixed focal length", or that at least it used to have this meaning before so many people started abusing the term. However, I did a little searching with google and had trouble finding anybody using it in this more precise manner.

It would be great if you could provide a link to a definitive source on this, e.g. somebody like Zeiss. I'd love to add it to my FAQ.

--
Ron Parr
FAQ: http://www.cs.duke.edu/~parr/photography/faq.html
Gallery: http://www.pbase.com/parr/
 
There is a huge difference between the two at full size. It isn't even close.
As some of you have noticed, Dave over at the imaging resource
goofed and posted a resolution chart with taken with a 28-70L
instead of 100mm fixed focal length lens. He now has both posted:

http://www.imaging-resource.com/PRODS/E10D/E10DPICS.HTM

Some people claim that the 28-70L is as sharp (or even sharper)
than a good prime. Still others have claimed that all lenses are
equally sharp when stopped down. I don't agree with either of
these statements and I don't think they're supported by the tests:
mine, photodo's, and now Dave's.

Here's a comparison of my 28-70L with my 50mm f/1.8 at F8. Last
time I posted this, I was told by some that I had a bad copy of the
28-70L. IMO, the softness in my 28-70L seems comparable to others
that I've seen.

http://www.pbase.com/parr/2870lvs50mm18

I really enjoy my 28-70L, but I've never seen an example where the
28-70L rivals a good prime.

--
Ron Parr
FAQ: http://www.cs.duke.edu/~parr/photography/faq.html
Gallery: http://www.pbase.com/parr/
Actually when you compare the original (not resized) versions of
these images there is very little difference. If we were all
shooting film we might notice the difference in sharpness, but with
digital and the advent of unsharp mask it would be very easy to
make a print from each lens and have one be just as sharp as the
other.

--
 
Wow. That looks pretty bad. Did it get dropped or something? You should send the lens back to Canon for service. It'll probably cost $150 or so (get a quote first), but they should be able to take it apart and clean it and make sure everything's aligned and working correctly.

--

I just want to take really spectacular pictures of my vacations, pets, family, projects, and drunk friends.
 
I'm not suggesting that an SLR is appropriate if the largest print size one would ever make is 5 by 7. I just think that recognizing that primes are better than zooms doesn't mean much to me without a reference to the print size.

I guess one advantage of a zoom is that one typically would not need as much cropping at whatever the print size (assuming that the "right" focal length is in between two primes).

So as I'm thinking through these trade offs, at what print size does prime vs. zoom make a meaningful difference?

--
Wayne
 
to consistently AF accurately provided it was properly adjusted at
the factory. I'm able to consistenly get tack sharp AF images of
small subjects even at f/1.4 but lets leave that for another time
and thread.
I would certainly like to see what you have to says about this...
 
As some of you have noticed, Dave over at the imaging resource
goofed and posted a resolution chart with taken with a 28-70L
instead of 100mm fixed focal length lens. He now has both posted:

http://www.imaging-resource.com/PRODS/E10D/E10DPICS.HTM

Some people claim that the 28-70L is as sharp (or even sharper)
than a good prime. Still others have claimed that all lenses are
equally sharp when stopped down. I don't agree with either of
these statements and I don't think they're supported by the tests:
mine, photodo's, and now Dave's.

Here's a comparison of my 28-70L with my 50mm f/1.8 at F8. Last
time I posted this, I was told by some that I had a bad copy of the
28-70L. IMO, the softness in my 28-70L seems comparable to others
that I've seen.

http://www.pbase.com/parr/2870lvs50mm18

I really enjoy my 28-70L, but I've never seen an example where the
28-70L rivals a good prime.

--
Ron Parr
FAQ: http://www.cs.duke.edu/~parr/photography/faq.html
Gallery: http://www.pbase.com/parr/
--

max..using a 1Ds it is my impression that my 16-35 is sharper than my 20 f2.8 and less so as compared with my sigma 50mm f2.8 macro. its colors are much warmer than either prime.
 
Some people claim that the 28-70L is as sharp (or even sharper)
than a good prime...

I really enjoy my 28-70L, but I've never seen an example where the
28-70L rivals a good prime.
...a lens which has no zoom and is fixed focal length such as the 50mm 1.8

They are far sharper than all but the very best (expensive) zooms and sharper even than those.
Regards,
--
DaveMart
 

Keyboard shortcuts

Back
Top