Here's a pretty deep dive by Tony Northrup on the specs of the 5D4. He's noted quite a few legit shortcomings of the camera, both in video and stills, including a few that I don't necessarily agree with. But still, it's a bit sobering to see all the limitations he's noted.
Wedding/Event and QuadCopter video shooters have used Mercalli for years to help with these issues. Best to rent the camera prior to purchase and process for yourself.
There is no dispute that the Canon 1D X Mark II is a better tool for frame grabs than is the 5D Mark IV, but the 1D X Mark II also crops the sensor for 4k video. (Approx 1.3x crop factor.)
I would be interested to see if the frame grab could be used to easily catch a baseball compressed on a bat.
Yes, I would probably wish to use a 4k video camera with around 240 fps and a global shutter for that particular situation. For many other, more common, situations, frame grabs from a 5D Mark IV might be quite nice.
I don't see how it could. You typically use shutterspeeds at double your frame rates to get smooth video. So how much action can you really freeze at 1/60th or even 1/120th? I hope I'm missing something here, but I don't see this working.
What's stopping you from shooting video with a shutter speed of 1/8000?
-Jack
An "accurate" reproduction of a scene and a good photograph are often two different things.
What's stopping you from shooting video with a shutter speed of 1/8000?
The rule of thumb is double your frame rate for best results. So 1/50th or 1/60th is pretty much the standard. Supposedly the video has a choppy feel when you exceed that. It makes sense if you think about it because you're capturing frozen milliseconds of time of much longer intervals, and they're sharp. So you're missing most of the video motion. I usually stick to the recommendations, although it would be interesting to play around with it. Different content might react differently to shutterspeeds.
The rule of thumb is double your frame rate for best results. So 1/50th or 1/60th is pretty much the standard. Supposedly the video has a choppy feel when you exceed that. It makes sense if you think about it because you're capturing frozen milliseconds of time of much longer intervals, and they're sharp. So you're missing most of the video motion. I usually stick to the recommendations, although it would be interesting to play around with it. Different content might react differently to shutterspeeds.
While it's true that you generally choose a video shutter speed of 1/2xFrame-Rate because of the 180 degree rule which related to the rotary shutters used in film movie cameras, it's a bit more freeing with today's video technology. Since there is no mechanical rotary shutter in the optical path, Video dSLR manufacturers allow much greater freedom for visual effects.
Want something more "dreamy" from your video captures? Let the shutter speed equal 1/Frame-Rate. The resulting motion blur frame to frame with look very smooth, to the point that most people don't like it for long. Still, there are times when it can work. Some cameras even use the capture buffer to allow very low shutter speeds, mostly because it allows more light to be captured when the available light is less than normal shutter speeds allow (at the current aperture and ISO). In this case the resulting video appears rather "ghostly" and slurred.
Going the other way with a very short shutter duration, the video takes on a more "staccato" effect and the frames feel more detached from each other.
... However, when you shoot video with the intent of extracting individual frames a shorter shutter speed for each frame is generally preferable, lighting and exposure permitting. (If you later want some motion blur introduced there are ways to do that in post-production so that the video can look more like "normal" video.)
Feel free to experiment and set the video capture shutter speeds according to your "artistic intent".
Ok, so for the purposes of frame grab, you could use a faster shutter speed. As for the rule of thumb of 1/2xframe rate, I dunno, seems slow, but I know nothing about video. Here's a little video I did recently of my kids and friends at a summer ski training camp at Lake Placid. I had the camera in aperture priority at I think f/8, ISO 200. It was a bright sunny day so the shutter speeds were way faster than that rule of thumb, like 1/640. Seems fine for a "home movie" to me.
"Compared to its predecessor, the 5D Mark III, this is a massive improvement. With the Mark III, one could not simply underexpose to protect highlights without paying a significant noise cost when boosting shadows in post, meaning you had to make your choices about which tones you wished to presere on the spot, at the time of the exposure. Even compared to the 5DS, the Mark IV shows vast improvements. It even slightly edges out the 1D X II. In a nutshell, the 5D Mark IV will be far more capable at dealing with high contrast scenes without the need for workarounds."
