lets end this whole evf lag now

[Lee Jay]

I just did some searching for DSLR Lag time.

It's effectively zero - the light exits the viewfinder within a nanosecond or so of when it enters the lens, only limited by the speed of light.

Yes, I realize this. But there are several factors that make up the DSLR lag time.

The only lag time we're talking about is viewfinder lag time, and that is made up only of the time of flight of the light.

They are measuring from the time the shutter is pressed until the image is captured.

Which is another topic, unrelated to the one in this thread.

The human reaction time is not part of this time.

Nor is that.

And by pre-focusing,

Nor is that.

the time for the lens to focus is also not part of this time. What remains is the time for the camera to respond to the shutter press, time to raise the mirror and time to capture the image. And this is the 49ms that is recorded. And this is the time on a camera that is faster than most other cameras

Which is not relevant to the topic of this thread.

And it is as hard to find as EVF lag time. Seems that they like to just lump it into Shutter Response Time.

No, those are two entirely separate things. Viewfinder lag can be zero and shutter lag could be anything (I've had it set to an hour once with a timer) or viewfinder lag could be long and shutter lag could be virtually zero. They are independent.

No, the same thing can be said for the mirrorless.

Mirrorless viewfinder lag can NEVER be zero.

The experiment being done here is showing how small the lag time really is on recent model cameras.

The experiment done in the OP is about viewfinder lag, not shutter lag, but it wasn't done well enough to determine viewfinder lag.

My own experiment with the A6300 was showing that the monitor and camera numbers were the same, but if I had done it more times, (I only did it about a dozen times), I would likely have seen them differ occasionally. And the resolution was 10ms, but the LCD scan rate was likely 16.67ms. So let's just say that the A6300 is no better than 16.67ms.

Probably around 25ms in good light.

This is a time that is less than 1/10 the human reaction time

Which is irrelevant.

of the fastest individuals. I am almost 70 now, and my time was more like 400ms. I suspect also, that the time from shutter pressed to image captured is not going to be much worse than the D500 time.

Which is irrelevant.

Let's now consider the Sony A9. It can do 20 frames per sec and can adjust focus on the fly too as it is bursting. This means that it is capturing an image every 50ms.

Which is irrelevant.

Now, if you go back in time, and consider the early mirrorless cameras, I agree that the lag time is going to be longer.

Which lag time?

But my contention is that the existing lag on recent cameras is much smaller.

Viewfinder lag hasn't decreased significantly. In fact, I have a camera that's close to 40 years old with a lag of 33ms.

And I don't care how fast you are, you just are not going to be able to discern a 20ms lag.

That's just flat out false.

Not when it takes someone with faster reflexes than my own to react in 200ms. You are not going to be able to discern it!

Yes, I can easily discern it, and it greatly affects my ability to track moving subjects, which is the important part of this discussion. As I've said repeatedly, when I switched from 0 lag (dSLR) to 25ms of lag (EVF - measured), tracking the same subject from the same place required me to reduce focal length from 600mm to 200mm, and that's a factor of 9 reduction in pixels on the target.

 

[PhotoTeach2]

I gave you a link to a millisecond clock and explained how to process the data, and you did neither one.

Looks like between 10 and 50ms to me, but you'd actually have to do the experiment right to figure it out.

you are right . im off to work but will shoot it with a faster shutter speed when i get home. 1/100 sec clock will do

If you average dozens of frames.

but will have a look at you 1/1000 later. interesting test though.

I've done it many times, and gotten numbers from 25ms to 130ms, depending on conditions and settings.

But Lee ... if you are still basing that on (dSLR) LV, (of older dSLR's); I suggest meaningless compared to more modern ML optimized EVF/LCD.

Especially if Canon/Nikon because they have proven they are not SOTA-ML.

 

[Donald B]

What exactly is the question being discussed here?

[…]

So exactly what is the debate of this thread?

When you see Donald B, you do not need to read the OP. It boils down to:

1. Mirrorless rules
2. In the mirrorless world, m43 rules
3. Whatever camera Donalnd B has right now, rules the m43 world
4. At the end of the thread, he will post spider pictures or photos of children dancing.

dont see you posting any images .cant you shoot a decent image ? i guess not nice image mate LOL

https://www.dpreview.com/galleries/8577108286/photos/3733671/

Don

 

[Donald B]

But as much time as you spend here arguing that you have the best camera, biggest yacht, and how DSLRs can't take pictures... there is no way you have any.

