Intresting comparison. Im thinking of replacing my 28-105 with the Tamron 28-75 for its f/2.8 which would also be useful on my digital bodies (Im only using my 28-105 on my film bodies at the moment).
mh2000 wrote:
ok, my guess is that the first is the L... is that right?
Correct. Notice how the light discs on the L are more uniform throughout, whereas the Tamron discs have a brighter edge? The Tamron does a very nice job with a smooth blur on the remainder of the background, although the edge definately goes to the L on the specular highlights. Note as well that here again, the Tamron is considerably sharper.
Will Patterson wrote:
This is a great test, and now I'd really like to see the Tamron 17-35 or 19-35 vs. the legendary 17-40L. I need a wide angle lens like nobody's business.
I had a Tamron 17-35 for a while, and compared to the 17-40 the results were very similar to Dave's test: the Tamron was sharper, particularly at f/4. It also had better (sharper) corners. The Canon's usability is definitely better. (The Tamron's moving focus ring was particularly bothersome to me, as it's hard NOT to touch it while holding the lens.)
Dave, the image on the right has more saturated reds, but the image on the left has nicer bokeh in the OOF lights (no donut-shaped rings). At first blush I'd pick the right image for "pop", because it's more saturated, but since I'd post-process both images anyway, I'd prefer the left image since saturation is easy to adjust, but ugly OOF lights are very difficult to fix.
I would pick the left photo for Canon 24-70L. The color is not washed out like the red. Background colors are very good. Also the reflective lights are not busy like Tamron. Also sharpness is there for Canon. Look at the hat. Much shaper with Canon.
A note on color: the bokeh scene uses the color from the Tamron, which was simply more accurate. The Canon is about 200 degrees cooler than the Tamron, and has slightly more red. To make them look equal, the Canon was bumped 200 degrees Kelvin, and the tint was reduced by 3.
Many people note that the Tamron produces color that is more "Yellow" than Canon lenses. This is true, however, I've found that the Tamron is usually more accurate. The fact is, most Canon lenses are on the cool side (more "blue"), which may or may not be pleasing to the eye depending on opinion. This may be one of the reasons that people (Me_Xman) say they prefer the skin tones from the Canon lenses. My point is, it is simple to give a Tamron shot that same look if shooting raw, by decreasing the color temp by a couple hundred degrees, and adding about +3 to the tint slider.
CKrueger wrote:
Dave, the image on the right has more saturated reds, but the image on the left has nicer bokeh in the OOF lights (no donut-shaped rings). At first blush I'd pick the right image for "pop", because it's more saturated, but since I'd post-process both images anyway, I'd prefer the left image since saturation is easy to adjust, but ugly OOF lights are very difficult to fix.
Either way, they're very close!
I agree with everything said here, I think you were typing this as I posted my reply above.
Me_XMan wrote:
I would pick the left photo for Canon 24-70L. The color is not washed out like the red. Background colors are very good. Also the reflective lights are not busy like Tamron. Also sharpness is there for Canon. Look at the hat. Much shaper with Canon.
Washed out red? I'm not seeing that.
Reflective lights, yes, that is the one area that the Canon has a clear advantage, as noted above.
The hat is not the focal point, the eye is. The Tamron image is considerably sharper, so I can't agree with you here
Do you consider yourself having "SHARP" copy's of all the lens you tested..
As it is well known that the 24-70 and 24-105 are copy dependent..
And I feel the Tamaron in your test has a more magenta HUE to them..
Snook
PS. And Tahnks for taking the time for the test...
Dave Jr wrote:
Many people note that the Tamron produces color that is more "Yellow" than Canon lenses. This is true, however, I've found that the Tamron is usually more accurate. The fact is, most Canon lenses are on the cool side (more "blue"), which may or may not be pleasing to the eye depending on opinion. This may be one of the reasons that people (Me_Xman) say they prefer the skin tones from the Canon lenses. My point is, it is simple to give a Tamron shot that same look if shooting raw, by decreasing the color temp by a couple hundred degrees, and adding about +3 to the tint slider....Show more →
It's funny how that works, eh? People often complain about the "Sigma yellow"... I wonder if SD10 users could mount Canon lenses if they would complain about "Canon blue"? Cause I see "correct" color somewhere in between, and both are close enough that either could be corrected with a tiny (and consistent) tweak.
Dave Jr wrote:
Does the Canon image have more pop?
I'd give the Tamron (right) image the nod for "pop" simply for the colors. But like I said, they're VERY close. Close enough that if you had posted just one of these images nobody would be able to tell you which was which.
