Happy Mother's Day! Happy Father's Day! Happy Graduation! Photo.net has great photography gift ideas for the Mom, Dad, or Grad in your life. Shop for camera bags, lenses, DSLRs, and more...
Clubs, bars, and small venues are the places where most concert and live music photographers get their start, the reason being that there are fewer restrictions since the performers are less likely to...
I see a lot of "users" commenting in on-line forums that the Canon EF 75-300
lenses are very soft, especially at 300mm, sometimes commenting that the lens is
"unusably soft". While I agree that they aren't the sharpest lenses, I often
wonder just what these users are seeing. I wonder if some of them are just
repeating "conventional wisdom", which in this case may not be all that wise, or
if the results they are seeing is more a consequence of how they use the lens
than the lens' intrinsic quality (or lack of it).
So, having both the EF300/4L USM and the EF75-300/4-5.6 IS USM available, I
decided to do some testing. I didn't have a non-IS version available for testing,
but a number of years ago I did test the IS lens against a non-IS version and I
really didn't see a significant difference. The Canon MTF plots for the IS and
non-IS versions of this lens seem to support the conclusion that their
performance is essentially the same
Test Charts
These images are 100% crops from images of a USAF 1951 resolution test chart,
shot using the 75-300/4-5.6 IS USM at 200mm and 300mm, and a 300/4L USM. For the
shots at 200mm the camera and lens were moved closer to the target to maintain
the image size. The camera and lens were supported on a sturdy tripod and head
(Bogen 3051 with Arca Swiss B1). Autofocus was used. The first set of crops are
from the center of the image.
From these images it seems the difference between these lenses
isn't huge (though see the "real world" images later for more on this). It's
quite clear that the 75-300 IS at 300mm improves as it's stopped down. Contrast
and sharpness are higher at f8 than f5.6. At f5.6 and f8 the 300/4L USM is
sharper and has higher contrast.
The second set of crops are taken from the corner of the EOS 20D
frame using the same lenses.
It's obvious here that the 300/4L shows less chromatic aberration
than the 75-300 IS. The 300/4L also has higher contrast.
Real World
Resolution charts are one thing, but the results of such tests don't always
easily translate to "real world" performance. For a real world test I took shots
of the scene shown below at various apertures with both lenses. The area outlined
in red is shown as a 100% crop in the next set of comparative images
It's fairly clear here that the 300/4L outperforms the 75-300 IS. In fact the
300/4L is better at f4 than the 75-300IS is at 300mm and f8. The difference on
real world targets is certainly more obvious than it was on the test charts. So
is the 75-300 "unusably soft"? Well, all is not lost. Digital processing come to
the rescue as shown below:
Here we see the results of some fairly aggressive unsharp masking
on the image from the 75-300 IS. As you can see, the 50% crops now look very
similar indeed. Some people might regard this as "cheating", but the results
speak for themselves I think. On my monitor (17", 1280x1024 resolution), the
crops above represent sections of a 20"x30" print.
Conclusion
Despite lots of negative comments on the Canon 75-300 lenses, I
think many of them may stem from users who simply don't know how to use a 300mm
lens. This could be because the 75-300 lenses are seen as "starter" lenses for
beginners, and so tend to be bought and used by less experienced photographers.
They may be shooting handheld at shutter speeds too low for sharp images, they
may be shooting moving subjects or they may not be optimizing their images with a
good image editor. I don't know. I do know that the EF 75-300/4-5.6 IS USM is a
decent lens. It's not as good as a 300/4L, it's not as sharp and it focuses
significantly slower, but when used properly and with the help of a little
digital magic, it's a lens capable of yielding pretty sharp images. This
especially appplies when shooting digital if, when at 300mm, you stop down a
stop, optimize the image in PhotoShop and don't try to make huge prints.
I've had one in my camera bag for 6 or 7 years now and I use it quite a lot.
It's not the only lens I have and it's not the lens I reach for first when I want
the highest possible quality at 300mm, but it's a lens that has proven both very
useful and capable of making images of surprisingly good quality when used within
its limitations.