It was Lee Jay claiming mirrorless cameras can't take photos of moving objects because of EVF lag.

And taking photos of a timer on a computer monitor sure put that to rest.....

Regardless, your comment has absolutely nothing to do with what I said.

Nowhere in this thread did the OP claim DSLRs can't take pictures. That is just something you made up.

This thread is about Lee Jay's claim that mirrorless cameras can't take photos of moving objects because of EVF lag...

Th OP never mentioned Lee Jay.

That is just something you made up.

Lee Jay's claim sparked a number of threads, including this one. If you follow the sequence, it is clear, even though he wasn't mentioned by name in the OP.

Lee Jay's claim was also based on entirely false premises, but that is another story.

What you and Kiwi have said about "Lee Jay's claim," is deliberately disingenuous and I think you probably know it. Lee Jay and others have addressed the issue of EVF lag causing problems in keeping fast moving subjects appropriately in the image frame. That is especially true when the fast moving subjects also change direction drapidly and unpredictably. No one denies the ability of EVF cameras to pan and follow moderately moving subjects with predictable trajectories.

Creating a disingenuous strawman argument does this discussion no good to anyone--not even the ML fanboys. It also makes you seem like you only want to engage in argumentation instead of addressing the actual issue of lag. That's also how I interpret your continual conflation of EVF lag with the very different effects of shutter lag, mirror blackout and other issues unrelated to tracking.

If one has a camera with an EVF, it helps one to keep the image within the image field to have a full frame sensor and to use a moderate focal length lens. That is simply overpowering the tracking problem by giving you a smaller image in a wider field but what we're really addressing here is tracking with otherwise similar framing not using different sized frames to assist a viewfinder lag.

Please understand that I do not wish to be argumentative. But I would appreciate your comments on the resolution and framing of the aircraft, birds and dragonfly images I posted.

I believe these cover most of the fast moving objects the vast majority of us wish to photograph.

No, I don't think they do cover the issue of fast moving or unpredictable. Gliding birds I've already commented on. They're easy. A taking off prop airplane is predictable, slow and easy to pan. Biplanes? Fast? Come on. A red-tailed hawk taking off is easy as is a red-tailed hawk cruising like the other birds you show. The osprey? It was nearly stationary just before your shot. That's not fast moving and unpredictable. I don't know what your dragon fly was doing but they hover a whole lot and that is easy to shoot.

I don't mean to be argumentative but I covered the issue of steady, predictable movement in a previous post. No, I think your images are fine and I have thousands of images of birds flying taken with my Canon SX50 but not doing the wild and crazy things they do in the more interesting situations.

I truly do not consider the EVF lag (which I readily accept is real, if small) to be a material issue for these subjects. Note also that I use (admittedly kind of “best in class” but still) relatively inexpensive bridge-style cameras and not the best mirrorless cameras.

I would suggest the results I get would be more than acceptable in by far the most common applications. In which case I would also suggest that, while real, the lag is sufficiently small to be irrelevant in all but highly specialised, esoteric applications.

Am I missing something?

Yes.

well put your money where your mouth is ? come on post one of your so called fast action images.

Don

 

[Donald B]

I just did some searching for DSLR Lag time. And it is as hard to find as EVF lag time. Seems that they like to just lump it into Shutter Response Time.

I did find this table on the Nikon D500 which is a very fast camera. The first time, 170ms, includes the time for the lens to focus, and different lenses are going to vary from this time. So I will not consider that time.

The next time involves having the lens in manual focus mode and to have it prefocused. We see here a time of 49ms. The next time is also 49ms where the shutter is half pressed to allow it to achieve focus before pressing the shutter all the way/

Included in these times would be the mirror up time. I did find reference to the fact that the exposure actually starts when the mirror is approx. 45 deg. But I assume this is dependent on factors such as shutter time.

The last time is the live view equivalent of the half press shutter method previously described. And this was 103ms.

https://www.imaging-resource.com/PRODS/nikon-d500/nikon-d500A6.HTM



I tried to find equivalent numbers for mirrorless cameras, and for slower DSLR cameras, but have not found them. I assume that the fastest mirrorless cameras are somewhere in the same vicinity as the D500.

they are actually faster , image resourse the em1mk2 or the em1x perfomance.

