Canon R6 Mark II Announced

[gt3rs]

I got maybe 5 frozen episodes with the R5 in 2 years all due to a crappy CFExpress card.

I used to move around R5 + adapter + 200-400 F4 (3.6 kg lens) by holding only the camera with one hand and never had issue with play… and this 1-2 times a weeks for 2h hockey matches.

 

[herein2020]

8 hours ago, Attila Bakos said:

But what if you have an Intel processor with 4:2:2 decoding support and a dedicated GPU? I was under the impression that they can work together.

I believe you are referring to Intel Arc, that's the only card I am aware of that can do that. The problem is if you already have a desktop that is not QS capable then the only way to get a QS capable system is to buy or build a new one.

 

12 hours ago, Django said:

EVF lag is a real thing and demands even further adaptation although it sounds like you are limited by your flash recycle time not necessarily the cameras fault?

As for the R5 freezing etc it kinda sounds like you got a lemon. I dunno since I've got an R6 and never experienced any type of freeze or shutter issue. Did they really cram too much into R5 to the point where it bugs out? If so, maybe the R5C is better at handling things with a totally separate OS that reboots in stills/video modes..?

It sounds frustrating but truth is other brands aren't immune to any of this either. I left Nikon after my D750 took months to repair because of a faulty part requiring a massive recall that took ages to arrive from China to Nikon EU service centre in Netherlands. Horrible customer service too compared to Canon who are based right outside of Paris. These are the type of things that make us pros stick or leave a brand. 

I've had several Sony Alphas ruined straight to the garbage dump due to their horrible weather sealing (just only recently improved).

But getting back to Canon, I originally joined them because of their reputation and hardware reliability. Was the only Japanese company with still Made in Japan products too (well there was Fuji too but that changed). But yeah I agree that the Canon R series and RF lenses don't have at all that same premium feel of the DSLRs/EF lenses. Maybe R3, I haven't tested it yet. But that's a $6K camera.  

Yes, not the camera's fault its a limitation of the technology. But coming from OVFs, I just want the camera to record what my eyes see in the viewfinder, didn't think that was asking for much until I switched to mirrorless.  All flashes have recycle times and I use some sort of additional lighting for nearly everything so it just is one more thing to try to adjust to. I try to predict now what will happen next but its a whole lot harder than just seeing it through an OVF.

5 hours ago, gt3rs said:

I got maybe 5 frozen episodes with the R5 in 2 years all due to a crappy CFExpress card.

I used to move around R5 + adapter + 200-400 F4 (3.6 kg lens) by holding only the camera with one hand and never had issue with play… and this 1-2 times a weeks for 2h hockey matches.

 

I don't doubt there are reliable well built R5's out there, what I do doubt is that I got one of them but no way to prove any of it and Canon's CPS already looked at it and said nothing is wrong with it.  I had the whole screen freezing problems when I first got it; that was fixed in FW 1.5, now with FW 1.6 the shutter button stopped working until a reboot.

There's no way I would put a 200-400 on my R5 body without a mono or tripod, not after the way it felt with the EF 70-200.  Maybe Canon had to source parts differently due to the pandemic, or maybe I got a bad copy.....who knows. In all the years I owned the 5DIV, it only ever locked up twice; once because the write protect switch got flipped on the SD card and once because the SD card was bad.

But anyway, this is way off topic, I am really impressed with the R7 so far as a hybrid camera, for me it has been more reliable and trouble free than the R5, so if the R6II is anything like the R7 for reliability then it is going to be a fantastic camera.  The R5II will probably fix all of the problems with the R5 similar to the R6II.

[Attila Bakos]

4 hours ago, herein2020 said:

I believe you are referring to Intel Arc, that's the only card I am aware of that can do that. The problem is if you already have a desktop that is not QS capable then the only way to get a QS capable system is to buy or build a new one.

