Sigma 120-300 f/2.8 vs. Canon 300 f/2.8L

Hey Mark,

One thing that I have not seen mentioned, however, is the focusing distance at which the tests were performed.

I said in my post the distances I shot at.

For example, it appears that Carl did his test indoors, focusing on some [lame] photograph on his wall

I hope you're not saying that you think a photograph my mother took is 'lame'??

As for your post, no surprise that a corporate lawyer would come to the support of a corporation.

I don't know the technical facets of your post, but I do know that if a lens is sold as a 300mm lens, and it offers about 2/3 the same focal length of other lenses sold as 300mm, then I feel the manufacturer is being extremely dishonest. It doesn't really matter what distance they were shot at, so lng as they were all shot at the same distance .. and if I have a cheap little 300 dollar zoom that can match the standard for taht focal length, then a "pro" lens like the Sigma should more than be able to do the same.

Cheers

Carl

I just checked the UK's Amateur Photographer mag review of the Sigma. They measure the lens' focal length as going from 120 to 296MM and point out it's a varifocal.

I think this means that at it's MFD it may well not be a 300 (we know it isn't ) but as it nears infinity it will be, but I might be wrong. Of course, for birder's trying to get close to the subject then this sort of thing defeats the purpose of the lens - as you get closer the bird gets smaller :roll:

It would be interesting to do a test at infinity focus to see.

the definition of varifocal is just that when you change the zoom, you have to refocus again. i don't think it means anything more than that...

I guess you're right - I'm no expert on this sort of thing. However, I do tend to trust Amateur Photographer's reviews (my middle name may be Gullible) and I do recall, somewhere, that they commented that a (different) zoom lens which was varifocal with an internal focussing mechanism caused the focal length to be different for close focus as opposed to infinity.

I've probably put 2 and 2 together and got 6 (not for the first time, either). Anyhow, repeating the test at infinity focus would settle it, I guess.

profotos wrote: I do recall, somewhere, that they commented that a (different) zoom lens which was varifocal with an internal focussing mechanism caused the focal length to be different for close focus as opposed to infinity.

I'm no optical expert, but my recollection is that this exact thing happens with internal focus macro lenses....They effectively lose focal length as you focus them in closer. Whether it becomes a factor with a relatively huge non macro lens like the Sigma is beyond what man (or at least this man) was meant to know!

Carl,

Sorry, but I don't know how you guys do the quote thing, so bear with me.

You are correct, you did mention the focusing distances you used, and it was something like 8 and 5 feet, near or at the minimum focusing distance of the lenses involved. I confirmed with an optical physicist that I (and you) know that indeed a lenses focal length will change depending on the focusing distance involved, and will be at its "shortest" at the closest focusing distance. This is an optical law of physics, or at least lens design. He also believes, but is not certain, that this mutable characteristic of lenses may be worse for zoom lenses, but is not certain, and it may depend on the optical design of the lens. Thus, none of the lenses involved in your tests was actually acting like a 300mm focal length lens in your test. Plainly, the Sigma had the shortest apparent focal length at the focusing distances you used. But, that does not mean that at infinity focus the lens would not be closer to the "marked" 300mm focal length.

I would never come to the support of a corporation, whether I am wearing my lawyer or photographer hat, unless the defense was justified. I was not suggesting that there is a justifiable defense in this case. Rather, I was trying to explain the results that you and EJ were coming up with and put them in proper context.

All corporations are guilty of giving only a single focal length for any lens they sell, based on measurements at infinity focusing. Canon's 300/2.8 will not have a 300 focal length when focused at its minimum focusing distance (although it may be closer to it than the Sigma at its minimum focusing distance). I am not a scientist and cannot explain the physics behind this characteristic. All companies are guilty in a similar manner as you are accusing Sigma. That's not a defense of Sigma, but rather a criticism of your selective accusation.

Two "tests" of the Sigma (Pop Photo and Amateur Photographer) apparently have the Sigma's focal length in the 290-296mm range, presumably when focused at infinity. I further assume that these measurements are based on some sort of scientific methodology and thus are somewhat more reliable indicators of the lens' focal length, at infinity focus, than your or EJ's tests. If those assumptions are correct, Sigma is not as guilty as you suggest (no more so than any company).

Finally, I've never to my knowledge seen a photograph that your mother took, and certainly would not call one lame without seeing it. At least I hope that's the case.

Hey Andy.. I'm no expert either.. I got the definition of varifocal from the web.. ha!

Hey Mark.. you sound like a lawyer
However I am still innocent (until proven otherwise) that varifocal may not the the term describing the behavior presented by you above, although the behavior does (assuming you are correct) exists in the lens(es)

vbpholaw wrote:Carl,
Sorry, but I don't know how you guys do the quote thing, so bear with me.

it is a little icon on the upper right hand of the post you want to quote

Maybe we need to invent a term for it. Varifocallengthdependingonhownearthefocusis gets my vote. Nice and pithy

profotos wrote:Maybe we need to invent a term for it. Varifocallengthdependingonhownearthefocusis gets my vote. Nice and pithy

I like it!

