AF Area, for 600 and 800 - What is root cause of this?

MAC

Forum Pro
Messages
20,182
Solutions
4
Reaction score
4,093
Location
US

Attachments

  • 4480984.jpg
    4480984.jpg
    1 MB · Views: 0
Last edited:
The F-11 min. aperture is likely the overall root cause, but the difference with the R7 would seem to be that the APC-C sensor area is somehow better "adapted" to the phenomenon?
 
The F-11 min. aperture is likely the overall root cause, but the difference with the R7 would seem to be that the APC-C sensor area is somehow better "adapted" to the phenomenon?
And I think being better adapted comes from simply being smaller - so the effective focus area of the APS-C sensor may be similar to more reduced area of the full frame sensor.

Possible a matter in the angle at which the effective f/16 or f/22 light is projected to the imager.

--

Kjeld Olesen
 
The F-11 min. aperture is likely the overall root cause, but the difference with the R7 would seem to be that the APC-C sensor area is somehow better "adapted" to the phenomenon?
And I think being better adapted comes from simply being smaller - so the effective focus area of the APS-C sensor may be similar to more reduced area of the full frame sensor.

Possible a matter in the angle at which the effective f/16 or f/22 light is projected to the imager.
thanks, it is interesting that the R6 and R5 had reduced AF area, and they corrected with the R6II and R5II

maybe the R7II without TC will be improved on this factor also
 
As you can see the R10 and R50 have already improved the situation, so R7 II will catch up.

Fascinating to see this. Things are not quite as rosy with limitless aperture AF as users have suggested.

Canon AF capability has always been reduced with the use of teleconverters, so it is interesting to see that RF is no different. Though RF cameras do not seem to be affected as much as EF ones.
 
As you can see the R10 and R50 have already improved the situation,
Really? All the crop sensor cameras , including the R7, are the same in that chart, (except for the R100, which can't take extenders with those lenses). Since the R10 was released at the same time as the R7, it would be very strange if it had somehow 'improved' the situation over the R7.
so R7 II will catch up.
To what?
Fascinating to see this. Things are not quite as rosy with limitless aperture AF as users have suggested.
Really? That's a strange conclusion to draw from this chart.
Canon AF capability has always been reduced with the use of teleconverters, so it is interesting to see that RF is no different.
What is interesting to see is how the converters affect AF with these two particular lenses.
Though RF cameras do not seem to be affected as much as EF ones.
It's a bit more than seeming. With DSLRs, a converter that reduced the maximum aperture to slower than F5.6 would make AF simply not work at AF points that were limited to F5.6, which was almost all of them on almost all DSLRs. A few higher end DSLRs had some AF points that would work at F8 (the central point on the 7DII, for example), but that was it. Put a 2X extender on the 400 F5.6, and it wouldn't AF at all. No seeming about it. The RF mount has revolutionized AF for Canon users (same with Nikon's mount for Nikon users). And things are exactly as rosy as many users, including myself, have always suggested. The AF area reductions involved with the 600 and 800 F11 lenses and extenders have been known since they were released. The fact that Canon has managed to improve on those reductions with the R5II and R6II over the R5 and R6 suggests, if anything, that things are even rosier than previously suggested.
 
. A few higher end DSLRs had some AF points that would work at F8 (the central point on the 7DII, for example), but that was it. Put a 2X extender on the 400 F5.6, and it wouldn't AF at all. No seeming about it. The RF mount has revolutionized AF for Canon users (same with Nikon's mount for Nikon users). And things are exactly as rosy as many users, including myself, have always suggested. The AF area reductions involved with the 600 and 800 F11 lenses and extenders have been known since they were released.
Quite. My 5Ds has 61 AF points but they are all in the central 50% of the image area, which is less than the area covered by those cheap 600mm & 800mm lenses at f/11. But only the central AF point of the 5Ds and four adjacent ones will AF at f/8 and none at all will work at f/11.
The fact that Canon has managed to improve on those reductions with the R5II and R6II over the R5 and R6 suggests, if anything, that things are even rosier than previously suggested.
--
“When I die, I want to go peacefully in my sleep like my grandfather. Not screaming in terror, like the passengers in his car.” Jack Handey
Alastair
http://anorcross.smugmug.com
Equipment in profile
 
"It is what it is".

