technic wrote:
IMHO the problem with the 80D and 90D is the very slow response of the PDAF system (slow to get initial lock, especially) and some other strange quirks like a strong preference for the back of the subject or the background, instead of the foreground which in most situations makes more sense (but this same preference is to a smaller extent present with DPAF).
This should be relatively easy to address in firmware if they want. But talking about tunable: not having to use MFA is a major advantage, given the MFA woes of my 80D with the 100-400II: there is no way I can get acceptable focus both for closeup and infinity using PDAF, it needs around 0 for infinity and around -20 at close up. ML has a big disadvantage there, but I'm curious about one of its main disadvantages which is focusing speed with long bright lenses from an initially strongly defocused situation (important for wildlife etc.).
If you add it all up, my impression continues to be that the M6II is a test bed for these technologies but deliberately "crippled" so as not to appeal to the real 7D3 type customers. Add a capable fixed EVF, bigger battery, some more AF parameters, some improvements for current video limitations, better software for handling burst mode images etc. and it could be a serious 7D3 alternative. Maybe Canon is waiting to see if they should introduce this in an M5II or a R7X
More AF parameters might require a stacked sensor, I doubt there is much more Canon can do in realtime with all this sensor data. Without this is would be difficult to compete with cameras like Sony A9, but maybe they don't have to when the current technology in M6II with some tweaks is good enough for most enthusiast action/wildlife shooters? In that case they could do with just one very expensive high end R action body....Show more →
Most of what you say makes very good sense. But, I don't have the AF problems with the 90D. The centre point locks on extremely rapidly in AI Servo, and it doesn't have the focus changing with distance, and I take close ups of dragonflies too at 1m.
technic wrote:
IMHO the problem with the 80D and 90D is the very slow response of the PDAF system (slow to get initial lock, especially) and some other strange quirks like a strong preference for the back of the subject or the background, instead of the foreground which in most situations makes more sense (but this same preference is to a smaller extent present with DPAF).
This should be relatively easy to address in firmware if they want. But talking about tunable: not having to use MFA is a major advantage, given the MFA woes of my 80D with the 100-400II: there is no way I can get acceptable focus both for closeup and infinity using PDAF, it needs around 0 for infinity and around -20 at close up. ML has a big disadvantage there, but I'm curious about one of its main disadvantages which is focusing speed with long bright lenses from an initially strongly defocused situation (important for wildlife etc.).
If you add it all up, my impression continues to be that the M6II is a test bed for these technologies but deliberately "crippled" so as not to appeal to the real 7D3 type customers. Add a capable fixed EVF, bigger battery, some more AF parameters, some improvements for current video limitations, better software for handling burst mode images etc. and it could be a serious 7D3 alternative. Maybe Canon is waiting to see if they should introduce this in an M5II or a R7X
More AF parameters might require a stacked sensor, I doubt there is much more Canon can do in realtime with all this sensor data. Without this is would be difficult to compete with cameras like Sony A9, but maybe they don't have to when the current technology in M6II with some tweaks is good enough for most enthusiast action/wildlife shooters? In that case they could do with just one very expensive high end R action body....Show more →
dunno, the small M6II mirrorless may not compete with a potential 7D3 but it certainly appears good enough for most enthusiasts
cpe1991 wrote:
Most of what you say makes very good sense. But, I don't have the AF problems with the 90D. The centre point locks on extremely rapidly in AI Servo, and it doesn't have the focus changing with distance, and I take close ups of dragonflies too at 1m.
Are you talking about using 90D PDAF with the 100-400II and about "action shots"? That would be weird, because I have verified the same behavior on three different 80D/ 100-400II combo's now (with both huge backfocus, plus focus shift and general IQ problems near wide open for closeups) and I haven't seen any combo without this problem. And while the PDAF is fine for static subjects, it simply doesn't work for action unless the subject is relatively big and far away from the camera (so the focus system and AF system have less work to do).