Don

--
Olympus EM5mk2 ,EM1mk2

http://www.dpreview.com/galleries/9412035244

past toys. k100d, k10d,k7,fz5,fz150,500uz,canon G9, Olympus xz1 em5mk1

 

[Donald B]

I just did some searching for DSLR Lag time.

It's effectively zero - the light exits the viewfinder within a nanosecond or so of when it enters the lens, only limited by the speed of light.

Yes, I realize this. But there are several factors that make up the DSLR lag time.

The only lag time we're talking about is viewfinder lag time, and that is made up only of the time of flight of the light.

They are measuring from the time the shutter is pressed until the image is captured.

Which is another topic, unrelated to the one in this thread.

The human reaction time is not part of this time.

Nor is that.

And by pre-focusing,

Nor is that.

the time for the lens to focus is also not part of this time. What remains is the time for the camera to respond to the shutter press, time to raise the mirror and time to capture the image. And this is the 49ms that is recorded. And this is the time on a camera that is faster than most other cameras

Which is not relevant to the topic of this thread.

And it is as hard to find as EVF lag time. Seems that they like to just lump it into Shutter Response Time.

No, those are two entirely separate things. Viewfinder lag can be zero and shutter lag could be anything (I've had it set to an hour once with a timer) or viewfinder lag could be long and shutter lag could be virtually zero. They are independent.

No, the same thing can be said for the mirrorless.

Mirrorless viewfinder lag can NEVER be zero.

The experiment being done here is showing how small the lag time really is on recent model cameras.

The experiment done in the OP is about viewfinder lag, not shutter lag, but it wasn't done well enough to determine viewfinder lag.

My own experiment with the A6300 was showing that the monitor and camera numbers were the same, but if I had done it more times, (I only did it about a dozen times), I would likely have seen them differ occasionally. And the resolution was 10ms, but the LCD scan rate was likely 16.67ms. So let's just say that the A6300 is no better than 16.67ms.

Probably around 25ms in good light.

This is a time that is less than 1/10 the human reaction time

Which is irrelevant.

of the fastest individuals. I am almost 70 now, and my time was more like 400ms. I suspect also, that the time from shutter pressed to image captured is not going to be much worse than the D500 time.

Which is irrelevant.

Let's now consider the Sony A9. It can do 20 frames per sec and can adjust focus on the fly too as it is bursting. This means that it is capturing an image every 50ms.

Which is irrelevant.

Now, if you go back in time, and consider the early mirrorless cameras, I agree that the lag time is going to be longer.

Which lag time?

But my contention is that the existing lag on recent cameras is much smaller.

Viewfinder lag hasn't decreased significantly. In fact, I have a camera that's close to 40 years old with a lag of 33ms.

And I don't care how fast you are, you just are not going to be able to discern a 20ms lag.

That's just flat out false.

Not when it takes someone with faster reflexes than my own to react in 200ms. You are not going to be able to discern it!

Yes, I can easily discern it, and it greatly affects my ability to track moving subjects, which is the important part of this discussion. As I've said repeatedly, when I switched from 0 lag (dSLR) to 25ms of lag (EVF - measured), tracking the same subject from the same place required me to reduce focal length from 600mm to 200mm, and that's a factor of 9 reduction in pixels on the target.

sorry lea but you havnt done the math, a guy has already done the math and the 6ms lag no one could ever track and lose the subject its that fast.

Don

 

[New Day Rising]

But as much time as you spend here arguing that you have the best camera, biggest yacht, and how DSLRs can't take pictures... there is no way you have any.

It was Lee Jay claiming mirrorless cameras can't take photos of moving objects because of EVF lag.

And taking photos of a timer on a computer monitor sure put that to rest.....

Regardless, your comment has absolutely nothing to do with what I said.

Nowhere in this thread did the OP claim DSLRs can't take pictures. That is just something you made up.

This thread is about Lee Jay's claim that mirrorless cameras can't take photos of moving objects because of EVF lag...

Th OP never mentioned Lee Jay.

That is just something you made up.

Lee Jay's claim sparked a number of threads, including this one. If you follow the sequence, it is clear, even though he wasn't mentioned by name in the OP.

Lee Jay's claim was also based on entirely false premises, but that is another story.

What you and Kiwi have said about "Lee Jay's claim," is deliberately disingenuous and I think you probably know it. Lee Jay and others have addressed the issue of EVF lag causing problems in keeping fast moving subjects appropriately in the image frame.