Canon produced a pair of unique long-reach lenses with exceptionally good optical quality, at extraordinarily low price points.

The compromise to achieve that affordability is a fixed F-11 aperture.

The notion that Canon engineering is applying great effort, or any effort at all, in redesigning any of the "R" series bodies AF areas for the sole purpose of accommodating these rather inexpensive lenses for use with teleconverters is a bit far-fetched.
 
Quite. My 5Ds has 61 AF points but they are all in the central 50% of the image area, which is less than the area covered by those cheap 600mm & 800mm lenses at f/11. But only the central AF point of the 5Ds and four adjacent ones will AF at f/8 and none at all will work at f_/_11.
It seems that f/9 may be the limit with DSLR AF. f/11 doesn't work, and there just aren't that many people shooting lenses that are f/10 wide open, to report on whether f/10 can work. Every attempt I've made to trick the camera body into thinking the TC wasn't there, f/11 would attempt to focus, but would hunt forever, or park way out of focus.
 
Quite. My 5Ds has 61 AF points but they are all in the central 50% of the image area, which is less than the area covered by those cheap 600mm & 800mm lenses at f/11. But only the central AF point of the 5Ds and four adjacent ones will AF at f/8 and none at all will work at f_/_11.
It seems that f/9 may be the limit with DSLR AF. f/11 doesn't work, and there just aren't that many people shooting lenses that are f/10 wide open, to report on whether f/10 can work. Every attempt I've made to trick the camera body into thinking the TC wasn't there, f/11 would attempt to focus, but would hunt forever, or park way out of focus.
And I presume that F9, with tricking the camera, will only work with central AF points? Or have you managed to get it to work with outer points?
 
I did not look clearly enough at the chart. My mistake. I was mainly looking at use with teleconverters.

One of the hypes around RF is the ability to AF given any lens maximum aperture. Now I see there can be performance reductions in the slower lenses.

Yes, the best DSLRs are restricted to f8, and there is a reduction in AF from f4. While I would think AF at f11 on RF would be worthwhile, I wonder if f16 would be reliable enough in challenging situations.
 
Last edited:
...

. A few higher end DSLRs had some AF points that would work at F8 (the central point on the 7DII, for example), but that was it. Put a 2X extender on the 400 F5.6, and it wouldn't AF at all. No seeming about it. The RF mount has revolutionized AF for Canon users (same with Nikon's mount for Nikon users). And things are exactly as rosy as many users, including myself, have always suggested. The AF area reductions involved with the 600 and 800 F11 lenses and extenders have been known since they were released.
Quite. My 5Ds has 61 AF points but they are all in the central 50% of the image area, which is less than the area covered by those cheap 600mm & 800mm lenses at f/11. But only the central AF point of the 5Ds and four adjacent ones will AF at f/8 and none at all will work at f/11.
that says a lot

in addition, my R8 is -6.5 EV working range and the 5Ds is only -2 EV ;

That is 4 and 1/2 stops

My R8 focuses in moonlight at F4L. Whereas the 5Ds will hunt badly in low light
The fact that Canon has managed to improve on those reductions with the R5II and R6II over the R5 and R6 suggests, if anything, that things are even rosier than previously suggested.
 
"It is what it is".

Canon produced a pair of unique long-reach lenses with exceptionally good optical quality, at extraordinarily low price points.

The compromise to achieve that affordability is a fixed F-11 aperture.

The notion that Canon engineering is applying great effort, or any effort at all, in redesigning any of the "R" series bodies AF areas for the sole purpose of accommodating these rather inexpensive lenses for use with teleconverters is a bit far-fetched.
More likely a byproduct of improving AF in some way, allowing a greater AF area to be used. Or they were just too conservative with the R5 and R6.

We'll probably never know because its not testable unless the firmware is ever hacked or the like.
 
I did not look clearly enough at the chart. My mistake. I was mainly looking at use with teleconverters.

One of the hypes around RF is the ability to AF given any lens maximum aperture. Now I see there can be performance reductions in the slower lenses.