When I'm using the 80D PDAF with other lenses (e.g. 100L macro, 2.8/200) I usually have little or no backfocus problems and AF is fine for still images, but totally useless for flying dragonflies. AF is simply too slow to get a lock on anything that moves at more than glacial speed; with PDAF active I'm 99% sure to get NO image of a flying dragonfly because the AF system tries to focus on the background first and by the time it finds out there is nothing there to lock on the subject is out of the frame ;( I also tried with the 100-400II for hovering dragonflies (moving quite slowly) and it is obvious that the PDAF is running 2-3 frames (1/4 sec or so) behind what is actually happening - the AF system reports focus lock for wings etc. that are not present in those frames.
CW100 wrote:
dunno, the small M6II mirrorless may not compete with a potential 7D3 but it certainly appears good enough for most enthusiasts
I don't doubt it is a nice camera for enthusiasts including those taking some nature/wildlife images from time to time. But for several reasons it seems pretty impractical to me for people who do lots of nature/wildlife stuff with fast action or long lenses (add-on EVF, small battery, small grip/body when using long lenses, rudimentary AF control options; lack of weather sealing can be another issue).
The issues for me with the M6II are not operational but ergonomic. If I'm going to be using a lens like the 100-400 for long periods of time, say wildlife oriented tours, I'd rather use an 80D like body. A big lens is manageable on the M but really does better with lenses in the 25 oz region rather than 40 to 50 ounces. Often 400 to 500 mm effective reach is enough so I am delighted the M has breathed new life into my 55-250.
Also there is the battery life consideration. I'm recently returned from four days hiking in the Havasupai region of the Grand Canyon. Took my 6D2 (no wildlife) and still had over half my battery capacity left when I got home. Not sure how long I can go on the M's SL2 sized battery with its increased sensor use. Still an experiment but fun to play with.
M6II + 55-250 or 70-300 + M11-22 + 35 STM macro + G1X3 = 4.25 to 5 lbs
technic wrote:
Are you talking about using 90D PDAF with the 100-400II and about "action shots"? That would be weird, because I have verified the same behavior on three different 80D/ 100-400II combo's now (with both huge backfocus, plus focus shift and general IQ problems near wide open for closeups) and I haven't seen any combo without this problem. And while the PDAF is fine for static subjects, it simply doesn't work for action unless the subject is relatively big and far away from the camera (so the focus system and AF system have less work to do).
When I'm using the 80D PDAF with other lenses (e.g. 100L macro, 2.8/200) I usually have little or no backfocus problems and AF is fine for still images, but totally useless for flying dragonflies. AF is simply too slow to get a lock on anything that moves at more than glacial speed; with PDAF active I'm 99% sure to get NO image of a flying dragonfly because the AF system tries to focus on the background first and by the time it finds out there is nothing there to lock on the subject is out of the frame ;( I also tried with the 100-400II for hovering dragonflies (moving quite slowly) and it is obvious that the PDAF is running 2-3 frames (1/4 sec or so) behind what is actually happening - the AF system reports focus lock for wings etc. that are not present in those frames....Show more →
This is with classical PDAF on the 90D, and with both my 100-400mm II. For flying objects, I use the central zone of 9 points and have been able, for example, to capture flying ducks against backgrounds as the points focus on the nearest object. With dragonflies, a simple background is important when they are small in comparison to reeds etc. Give the 90D a try
Jeff Nolten wrote:
The issues for me with the M6II are not operational but ergonomic. If I'm going to be using a lens like the 100-400 for long periods of time, say wildlife oriented tours, I'd rather use an 80D like body. A big lens is manageable on the M but really does better with lenses in the 25 oz region rather than 40 to 50 ounces. Often 400 to 500 mm effective reach is enough so I am delighted the M has breathed new life into my 55-250.