Others have addressed the issue of EVF lag causing problems in keeping fast moving subjects appropriately in the image frame.

Lee Jay purported over multiple posts to be addressing the issue of EVF lag causing problems in keeping fast moving subjects appropriately in the image frame. It turns out, after much probing by me and evasive answers by him, that he was actually talking about the issue of the lag from the LCD screen on his DSLR causing problems in keeping fast moving subjects appropriately in the image frame.

Lee Jay claims that an LCD screen on a DSLR is exactly the same as an EVF on a mirrorless camera and that by stating he was using an EVF he was not misleading anyone (despite multiple other members obviously gaining the impression that he was using an EVF on a mirrorless camera). He may be right and it may be that they are one and the same for all intents and purposes, but nobody has demonstrated that.

That is especially true when the fast moving subjects also change direction drapidly and unpredictably. No one denies the ability of EVF cameras to pan and follow moderately moving subjects with predictable trajectories.

Creating a disingenuous strawman argument does this discussion no good to anyone--not even the ML fanboys.

I wasn't the one that created the disingenuous strawman.

I own and have used both DSLRs and mirrorless cameras. Each has its pros and cons and there is no doubt that EVF lag is an issue. The question is whether it is reasonable to make statements about EVF lag based on your experience with an LCD screen on a DSLR (while continuing to call it an EVF, which I believe has a generally accepted meaning that is distinct from an LCD).

It also makes you seem like you only want to engage in argumentation instead of addressing the actual issue of lag.

I am happy to address the actual issue of lag. I just want it to be clear what lag I am talking about.

That's also how I interpret your continual conflation of EVF lag with the very different effects of shutter lag, mirror blackout and other issues unrelated to tracking.

I have never conflated those and have actually commented on others conflating what are obviously different issues. You are addressing your comments to the wrong person.

If one has a camera with an EVF, it helps one to keep the image within the image field to have a full frame sensor and to use a moderate focal length lens. That is simply overpowering the tracking problem by giving you a smaller image in a wider field but what we're really addressing here is tracking with otherwise similar framing not using different sized frames to assist a viewfinder lag.

All of these threads are based on Lee Jay's claims about having to use different sized frames to assist a viewfinder (in fact, LCD) lag. You are addressing your comments to the wrong person.

 

[PerL]

Don't have the slightest idea what this is trying to show. Just look through the viewfinder and pan side to side quickly. There it is and if you say you can't see it, you are not telling the truth. You will never see it with a motionless camera and subject.

Yes, pan the camera fast back and forth - and push the camera processing at the same time by setting the camera to AF-C and press the shutter, taking series while the AF is working. Do this is good light and in bad light and with autoexposure.

You have just added another variable to this test, and that is lens focus time. The lens must be in manual mode in order to judge the EVF or LCD lag.

Not if you shoot in release priority. EVF lag while focusing and shooting and saving images is a real world scenario where the lag actually matters.

I have lenses that focus real fast. And I have lenses that take their good old time.

 

[Lee Jay]

sorry lea but you havnt done the math, a guy has already done the math and the 6ms lag no one could ever track and lose the subject its that fast.

My name isn't "lea" and you should try punctuation and capitalization. It won't help the fact that you're clueless about control loops but it might make you look less stupid.

 

[Bob A L]

Your images are fine. I have never said mirrorless doesn't make great images. My point is how hard it is to get those images in the smeary laggy electronic viewfinder. My hats off to your skill.

 

[PhotoTeach2]

I gave you a link to a millisecond clock and explained how to process the data, and you did neither one.

Looks like between 10 and 50ms to me, but you'd actually have to do the experiment right to figure it out.

 

[fferreres]

I am NOT saying that cameras have A or B or C. I am saying, I would NOT dismiss people that say they are affected. I'd give them the benefit of the doubt,

I agree. In fact I don't even think there is any doubt. There is a lag in EVF's. In reality it is small but for some it would be unacceptable. That was what I was trying to emphasise in an earlier post when I wrote "Different? Absolutely. I believe we are all comfortable with what we are comfortable with."

Yes, I agree.

I am comfortable with whatever the lag is on my RX10 IV and I get the captures I desire. Others would not be satisfied: all good.