I don't know what the phenominon is called but I've seen it and I own a Sigma 120-300 so I'm going to test it this weekend just for curiosity. I have used lenses that remain focused regardless of zoom and others that require refocus after a zoom is made, like the Sigma. In the latter I have noticed that when you change zoom and then refocus (at least with some lenses), the apparent size of the image in the viewfinder also changes as you focus. I just don't recall if it gets larger or smaller as the distance grows.

As for the Sigma 120-300, I love it, whatever actual focal length it is! I have found that most birds don't conform to my preferred perch and pose anyway. I still shoot them whether they land closer or farther away than intended and I crop for comp (you can do that with a good sharp image like the Sigma can produce). The big benefit to me is the zoom. I lost some opportunities on elk with my 300 prime because the subject would get too close and I had no opportunity to back up. My wife got the shots with her 120-300! My prime was replaced shortly thereafter.

Hey Mark,

My point is that I tested the lens against 2 other lens which both yielded remarkably similar results. The Sigma lens yielded remarkably (IMO) different and poor results .. I really don't care why that is so - I know that a cheap little 300mm zoom from Nikon gave me a focal length half again of what the Sigma lens did. I disagree that this means I'm guilty of 'selective accusation'. The results selected themselves. If all the lenses yielded a similr result, then it would be a matter of physics .. if the Sigma yields, as it does, an equivalent of other lenses at or about 200m, then I think they're guily of malfeasance. It may still be a matter of physics, but it's a matter of the Sigma lens being physically lamer than the Nikon lenses.

Focal length, for all the nonsense about physics, is really only important as a relative measure .. There's nothing magical about 300mm or any other focal length. What there is, is a standard that this lens appears to fall far short of at a particular focusing distance.

I'll try it again when I get some time, and shoot at a longer focal length. Maybe it'll be closer to the other lenses, maybe it'll be longer, I don't know. But that doesn't mean that it's not lame that at a close focusing distance, the Sigma lens falls far short of comparative products.

Finally, I've never to my knowledge seen a photograph that your mother took, and certainly would not call one lame without seeing it. At least I hope that's the case.

True - but at the moment you're uncertain, at best.

Cheers

Carl

vbpholaw wrote:Two "tests" of the Sigma (Pop Photo and Amateur Photographer) apparently have the Sigma's focal length in the 290-296mm range, presumably when focused at infinity. I further assume that these measurements are based on some sort of scientific methodology and thus are somewhat more reliable indicators of the lens' focal length, at infinity focus, than your or EJ's tests. If those assumptions are correct, Sigma is not as guilty as you suggest (no more so than any company).

Knowing someone who reviews equipment for a magazine I wouldn't trust any magazine review. On more than one occasion the person has been told he needn't return the high end equipment if he liked it (and by inference, gave it a good review).

And if the lens is close to 300mm at infinity, with an objective diameter of 96mm the lens can be at best f3.1 at the long end of it's zoom range. And as it's a zoom and won't get anywhere near 100% transmission it's probably closer to f3.5. So either way Sigma are cheating their figures.

This is all very fascinating and I do agree with Carl. "It may still be a matter of physics, but it's a matter of the Sigma lens being physically lamer than the Nikon lenses."

"Varifocallengthdependingonhownearthefocusis"

Darth wrote:Knowing someone who reviews equipment for a magazine I wouldn't trust any magazine review. On more than one occasion the person has been told he needn't return the high end equipment if he liked it (and by inference, gave it a good review).

And if the lens is close to 300mm at infinity, with an objective diameter of 96mm the lens can be at best f3.1 at the long end of it's zoom range. And as it's a zoom and won't get anywhere near 100% transmission it's probably closer to f3.5. So either way Sigma are cheating their figures.

I know what you mean. One magazine compared one of my products to similar offerings from Adobe (their Album product MK1) & ACDSee - mine outscored them on all but one aspect (image sharing) because the reviewer 'overlooked' the fact that mine too could produce web albums. Result - Adobe won by a whisker and got their recommendation. I feel sure that review cost me sales. They never replied to my or some of my customer's enquiries as to why they did that, but guess whose product was on their coverdisk the next month...

Not that I'm implying anthying, of course

When it comes to their lens reviews, I have always found Amateur Photographer's reviews to be pretty thorough and they have marked lenses down (sometimes quite severely) when they haven't lived up to expectations. I'm not sure they test the accuracy of the aperture.

To answer a previous post re EXIF accuracy, my guess is that the EXIF records the focal length and aperture that the lens reports to the camera - there's no other way I think it could deduce it. I suspected that some lenses (like the Tamron 200-500 I used to own) would 'lie' to my Canon 300D to ensure it's AF worked with the Kenko 1.4 extender, despite being a F6.3 when at 500MM, but I might be wrong here too

profotos wrote:I know what you mean. One magazine compared one of my products to similar offerings from Adobe (their Album product MK1) & ACDSee - mine outscored them on all but one aspect (image sharing) because the reviewer 'overlooked' the fact that mine too could produce web albums. Result - Adobe won by a whisker and got their recommendation. I feel sure that review cost me sales. They never replied to my or some of my customer's enquiries as to why they did that, but guess whose product was on their coverdisk the next month...