Also there is the battery life consideration. I'm recently returned from four days hiking in the Havasupai region of the Grand Canyon. Took my 6D2 (no wildlife) and still had over half my battery capacity left when I got home. Not sure how long I can go on the M's SL2 sized battery with its increased sensor use. Still an experiment but fun to play with.
M6II + 55-250 or 70-300 + M11-22 + 35 STM macro + G1X3 = 4.25 to 5 lbs ...Show more →
Agree about those concerns. For me the 100-400II is already too inconvenient for prolonged use like holding up for several minutes until something happens ... and too heavy for tracking a fast moving dragonfly. No doubt with M6II body instead of 80D it would be worse.
---------------------------------------------
cpe1991 wrote:
This is with classical PDAF on the 90D, and with both my 100-400mm II. For flying objects, I use the central zone of 9 points and have been able, for example, to capture flying ducks against backgrounds as the points focus on the nearest object. With dragonflies, a simple background is important when they are small in comparison to reeds etc. Give the 90D a try
Weird that you don't have the 100-400II closeup backfocus issue with PDAF, because it has been confirmed on several bodies (but seems worse on APS-C like the 70D/80D and some Rebels). It's possible that the 90D is different but I really doubt it because everything suggests the PDAF unit is exactly the same, the only difference is the new pixel sensor for improved tracking in some situations. For PDAF I also use central zone with 9 points as standard setting, but it only works on big BIFs with far away backgrounds (and only in good light).
I have no easy way to try a 90D and will more likely try how the M6II works in similar situations (when the price with EVF and EF adapter comes down a bit, and I haven't purchased a D500 by then)
technic wrote:
Agree about those concerns. For me the 100-400II is already too inconvenient for prolonged use like holding up for several minutes until something happens ... and too heavy for tracking a fast moving dragonfly. No doubt with M6II body instead of 80D it would be worse.
---------------------------------------------
Weird that you don't have the 100-400II closeup backfocus issue with PDAF, because it has been confirmed on several bodies (but seems worse on APS-C like the 70D/80D and some Rebels). It's possible that the 90D is different but I really doubt it because everything suggests the PDAF unit is exactly the same, the only difference is the new pixel sensor for improved tracking in some situations. For PDAF I also use central zone with 9 points as standard setting, but it only works on big BIFs with far away backgrounds (and only in good light).
I have no easy way to try a 90D and will more likely try how the M6II works in similar situations (when the price with EVF and EF adapter comes down a bit, and I haven't purchased a D500 by then)...Show more →
cpe1991 wrote:
What lens will you use with a D500?
In that case I would buy the 300PF, 1-2 general lenses like the 16-80VR and a small prime like the 1.8/35 and maybe the 500PF. The 300PF is ideal for my dragonfly photography. Not sure if I can accept the Nikon Liveview though as the Canon DPAF despite some flaws is much faster, and I use that a lot for moving macro subjects when I cannot look through the OVF. So for that I may have to keep a Canon body with the 100L macro.
technic wrote:
In that case I would buy the 300PF, 1-2 general lenses like the 16-80VR and a small prime like the 1.8/35 and maybe the 500PF. The 300PF is ideal for my dragonfly photography. Not sure if I can accept the Nikon Liveview though as the Canon DPAF despite some flaws is much faster, and I use that a lot for moving macro subjects when I cannot look through the OVF. So for that I may have to keep a Canon body with the 100L macro.
They are the lenses I would get if I were to add Nikon to my armoury. Unfortunately, I do not like their telephoto zooms.
cpe1991 wrote:
They are the lenses I would get if I were to add Nikon to my armoury. Unfortunately, I do not like their telephoto zooms.
In general Canon has a better lens line, and often a bit cheaper as well. But Nikon has a few gems for which Canon has no good alternative, and those are exactly the lenses I would use most. The only lens missing is a compelling macro lens like a 4/200 with IS (same story for Canon) - but that may be in the Nikon pipeline, and for macro I would be happy with third party lens like Sigma 2.8/150.