The slight ring to the Tamron bokeh seems to only be noticeable on the point light blobs... the rest of the background looks identical between frames.
I think it's safe to say that image quality is much less a factor than price vs build when comparing these lenses. Only people obsessively critical about sharpness or bokeh would choose these lenses solely for image quality.
So, USM and weather sealing, or price and size/weight?
snook wrote:
Do you consider yourself having "SHARP" copy's of all the lens you tested..
As it is well known that the 24-70 and 24-105 are copy dependent..
First off, I'd say that I have not heard of much copy variation with the 24-105L like I have with the 24-70L. Then, I'd say yes, except for the 24-70, which I just received, and ascertaining it's sharpness was one of the reasons for this test. I would say it (the 24-70) is not great at 2.8 at 24mm and 70mm, but not bad elsewhere. I just received the 16-35L as well, and I would say it is amazing. The Tamron, 24-105L, 35/2.0 and two 70-200's are very good and have been thoroughly tested.
snook wrote:
And I feel the Tamaron in your test has a more magenta HUE to them..
The outdoor shots? Yes, maybe just a bit, like + one or two on the tint slider, although I did not do any color work on those.
snook wrote:
Snook
PS. And Tahnks for taking the time for the test...
CKrueger wrote:
I'd give the Tamron (right) image the nod for "pop" simply for the colors. But like I said, they're VERY close. Close enough that if you had posted just one of these images nobody would be able to tell you which was which.
The slight ring to the Tamron bokeh seems to only be noticeable on the point light blobs... the rest of the background looks identical between frames.
I think it's safe to say that image quality is much less a factor than price vs build when comparing these lenses. Only people obsessively critical about sharpness or bokeh would choose these lenses solely for image quality.
So, USM and weather sealing, or price and size/weight? ...Show more →
I tested all lenses for bokeh, but handheld, quickly, so they should not be used to judge sharpness, nor color. These are better than the other sample because they are closer.
heehee... glad I was right (now I can feel good about keeping my 24-70
Seriously, the Tamron looks like a really good lens, in fact, had it not been new, on rebate and backordered everywhere at the time that I needed one I would own that instead of the 24-70 today... but more than just the outlined highlights I find other subtle rendering qualities of the 24-70L to be slightly more pleasing to me... I don't really worry about slight sharpness differences.
Thanks a lot for posting the tests! Nice useful work!
Dave Jr wrote:
Correct. Notice how the light discs on the L are more uniform throughout, whereas the Tamron discs have a brighter edge? The Tamron does a very nice job with a smooth blur on the remainder of the background, although the edge definately goes to the L on the specular highlights. Note as well that here again, the Tamron is considerably sharper.
1) The 16-35 looks like a donut factory! Not a significant downside for a lens of this type, but it looks much like some mirror lenses I've seen.
2) The 70-200/2.8 has the same football-shaped bokeh that I'm used to from my 50/1.8 and 35/2. That surprises me because I thought the 70-200's had circular apertures.
I'm not that surprised. Back a few years ago I went through 3 copies of 24-70 2.8L and ALL had excessive CA and weren't that sharp. I tried a Tamron 28-70 2.8 at the urging of the dealer. It BLEW all of the L's away and had NO CA at all. Once in a while Tamron really get's things right. If they had the equivalent of the Sigma HSM motor they'd have it licked.......
CKrueger wrote:
Wow, that's an extensive test! Thanks, Dave!
Two things surprise me in that test:
1) The 16-35 looks like a donut factory! Not a significant downside for a lens of this type, but it looks much like some mirror lenses I've seen.
That surprised me as well. It's not always that bad, really depends on the distance between subject and background.
CKrueger wrote:
2) The 70-200/2.8 has the same football-shaped bokeh that I'm used to from my 50/1.8 and 35/2. That surprises me because I thought the 70-200's had circular apertures.
I think that might have to do with the angle. I had to stand much further back, and there was a wall in the way, so I had to move left. Usually, they have round highlight discs.
Man, Dave awesome body of work put in by you, kudos man.
I found the same to be true and got beat down for saying the same.
The Tamron 28-75 & 17-35 beat all the copies of the 24-70 (2), 17-40 (2) & 16-35 (3) I could find. Speed was not the issue 2.8 and 4.0 were for me. I tried the Tamrons first because of money issues, then tried to replace them when money wasn't an issue. The Tamrons more than held there own, they had plenty of Snap, Crackle and Pop. Again Good work Dave, U the Man......