"The Canon 5D Mark IV, like the 1D X II, represents a significant step forward for Canon with respect to dynamic range. We can confidently say these cameras bring the massive improvement many Canon shooters had wished for in this department. The increased exposure latitude, and ability to shoot in an (almost) ISO-invariant manner frees up the photographer to make unconventional exposure decisions in challenging scenes to retain more tones than might have previously been possible. Ironically, this may have more of an impact on fast-paced photography than landscapes, the latter affording more time to work around camera limitations using techniques like filters, and bracketing."
Yeah, that's great, although honestly it is rare that I feel limited by my 5D3's DR. Too bad the 5DS/R just missed out on this technology, that's my next camera.
-Jack
An "accurate" reproduction of a scene and a good photograph are often two different things.
I read that DPR review about its DR and quite frankly, the Nikon D810 still blows it out of the water -- a 3 year old camera.
I wouldn't say blows it out of the water. Anyway, needing to push shadows more than 3 stops is only due to user error.
As for Tony Northrup, I'm not going to watch his ego for 35 minutes when he should have just typed it out, but I've read that he hasn't even touched the camera yet.
-Jack
An "accurate" reproduction of a scene and a good photograph are often two different things.
I read that DPR review about its DR and quite frankly, the Nikon D810 still blows it out of the water -- a 3 year old camera.
Tell DPR to compare the Nikon D5, the current flagship in the Nikon line, with the Canon 5D Mark IV. (I have absolutely nothing against the Nikon D5 and it really does shine in many ways, but the DPR testing and assessment seems a bit biased against Canon.)
Also remind DPReview that the Canon 5D Mark IV is not even in production yet (they are reviewing a pre-production body) and that they are using a "beta" version of ACR. (No finalized version of ACR which supports 5D Mark IV Raw files exists.)
I wouldn't say blows it out of the water. Anyway, needing to push shadows more than 3 stops is only due to user error.
As for Tony Northrup, I'm not going to watch his ego for 35 minutes when he should have just typed it out, but I've read that he hasn't even touched the camera yet.
Not always, with my D750 I took a photo of a couple in the shade with a brightly lit background and I was able to expose for that and then bring up the shadows no problem.
Previously I would have needed a fill flash.
I'm really happy that Canon is stepping up their game in this respect with this camera and the 80D, it really transforms how you shoot. Using that DPReview tool it looks like the 5D MkIV is closer to the D7200 than the D810/750 but that's not to trash the camera...I've honestly been shocked at what the D7200 can do. And having worked with Canon files and their base ISO noise this is HUGE at least in my opinion.
Yes, I've used Magic Lantern on my Canon 5D mark II, 7D and 5D Mark III with no real issues.
Yes, you do have to be careful and follow every step for installation with precision.
Sorry, I don't have anything worthy of showing which required Magic Lantern Dual ISO, but others do have pretty incredible stuff to show (using a Canon 5D series body with Magic Lantern, which "may" include Dual ISO).
"Frames of Life" was shot on a Canon 5D Mk III RAW mode with Magic Lantern (exclusively).
To get the discussion back to the new Canon 5D Mark IV, the new camera feature of HDR Video on the 5D Mark IV was directly inspired by the (now) nearly 5-year-old Magic Lantern HDR Video feature. The difference is that the ML version only worked at 15 fps (output) and was a pain to use.
It appears that the Canon 5D Mark IV acquires video at 60 fps in HDR Video mode and then combines two adjoining frames, each captured at a different ISO, producing 30 frames-per-second 1080P HDR Video "in the camera". I guarantee you that this has never been done before in a dSLR.
We will have to wait to see the resulting quality because I can't find any demo videos using the 5D Mark IV and the HDR Video yet.
Check out the Canon USA channel on YouTube for some early demo videos, starting with: https://youtu.be/nYzp5sGqMVw
Tell DPR to compare the Nikon D5, the current flagship in the Nikon line, with the Canon 5D Mark IV. (I have absolutely nothing against the Nikon D5 and it really does shine in many ways, but the DPR testing and assessment seems a bit biased against Canon.)
Also remind DPReview that the Canon 5D Mark IV is not even in production yet (they are reviewing a pre-production body) and that they are using a "beta" version of ACR. (No finalized version of ACR which supports 5D Mark IV Raw files exists.)