Anyway, the shift to mirrorless makes jumping ship and buying new lenses more risky now, because neither of them has a compelling ML wildlife/action body yet and the current DSLRs may be a dead end in a few years.
seems to have better tracking AF than the 90D which disappointingly rehashes the 80D AF.
Jeff Nolten wrote:
Pixel notes the M6II
But this is comparing the 90D viewfinder AF with the M6II's exclusively sensor based AF. The 90D has this AF but you need to use live view to use it. It seems to me that trying to do face detect with 45 AF points is a bit of a stretch. Using the full sensor to do image processing like this seems more logical.
I've owned the M6II for about a week now and can report that its tracking AF seems to work pretty well. The AF in general seems very good. I was disappointed in the AF of the 55-250 STM on the SL2 and 77D since it seemed to hunt acquiring initial focus and the STM drive is slow compared to USM. The 80D is a bit better but the M6II seems much more sure footed. I haven't been able to get it to focus search yet. It always goes the right direction and achieves focus as quickly as STM can. The 70-300 IS II nano-usm is blazingly fast as are both the Sigma and Canon 100-400s. The sensor based focus bracketing/stacking feature also seems to work well with all the lenses I've tried. The IQ is good, a bit higher res than the 80D but about the same for noise and DR.
I agree with Pixel that it would have been nice if the M6II used the M5 form factor. The add-on EVF sticks out like a sore thumb but is essential to use the M6II for any lenses bigger than, well, M lenses. I think what this sensor and AF system needs is a mirrorless 7D3 with the same body and lens mount as the 7D2. The M6II is not robust enough to really handle the 100-400 II + 1.4xIII. Its battery needs to be bigger. It needs a real field worthy body. ...Show more → I'll add that the M6II has no focus parameter adjustment. What it is is what you get. You can turn off eye detect but thats it. I don't know if the sensor based AF is even potentially tunable but I think there is some untapped potential here.
technic wrote:
IMHO the problem with the 80D and 90D is the very slow response of the PDAF system (slow to get initial lock, especially) and some other strange quirks like a strong preference for the back of the subject or the background, instead of the foreground which in most situations makes more sense (but this same preference is to a smaller extent present with DPAF).
This should be relatively easy to address in firmware if they want. But talking about tunable: not having to use MFA is a major advantage, given the MFA woes of my 80D with the 100-400II: there is no way I can get acceptable focus both for closeup and infinity using PDAF, it needs around 0 for infinity and around -20 at close up. ML has a big disadvantage there, but I'm curious about one of its main disadvantages which is focusing speed with long bright lenses from an initially strongly defocused situation (important for wildlife etc.).
If you add it all up, my impression continues to be that the M6II is a test bed for these technologies but deliberately "crippled" so as not to appeal to the real 7D3 type customers. Add a capable fixed EVF, bigger battery, some more AF parameters, some improvements for current video limitations, better software for handling burst mode images etc. and it could be a serious 7D3 alternative. Maybe Canon is waiting to see if they should introduce this in an M5II or a R7X
More AF parameters might require a stacked sensor, I doubt there is much more Canon can do in realtime with all this sensor data. Without this is would be difficult to compete with cameras like Sony A9, but maybe they don't have to when the current technology in M6II with some tweaks is good enough for most enthusiast action/wildlife shooters? In that case they could do with just one very expensive high end R action body. ...Show more → cpe1991 wrote:
Most of what you say makes very good sense. But, I don't have the AF problems with the 90D. The centre point locks on extremely rapidly in AI Servo, and it doesn't have the focus changing with distance, and I take close ups of dragonflies too at 1m.
technic wrote:
Are you talking about using 90D PDAF with the 100-400II and about "action shots"? That would be weird, because I have verified the same behavior on three different 80D/ 100-400II combo's now (with both huge backfocus, plus focus shift and general IQ problems near wide open for closeups) and I haven't seen any combo without this problem. And while the PDAF is fine for static subjects, it simply doesn't work for action unless the subject is relatively big and far away from the camera (so the focus system and AF system have less work to do).