No, I'm referring to NVIDIA/AMD. I just did a test for you. I have a Lenovo Legion 5i Pro laptop with an i7-12700h CPU and a 3070Ti GPU. I loaded a 2160p 4:2:2 HEVC F-Log clip to Resolve, added a LUT and some tweaking.
With hybrid mode enabled (that is Resolve can use both internal GPU and the dedicated one), I get smooth 25fps playback with low CPU usage and rendered out the clip with the H.265 Master preset in 18 seconds.
Now if I switch to dedicated GPU only mode, I get 21-23 fps playback and high CPU usage and the clip rendered out in 41 seconds.
So basically with 10th or higher gen Intel CPU I get 4:2:2 support in addition to what the GPU already offers, no need for a Mac (for this).

[Attila Bakos]

Forgot to mention that in this test in hybrid mode the decoding was done by the iGPU and the encoding was done by the dedicated GPU (as I encoded to 4:2:0), so they can really work together.

[Django]

10 hours ago, herein2020 said:

Yes, not the camera's fault its a limitation of the technology. But coming from OVFs, I just want the camera to record what my eyes see in the viewfinder, didn't think that was asking for much until I switched to mirrorless.  All flashes have recycle times and I use some sort of additional lighting for nearly everything so it just is one more thing to try to adjust to. I try to predict now what will happen next but its a whole lot harder than just seeing it through an OVF.

Yeah looking it up on forums it seems the R5 EVF lag is worst on some cameras (one guy who has two even claims they perform differently). I find that hard to believe but it is starting to sound like maybe there are quite a few lemons out there with the freeze issues etc. I feel like maybe Canon crammed way too much inside the R5 and its basically taking firmware after firmware to fix issues if there aren't hardware related problems as well. 

Maybe why there are so many second hand R5s on the market versus any other R model. I've always been kinda weary of the R5 since the start which is why I never felt comfortable investing so much money in one and even though you can get one used for under $3K now which is tempting, I'm starting to feel it may be more prudent to simply skip that first gen series. R5C, R6ii, R7 are probably all wiser choices at this point, and of course an R5ii is probably going to hit sooner than later to compete with the A7RV.

Getting back to EVF vs OVF again it seems stacked sensors really minimise lag while improving AF performance but there are still some improvements to reach true OVF experience. I thought it was interesting R6ii has this new "OVF simulation" mode that is supposed to enhance DR or something, maybe its just a gimmick or a step in the right direction.

[gt3rs]

I would not take the R5 on sales as a metric as I would guess that the R5 is the most sold R series model, so you get more on the used marked.

Regarding the EVF latency I'm really not sure what to say and it is probably that some people adapt better than the others. One thing is for sure my two R5 and my one R5C behave exactly the same, but they are also configured the same. 

I was surprised how quickly I adjusted from 1Dx III OVS/mirror blackout to R5 EVF 20 fps (I use 90% ES). I have more keepers while tracking & panning with the R5 than the 1Dx III and this it is what it counts for me.

This is why I never moved to the R3 as I prefer to have the 45mpix than the better EVF....

 

 

[Django]

2 hours ago, gt3rs said:

I would not take the R5 on sales as a metric as I would guess that the R5 is the most sold R series model, so you get more on the used marked.

Regarding the EVF latency I'm really not sure what to say and it is probably that some people adapt better than the others. One thing is for sure my two R5 and my one R5C behave exactly the same, but they are also configured the same. 

I was surprised how quickly I adjusted from 1Dx III OVS/mirror blackout to R5 EVF 20 fps (I use 90% ES). I have more keepers while tracking & panning with the R5 than the 1Dx III and this it is what it counts for me.

This is why I never moved to the R3 as I prefer to have the 45mpix than the better EVF....

 

Actually R6 is the best selling R series model. That category and price point is where sales max out just like A7III/IV is the best selling FF Alpha. That is also why Canon quickly released a R6 MK2 to counter A7IV sales. So no, that's not a valid excuse.. R5C however could explain it as it fixes overheating & adds cine os (its already a sort of R5 mk2).