Where to buy
Purchasing from these vendors via these links helps support photo.net
Bob! I must say this was a brave test. I can hear the pixel peepers screaming from my desk as I type. I had the 75-300 IS and liked it; I sold it and bought the 70-200 2.8 because I needed the speed. I then traded that and cash for the 70-300 DO (because I no longer shoot weddings and don?t need the speed). I like your use of the phrase "real world" I cant remember the last time I saw gallery show with computer monitors set to show the image at 200%. If some of these pixel peepers would spend more time taking photos and less time looking at their images at 500% they would be much happier. I am fairly certain if the 75-300 IS lens is used by you for work, its good enough for us mere mortals!
Bob,
Thanks taking the time to do the test. I happen to be looking for a mid range tele zoom that's light enough for hiking and travel (I have been using the 70-200L 2.8 with the 1.4x, but need something lighter to lug around all day and not draw attention to itself.) There's a positive review on this site on the discontinued 70-210 3.5-4.5 USM, which I'm considering along with the 100-300 4.5-5.6 USM, and of course the various versions of your 75-300. It seems these are all f8 lenses, at least at the long end (the 1.6x digital crop sure helps). Which is fine, and if I choose one of them it's for the focusing speed. You mentioned somewhere that you had the first generation (clunky) 100-300L 5.6. Do you remember much about weight, operation, and wide open performance of this lens? Thanks again.
Thanks for sharing this! As always, you manage to set a new perspective on the same wisdom one gets to hear over and over!
One thing I wanted to add is the price diffrence between those two lenses... Here in Europe (at least where I live), the 75-300IS goes for 474 euros, whereas the 300f4 goes for 1188...
Personally I'd prefer having a not so sharp shot with the zoom over NOT having a sharp shot with a much more expensive lens I will never buy because it is a lot more expensive and the focal length is not that useful to me anyway!
Technical issues are fine and important, but in the end the most important to me is getting the shot, no matter what I use to get it!
Remember the IS-feature. With long lenses, shot hand held, the most common unsharpness is related to shake - not optical quality. And even if the EF 75-300 is optically soft, in practise, if there is even a slight lack of light, the IS feature totally outruns the importance of the optical quality. It really is worth two stops or even more. You can get "sharp" shots with 1/60s using full length! I had an EF 75-300 IS before my state of the art EF 70-200/2.8 IS L. I still think the EF 75-300 was more value for money, despite the fact that the new top lense is superior in optical quality. In practise the difference in photo quality is not that big, but the difference in price is huge - about four fold. I've long considered the EF 28-135 IS and the EF 75-300 IS to be the best value for money in the Canon zoom-lense range. Reasonably priced and yet you get the job done - usually well.
Nevertheless, the EF 300/4 L is great lense too - and quite a lot more expensive than the Ef75-300 IS.
I did have the 100-300/5.6L at one time, but I sold it soon after I bought a 300/4L which must have been 10 years ago. I remember testing it against the 75-300 and the 300/4 and as I recall it was better then the 75-300, but not as good as the 300/4. It's a bit of a clunky lens, slow and noisy focus and only f5.6 at 100mm, but if you can find a used one cheap and you don't mind the ergonomics, it could be your best sharpness bang for the buck.
I still consider the original EF300/4L to be the overall best choice 300mm lens from Canon (or anyone else). It's sharp, small, light, fast, not too expensive ($700 used) and works very well indeed with the Canon 1.4x TC and passably well with the 2x (especially if you stop down to f11!). The IS version is also very good, though a bit more expensive of course.
I had the 75-300 for 6mths before it broke, it's fine if you don't start cropping, but many dslr users will likely try to start shooting little critters, birds and things that are just too far away with this lense, and you're likely to be cropping down to less than 1/4 frame and then the softness shows. If they were able to "fill the frame", then you wouldn't see this complaint as often. So it's probably the same old problem... "you're not close enough".