Spoiler alert:
I also use Magic Lantern for some features
the 5D4 looks like a good upgrade (not that I'm in the market)
Ultimately it's about the image quality and that's why most pros (and 'enthusiasts' !) shoot with Canon
I'm extremely proud of Canon's AF section in the higher tier of bodies (enthusiast, semi-pro and professional). Competent AF is what separates the better and best dSLRs from any manufacturer. The latest Canon USM lenses partner with the bodies to produce an even better result.
Nikon dSLRs are at a similar level of AF competence and the majority of professional photographers shoot with one of those two brands.
Now that Canon is making the effort to compete in base-ISO dynamic range (1D X Mark II, 80D and now the 5D Mark IV), that part of the image quality puzzle should make "staying with one brand" easier for most shooters.
Tell DPR to compare the Nikon D5, the current flagship in the Nikon line, with the Canon 5D Mark IV. (I have absolutely nothing against the Nikon D5 and it really does shine in many ways, but the DPR testing and assessment seems a bit biased against Canon.)
Also remind DPReview that the Canon 5D Mark IV is not even in production yet (they are reviewing a pre-production body) and that they are using a "beta" version of ACR. (No finalized version of ACR which supports 5D Mark IV Raw files exists.)
Spoiler alert:
In my reading they were rightfully critical of the D5. Basically it's better at high ISO and the AF is a big improvement but the low ISO dynamic range took a hit compared to the D4s which already was a step behind the D750/D810. Using the D7200 as a benchmark again it clearly beats the D5 for shadow recovery at base ISO. And they also said the 1Dx II is superior to the D5 in this respect in their review:
The Canon EOS-1D X Mark II shows very similar amounts of noise to the excellent sensor in the Sony a7R II up until a 3EV push, with the Canon dropping behind after a 4EV push. It's a similar story against the likes of the Nikon D750 or D810. This means that the darker shadows in a processed image would be slightly cleaner in images from these cameras, after contrast adjustments or a less extreme push.
However, this performance is noticeably better than the Canon EOS 5DS R and, significantly, better than the 1D X II's most direct rival: the Nikon D5.
I'm extremely proud of Canon's AF section in the higher tier of bodies (enthusiast, semi-pro and professional). Competent AF is what separates the better and best dSLRs from any manufacturer. The latest Canon USM lenses partner with the bodies to produce an even better result.
Nikon dSLRs are at a similar level of AF competence and the majority of professional photographers shoot with one of those two brands.
.
right, Canon is pretty much blowing away the competition in the "upper tiers' and middle tier cameras. The lower tier cameras ... the Rebels and mirrorless are holding their own. Nikon, Sony and Olympus are not far behind
Not always, with my D750 I took a photo of a couple in the shade with a brightly lit background and I was able to expose for that and then bring up the shadows no problem.
Previously I would have needed a fill flash.
In Canon land at least, it would be better to expose for the people and then bring the highlights down. Or maybe expose for a compromise between the people and the background, more towards the people, and then adjust to taste. Exposing for the highlights and then bringing up the shadows in post is the worst option. I have to assume this is probably true for Nikon/Sony sensors too, but people seem to like to do that just because they can. I'm not sure they actually should.
At any rate, the DR issue with Canon is there, but it's overblown. Most often you have to view at 100% to see it, and that's only relevant when you want to print big. I was shooting high school football on Friday and these two players grabbed me for a shot before the game. I wasn't expecting that, and I was in aperture priority, and my 5D3 underexposed as it notoriously does whenever bright sky is involved. They were basically silhouettes before processing:
I should work on the kid-on-the-right's face with a brush some more, but you get the idea.
-Jack
An "accurate" reproduction of a scene and a good photograph are often two different things.
In Canon land at least, it would be better to expose for the people and then bring the highlights down. Or maybe expose for a compromise between the people and the background, more towards the people, and then adjust to taste. Exposing for the highlights and then bringing up the shadows in post is the worst option. I have to assume this is probably true for Nikon/Sony sensors too, but people seem to like to do that just because they can. I'm not sure they actually should.