As for OVF/EVF well if you're using ES then of course blackout disapears and you might have more keepers but that doesn't eliminate EVF lag, adds RS artefacts and you lose 14-bit IQ. That probably doesn't matter as much to you then nailing the shot given your type of extreme action work (impressive shots btw). R3 would be better for the EVF but also 50% more fps (30fps vs 20fps) in full 14-bit RAW and lower RS. But I understand 45mp may ultimately trump those specs/IQ if you crop a lot.

Lastly, I've found more info about the R6 mk2 OLED EVF and its OVF simulation mode and it looks like it is directly lifted off the R3 so another huge bonus win for the R6ii:

The R3 goes beyond the EVF of the R5, though, utilizing an all-new, Canon-designed OLED panel in this EVF. The EVF includes a setting called "OVF Simulation" that is said to mimic the natural viewing experience of the 1D X's optical viewfinder thanks to the EVF display's high brightness performance and Canon's HDR image processing technology. By default, OVF Simulation is turned off, which means you can see a live preview of the exposure and image-processing picture profile settings as you shoot -- just like with most mirrorless cameras (arguably one of the qualities I enjoy from an EVF over an optical viewfinder). 

However, for those that prefer the look at the behavior of an optical viewfinder, the OVF Simulation works quite well. But, the OVF Simulation goes beyond just making things look like an optical viewfinder and turning off the exposure preview simulation, for example. The OLED panel can get incredibly bright, and with Canon's HDR technology, you essentially have a natural-looking, HDR-based view when the OVF Simulation mode is enabled. You can see more highlight and shadow detail than in the standard EVF mode, as well as get a very bright and crisp view of your scene.

 

[Ty Harper]

5 hours ago, Django said:

Maybe why there are so many second hand R5s on the market versus any other R model.

I'll add that anecdotally, I have seen more used R5s going up for sale this past weekend in my neck of the woods, than I have ever recalled seeing before.

[Kisaha]

These recent releases seem more mature to me. It is the 3rd wave of RF. Although the R7 and R6mkII sensors are not that hi end, Canon squeezes every drop of performance out of them and achieves great things without too much fanfare.

[kye]

21 hours ago, gt3rs said:

Regarding the EVF latency I'm really not sure what to say and it is probably that some people adapt better than the others.

Getting used to there being a delay in something is absolutely a thing.  

I've had the experience several times where I went from something that had a delay to something that had less delay and the sensation that you get is that the thing happens before you hit the button!  
Obviously it wasn't beforehand, but I was so used to the delay being there that my brain had eliminated it.

Some people may be better at adapting to small delays than others of course.

[gt3rs]

23 hours ago, Django said:

As for OVF/EVF well if you're using ES then of course blackout disapears and you might have more keepers but that doesn't eliminate EVF lag, adds RS artefacts and you lose 14-bit IQ. That probably doesn't matter as much to you then nailing the shot given your type of extreme action work (impressive shots btw). R3 would be better for the EVF but also 50% more fps (30fps vs 20fps) in full 14-bit RAW and lower RS. But I understand 45mp may ultimately trump those specs/IQ if you crop a lot.

 

The 14-bit gives you max 2/3 of a stop and only up to iso 160 from 200 on no difference: https://www.photonstophotos.net/Charts/PDR.htm#Canon EOS R3,Canon EOS R3(ES),Canon EOS R5,Canon EOS R5(ES). 
This is why I shoot in ES most of the time other than the few times I need max DR and I can use ISO 100 or ISO 400 (btw the R3 has less DR than R5 at iso 400 mechanical)


The key differentiator of the R3 imo is the very fast rolling shutter so for somebody shooting any ball related sports it may make a big difference. My grip with the R3 is the price, it should be more on the A9 range than on the A1/Z9 range....

Overall price performance the R6 II is much better than the R3, the same for R5 vs R3....

R3 should have had the same spec but with a 45 Mpix sensor then would be worth the price imo.... R1?