All good for me too. One concern, unrelated to this Jay is...do you want to see the real light? Or the sensor interpretation? I am ok with the sensor+EVF interpretation. It's unrelated. With EVF you are working mostly with the outcome. With DSLR you are working with the source.

Where I start getting concerned is when someone makes a blanket statement such as "Since EVFs force a 3x reduction in focal length for the same subject, I'd need 9x more pixels to cover the same pixels-on-target." (Direct quote from Lee Jay's post HERE). I consider that a broad generalisation and a significant exaggeration. There are countless examples that disprove that. I personally have literally thousands of images similar to the ones accompanying the above statement, all of them in a burst sequence showing 5 - 10 well framed aircraft.

If there could be fines for misleading so grossly. I can see the uproar. A bad seed an social media and these grow like mushroom in the mind. It is a problem.

But I am NOT saying everybody can comfortably do that or be happy with the EVF performance. I am, others might not be.

agree that having benchmarks of EVF performance, that these need to measure 1) what are the ms delay 2) how is the variance in demanding scenarios (AF-C or other cases) 3) How many frame and to what degree show a frame that is out of sync with the correct timeline.

I agree it would be good to have a standard measurement. Where I see a problem is the practical interpretation of the data. How much out of sync is OK and how much isn't? My own experience is that my own reaction time and co-ordination / fine motor skills to keep a fast moving subject in the frame is by far the biggest limiting factor, not a 10 to 50ms EVF delay. Aeroplanes and decent sized birds I have no problem with using an EVF. The crazy jinks of a small swallow? No chance in a million years. And I have grave doubts whether having a lagless OVF would do much to materially improve that: I am just not good enough for that. And I am not sure how many people ARE good enough to manage that, regardless equipment.

Sadly, nothing is ever simple.

I think this will sort itself out, as Sports shooter will chose cameras with systems that don't degrade framerate or lock processes (out of timing frame). For example, I don't know if OMD EM1X with the two separate cores helps. Remember when computers had 1 core. And they also share many other resources (cache, etc).

For me, right now, EVF has so many additional benefits. The on screen aids (peaking, cues, magnify, etc) are all things that far far faaaar outweight anything else. And this is going to continue to increase.

 

[robgendreau]

I am NOT saying that cameras have A or B or C. I am saying, I would NOT dismiss people that say they are affected. I'd give them the benefit of the doubt,

I agree. In fact I don't even think there is any doubt. There is a lag in EVF's. In reality it is small but for some it would be unacceptable. That was what I was trying to emphasise in an earlier post when I wrote "Different? Absolutely. I believe we are all comfortable with what we are comfortable with."

Yes, I agree.

I am comfortable with whatever the lag is on my RX10 IV and I get the captures I desire. Others would not be satisfied: all good.

All good for me too. One concern, unrelated to this Jay is...do you want to see the real light? Or the sensor interpretation? I am ok with the sensor+EVF interpretation. It's unrelated. With EVF you are working mostly with the outcome. With DSLR you are working with the source.

Where I start getting concerned is when someone makes a blanket statement such as "Since EVFs force a 3x reduction in focal length for the same subject, I'd need 9x more pixels to cover the same pixels-on-target." (Direct quote from Lee Jay's post HERE). I consider that a broad generalisation and a significant exaggeration. There are countless examples that disprove that. I personally have literally thousands of images similar to the ones accompanying the above statement, all of them in a burst sequence showing 5 - 10 well framed aircraft.

If there could be fines for misleading so grossly. I can see the uproar. A bad seed an social media and these grow like mushroom in the mind. It is a problem.

But I am NOT saying everybody can comfortably do that or be happy with the EVF performance. I am, others might not be.

agree that having benchmarks of EVF performance, that these need to measure 1) what are the ms delay 2) how is the variance in demanding scenarios (AF-C or other cases) 3) How many frame and to what degree show a frame that is out of sync with the correct timeline.

I agree it would be good to have a standard measurement. Where I see a problem is the practical interpretation of the data. How much out of sync is OK and how much isn't? My own experience is that my own reaction time and co-ordination / fine motor skills to keep a fast moving subject in the frame is by far the biggest limiting factor, not a 10 to 50ms EVF delay. Aeroplanes and decent sized birds I have no problem with using an EVF. The crazy jinks of a small swallow? No chance in a million years. And I have grave doubts whether having a lagless OVF would do much to materially improve that: I am just not good enough for that. And I am not sure how many people ARE good enough to manage that, regardless equipment.