At any rate, the DR issue with Canon is there, but it's overblown. Most often you have to view at 100% to see it, and that's only relevant when you want to print big. I was shooting high school football on Friday and these two players grabbed me for a shot before the game. I wasn't expecting that, and I was in aperture priority, and my 5D3 underexposed as it notoriously does whenever bright sky is involved. They were basically silhouettes before processing:
I should work on the kid-on-the-right's face with a brush some more, but you get the idea.
Oh you'd still bring the highlights down some, but you'd do it within reason not like my old Fuji S5 where you'd push them big time knowing that's where your extra DR is. With the Sony sensors you can bring the shadows up without much of an issue, ISO invariance is a game changer in many ways. Basically what it means is that you can shoot something at ISO 100 and ISO 800 with the same aperture/shutter speed and when the ISO 100 image is brought up it's almost identical.
If you check out the link Ziggy posted you can do it for 2 stops no problem with the 5D MkIV and 3 stops isn't too big of an issue and that's going to ISO 6400:
But with a Sony/Nikon sensor, is it still better to expose for the midtones and recover highlights and shadows, than to expose for the highlights and just lift shadows? I am not knowing.
-Jack
An "accurate" reproduction of a scene and a good photograph are often two different things.
Now that Canon is making the effort to compete in base-ISO dynamic range (1D X Mark II, 80D and now the 5D Mark IV), that part of the image quality puzzle should make "staying with one brand" easier for most shooters.
It's usually the amount of lenses that keep people locked into one manufacturer for the long haul.
Sure there are some nitpicky pixel peeping photographers out there that may flip flop on the incremental upgrades, but I still don't see anything with this latest version that will cause long time users to switch or upgrade based solely on the changes over the last 5d.
In the studio mirrorless is a true game changer and I haven't seen anything like that here.
I can see the argument for the upgrade in video, but for those who are heavy into that, a dedicated video system would make more sense.
I still see this more as jumping in point for those with older consumer cameras looking to upgrade into a more serious system. I just don't see enough here for major Nikon users or current 1d/5d users to run out and buy this one.
But with a Sony/Nikon sensor, is it still better to expose for the midtones and recover highlights and shadows, than to expose for the highlights and just lift shadows? I am not knowing.
It is situation dependent, at or near base ISO exposing for the highlights and underexposing the midtones roughly 2-3 stops doesn't matter but you take more of a hit as you go from there. Those areas that are lifted essentially get higher effective ISOs. So for example the midtones at ISO 800 lifted 2 stops look like they came from an ISO 3200 shot which could become an problem depending on your final output so it might be better there to deal with the blown out highlights VS shooting this way.
JMP so I've never done a head to head comparison so this morning to procrastinate from some design work I did a quick one. I only have a Canon G1X on hand but it's a 7D essentially and while more pronounced has the same dynamic range general characteristics as Canon DSLR's until their most recent ones. For the other camera I used the D7200 because it's ability to push images looks to be closest to the 5D mkIV. And these are at F2.8 so forgive the softness, also no noise reduction was used for any of these shots and the first 4 are 100% crops:
D7200 @ ISO 100:
D7200 @ ISO 100 shot 3 stops faster shutter speed and pushed in the shadows:
G1X @ ISO 100:
G1X @ ISO 100 shot 3 stops faster shutter speed and pushed in the shadows:
One other thing this test showed is how much dynamic range these sensors have compared to new ones. The only editing I did was a little contrast, but to get that to look the same it took over twice the amount on the D7200 images. Since upgrading to the 750/7200 there have been many times I have more dynamic range than I want for an image, though that's a good problem to have and the more I read up on the 5D mkIV I'm ecstatic that Canon has moved into a similar space.
Here's what the original pushed D7200 file looked like for context:
Not really a fair fight. The G1X and 7D sensors are quite old now and even then the 7D was much worse than the 5D2, I had both.
Oh I know, it's just all I got and while it's worse than the 5D II & III they share the same general characteristic just to a lesser extent. But the D7200 shots are the important part here since they show you roughly how you'll be able to shoot with the 5D mkIV and have relatively little penalty in terms of shadow noise.
And I have one of those old, old 60Ds
But I expose for the faces and "lift the shadows" and the image always seems better than the other brands ! ...
