Myth or fact??

I quite often see comments bandied around like 'Sigma lenses' don't nail focus as well as 'L' lenses, but, it's the camera that does the focusing, not the lens? I can understand the comment that maybe they focus faster due to better motors or lighter construction, or that overall they are sharper, but to imply the lens itself doesn't carry out the process of focusing very well and often misses, that I don't understand.......

One can find a number of posts on POTN about folks who have trouble getting accurate focus with the Sigma lens on their Canon camera. They would send it in, a number of times sending it back multiple times -- before it came back able to more consistently and accurately get focus. In a few cases, folks gave up and sold the lens, to buy something else. It seems the issue was more often focus plane adjustment (e.g. MFA not being sufficient)

BTW, I don't have a bias against Sigma products myself, I am merely repeating what I have observed on the basis of POTN posts. One has to keep in mind, though, that often all the satisfied customers say nothing and it a vocal minority doing the complaining. So one needs to take casual observations based upon posts with a grain of salt!

Purely used Sigma as an example, you can easily replace Sigma with Tamron etc. The point is, the comparison is made that Canon lenses focus better than 3rd part lenses, I just don't understand why. Is this a canon fan boy knee jerk reaction, or, is there something behind these statements?

Well I will admit that my Tamron focuses just a bit slower than my Canon zoom, and the Tamron zoom hits a can't-focus-in-low-light at a 1EV brighter level than the Canon zoom (when both at the same aperture and FL) when mounted on the same body. So the issues can indeed be valid ones.

You must also remember that Canon doesn't share its algorithms/tech with third parties. Competitors must reverse engineer, which often results in their products performing inferiorly compared to Canon's own.

The AF system works to some extent open loop. The camera commands the lens to go to a spot, the lens moves and reports when ready.

This control system works better with both sides being engineered by one team. Also, there is quite a bit of evidence that Canon has introduced some degree of closed-loop controls from about the 70-200/2.8 IS II and newer. These newer lenses seem to be a step better in AF performance, which will be hard to match for 3rd parties.

In use, I find the latest Canon lenses (for me, the 100-400 L II, the 70-200/2.8 IS II and the 24-70 II) to be unmatched in AF performance.

I have and use a Sigma 120-300/2.8 Sport which works quite well with my 5D3. It isn't as good as the above lenses, but it is pretty good. It did not work very well with my 1D4 though. No idea why.

Alveric wrote in post #18277784
You must also remember that Canon doesn't share its algorithms/tech with third parties. Competitors must reverse engineer, which often results in their products performing inferiorly compared to Canon's own.

I would expect, as well, that the actual focusing firmware in the 3rd party lenses is common between their lenses for different cameras. For instance, I expect that deep down inside, Sigma uses the same native Sigma focusing firmware for a lens whether it's in Canon, Nikon, or Sony mount, but that there is upper level firmware that translates from Sigma to the other cameras.

If this is the case, it would be a point of a smidgen of delay and possible error.

Also, sometimes even Canon lenses have trouble matching Canon cameras...which is what micro adjustment is fore.

RDKirk wrote in post #18277847
I would expect, as well, that the actual focusing firmware in the 3rd party lenses is common between their lenses for different cameras. For instance, I expect that deep down inside, Sigma uses the same native Sigma focusing firmware for a lens whether it's in Canon, Nikon, or Sony mount, but that there is upper level firmware that translates from Sigma to the other cameras.

If this is the case, it would be a point of a smidgen of delay and possible error.

Also, sometimes even Canon lenses have trouble matching Canon cameras...which is what micro adjustment is fore.

The firmware might be common but each manufacturer has a different interface and different protocols. I could see 3rd party firmware having built in all manufacturers but when mounted to a camera it would have to use the sections of the firmware pertaining to that brand.

Fact. Sorry, you usually get what you pay for.

To add one more reason to the ones already cited by others: Canon lenses usually trigger better focus point modes, even at the same max lens aperture.

gjl711 wrote in post #18277857
The firmware might be common but each manufacturer has a different interface and different protocols. I could see 3rd party firmware having built in all manufacturers but when mounted to a camera it would have to use the sections of the firmware pertaining to that brand.

That's what I said.

RDKirk wrote in post #18277847
Also, sometimes even Canon lenses have trouble matching Canon cameras...which is what micro adjustment is fore.

Actually, MA is meant to allow Canon to save money with less strict tolerances and lower QA work. If your lens and camera tolerance errors sum up, then you will "fix" the problem by means of MA, while Canon is happily counting your money.

MA is about accuracy, not precision nor performance. And the problem with third-party lenses is mostly precision and performance.

Again, who cares why we have AFMA, we are happy to have it. It was terrible before they finally gave it to us. Canon QA is no worse than before, in fact better. They have tighter tolerances, better builds, longer shutter life, more seals, etc than years prior. Call AFMA what you will, it matter little, it serves a very useful purpose. And today's FF offerings like the 5D4 is better than the 5D, while being close to the same price point as the 5D, so we are actually getting much more for our money than previously.

I've had two Sigma lenses, three Tamron lenses, and at least one Tokina lens.. Over the years, I've probably had 30 Canon lenses. Here is a list of lenses I've had focus problems with:

Canon 135L - first one would not focus accurately under 20' on my 500D. Worked fine on other bodies. Second copy is perfect.
Canon 85 1.8 - I bought it used; it was likely worn out. Very inconsistent. Second copy requires -10 MFA on my 1DIV and works fine on everything else.
Canon 50 1.4 - I bought it used and focus accuracy/consistency faded over time.
Canon 50 1.8 II x2 (1 used, 1 refurb) - I couldn't get either one to focus accurately in any light. (Bought 50 STM, It works.)

I've never had focus problems with anything BUT Canon. Not knocking Canon. I play in the used market. It is what it is.

Bassat wrote in post #18277973
I've never had focus problems with anything BUT Canon. Not knocking Canon. I play in the used market. It is what it is.

Again, we are not talking about defective lenses here.
We are talking about precision and performance issues by design.
In that sense, I only had problems with Sigma (never used Tamron or other AF 3rd party lens).

I've had issues with the 35mm and 50mm Sigma Art lenses. The focusing was very inconsistent. Sometimes they would be dead on, however other times they couldn't decide if they wanted to front focus or back focus which eliminated AFMA as an option. The 20mm Art however has preformed fantastically for me every time I've used it.

One of the photographers I shoot with regularly also has the 35mm Art lens and he hasn't had any focusing issues with his. I haven't tried his copy yet, but it would be interesting to see how it works on my bodies.