[Django]

49 minutes ago, gt3rs said:

The 14-bit gives you max 2/3 of a stop and only up to iso 160 from 200 on no difference: https://www.photonstophotos.net/Charts/PDR.htm#Canon EOS R3,Canon EOS R3(ES),Canon EOS R5,Canon EOS R5(ES). 
This is why I shoot in ES most of the time other than the few times I need max DR and I can use ISO 100 or ISO 400 (btw the R3 has less DR than R5 at iso 400 mechanical)


The key differentiator of the R3 imo is the very fast rolling shutter so for somebody shooting any ball related sports it may make a big difference. My grip with the R3 is the price, it should be more on the A9 range than on the A1/Z9 range....

Overall price performance the R6 II is much better than the R3, the same for R5 vs R3....

R3 should have had the same spec but with a 45 Mpix sensor then would be worth the price imo.... R1?

As recently discussed with you in a parallel thread, increasing MPix will inevitably decrease RS and overall read out among other things. That is why R3 has much better RS in video than Z9 or A1. Or can do 195fps max burst rate at full resolution in FF vs 120fps in 11MP crop mode for Z9. It's always a give & take. And depending on the user that will tilt balance in either direction.  

Again I think Fuji did the smart thing with XH2/XH2S giving the user the option of super fast readout or high resolution 45MP/8K.

Finally I think the R3 sits just fine against the A1/Z9 and given the choice I'd pick it over the other two. But that's because I value read-out speed in both photo/video over super high resolution. 

In the end I'm not a sports/action shooter though so no real need to invest in such cameras. I just wished there was a fast stacked sensor in a non flagship action/sports oriented body. XH2S fits the bill but I'm not sure I wanna head back to Fuji system for other reasons.

[herein2020]

On 11/7/2022 at 2:16 AM, Attila Bakos said:

No, I'm referring to NVIDIA/AMD. I just did a test for you. I have a Lenovo Legion 5i Pro laptop with an i7-12700h CPU and a 3070Ti GPU. I loaded a 2160p 4:2:2 HEVC F-Log clip to Resolve, added a LUT and some tweaking.
With hybrid mode enabled (that is Resolve can use both internal GPU and the dedicated one), I get smooth 25fps playback with low CPU usage and rendered out the clip with the H.265 Master preset in 18 seconds.
Now if I switch to dedicated GPU only mode, I get 21-23 fps playback and high CPU usage and the clip rendered out in 41 seconds.
So basically with 10th or higher gen Intel CPU I get 4:2:2 support in addition to what the GPU already offers, no need for a Mac (for this).

 

Oh ok, I was referring to the link between the GPU and CPU in the hardware not software which so far is proprietary to Intel Arc. Based on your tests it does sound like there is a CPU/GPU combination that will work for the footage coming out of the Canon cameras, but it does not help those of us that do not have a QS capable CPU.  I still think NVIDIA, AMD, or Intel should be able to offer a GPU that handles 10 bit 4:2:2 footage natively. 

On 11/7/2022 at 1:07 PM, Django said:

Actually R6 is the best selling R series model. That category and price point is where sales max out just like A7III/IV is the best selling FF Alpha. That is also why Canon quickly released a R6 MK2 to counter A7IV sales. So no, that's not a valid excuse.. R5C however could explain it as it fixes overheating & adds cine os (its already a sort of R5 mk2).

As for OVF/EVF well if you're using ES then of course blackout disapears and you might have more keepers but that doesn't eliminate EVF lag, adds RS artefacts and you lose 14-bit IQ. That probably doesn't matter as much to you then nailing the shot given your type of extreme action work (impressive shots btw). R3 would be better for the EVF but also 50% more fps (30fps vs 20fps) in full 14-bit RAW and lower RS. But I understand 45mp may ultimately trump those specs/IQ if you crop a lot.

Lastly, I've found more info about the R6 mk2 OLED EVF and its OVF simulation mode and it looks like it is directly lifted off the R3 so another huge bonus win for the R6ii:

The R3 goes beyond the EVF of the R5, though, utilizing an all-new, Canon-designed OLED panel in this EVF. The EVF includes a setting called "OVF Simulation" that is said to mimic the natural viewing experience of the 1D X's optical viewfinder thanks to the EVF display's high brightness performance and Canon's HDR image processing technology. By default, OVF Simulation is turned off, which means you can see a live preview of the exposure and image-processing picture profile settings as you shoot -- just like with most mirrorless cameras (arguably one of the qualities I enjoy from an EVF over an optical viewfinder). 

However, for those that prefer the look at the behavior of an optical viewfinder, the OVF Simulation works quite well. But, the OVF Simulation goes beyond just making things look like an optical viewfinder and turning off the exposure preview simulation, for example. The OLED panel can get incredibly bright, and with Canon's HDR technology, you essentially have a natural-looking, HDR-based view when the OVF Simulation mode is enabled. You can see more highlight and shadow detail than in the standard EVF mode, as well as get a very bright and crisp view of your scene.

 

I did read or hear something about that in one of the videos but didn't pay close attention to it. If that does indeed come a lot closer to the OVF experience, then that really is good news. I keep getting the feeling that I should have trusted my gut instinct and skipped the R5 and waited for the R5II which I am sure will fix almost if not every problem I have. One question I would have though is how much does that add to the battery drain when that mode is enabled.

I have noticed with the R7 that if I turn on the high FPS mode the battery life seems to drop by at least 30%.

 

12 hours ago, kye said:

Getting used to there being a delay in something is absolutely a thing.  

I've had the experience several times where I went from something that had a delay to something that had less delay and the sensation that you get is that the thing happens before you hit the button!  
Obviously it wasn't beforehand, but I was so used to the delay being there that my brain had eliminated it.

Some people may be better at adapting to small delays than others of course.

 

I think that would work fine for something with a predictable trajectory like a person walking or in a situation like the ballerina where you can just hold down the shutter and pick the best one later (if you are using continuous lighting); but it is impossible to predict something that happens with completely unpredictability when combined with flash synch constraints as well as an EVF which is only showing you something that already happened. There is no way to predict a quick dress flip, hair toss, over the shoulder glance, etc. when the EVF is so far behind you don't even see it until it is over.

It is kind of ironic because my keeper rate for sharp images went up due to the improved AF of the R5, then went back down due to the EVF lag so I ended up around where I started. I would actually prefer an occasional out of focus image to what I am dealing with now which is missing peak action moments.

I have to imagine that most people who are fine with the EVF lag aren't filming the type of subject matter that I am where I am so dependent on flash recycle times.  Don't get me wrong though, for pretty much everything else I shoot (headshots, portraits, events, etc. etc.) the R5 is great since I can either redo, everything is predictable, or there is continuous lighting (the sun).  But it really irks me when I get in situations where an OVF would be far more reliable after paying so much for the R5.

[Django]

10 hours ago, herein2020 said:

I did read or hear something about that in one of the videos but didn't pay close attention to it. If that does indeed come a lot closer to the OVF experience, then that really is good news. I keep getting the feeling that I should have trusted my gut instinct and skipped the R5 and waited for the R5II which I am sure will fix almost if not every problem I have. One question I would have though is how much does that add to the battery drain when that mode is enabled.

I have noticed with the R7 that if I turn on the high FPS mode the battery life seems to drop by at least 30%.

 

I think it brings you closer to the OVF viewing experience but its unclear if it eliminates EVF lag. For that I think you really need a stacked sensor (or faster processor) which improves refresh rates but who knows it may still be improved from R5. Kind of a shame there is no data on this issue. As for battery drain, I think I read it didn't drain any more but again no hard evidence.

10 hours ago, herein2020 said:

have to imagine that most people who are fine with the EVF lag aren't filming the type of subject matter that I am where I am so dependent on flash recycle times.  Don't get me wrong though, for pretty much everything else I shoot (headshots, portraits, events, etc. etc.) the R5 is great since I can either redo, everything is predictable, or there is continuous lighting (the sun).  But it really irks me when I get in situations where an OVF would be far more reliable after paying so much for the R5.

I guess this is where your 5D IV still holds a valuable purpose? I can really see how in run & gun flash photography EVF lag would be a serious handicap.

[kye]

11 hours ago, herein2020 said:

I think that would work fine for something with a predictable trajectory like a person walking or in a situation like the ballerina where you can just hold down the shutter and pick the best one later (if you are using continuous lighting); but it is impossible to predict something that happens with completely unpredictability when combined with flash synch constraints as well as an EVF which is only showing you something that already happened. There is no way to predict a quick dress flip, hair toss, over the shoulder glance, etc. when the EVF is so far behind you don't even see it until it is over.

It is kind of ironic because my keeper rate for sharp images went up due to the improved AF of the R5, then went back down due to the EVF lag so I ended up around where I started. I would actually prefer an occasional out of focus image to what I am dealing with now which is missing peak action moments.

I have to imagine that most people who are fine with the EVF lag aren't filming the type of subject matter that I am where I am so dependent on flash recycle times.  Don't get me wrong though, for pretty much everything else I shoot (headshots, portraits, events, etc. etc.) the R5 is great since I can either redo, everything is predictable, or there is continuous lighting (the sun).  But it really irks me when I get in situations where an OVF would be far more reliable after paying so much for the R5.

Yes, being able to anticipate the moment and compensate for it by hitting the button early would be highly dependent on how predictable the perfect moment is, how long the delay time is, and how good you are at compensating for the delay.

It seems like you have particularly good timing and are operating in less predictable scenarios so the performance on offer isn't sufficient to meet your needs.  Pity.  Ultimately though, you have to go with what gets you the most keepers.  After all, cameras are tools, not toys, right 🙂 

[herein2020]

11 hours ago, Django said:

I think it brings you closer to the OVF viewing experience but its unclear if it eliminates EVF lag. For that I think you really need a stacked sensor (or faster processor) which improves refresh rates but who knows it may still be improved from R5. Kind of a shame there is no data on this issue. As for battery drain, I think I read it didn't drain any more but again no hard evidence.

I guess this is where your 5D IV still holds a valuable purpose? I can really see how in run & gun flash photography EVF lag would be a serious handicap.

I know a lot of buyers of the EOS R really complained about the lag. They had one on display in a store once and I picked it up and just panned from side to side while looking through the viewfinder; the lag was so bad it was the final nail in the coffin for the EOS R for me. I was truly in disbelief that anyone was willing to put up with it.

I think at the end of the day I am probably the only one that notices the shots I missed or didn't get perfect while knowing the lag caused it. The only time when the client noticed as well was the ballerina shoot. She was giving me looks that made it clear she didn't know why I wasn't nailing her leg and arm placement mid leap. 

10 hours ago, kye said:

Yes, being able to anticipate the moment and compensate for it by hitting the button early would be highly dependent on how predictable the perfect moment is, how long the delay time is, and how good you are at compensating for the delay.

It seems like you have particularly good timing and are operating in less predictable scenarios so the performance on offer isn't sufficient to meet your needs.  Pity.  Ultimately though, you have to go with what gets you the most keepers.  After all, cameras are tools, not toys, right 🙂 

Honestly, I am probably just making it more than it is.  If I did not rely on flash so heavily or shoot so many specific situations that require both flash lighting as well as precision then the lag wouldn't bother me as much. With sports you can just hold down the shutter so mirrorless is definitely capable of capturing peak action even with the lag. At events people are used to redos over someone blinking so that's not a big deal either.

For 99% of my work, only I will know when I missed the exact moment I wanted due to the EVF lag so I can live with that.

[Django]

8 hours ago, herein2020 said:

I know a lot of buyers of the EOS R really complained about the lag. They had one on display in a store once and I picked it up and just panned from side to side while looking through the viewfinder; the lag was so bad it was the final nail in the coffin for the EOS R for me. I was truly in disbelief that anyone was willing to put up with it.

I think at the end of the day I am probably the only one that notices the shots I missed or didn't get perfect while knowing the lag caused it. The only time when the client noticed as well was the ballerina shoot. She was giving me looks that made it clear she didn't know why I wasn't nailing her leg and arm placement mid leap. 

Your issue prompted me to investigate on my end so I popped on my Speedlite flash last night on the R6 having my nephew around and tried to catch random split second movements and to my surprise I found it was pretty easy to nail shots. I also couldn't notice any significant lag from the EVF when set to smooth. So I don't know if its a user case thing,  if R5 and its 45MP somehow increases the lag, if you got a lemon or wether ballerinas are just much harder to shoot than an 8 year-old (I'd assume so!) but for my expectations the R6 works just fine even for flash photography.

Since you've also got an R7 have you noticed any difference concerning EVF lag with your R5? 

[gt3rs]

7 hours ago, Django said:

Your issue prompted me to investigate on my end so I popped on my Speedlite flash last night on the R6 having my nephew around and tried to catch random split second movements and to my surprise I found it was pretty easy to nail shots. I also couldn't notice any significant lag from the EVF when set to smooth. So I don't know if its a user case thing,  if R5 and its 45MP somehow increases the lag, if you got a lemon or wether ballerinas are just much harder to shoot than an 8 year-old (I'd assume so!) but for my expectations the R6 works just fine even for flash photography.

Since you've also got an R7 have you noticed any difference concerning EVF lag with your R5? 

I also never had a problem with lag on R5 and strobes, the only thing I need is to disable exposure simulation as it would be all black in scenarios like this one

[Kisaha]

pre-order from the official photo-shops in Greece 3.120euros. I hope it will be stabilized lower in time..

[herein2020]

12 hours ago, Django said:

Your issue prompted me to investigate on my end so I popped on my Speedlite flash last night on the R6 having my nephew around and tried to catch random split second movements and to my surprise I found it was pretty easy to nail shots. I also couldn't notice any significant lag from the EVF when set to smooth. So I don't know if its a user case thing,  if R5 and its 45MP somehow increases the lag, if you got a lemon or wether ballerinas are just much harder to shoot than an 8 year-old (I'd assume so!) but for my expectations the R6 works just fine even for flash photography.

Since you've also got an R7 have you noticed any difference concerning EVF lag with your R5? 

I tested with the R7 after you mentioned it and it does seem to be much better than the R5 but until a challenging shoot I won't really know for sure. To see the latency, you would need to specifically test for it. A good test would be to attempt to film someone throwing a ball in the air directly in front of them with the lens set to around 50mm while standing close enough to frame just their chest and their chin and try to capture the ball right as it leaves their hand and is midway in their chest but before it is out of frame. 

Or if you are filming a kid running around specifically try to capture when their right or left foot just touches the ground but before the heel of their foot touches the ground. 

In the ball scenario the ball will probably be out of the frame in real life but still show in the EVF and in the kid scenario the ball of their foot will probably be on the ground while the EVF still shows their heel in the air. Both of these would be easy to do with an OVF.

 

5 hours ago, gt3rs said:

I also never had a problem with lag on R5 and strobes, the only thing I need is to disable exposure simulation as it would be all black in scenarios like this one

 

Even a jumping horse with strobes would be easy enough with my current R5 or R7.  It is when you are focusing on something very specific; a specific detail that you know is there in the EVF but does not show up in the final image that is gets frustrating. 

I notice it most at runway shows, where I miss the peak of a hair toss, a split second over the shoulder look, a heel that hasn't lifted back off the ground yet, etc.  The ballerina was a great example. In the EVF her legs would be perfectly straight, the hair would be at the perfect angle, she would be at the peak of the jump with perfect form; but in the captured image the hair would have fallen a bit, the knees are bent, she is already starting her descent back to the ground.  With an OVF capturing those moments would have been effortless.

Like I mentioned, only I will know what I wanted to capture vs what ended up on the memory card, but to me that's enough to matter.