Sigma Video on their lenses adapted to the RF mount

[Tronhard]

Anyone following this site for any length of time will be aware of the controversy among some about the performance of Sigma lenses, and in particular the 150-600 Contemporary, working (or not) on the R-series bodies.

This has led to calls for Canon to modify their cameras to accommodate the perceived issues with third-party lenses.  As I have had to respond ad nauseam, no OEM camera maker takes any responsibility for 3rd party equipment attached to their bodies - be it Nikon, Canon or any other, whether it be lenses or flashes.

Well, about six months ago, SIGMA released the following video on using their current EF lenses on the R-series bodies. 

What is significant to me is their strong statement that their lenses perform well on the R-series  bodies - so obviously they see no issue to be fixed.  While I have faith that Sigma and Tamron are working on RF native lenses and will hopefully be able to release them in the foreseeable future, this video lends weight to the argument that it is more likely that it is how the cameras are set up, or operated. 

I observe that the bodies I have seen most people claim flawless operation are the full-frame bodies, for which the Sigma lenses were designed. Most issues seemed to happen with the APS-C R7 tracking system, which has also had issues reported with Canon EF legacy glass and even RF glass, so I wonder if the focusing system of that unit is worthy of consideration as a factor?  I don't have one so I am only speculating.

Another variable might be the firmware version of the Sigma lenses and their customization settings. There may also be issues with technique - using a tripod vs hand-held.  Finally (tongue in cheek) I observe that that presenter from Sigma is shooting gannets, where I have taken a lot of my shots of the same moving birds without any issue, so maybe someone will suggest that it's all about the subject! 🙂

[shadowsports]

Nice post @Trevor.  I've had a similar experience to yours.  My adapted Sigma glass has performed extremely well for stills on my R5 C.  I do agree with your statement regarding the R7 (APS-C) AF Tracking performance.  I noted that Duade Patton put out a supplemental video to his initial review of the 150-600c where he demonstrated focus pulsing.  He provided some recommended settings changes in a subsequent video.  They appear to have improved subject tracking to about 90%.  Sharing if you'd like to watch.

[Waddizzle]

Just because you’re a YouTube content creator does not mean you’re a professional.  Duane makes no pretense or claim that he is a professional wildlife photographer.  He is an amateur, who is still learning his gear.  

I’ve posted comments telling him that his issues are a product of his settings and shooting conditions.  Of course, he disagreed.  Months later, he posted the video that you linked.  The solution is simple.  Turn off OS in the Sigma lenses.  Sigma OS and Canon AF have had a long history of getting into a tug-of-war.

His videos about “focus pulsing” were bad science at its ugly worst.  He was focusing on subjects that were either too far away, or subjects that were inside the MFD.  Here is a screenshot from another “focus pulsing” video.  

Can you see the birds that he is trying to photograph?  He was complaining about focus not staying locked on the subjects.  His subjects are on the other side of the lake, hidden in shadows.

[Tronhard]

Hi Rick!
Yes, I have watched Duade's videos, especially the ones on the Sigma and actually had some long email conversations with him.  I explained I had great results with the Sigma on my R6 MkI and MkII, and R5.  I suggested that his experiments and conclusions could not be held as rigorous because:
1.  He tested only one sample of the R7 and the Sigma, which he had borrowed.  I suggested he needed other units to eliminate individual issues and variations.
2. He did not specify if the Sigma had IS turned on or off, but should have done tests with both
3. He could not specify if the two were using the latest firmware and if the Siggy had been customized using the dock
4.  At the time he was using the Sigma he did so from a tripod but not hand-held - both needed to be tested.
5.  I questioned the configuration of the R7 focusing - exactly what setup scenario(s) he used.  I noted that the R7, being an APS-C with a very high pixel density, was using a modified version of the R3 focus and tracking and issues had been seen with Canon native lenses, so there was a risk factor there.  I explained what focusing configuration I was using on my units and that the were not exhibiting pulsing, and encouraged him to try that, but got no response.