Yes, the best DSLRs are restricted to f8, and there is a reduction in AF from f4. While I would think AF at f11 on RF would be worthwhile, I wonder if f16 would be reliable enough in challenging situations.
My RF 800 F11 with 1.4X extender has never let me down. So yes, AF at F16 is plenty reliable enough. I could never go back to DSLR AF. The RF system is every bit the game changer that people say it is for Canon shooters (and I'm sure the Nikon mirrorless is equally superior to Nikon DSLR).
 
I did not look clearly enough at the chart. My mistake. I was mainly looking at use with teleconverters.

One of the hypes around RF is the ability to AF given any lens maximum aperture. Now I see there can be performance reductions in the slower lenses.

Yes, the best DSLRs are restricted to f8, and there is a reduction in AF from f4. While I would think AF at f11 on RF would be worthwhile, I wonder if f16 would be reliable enough in challenging situations.
My RF 800 F11 with 1.4X extender has never let me down. So yes, AF at F16 is plenty reliable enough.
I could never go back to DSLR AF.
me neither, hardly ever take my 7d2

Even my RP anf M6II are better but getting old in the tooth with Digic 8
The RF system is every bit the game changer that people say it is for Canon shooters
indeed, it is the game changer from the overall system perspective, with not only improvements in mirrorless technology, but improvements in Digic X, and improvements in mirrorless lenses.

BIF and low light AF are some of the torture tests where my R8 system shines

I'm happy camper that my F4L can now focus in moonlight and I got a $499 - 800 mm refurb that can shoot BIF at 20 and even 40 fps !!!

Not only AF, but exposure simulation of mirrorless has also become a game changer for me!

And DXO PL has also been a game changer for me - I feel like I gain at least a stop or two in NR with their AI and I don't pay them rent and greatly prefer their faster un-bloated software to my long use of LR before they started charging rent. I've transitioned to Faststone for culling, DXO PL7, Filmpack, Paintshop Pro, all of which played a part for me in below below photograph for my wall.













(and I'm sure the Nikon mirrorless is equally superior to Nikon DSLR).
 
Quite. My 5Ds has 61 AF points but they are all in the central 50% of the image area, which is less than the area covered by those cheap 600mm & 800mm lenses at f/11. But only the central AF point of the 5Ds and four adjacent ones will AF at f/8 and none at all will work at f_/_11.
It seems that f/9 may be the limit with DSLR AF. f/11 doesn't work, and there just aren't that many people shooting lenses that are f/10 wide open, to report on whether f/10 can work. Every attempt I've made to trick the camera body into thinking the TC wasn't there, f/11 would attempt to focus, but would hunt forever, or park way out of focus.
And I presume that F9, with tricking the camera, will only work with central AF points? Or have you managed to get it to work with outer points?
I just tried to reproduce my memory with the 6D, and a Kenko Pro 300 G 1.4, and the G1 Tamron 150-600, and it just hunted. I am 100% certain I've used AF successfully at f/9 with at least one of my DSLRs; I'll have to try the 7D and 7D2 when I get a chance.

I do now remember that I was using all the AF points on the 6D (but they're not spread out very far on that camera) at f/8, with the tricky Kenko Pro 300 DGX 1.4 and f/5.6 lenses, which even reported EXIF correctly. That DGX was like a miracle with the 6D and 7D, but when I upgraded to the 7D2, it crashed the camera.
 
I did not look clearly enough at the chart. My mistake. I was mainly looking at use with teleconverters.

One of the hypes around RF is the ability to AF given any lens maximum aperture. Now I see there can be performance reductions in the slower lenses.

Yes, the best DSLRs are restricted to f8, and there is a reduction in AF from f4. While I would think AF at f11 on RF would be worthwhile, I wonder if f16 would be reliable enough in challenging situations.
My RF 800 F11 with 1.4X extender has never let me down. So yes, AF at F16 is plenty reliable enough.

I could never go back to DSLR AF.
me neither, hardly ever take my 7d2

Even my RP anf M6II are better but getting old in the tooth with Digic 8
The RF system is every bit the game changer that people say it is for Canon shooters
indeed, it is the game changer from the overall system perspective, with not only improvements in mirrorless technology, but improvements in Digic X, and improvements in mirrorless lenses.