I had the 75-300 MKII USM and I found it soft. I would often try and use it wide open for effect but even stopped down on a tripod I thought it wasn't as good as I expected. I sold it after a couple of months. 18 months ago i bought the dinosaur 100-300 f/5.6L and was so impressed on the images from this lens over the USM lens. They were very detailed and sharp. However the 100-300 f/5.6L lens only performs well(ish) if you have plenty of light to maintain shutter speed and focusing. Focusing is slow at the best of times but I found on the EOS3 that it was very useable with shutter speeds of 1/500 @ ISO 320. On lesser bodies it may not be so useable. On my 10D it is pretty unusable due to the 10D's poor AF and also the extra shutter speed required for the digital crop factor. I still have the L lens and may use it on my EOS 50e (elanII) this summer - I can't afford the current fast L lenses.
I've never noticed excessive flare with the 75-300 IS. I tend not to use filters, which helps, but I also tend not to use a lens hood either, which doesn't help.
I'd guess you'd see more flare and ghosting shooting directly into the sun since the 75-300 IS has a lot more elements than the 300/4L.
If I get a chance, I'll take a couple of test shots, but I can say that flare has not been a characteristic I've associated with 75-300.
thank you for your test. I often used the 75-300mm IS lens, now it often sticks in the camera bag because for most times the Sigma 18-125mm is long enough with the EOS 300D for most of the images I take.
What I found with the 75-300mm lenses (I had the I-Version of the 75-300mm before) was a soft and often greyish appearence of images taken at 300mm wide open. The main cause for that is probably the fact, that these lenses do not have a very good anti-reflective coating on the optical elements and that some stray light might get scattered inside the lens tube. Using a lens hood helps much, but with digital imaging you can easily remove the resulting gray level of the image to a real black. This is as much "cheating" as the unsharp masking you performed on the test image.
Nevertheless, the better the lens is, the less artifacts you will get and the better the results are. But everyone has to check if it is worth the 500 or more than 1000$ extra for an EF 100-400mm IS or an EF 70-300mm DO IS. For me, it was not and so I am very happy with my EF 75-300mm IS. Plus if it is worth the extra weight in case of an EF 100-400mm lens.
I have tested another Canon consumer tele-zoom, the EF100-300mm/4.5-5.6 USM using a 055CL/486CR2 - MLU and self-timer combination with my XT/350D (meaning 1.6xcrop), and found out that this lens should be seen and used as (my arbitrary minimum standard being able to get decent 8x10 enlargements):
- At 100mm: a f/5.6 lens
- At 200mm: a f/8 lens
- and at 300mm: an f/11 lens.
Within these limits, it gives "very good" performance for the money. Wide open, especially at the long end, it is JUNK - joke of a lens if the ability shoot wide open is what you are after. You do not need 8x10s to see what I mean, you will see it I by looking at 4x6s.. That bad.
Now, of course, those parameters will force you to use this lens @ ISO 400 and ISO 800 (the latter more likely) or even ISO 1600 if you want to handhold and stick to the 1/focal length seconds formula at the same time. Using a monopod shaves off 2 extra-stops for me, and gives me flexibility to move around.
If you are O.K. with using this lens on a tripod, ideally with MLU, then lower ISO levels might be possible, and the results are even better.
A real advantage of this lens vis-a-vis the 75-300 IS reviewed here (or the non IS 75-300 lenses for that matter) is the awsome focus speed, even on entry level bodies (my references are to Rebel G, Elan 7 and XT/350d). Add to that the better build quality, and full-time manual focusing (although, I would not dare use entry level viewfinders for manual focusing - ever...), I think this lens comes a head of the rest of the cheap Canon tele-zoom pack, including the discontinued EF 70-210/3.5-4.5 USM.
So, I guess I am one of those guilty of constantly whining about the quality of Canon's consumer tele-zooms, but I have tested them out for myself, and have seen that within certain limitations, they give pretty good results, and 'very good' results if you factor in their price levels.
I owned an EF 75-300 II but in those days i shoot on film. The lens was much below my usual standards - extremely soft and washed out colors even with the hood allways on it. I owned it for three months than give it away. But i guess that was before "Photoshop magic".