Generally, I think Duade was genuinely trying to do the right thing, but his methodology was flawed.  Other You Tubers took his video and extended his conclusions out of context - leading to the demands for Canon to modify their gear - which I find totally perplexing and ignorant.  That is why I posted a series of images of wildlife using different legacy lenses on this site to try to indicate that there were other truths to this whole debate.

[Waddizzle]

“ Generally, I think Duade was genuinely trying to do the right thing, but his methodology was flawed. “

That’s the “bad science” that I referred to.  He really does mean well.  But he is an amateur with a lot to learn about photography and scientific methods.  Rule #1, you have to detail how others can reproduce your results.  He thinks that box is checked.

I’ve noticed that the image stabilization is turned on when he shows video captured from the HDMI port.  Look in the top left corner.

I failed to make him or his friend understand that MFD applies at the minimum zoom setting, 150mm.  You cannot zoom out to the max, 600mm, and then expect the MFD to be the same.  It is not.  It will be longer by some unspecified amount.  I’ve been argued down.  I gave up on them.

[Tronhard]

Yep.  He's a nice guy, very genuine and I have really enjoyed his laid back videos on bird photography, at which he is very talented.   Nice when you own a huge piece of bush to be able to get out and shoot bird in their habitat when you like!  Still, due to his You Tube success he has branched out into the business of comparing technology and doing reviews as a major focus, rather than a sideline, and I am a bit disappointed in that, but if it makes him money...

[ctitanic]

I explained what focusing configuration I was using on my units and that the were not exhibiting pulsing, and encouraged him to try that, but got no response.

---

Could you tell us what configuration did you recommend him, please?

[Tronhard]

I got a response from writing to Duade Paton but a very vague one.

Settings:
I shoot using back button focus and back button exposure, with the shutter button purely there to trigger the shutter which is set at high speed continuous.   I use single-point focus located in the centre of the viewfinder, and exposure with servo but not continuous focus, and eye tracking for animals, humans and other -each assigned to a C1 - C3 mode setting.   I pretty well always shoot hand-held in available light.  I leave the focus tracking on the default with no tweaking.

Technique:
I point at the eye of the subject, and press BBF to lock on, if the focus prefers some other point, I use the centre button on the rear wheel to force the focus to centre.  From there it tracks,  Similarly, I point to whatever object I consider mid-tonal, and lock exposure on that.  It is how I have shot for a long, long time, so I can do it very quickly without having to think - it will take a while for someone starting to make it subconscious.

I shoot mechanical shutter always and therein may lie another issue.  The electronic shutters are fast - the R7 can get to 30FPS, while the R6II can get to 40.  Personally, I have no need for such speed, and I don't want to wade through thousands of images later. Given that the R7 in particular seems to show issues with pulsing, I wonder if the shutter rate, combined with massive pixel count and what I consider is a deeply flawed digital bus - pushing 40FPS through a very dense sensor that is BSI/stacked, to an inadequate buffer and slow cards is not a good balance.  So, since the sensor and data transfer are so integral in these R-series bodies, I wonder if there are delay factors also impacting the focusing process.

Conclusion:
I think the issue is a very complex one, involving the technology, configuration and operation of a much more complex array of factors impacting focus and throughput.  Thus, I think much more, carefully planned testing needs to be done if one is going to pinpoint where an issue lies.

I would also say that this has been, to a degree taken on as a cause celebre by those who are trying to pressure Canon to allow third party RF lenses - it makes grist to their mills to point to otherwise respected lenses not apparently functioning on the R-series bodies, and that Canon is the evil empire in denying open access to their R mount technology.

[ctitanic]

I don't understand what do you mean by "exposure with servo but not continuous focus".

[Tronhard]

This feature may not  be available on your camera model - it's on the R5 and R6 MkI but not on the R6II.  On the cameras that do, on the first focus screen there is an option for Servo mode and another for continuous focus.  It was very confusing to everyone, so Canon removed it.  Servo would allow focus to follow moving objects if the button for focus was held down, but not if it was not.  Continuous would focus all the time whether the button was held or not - which is a great way to drain the battery!