Quality skin tone rendition has long been a trait of Canon cameras, but the latest Nikon bodies do compare extremely well. (My Nikon D7100 and D7200 compare very well with my Canon 5D Mark III, 5D Mark II and 7D, for instance. I use Capture One 9.1 for RAW processing and that may explain some of the similarities. [... Compared with ACR, which required more user intervention, IME.])
Other colors and tones are more than acceptable, and lens choice often plays a greater role in subtle color tinting than the choice of body.
What's stopping you from shooting video with a shutter speed of 1/8000?
To get back to this question, it appears that the EOS 5D Mark IV will do 1/8000th in still mode, and up to 1/4000th in video mode. Still pretty fast for a video shutter speed. Unsure whether that 1/4000th is available in all video modes.
Comments
Link to my Smugmug site
Extremely fast action plus fast panning the camera can indeed ruin video footage used for still grabs.
Not quite a miracle software, this video demonstrates some considerable "jello and skew" software remediation: CMOS Rolling Shutter Gone with Mercalli v4
Wedding/Event and QuadCopter video shooters have used Mercalli for years to help with these issues. Best to rent the camera prior to purchase and process for yourself.
There is no dispute that the Canon 1D X Mark II is a better tool for frame grabs than is the 5D Mark IV, but the 1D X Mark II also crops the sensor for 4k video. (Approx 1.3x crop factor.)
Yes, I would probably wish to use a 4k video camera with around 240 fps and a global shutter for that particular situation. For many other, more common, situations, frame grabs from a 5D Mark IV might be quite nice.
Moderator of the Cameras and Accessories forums
What's stopping you from shooting video with a shutter speed of 1/8000?
An "accurate" reproduction of a scene and a good photograph are often two different things.
Link to my Smugmug site
While it's true that you generally choose a video shutter speed of 1/2xFrame-Rate because of the 180 degree rule which related to the rotary shutters used in film movie cameras, it's a bit more freeing with today's video technology. Since there is no mechanical rotary shutter in the optical path, Video dSLR manufacturers allow much greater freedom for visual effects.
Want something more "dreamy" from your video captures? Let the shutter speed equal 1/Frame-Rate. The resulting motion blur frame to frame with look very smooth, to the point that most people don't like it for long. Still, there are times when it can work. Some cameras even use the capture buffer to allow very low shutter speeds, mostly because it allows more light to be captured when the available light is less than normal shutter speeds allow (at the current aperture and ISO). In this case the resulting video appears rather "ghostly" and slurred.
Going the other way with a very short shutter duration, the video takes on a more "staccato" effect and the frames feel more detached from each other.
... However, when you shoot video with the intent of extracting individual frames a shorter shutter speed for each frame is generally preferable, lighting and exposure permitting. (If you later want some motion blur introduced there are ways to do that in post-production so that the video can look more like "normal" video.)
Feel free to experiment and set the video capture shutter speeds according to your "artistic intent".
https://vimeo.com/blog/post/frame-rate-vs-shutter-speed-setting-the-record-str
Moderator of the Cameras and Accessories forums
http://www.jmphotocraft.com/Sports/Skiing-and-Snowboarding/Water-Ramps-2016/i-dkkL63m/A
An "accurate" reproduction of a scene and a good photograph are often two different things.
"Compared to its predecessor, the 5D Mark III, this is a massive improvement. With the Mark III, one could not simply underexpose to protect highlights without paying a significant noise cost when boosting shadows in post, meaning you had to make your choices about which tones you wished to presere on the spot, at the time of the exposure. Even compared to the 5DS, the Mark IV shows vast improvements. It even slightly edges out the 1D X II. In a nutshell, the 5D Mark IV will be far more capable at dealing with high contrast scenes without the need for workarounds."
"The Canon 5D Mark IV, like the 1D X II, represents a significant step forward for Canon with respect to dynamic range. We can confidently say these cameras bring the massive improvement many Canon shooters had wished for in this department. The increased exposure latitude, and ability to shoot in an (almost) ISO-invariant manner frees up the photographer to make unconventional exposure decisions in challenging scenes to retain more tones than might have previously been possible. Ironically, this may have more of an impact on fast-paced photography than landscapes, the latter affording more time to work around camera limitations using techniques like filters, and bracketing."