Sadly, nothing is ever simple.

I think this will sort itself out, as Sports shooter will chose cameras with systems that don't degrade framerate or lock processes (out of timing frame). For example, I don't know if OMD EM1X with the two separate cores helps. Remember when computers had 1 core. And they also share many other resources (cache, etc).

For me, right now, EVF has so many additional benefits. The on screen aids (peaking, cues, magnify, etc) are all things that far far faaaar outweight anything else. And this is going to continue to increase.

Nice response. Most of this discussion is angels dancing on pin and whether they are .01 seconds behind or not. For most of us it doesn't matter (if it can matter) and we choose EVF or OVF, or rear display vs OVF, based on many very much more important factors, like those you mention.

Things like low light, video, and even edge cases like infrared. I use both EVF and OVF on all kinds of subjects, fast and static, and can't say that any supposed shutter lag has played a role in that decision in years. If there's a lag in hummingbird wings I can't discern it, and I'm far more interested in whether the camera can snap bursts fast enough to grab what I want. Or whether say the video frame rate can be fast enough to catch the action so I can do slo mo. As with many situations with photography, WYS isn't WYG necessarily; fast action is just one example. Maybe those of us who grew up with film and rangefinders learned this more easily, I don't know. But for me the human eye is more often a bottleneck than the VFs.

 

[AlwynS]

But as much time as you spend here arguing that you have the best camera, biggest yacht, and how DSLRs can't take pictures... there is no way you have any.

It was Lee Jay claiming mirrorless cameras can't take photos of moving objects because of EVF lag.

And taking photos of a timer on a computer monitor sure put that to rest.....

Regardless, your comment has absolutely nothing to do with what I said.

Nowhere in this thread did the OP claim DSLRs can't take pictures. That is just something you made up.

This thread is about Lee Jay's claim that mirrorless cameras can't take photos of moving objects because of EVF lag...

Th OP never mentioned Lee Jay.

That is just something you made up.

Lee Jay's claim sparked a number of threads, including this one. If you follow the sequence, it is clear, even though he wasn't mentioned by name in the OP.

Lee Jay's claim was also based on entirely false premises, but that is another story.

What you and Kiwi have said about "Lee Jay's claim," is deliberately disingenuous and I think you probably know it. Lee Jay and others have addressed the issue of EVF lag causing problems in keeping fast moving subjects appropriately in the image frame. That is especially true when the fast moving subjects also change direction drapidly and unpredictably. No one denies the ability of EVF cameras to pan and follow moderately moving subjects with predictable trajectories.

Creating a disingenuous strawman argument does this discussion no good to anyone--not even the ML fanboys. It also makes you seem like you only want to engage in argumentation instead of addressing the actual issue of lag. That's also how I interpret your continual conflation of EVF lag with the very different effects of shutter lag, mirror blackout and other issues unrelated to tracking.

If one has a camera with an EVF, it helps one to keep the image within the image field to have a full frame sensor and to use a moderate focal length lens. That is simply overpowering the tracking problem by giving you a smaller image in a wider field but what we're really addressing here is tracking with otherwise similar framing not using different sized frames to assist a viewfinder lag.

Please understand that I do not wish to be argumentative. But I would appreciate your comments on the resolution and framing of the aircraft, birds and dragonfly images I posted.

I believe these cover most of the fast moving objects the vast majority of us wish to photograph.

No, I don't think they do cover the issue of fast moving or unpredictable. Gliding birds I've already commented on. They're easy. A taking off prop airplane is predictable, slow and easy to pan. Biplanes? Fast? Come on. A red-tailed hawk taking off is easy as is a red-tailed hawk cruising like the other birds you show. The osprey? It was nearly stationary just before your shot. That's not fast moving and unpredictable. I don't know what your dragon fly was doing but they hover a whole lot and that is easy to shoot.

Whatever.

I don't mean to be argumentative but I covered the issue of steady, predictable movement in a previous post. No, I think your images are fine and I have thousands of images of birds flying taken with my Canon SX50 but not doing the wild and crazy things they do in the more interesting situations.

I truly do not consider the EVF lag (which I readily accept is real, if small) to be a material issue for these subjects. Note also that I use (admittedly kind of “best in class” but still) relatively inexpensive bridge-style cameras and not the best mirrorless cameras.

