The Slow Mo Guys do just what their name implies - record slow motion videos. In a new episode they use a Phantom Flex to capture the Canon 7D's mirror box and shutter mechanism being triggered at a variety of shutter speeds (1/4000s, 1/8000s, etc). The footage itself was recorded at 10,000fps, and the result is a look at a shutter in action - something that would otherwise be nearly invisible to the naked eye.
Exactly why the Optical View Finder & the SLR is so completely screwed up in this digital world.
Its a century old "periscope" technology invented for the first submarines and trench warfare in WWI.
It was built for the film era when you couldn't see what the film was gonna see, so you would settle for the next best thing & see what the lens is seeing.
I've been saying this since the Sony A99 that bringing professional quality electronic view finders to the digital sensor is the most important development in photography.
Its about seeing what the sensor is seeing, even before taking the shot. My LCD panel is permanently turned inwards now. I don't chimp like a chimpanzee after each shot anymore because if the EVF looks right, the sensor has captured it right.
The LCD display is opened for waist /floor/overhead shots only!
Look at the unnecessary complexity of the operation! Look at how the mirror & shutter bounces several times before coming to rest! Vibration blur geddit!
Not to take away from the cool video, but I can't find any info that suggests that the Phantom Flex uses film or that anyone would use it to record to digital tape. So I am confused as to how they "filmed" anything or obtained "footage". I would have not imagined on a photographic website that such incorrect capture terms would be used. Especially when there are artsy hipsters re-newing the use of analog equipment. Sorry, but some folks do still film and get some great footage. Thanks for your time.
That's as bad as 'us reenactors' pointing out that you don't 'fire' an arrow from a longbow. Who really cares though? It's just pedantry.
'Filming' is used as a generic term these days to indicate the recording of a motion picture, regardless of the medium. If you don't like the way language changes in use over time, I'm sure you can build yourself a 'wicked' time machine and head back to your glory days. That'd be 'cool'. Perhaps you can tell the people back then to stop referring to vacuum cleaners as 'hoovers' and ballpoint pens as 'biros' while you're at it.
I did not mean to provoke anyone. It's just that I've been associated with too many activities where death is certain if what you say is not exactly what you mean, clearly and specifically communicated without need of implication or inference. But, again, some images are indeed "filmed", "caught on tape", or have "footage" of.
@dr8: I'm pretty sure if you say "filmed" instead of "acquired digital motion capture" that nobody is going to die.
And drop the "hipster" pejorative crap. They aren't the only people in the world using film. In fact If you actually left your house you'd probably find that film users are typically older than your average "hipster". I'm not even sure you'd recognize a hipster if you saw one.
As a matter of fact since you seem to be the "semantics police", the term hipster was coined in the 40's. It was a term that described a person that was into the "hot jazz"scene. So technically speaking if you're talking about hipsters then you mean someone who was around 20ish back in the 40's, so I would guess the average age of a real hipster would be somewhere around 90 years of age.
So if you "said what you meant" about hipsters, then you were completely wrong. I hope nobody died because of your incorrect usage of the word "hipster".
Even the DSLR haters have to respect such precise mechanics, especially if it works flawlessly after such abuse as this guy puts it through. I really like DSLRs.
Well practicality is relative. For example something based on mechanics requires less or no electricity. Hence you can look through the DSLR viewfinder waiting for the shot without having to worry about battery life. Not so with EVF. And the look and feel of OVF vs EVF is a personal choice too.
And don't forget that digital cameras still need to rely on mechanics - a lens and mounts cannot be just cobbled together. Apertures and their control, image stabilization systems and most shutters are mechanical too.
But regardless I was talking about respect to such mechanics, not practicality.
Flash sync is easy: you just witnessed how a shutter handles fast shutter speeds, with a moving slit that exposes the sensor behind it. Flash sync is ANY speed - up to the fastest for the camera - where there is NO slit, the shutter opens fully and completely before closing again.