Moderator of the Cameras and Accessories forums
An "accurate" reproduction of a scene and a good photograph are often two different things.
"You miss 100% of the shots you don't take" - Wayne Gretzky
I wouldn't say blows it out of the water. Anyway, needing to push shadows more than 3 stops is only due to user error.
As for Tony Northrup, I'm not going to watch his ego for 35 minutes when he should have just typed it out, but I've read that he hasn't even touched the camera yet.
An "accurate" reproduction of a scene and a good photograph are often two different things.
Tell DPR to compare the Nikon D5, the current flagship in the Nikon line, with the Canon 5D Mark IV.
Also remind DPReview that the Canon 5D Mark IV is not even in production yet (they are reviewing a pre-production body) and that they are using a "beta" version of ACR. (No finalized version of ACR which supports 5D Mark IV Raw files exists.)
Spoiler alert:
Moderator of the Cameras and Accessories forums
Not always, with my D750 I took a photo of a couple in the shade with a brightly lit background and I was able to expose for that and then bring up the shadows no problem.
Previously I would have needed a fill flash.
I'm really happy that Canon is stepping up their game in this respect with this camera and the 80D, it really transforms how you shoot. Using that DPReview tool it looks like the 5D MkIV is closer to the D7200 than the D810/750 but that's not to trash the camera...I've honestly been shocked at what the D7200 can do. And having worked with Canon files and their base ISO noise this is HUGE at least in my opinion.
I also use Magic Lantern for some features
the 5D4 looks like a good upgrade (not that I'm in the market)
Ultimately it's about the image quality and that's why most pros (and 'enthusiasts' !) shoot with Canon
www.flickr.com/photos/mmirrorless
Nikon dSLRs are at a similar level of AF competence and the majority of professional photographers shoot with one of those two brands.
Now that Canon is making the effort to compete in base-ISO dynamic range (1D X Mark II, 80D and now the 5D Mark IV), that part of the image quality puzzle should make "staying with one brand" easier for most shooters.
Moderator of the Cameras and Accessories forums
In my reading they were rightfully critical of the D5. Basically it's better at high ISO and the AF is a big improvement but the low ISO dynamic range took a hit compared to the D4s which already was a step behind the D750/D810. Using the D7200 as a benchmark again it clearly beats the D5 for shadow recovery at base ISO. And they also said the 1Dx II is superior to the D5 in this respect in their review:
https://www.dpreview.com/reviews/canon-eos-1d-x-mark-ii/8
right, Canon is pretty much blowing away the competition in the "upper tiers' and middle tier cameras. The lower tier cameras ... the Rebels and mirrorless are holding their own. Nikon, Sony and Olympus are not far behind
In Canon land at least, it would be better to expose for the people and then bring the highlights down. Or maybe expose for a compromise between the people and the background, more towards the people, and then adjust to taste. Exposing for the highlights and then bringing up the shadows in post is the worst option. I have to assume this is probably true for Nikon/Sony sensors too, but people seem to like to do that just because they can. I'm not sure they actually should.
At any rate, the DR issue with Canon is there, but it's overblown. Most often you have to view at 100% to see it, and that's only relevant when you want to print big. I was shooting high school football on Friday and these two players grabbed me for a shot before the game. I wasn't expecting that, and I was in aperture priority, and my 5D3 underexposed as it notoriously does whenever bright sky is involved. They were basically silhouettes before processing:
I should work on the kid-on-the-right's face with a brush some more, but you get the idea.
An "accurate" reproduction of a scene and a good photograph are often two different things.
Oh you'd still bring the highlights down some, but you'd do it within reason not like my old Fuji S5 where you'd push them big time knowing that's where your extra DR is. With the Sony sensors you can bring the shadows up without much of an issue, ISO invariance is a game changer in many ways. Basically what it means is that you can shoot something at ISO 100 and ISO 800 with the same aperture/shutter speed and when the ISO 100 image is brought up it's almost identical.