BIF and low light AF are some of the torture tests where my R8 system shines

I'm happy camper that my F4L can now focus in moonlight and I got a $499 - 800 mm refurb that can shoot BIF at 20 and even 40 fps !!!

Not only AF, but exposure simulation of mirrorless has also become a game changer for me!

And DXO PL has also been a game changer for me - I feel like I gain at least a stop or two in NR with their AI and I don't pay them rent and greatly prefer their faster un-bloated software to my long use of LR before they started charging rent. I've transitioned to Faststone for culling, DXO PL7, Filmpack, Paintshop Pro, all of which played a part for me in below below photograph for my wall.


(and I'm sure the Nikon mirrorless is equally superior to Nikon DSLR).
Beautiful shot.

--
“When I die, I want to go peacefully in my sleep like my grandfather. Not screaming in terror, like the passengers in his car.” Jack Handey
Alastair
Equipment in profile
 
I did not look clearly enough at the chart. My mistake. I was mainly looking at use with teleconverters.

One of the hypes around RF is the ability to AF given any lens maximum aperture. Now I see there can be performance reductions in the slower lenses.

Yes, the best DSLRs are restricted to f8, and there is a reduction in AF from f4. While I would think AF at f11 on RF would be worthwhile, I wonder if f16 would be reliable enough in challenging situations.
My RF 800 F11 with 1.4X extender has never let me down. So yes, AF at F16 is plenty reliable enough.

I could never go back to DSLR AF.
me neither, hardly ever take my 7d2

Even my RP anf M6II are better but getting old in the tooth with Digic 8
The RF system is every bit the game changer that people say it is for Canon shooters
indeed, it is the game changer from the overall system perspective, with not only improvements in mirrorless technology, but improvements in Digic X, and improvements in mirrorless lenses.

BIF and low light AF are some of the torture tests where my R8 system shines

I'm happy camper that my F4L can now focus in moonlight and I got a $499 - 800 mm refurb that can shoot BIF at 20 and even 40 fps !!!

Not only AF, but exposure simulation of mirrorless has also become a game changer for me!

And DXO PL has also been a game changer for me - I feel like I gain at least a stop or two in NR with their AI and I don't pay them rent and greatly prefer their faster un-bloated software to my long use of LR before they started charging rent. I've transitioned to Faststone for culling, DXO PL7, Filmpack, Paintshop Pro, all of which played a part for me in below below photograph for my wall.


(and I'm sure the Nikon mirrorless is equally superior to Nikon DSLR).
Beautiful shot.
Thanks Alastair, inspiring lens with new views for me.
 
I did not look clearly enough at the chart. My mistake. I was mainly looking at use with teleconverters.

One of the hypes around RF is the ability to AF given any lens maximum aperture. Now I see there can be performance reductions in the slower lenses.

Yes, the best DSLRs are restricted to f8, and there is a reduction in AF from f4. While I would think AF at f11 on RF would be worthwhile, I wonder if f16 would be reliable enough in challenging situations.
That depends on the AF sensitivity of the body, but subject illumination and contrast play a very big role, too, as well as the brightness of the EVF live view, which is variable with EVF settings. The R7 has lower AF sensitivity than all the R-series FF cameras from the R5 and later, but it can still focus accurately and reasonably fast at f/22 with the RF800/11 and RF2x in bright daylight with good contrast in the subject. Walk into the forest though, and you will take the TC off in a heartbeat, and even at f/11, focus will be slow because of low projected light onto the sensor, plus the fact that the light is flat and has no micro-shadows in textures like hair, feathers, skin, stone, bark, etc.

I can't reproduce everything I've tried successfully in the past, because my non-OEM Kenko TCs seem to change in compatibility with body firmware upgrades, especially when stacked, but when I first got my R5, I was using the 400DO II at 2400/24 with the EF2xIII, plus a Kenko Pro 300 DG 3x, and it AFed reasonably well in bright and medium light. In fact, the first time I tried the 2400/24 combo, it was indoors in a room with closed window blinds on a bright day, and it focused accurately, but slowly.

Also, the EVF brightness settings play a role in AF, so the bodies are clearly relying on the clog video used for live view for some of the AF intelligence, and so ETTR or ETTL combined with Exposure Simulation can interfere with AF ability.
 

Keyboard shortcuts

Back
Top