I would suggest the results I get would be more than acceptable in by far the most common applications. In which case I would also suggest that, while real, the lag is sufficiently small to be irrelevant in all but highly specialised, esoteric applications.

Am I missing something?

Yes.

We differ philosphically: it happens.

And FWIW: the reason why I posted the images of prop planes and birds was in response to someone else's comment on aeroplanes against a background of trees.

--
Cheers
Alwyn

 

[fferreres]

But as much time as you spend here arguing that you have the best camera, biggest yacht, and how DSLRs can't take pictures... there is no way you have any.

It was Lee Jay claiming mirrorless cameras can't take photos of moving objects because of EVF lag.

And taking photos of a timer on a computer monitor sure put that to rest.....

Regardless, your comment has absolutely nothing to do with what I said.

Nowhere in this thread did the OP claim DSLRs can't take pictures. That is just something you made up.

This thread is about Lee Jay's claim that mirrorless cameras can't take photos of moving objects because of EVF lag...

Th OP never mentioned Lee Jay.

That is just something you made up.

Lee Jay's claim sparked a number of threads, including this one. If you follow the sequence, it is clear, even though he wasn't mentioned by name in the OP.

Lee Jay's claim was also based on entirely false premises, but that is another story.

What you and Kiwi have said about "Lee Jay's claim," is deliberately disingenuous and I think you probably know it. Lee Jay and others have addressed the issue of EVF lag causing problems in keeping fast moving subjects appropriately in the image frame. That is especially true when the fast moving subjects also change direction drapidly and unpredictably. No one denies the ability of EVF cameras to pan and follow moderately moving subjects with predictable trajectories.

I can see why, as it'd be the same case as with fast paced online games. Timing of frames, dropped frames, etc. may completely confuse our own tracking (for good reason). This may also be case in sports. I think it may represent 0.01% of cases (that level of unpredictable speed when having the button pressed for a while in AF-C) and different cameras will have different effects, so the problem is hen there is a broad generalization.

Creating a disingenuous strawman argument does this discussion no good to anyone--not even the ML fanboys. It also makes you seem like you only want to engage in argumentation instead of addressing the actual issue of lag. That's also how I interpret your continual conflation of EVF lag with the very different effects of shutter lag, mirror blackout and other issues unrelated to tracking.

We lack the benchmarks. The problem is when he says he needs 9x the pixels, or when he is generalizing from one camera to the entire world of EVF cameras.

If one has a camera with an EVF, it helps one to keep the image within the image field to have a full frame sensor and to use a moderate focal length lens. That is simply overpowering the tracking problem by giving you a smaller image in a wider field but what we're really addressing here is tracking with otherwise similar framing not using different sized frames to assist a viewfinder lag.

I think unless you enrich your understanding of lag, it creates flame wars. The 5ms lag of many modern EVF systems is not enough to cause anything. It's too short.

1) the lag needs to be small

2) the frame updates need to be as consistent (dropped frames)

3) The timing needs to be perfect (out of time)

The reading of a sensor 18 times per second at full resolution, etc. can quickly bottleneck the process, in addition to introducing blackout times, and can also cause delays. The same problems arise with DSLRs. When the photo is being taken, the mirror is on the way. if you take 18 shots in a second, the sensor needs to be read, but this is alleviated as the light is routed through the mirror and prism, even if the camera core processing it is saturated. Ultimately, the EVF will WIN because it will be able to continue to show the image with incredibly low delay 100% of the time. But, this is not easy when the readouts create locking or delays or unresponsiveness. For some people, the mirror may be a big issue too. But right now, I understand if the system is not designed for these cases, the processing of the information may be affecting EVF. And I think anyone not taking AF-C full res 15 or 30 shoots, may not notice.

The other aspect is that people that trained their targeting with OVF do have to adjust a little. They will need tocope with 15ms, or 4ms or X ms delay. All the tracking you do is not conscious. Has been automated. So it's normal if at the beggining you are off by some 50 pixels or so. When I played games, I adjusted to the lag, my brain would for example aim 4 pixels to the right or left, knowing there is some slight delay. Literally, if you are always looking 15ms into the past, in a highly dynamic shooting, you will be off certain amount of pixels IF you have trained yourself only on 0ms delay. But for anyone shooting with EVF at 15ms, they will always get it mostly right (except when the subject takes unexpected turn, where you will have, if you are used to OVF and are now trying EVF:

Thing travels fast to right, you track. You are 15ms behind without knowing.