When I'm using the 80D PDAF with other lenses (e.g. 100L macro, 2.8/200) I usually have little or no backfocus problems and AF is fine for still images, but totally useless for flying dragonflies. AF is simply too slow to get a lock on anything that moves at more than glacial speed; with PDAF active I'm 99% sure to get NO image of a flying dragonfly because the AF system tries to focus on the background first and by the time it finds out there is nothing there to lock on the subject is out of the frame ;( I also tried with the 100-400II for hovering dragonflies (moving quite slowly) and it is obvious that the PDAF is running 2-3 frames (1/4 sec or so) behind what is actually happening - the AF system reports focus lock for wings etc. that are not present in those frames. ...Show more → cpe1991 wrote:
This is with classical PDAF on the 90D, and with both my 100-400mm II. For flying objects, I use the central zone of 9 points and have been able, for example, to capture flying ducks against backgrounds as the points focus on the nearest object. With dragonflies, a simple background is important when they are small in comparison to reeds etc. Give the 90D a try
The amount of focus shift varies with the copy. I have one copy with the significant focus shift (#2)that is slightly sharper than the other copy (#1) which has minimal focus shift.
The amount of focus shift varies with the copy. I have one copy with the significant focus shift (#2)that is slightly sharper than the other copy (#1) which has minimal focus shift.
EBH
Interesting; that's possible of course, probably very difficult/expensive to adjust. Are those two lenses from very different batch/purchase date? The ones I checked were all from 2-3 years ago and all three need something around AFMA -20 near MFD (at 400mm, but from experience with my copy probably similar for shorter focal length).
technic wrote:
Interesting; that's possible of course, probably very difficult/expensive to adjust. Are those two lenses from very different batch/purchase date? The ones I checked were all from 2-3 years ago and all three need something around AFMA -20 near MFD (at 400mm, but from experience with my copy probably similar for shorter focal length).
I have had 3 good copies. Bought one of the very first batch, which was good. Bought a second for my wife. Sold the first, which was a mistake, when I bought a Sigma 150-600mm. Then bought a replacement as I got fed up carrying the Sigma (which has brilliant IQ). The two 100-400mm IIs I have now do not have any noticeable focus shift.
technic wrote:
Are you talking about using 90D PDAF with the 100-400II and about "action shots"? That would be weird, because I have verified the same behavior on three different 80D/ 100-400II combo's now (with both huge backfocus, plus focus shift and general IQ problems near wide open for closeups) and I haven't seen any combo without this problem. And while the PDAF is fine for static subjects, it simply doesn't work for action unless the subject is relatively big and far away from the camera (so the focus system and AF system have less work to do).
When I'm using the 80D PDAF with other lenses (e.g. 100L macro, 2.8/200) I usually have little or no backfocus problems and AF is fine for still images, but totally useless for flying dragonflies. AF is simply too slow to get a lock on anything that moves at more than glacial speed; with PDAF active I'm 99% sure to get NO image of a flying dragonfly because the AF system tries to focus on the background first and by the time it finds out there is nothing there to lock on the subject is out of the frame ;( I also tried with the 100-400II for hovering dragonflies (moving quite slowly) and it is obvious that the PDAF is running 2-3 frames (1/4 sec or so) behind what is actually happening - the AF system reports focus lock for wings etc. that are not present in those frames....Show more →
Continues Focus search "off" should stop focus to back ground.
Interesting; that's possible of course, probably very difficult/expensive to adjust.
-------------------------
Focus shift is with a lens for life and never altered.
I would stick with a good DSLR for wildlife photography like the 80D/90D or the 5D Mark III/5D Mark IV, instead of switching over to a mirrorless system.