If you check out the link Ziggy posted you can do it for 2 stops no problem with the 5D MkIV and 3 stops isn't too big of an issue and that's going to ISO 6400:
https://www.dpreview.com/news/3229755227/canon-5d-mark-iv-brings-dramatic-dynamic-range-improvements-to-the-5d-line/2
This isn't to say that older Canon's have some huge issue due to the DR difference, just you have to be a bit more careful in your exposure.
An "accurate" reproduction of a scene and a good photograph are often two different things.
It's usually the amount of lenses that keep people locked into one manufacturer for the long haul.
Sure there are some nitpicky pixel peeping photographers out there that may flip flop on the incremental upgrades, but I still don't see anything with this latest version that will cause long time users to switch or upgrade based solely on the changes over the last 5d.
In the studio mirrorless is a true game changer and I haven't seen anything like that here.
I can see the argument for the upgrade in video, but for those who are heavy into that, a dedicated video system would make more sense.
I still see this more as jumping in point for those with older consumer cameras looking to upgrade into a more serious system. I just don't see enough here for major Nikon users or current 1d/5d users to run out and buy this one.
Website
It is situation dependent, at or near base ISO exposing for the highlights and underexposing the midtones roughly 2-3 stops doesn't matter but you take more of a hit as you go from there. Those areas that are lifted essentially get higher effective ISOs. So for example the midtones at ISO 800 lifted 2 stops look like they came from an ISO 3200 shot which could become an problem depending on your final output so it might be better there to deal with the blown out highlights VS shooting this way.
D7200 @ ISO 100:
D7200 @ ISO 100 shot 3 stops faster shutter speed and pushed in the shadows:
G1X @ ISO 100:
G1X @ ISO 100 shot 3 stops faster shutter speed and pushed in the shadows:
One other thing this test showed is how much dynamic range these sensors have compared to new ones. The only editing I did was a little contrast, but to get that to look the same it took over twice the amount on the D7200 images. Since upgrading to the 750/7200 there have been many times I have more dynamic range than I want for an image, though that's a good problem to have and the more I read up on the 5D mkIV I'm ecstatic that Canon has moved into a similar space.
Here's what the original pushed D7200 file looked like for context:
And after the shadows were pushed:
If you want to check out the full resolution files for any of these I put them up at: http://www.jonathanbeckley.com/Other/Iso-Invariance-Test/n-8fq57X/
An "accurate" reproduction of a scene and a good photograph are often two different things.
Oh I know, it's just all I got and while it's worse than the 5D II & III they share the same general characteristic just to a lesser extent. But the D7200 shots are the important part here since they show you roughly how you'll be able to shoot with the 5D mkIV and have relatively little penalty in terms of shadow noise.
And I have one of those old, old 60Ds
But I expose for the faces and "lift the shadows" and the image always seems better than the other brands !
Quality skin tone rendition has long been a trait of Canon cameras, but the latest Nikon bodies do compare extremely well. (My Nikon D7100 and D7200 compare very well with my Canon 5D Mark III, 5D Mark II and 7D, for instance. I use Capture One 9.1 for RAW processing and that may explain some of the similarities. [... Compared with ACR, which required more user intervention, IME.])
Other colors and tones are more than acceptable, and lens choice often plays a greater role in subtle color tinting than the choice of body.
Moderator of the Cameras and Accessories forums
To get back to this question, it appears that the EOS 5D Mark IV will do 1/8000th in still mode, and up to 1/4000th in video mode. Still pretty fast for a video shutter speed. Unsure whether that 1/4000th is available in all video modes.
For a better "visual feel" for what the 5D Mark IV video is like, the Canon authorized video called, "Introducing the Video Features of the Canon EOS 5D Mark IV" shows much.
Moderator of the Cameras and Accessories forums
http://www.digitaltrends.com/photography/dual-pixel-coming-to-lightroom/
Compared to the 1D X Mark II:
https://www.dxomark.com/Cameras/Compare/Side-by-side/Canon-EOS-5D-Mark-IV-versus-Canon--EOS-1D-X-Mark-II___1106_1071
... they look very similar and very similar to Sony/Nikon Dynamic Range at low ISOs too.
Moderator of the Cameras and Accessories forums
Nikon D810 has slightly more than a stop more DR, not that I care.
An "accurate" reproduction of a scene and a good photograph are often two different things.