Subject abruptly changes direction. After 7 ms subject is well aimed. Now you depend on the refresh to know it turned. May not show until 15 ms later. Now you are ˜23 ms apart, and only now you notice it...you still haven't even started processing, and may need a few more frames. Plus, add reaction time? 20ms? You are still going to the right for 20 ms more time. So the math may be off and not perfect, it's an example, when things compound in this way, you may well have an object that traveled 30ms left while you traveled 30 ms right. Now, you will get all the shots, ut the framing may be affected a lot more than with OVF. Especially, if the frame rate changes, the camera is not very good in AF-C, any timing issue, and higher baseline ms delay. The first thing that HELPS is try different cameras. The second is that, yes, you need to adjust to the delay. If you know it's 20ms, you want to always be 20ms AHEAD in your aim. And when you see the object change the direction, you need to OVERCOMPENSATE at first. This is how it works when you chance something with a little lag. And at this level of FL and kind of photography, and when you factor in our reaction time and other things, it's easy to blame de EVF when in reality it was already hard to get any good framing in these cases. But if you do this for a living, you can see that maybe a lot more shots are a little more off in the framing. And now there's the temptation to always blame the camera as well.

But the problem comes that this experience is not shared... This is an extreme case, and those not being listened to then feel the need to generalize from one camera to another, or make grossly overstated statements that just annoy the entire universe.

 

[mamallama]

Your images are fine. I have never said mirrorless doesn't make great images. My point is how hard it is to get those images in the smeary laggy electronic viewfinder. My hats off to your skill.

Not as hard as those who want to keep making EVF lag a whipping boy like to think it is.

 

[Bob A L]

It was for me - I put up my dslr's for an entire year trying to like mirrorless before finally giving up in frustration. And finally got the mirrorless sold just two weeks ago after setting in a drawer for the past year or so. I don't generally sell my old gear, but it had to go. Maybe they work fine for you - it' a highly individual thing - have fun with yours, but I won't be trying again till I can't find a dslr to use for my main machine. Suppose I am not completely right here though, my second most used camera actually has a evf, that I absolutely hate, but until they come out with a dslr with a 1/2.3 sensor it'll have to work.

 

[mamallama]

It was for me - I put up my dslr's for an entire year trying to like mirrorless before finally giving up in frustration. And finally got the mirrorless sold just two weeks ago after setting in a drawer for the past year or so. I don't generally sell my old gear, but it had to go. Maybe they work fine for you - it' a highly individual thing - have fun with yours, but I won't be trying again till I can't find a dslr to use.

If the lag on your EVF is too much for the type of shooting you do there are workarounds like an external optical viewer. Optical viewers are not rocket science and they are inexpensive. That is what my friend attachs to his GH5 when he goes out birding. Also there is pre-burst shooting he uses in challenging situations.

 

[fferreres]

It was for me - I put up my dslr's for an entire year trying to like mirrorless before finally giving up in frustration. And finally got the mirrorless sold just two weeks ago after setting in a drawer for the past year or so. I don't generally sell my old gear, but it had to go. Maybe they work fine for you - it' a highly individual thing - have fun with yours, but I won't be trying again till I can't find a dslr to use.

I can't focus a manual lens at 200mm precisely enough on an OVF that has no magnification at all. Actually, the higher the resolution one can achieve, the more the OVF is inadequate, and the more you need to blindly trust some AF algorithm, that when I test my hand vs the AF apparatus, I always seem to know better what is better focused and what I wanted that the random algorithms in the AF system.

This creates a simple rule that for static subjects, EVF are BETTER suited, where extreme high speed post or bird chasers, the OVF may be need, since in these last cases I anyway need to rely on AF, but on the former, I not only cannot rely on AF but also, cannot rely on my bare eyes at the null magnification of the OVF.

 

[Jonsi]

It was for me - I put up my dslr's for an entire year trying to like mirrorless before finally giving up in frustration. And finally got the mirrorless sold just two weeks ago after setting in a drawer for the past year or so. I don't generally sell my old gear, but it had to go. Maybe they work fine for you - it' a highly individual thing - have fun with yours, but I won't be trying again till I can't find a dslr to use.

If the lag on your EVF is too much for the type of shooting you do there are workarounds

'Cuz everyone wants a camera they need "workarounds" to use.