Not that I'm implying anthying, of course

Well your advertising budget is probably a 'little' smaller than Adobes and magazines have to keep their advertisers happy.

I once freelanced for a motorcycle magazine that gave the original Gold Wing a very bad review/road test, re-naming it the Lead Wing if my memory's correct. I seem to recall Honda pulled all their advertising from the magazine for 12 months in 'protest'.

profotos wrote:When it comes to their lens reviews, I have always found Amateur Photographer's reviews to be pretty thorough and they have marked lenses down (sometimes quite severely) when they haven't lived up to expectations. I'm not sure they test the accuracy of the aperture.

To answer a previous post re EXIF accuracy, my guess is that the EXIF records the focal length and aperture that the lens reports to the camera - there's no other way I think it could deduce it. I suspected that some lenses (like the Tamron 200-500 I used to own) would 'lie' to my Canon 300D to ensure it's AF worked with the Kenko 1.4 extender, despite being a F6.3 when at 500MM, but I might be wrong here too

With an objective diameter of 96mm it's impossible for the lens to be a 300mm f2.8. If it's a 300mm lens the aperture can't be any faster than f3.125. Alternatively if the aperture is f2.8 the focal length can't be more than 268.8mm.

profotos wrote:I suspected that some lenses (like the Tamron 200-500 I used to own) would 'lie' to my Canon 300D to ensure it's AF worked with the Kenko 1.4 extender, despite being a F6.3 when at 500MM, but I might be wrong here too

I think you'd find it's the TC at fault in this case, not the lens. The cheaper TC's don't fully implement the EF protocol and block out the part of the info that indicates the presence of the TC. There's nothing spooky about it, the body just thinks the lens is on there by itself and so AF or whatever else continues to operate (or attempt to) as usual. At least TTL metering works properly taking the TC's effect into account.

I have a cheap Tamron 1.4X (maybe identical to the Kenko) that I use for exactly this reason. It's optical quality is good enough for some of my needs, and my 10D's continue to AF with it on because they don't see it. When I want better optics I put on the Canon TC's instead and drop down to manual focus. (I prefer not to mess with the pin taping trick.)

On the main subject of this thread, I found out today that my favorite local camera shop has the Sigma 120-300mm f/2.8 in their rental pool, in both Canon and Nikon mounts. When I get done with my soul-devouring home reno project in a couple of weeks, I'm going to rent the lens and take it out for a weekend. I played around with a display copy today in the shop, and it seems like a tidy package. Not as heavy to hold as I thought it might be.

I don't have any 300mm primes, but I do have the Canon 100-400mm and Sigma 300-800mm. So I can do some head-to-head comparisons for focal length parity (among other things) at 300mm. Even with the focal length issue discussed here, if the lens performs at all well I imagine I'll use it to replace my unsatisfactory Canon 100-400mm...

Well, I shouldn't have been so lazy and not included the first part of John Shaw's footnote:

An f-number is a value representing the quantity of light passing through the diaphragm when the lens is focused at infinity. The "f" is an abbreviation for "factor" and describes the mathematical ratio of the focal length divided by the diameter of the effective aperture, which may not be the same as the physical diameter because the light rays entering the front element of the lens are converged towards the diaphragm. Got that?

Butting into the math fray once more, and taking into account the possibility that Sigma has fudged its figures, I love the Sigma 120-300.....or whatever it really is in focal length! It allows for very sharp images, and that'[s the bottom line for me.

Here's a link regarding calculating F stops:

http://www.uscoles.com/fstop.htm

To quote a bit from the site:

"(Question)I took my lens apart. The aperture is nowhere near as big as the calculation shows. What's up?

You're right. I had an email from a guy who had taken apart a Rokkor 300mm f/4.5 (for other reasons, not to check my measurements) and he said the diameter of the f/stop blades was way smaller than the calculation would indicate. The calculations above would be accurate if the aperture blades were mounted right in front of the front element. In fact, they're buried in the lens somewhere and, on the Nikkor 300mm f/4.5 IF-ED I used to own, were actually located behind all the lens elements. They still have the same relationship but the manufacturer can make the aperture blades way smaller in the light path partway back. However, the relationship is the same between each of the adjacent stops. "

------------------------------

If he's right then maybe Sigma aren't deceiving us. But he could be wrong.

Darth wrote: Well your advertising budget is probably a 'little' smaller than Adobes and magazines have to keep their advertisers happy.

I once freelanced for a motorcycle magazine that gave the original Gold Wing a very bad review/road test, re-naming it the Lead Wing if my memory's correct. I seem to recall Honda pulled all their advertising from the magazine for 12 months in 'protest'.

I understand this - it's morally wrong that such considerations bias/affect the objectivity of the review/reviewer IMHO. Such are the realities of life :roll: