Will there be a 7D3 or 5DV?

russbecker wrote in post #19017880
IMO there will be no DSLR replacements for either the 7D2 or the 5D4; both the announcement of the R5 and my experience with both the EOS R and the Fuji X-H1 have convinced me that MILC is where we are going to be going. For the most part, all that really needs serious addressing is the EVF in both quality of image and lag in updates (shooting BIF with the X-H1 can be somewhat disconcerting although the results are excellent). They key here is sensor readout speed, the newest BSI Sony type sensors with embedded DRAM try to address this.

I have, in some past posts, pointed out a theoretical EFV issue, even at 120 fps update rate of the viewfinder LED of cameras like the Sony A9, that the display can present information delayed from the real position of the subject. At 120 fps, that computes to (at best) an 8.33ms delay in real location vs. displayed view, which can present a significant error if tracking a very high speed object while zoomed in tight...like tracking a Blue Angel, flying the relatively sedate pace of an airshow travelling 350mph, is moving 513 ft/sec... 120 Hz EVF refresh misrepresents subject position by 51 inches (at the show max speed of 700mph, that positional error becomes 102 inches!)

The XH-1 is claimed to have a display time lag of a mere 0.005 seconds, with an 100 fps EVF update rate...5 ms claimed with 100 fps (vs. 8.3ms theoretical at 120 fps) So I am interested to hear what your experiences have been with the X-H1 that results in your above assessment. Tracking a bird flying and rapidly changing direction is obviously very different from tracking a 40000 lb. jet

The distance offset is not the issue, it is the angular offset that matters. How close is the 51 or 104 inches? That will determine how far across the frame the object will have moved.

The human vision system also has an effective frame rate and there is a lag in the human vision system that has been measured to be at least 40ms, as reactions to sound have been found to be this much faster than reaction to light. How this frame rate and lag combine with the lag from the EVF, I don’t know. However, the human lag is greater than that of a 120 FPS screen. Having said that humans can detect the difference in smoothness between 30, 70 and 120 FPS. Human vision is a complex system.

Honey Monster wrote in post #19018584
The distance offset is not the issue, it is the angular offset that matters. How close is the 51 or 104 inches? That will determine how far across the frame the object will have moved.

The human vision system also has an effective frame rate and there is a lag in the human vision system that has been measured to be at least 40ms, as reactions to sound have been found to be this much faster than reaction to light. How this frame rate and lag combine with the lag from the EVF, I don’t know. However, the human lag is greater than that of a 120 FPS screen. Having said that humans can detect the difference in smoothness between 30, 70 and 120 FPS. Human vision is a complex system.

You can't threshold equipment delays against empirical human reaction times, though. Humans are self-calibrated to take everything into account when reacting to things where the speed of light is the only delay. If that were not so, you'd see a lot more baseball batters striking out.

Back to what Wilt was saying, I do not trust the numbers quoted for equipment delays. There are so many ways to use weasel words to express delays that are not the true extra perceptual delays of every part of the frame.

Ideally, if you have a rolling electronic shutter, you also have a rolling EVF display at the same framerate and speed, and several lines are read, converted, and put to the display. That is the best case scenario, and the minimum delay time is going to be half the exposure time, plus the wait for enough read sensor lines to demosaic and convert, and then half the EVF refresh period. I am skeptical that even if existing systems are doing this, that they are getting consistent delays of only 5 or even 8.33 ms as we have seen quoted. Delay is not just the rolling shutter period.

,

Honey Monster wrote in post #19018584
The distance offset is not the issue, it is the angular offset that matters. How close is the 51 or 104 inches? That will determine how far across the frame the object will have moved.

The human vision system also has an effective frame rate and there is a lag in the human vision system that has been measured to be at least 40ms, as reactions to sound have been found to be this much faster than reaction to light. How this frame rate and lag combine with the lag from the EVF, I don’t know. However, the human lag is greater than that of a 120 FPS screen. Having said that humans can detect the difference in smoothness between 30, 70 and 120 FPS. Human vision is a complex system.

If you are zoomed in on the cockpit of the F-18 with your 100-400mm lens, the fact that you think you are pointed 'at the cockpit', any time lag in the presentation of the image to the EVF does indeed matter, because the jet has actually moved 50 inches or 100 inches out of frame before you see it in the EVF...and in that case the angular displacement doesn't matter.

But 100 inches at 100 metres has a very different effect on the sensor than 100 inches at 100 cm. It is the angular difference that is important.

Honey Monster wrote in post #19018764
But 100 inches at 100 metres has a very different effect on the sensor than 100 inches at 100 cm. It is the angular difference that is important.

If your tele lens at 400mm frames a horizontal of 200 inches when the target is 2300' away, 100 inches is half way across the frame.

If your 24mm wide angle lens frames a horizontal of 200 inches when the target is 140' away, 100 inches is half way across the frame.

The angles involved may be very different, but based upon a horizontal 200" framed, a 100" error is nevertheless half of the frame!

John Sheehy wrote in post #19018719
You can't threshold equipment delays against empirical human reaction times, though. Humans are self-calibrated to take everything into account when reacting to things where the speed of light is the only delay. If that were not so, you'd see a lot more baseball batters striking out.

The EVF delay of less than 10 msec is nothing compared to our reaction times. Hitting a fastball, the hitter has about 400 msec to respond to a regular moving object; this is something they have practiced many times.

We take 150 to 300 msec to respond to a visual stimulus.
https://www.scientific​american.com …ience-home-reaction-time/
https://spectrum.ieee.​org …hout-feeling-like-a-robot

The first link is interesting to anyone who likes to experiment, and posits some reasons for reaction time variance.

John Sheehy wrote in post #19018719
Back to what Wilt was saying, I do not trust the numbers quoted for equipment delays. There are so many ways to use weasel words to express delays that are not the true extra perceptual delays of every part of the frame.

Ideally, if you have a rolling electronic shutter, you also have a rolling EVF display at the same framerate and speed, and several lines are read, converted, and put to the display. That is the best case scenario, and the minimum delay time is going to be half the exposure time, plus the wait for enough read sensor lines to demosaic and convert, and then half the EVF refresh period. I am skeptical that even if existing systems are doing this, that they are getting consistent delays of only 5 or even 8.33 ms as we have seen quoted. Delay is not just the rolling shutter period.

,

Why so? The "extra perceptual" delays shouldn't count, anyway, as those are our delays, not that of the EVF. Why do you think the manufacturers are lying? You have posted this stuff earlier but never an iota of evidence.

russbecker wrote in post #19017880
IMO there will be no DSLR replacements for either the 7D2 or the 5D4; both the announcement of the R5 and my experience with both the EOS R and the Fuji X-H1 have convinced me that MILC is where we are going to be going. For the most part, all that really needs serious addressing is the EVF in both quality of image and lag in updates (shooting BIF with the X-H1 can be somewhat disconcerting although the results are excellent). They key here is sensor readout speed, the newest BSI Sony type sensors with embedded DRAM try to address this.

The problem I have with acquiring an Nikon D500 is that camera is already nearly four years old; not a good idea to sink money into old tech. The real attraction for that system is the 500mm PF lens but you would hate to be stuck with an orphaned system that set you back $5K (D500 plus 500 PF). Nikon is not likely to bring out a D500 successor IMO, so you would be moving to Nikon MILC with an adapted lens.

I am more interested in seeing how the competing MILC sensor architectures play out: right now we have two distinct designs, one from Sony and one from Canon. The Sony type sensors use embedded PDF pixels (which then have to be compensated for in the RAW data) and PD/CD hybrid AF. Canon has bet the farm on their DPAF architecture which I suspect is the primary reason why their sensors have a bit less DR than the Sony sensors, mostly due to the greater actual photo diode count. The DPAF may also be why Canon has not abandoned low-pass AA filters, rolling off the high spatial frequencies may be needed for DPAF reliability, although this is pure speculation on my part.

Hey Russ,I see your in Central PA like me!Since I shoot mostly sports/ wildlife I love having a cropper for those daylight sports and the D500 is still at the top of the heap and will probably last me until I eventually shot shooting action stuff.IMHO mirrorless ( Canon / Nikon) can’t quite meet my needs yet.

THE most important question is, "How much longer will new 5D4s be available?"

If I have to replace either of my two DSLRs, I would be very happy to get a 5D4, especially with the price slowly falling.

That's if I continue with large cameras. The arthritis gets to make that decision.

Doesn't appear there will be a 5DV.

The R5 will be a mirrorless 5D

https://www.dpreview.c​om …m_campaign=traf​fic_source

Spencerphoto wrote in post #19019003
THE most important question is, "How much longer will new 5D4s be available?"

If I have to replace either of my two DSLRs, I would be very happy to get a 5D4, especially with the price slowly falling.

That's if I continue with large cameras. The arthritis gets to make that decision.

Sorry to hear about the arthritis Spencer - can only imagine how painful that must be. I haven't been hit by that yet but each week I seem to invent a new ache or pain.

My 7 year old 5D3 is still going strong after I don't know how many hundreds of thousands of shots as well as a major repair - tried to drown it about a year ago - and it's been a wonderful machine. Whatever I get next though will set off a chain reaction though and I'll have to decide which direction I want to go. I'm still using a standalone copy of CS5 that I own, which won't work with any of the advanced cameras (can't read the .CR2 files) so I'll have to move up to the Cloud. Things are functioning perfectly now so I'm always loathe to fix what ain't broken .

Of course with the larger file sizes of a newer body I'll probably want to upgrade from my 8-year old PC, which works just fine with the current gear. It's a never ending cycle......

sapearl wrote in post #19023505
Sorry to hear about the arthritis Spencer - can only imagine how painful that must be. I haven't been hit by that yet but each week I seem to invent a new ache or pain.

My 7 year old 5D3 is still going strong after I don't know how many hundreds of thousands of shots as well as a major repair - tried to drown it about a year ago - and it's been a wonderful machine. Whatever I get next though will set off a chain reaction though and I'll have to decide which direction I want to go. I'm still using a standalone copy of CS5 that I own, which won't work with any of the advanced cameras (can't read the .CR2 files) so I'll have to move up to the Cloud. Things are functioning perfectly now so I'm always loathe to fix what ain't broken .

Of course with the larger file sizes of a newer body I'll probably want to upgrade from my 8-year old PC, which works just fine with the current gear. It's a never ending cycle......

My opinion

Im a wait and see

the R5 has to be better than my current 5D4 ir there is no need for me to buy one

5D4 is a great camera and the 1Dx2 is even better