Hi Bob. Thanks for the comparison article. I always enjoy your perspectives on different lenses.
I own two IS lenses, the 28-135mm IS and the 70-200mm IS. I think they are both spectacular. However, there is one thing I'd like to suggest about the IS lenses that I don't ever seem to see mentioned. That is that the IS feature does take some time to stabilize. If users expect to rattle off pictures here, there and everywhere without letting the IS do its thing the result will be soft pictures. I do a lot of bird shooting with a 400mm with no IS and, if I'm not careful, when I go back to my 28-135, my everyday lens, I'll end up with soft pictures. I suspect this is the biggest softness issue for many 75-300mm IS users. Just my 2 cents.
I agree that the unsharp-masked 70-300 birdhouse is comparable to the 300/4 in your crop, but the background of the 75-300, in my opinion, is distractingly distorted.
I don't see how the claims are not true if USM and f/8 have to be used to produce acceptable results where another lens can do it without. I bought the 75-300 IS after reading an earlier review/comparison on IS.
For me the softness at 300mm is very apparent in the majority of shots, because not always am I able to use f/8, or shoot a perfectly still subject. Indoor catwalks and moving models come to mind. This lens is NOT built for certain things. And the focussing speed really can be a bother. I'm not enitrely unhappy with my purchase though, because I got it used and the price suit me fine.
How can you tell? I haven't posted anything from which bokeh could be objectively judged.
I'd also add that 95% of photographers interested in the 75-300 IS probably don't even know what Bokeh is and probably most them wouldn't care about it if they did.
I'm not saying the bokeh is great, just that there's insufficient data here to draw any real conclusions.
I am surprised at how well the zoom held up against the pro prime. I was personaly expecting the 75-300 to do worse than it did. Although some people might decry tests like these irrelevant, I find them very useful because they help dispel alot of disinformation that spreads around on the net about cameras. Great write up Bob.
I also used the 75-300 IS for several years before moving on to the 70-200 2.8/IS with the 1.4 TC combo. Aside from the slow zoom (a fatal flaw shooting action), lack of manual focusing while switched to auto focus, and a revolving front lens element (making the use of a polarizer a PITA), I simply could not accept the degree of sharpness and color washout that I experienced with the 75-300 IS fully extended. But then, while I have been very pleased with the performance of the 70-200 IS, particularly in the color rendition and bokeh, I was quite surprised to see the extent of image degradation caused by the 1.4 TC (see http://photo.net/equipment/canon/tc1/). That comparison has now caused me to re-evaluate whether, on the basis of image quality alone, the 75-300 IS extended out to 280mm may be a match for my 70-200 IS/1.4 TC combo fully extended. I would still not choose the 75-300 IS over the 70-200 IS for the majority of my picture taking (if only for the other reasons noted above), but for travel/family photos, I may need to rediscover the merits of the 75-300 IS that's been accumulating dust for the past year.
Hi
Having just got home with a 75-300 III USM, I got down to reasearching this lens a bit more. I had read this page before but reading it all over makes me feel good. Now to go out and develop a good technique for taking good pics with the lens. Am really happy with the out of focus blur. Thanks a lot of this article, Bob. This is good for people like me who can neither afford the 300 prime nor the IS zoom.
I don't think the bokeh is so much of a problem as is the CA, if you look at the pic of the birdhouse with the chain link fence behind it, the patches of soil and sand are red and green, where as in the prime lens the soil is nicely coloured. I've got the same problem with an old cheap manual focus zoom, and it drives me crazy, I would like to see a pic of bare tree brances in close against an overcast sky.
Bob, I completely agree with your review. I own this lens and have had for almost 5 years now. I have used it frequently with my Digital rebel and have gotten excellent results. The pictures are sharp and the bokeh is very good. I recently upgraded my body to a XSi and have noticed the results are not as sharp. It may have to do with calibration of the lens or is it something about the XSi that degrades the picture quality ? My 50mm 1.8 works just fine. Ofcourse, it could be me since I have not done any systematic testing. Here's a picture I took with my digital rebel and the 75-300.