Why? Because the flash pop is a single, very quick event - the flash pop is over before a slit travels all the way across the sensor. That would leave any spot on the sensor NOT exposed by the slit, in time, still dark.
correct me if im wrong but also a brighter flash is a longer burn time of the flash... its not more intense. change in intensity is a change of 'temperature' which would yield different WB results.
nstam - I think you are wrong (not the expert myself).
You usually want a short burn-rate of the flash, to freeze the action. And there really can be a slight change in temperature, cause of different flash-intensities.
What will use a longer burn-time is high-speed-sync... which in fact really lets the flash light up as long as the shutter slit is moving across the whole sensor. That's why HSS drains so much battery and may cause heat issues with your flash.
Karl Günter - so can you explain it right then please?
At least for HSS I'm pretty sure how it works (ok - it could strobe... but that's even more complicated to sync with the curtains)
And also regarding burn time vs. burn intensity I found information, that at least the better flashes really vary the intensity (capacitor charge) and not the burn time...
There is no hard and fast rule in regards to flash output brightness vs. speed. IN GENERAL a flash that puts out more lumens (brightness) makes the flash 'pop' last longer, but for example that can be cured by using a higher output flash tube set to a lower (equivalent) power.
Flash tubes can indeed change color when their power is turned down, due to an increased time duration between the command for the flash tube to turn off and the time it takes for the flash tube to actually dim to a point where it is no longer making light at a usable level (the "T1" time). The use of IGBT (insulated gate bipolar transistor) fixes this - the power cutoff is sharp and fast, allowing the tube to stay at a very bright, known color point output and then have the power chop so suddenly that the bulb goes dim quickly, leading to very little off-color light.
HSS works by having lots of little flashes that lasts as long as it takes for the shutter to complete its task.
@ThorstenMUC: Hi-speed sync does not work that way. It does not make the flash to light up for an extended "burn time". It is technically possible to make the flash burn continuously, but it would kill the flash pretty quickly.
Hi-speed sync make the flash to work in strobe mode: it fires a long sequence of very short low-power light pulses. The pulse frequency is high enough to make it work as of if the flash were burning continuously even for the fastest shutter speeds.
A side-effect of that approach is that the integrated light output is smaller than would be in case of a single regular pulse.
The video of fast shutter speeds prompted me to contemplate diffraction effects. Clearly, shorter exposures produce increasing shutter diffraction with commensurate increased blur and decreased contrast.
Diffraction is a wave phenomenon that has nothing to do with "in focus". For example, the well known star spike (crosses) in astro-images are due to secondary support vanes that cross the light path in front of all optical elements (and are thus completely "out of focus").
The shutter blades form 2 straight-edge diffractors, each scattering light. These diffrators affect the image in proportion to the amount of time that each pixel is subjected to the scatter. Shorter exposures result in a higher proportion of time that is affected by the scattering.
If interested, you can research diffraction via Wikipedia, etc.
To have visible diffraction effect at the sensor position, the aperture and the sensor have to be apart far enough to create enough phase difference between the light coming from both ends of the aperture.
This shutter effect is not an aperture. Diffraction scatter occurs at a single edge. It is true that the blurring effect depends on angles of scatter (including f-ratio related) and the distance from the diffraction edge to the sensor. I haven't done the calculations but it is possible that much of this scatter is on the order of or less than the AA filter.
In practice, this is effect may never be significant because short shutter exps are generally accompanied by lower S/N (higher ISO) and faster f-ratio with less than perfect optics, both of which degrade the MTF.
But it is an interesting question. A well controlled empirical test is not as simple at one might first think. It is necessary to hold constant the lens (incl f-ratio) and true S/N while varying exposure times. This precludes any change of ISO and discourages filters (neutral density). It is necessary to vary the light source intensity. I may try to detect and measure this effect...
I am glad "Slow Mo Guy" created and posted the video. I used to repair cameras in the mid-1960s to mid-1970s but I never tire appreciating videos showing the precision of cameras.
For those who ever used Canon Pellix, in the mid-1960s, the mirror is a fixed pellicle (two-way mirror), hence, is quieter and less vibration. The comments touting the advantage of mirrorless may not be aware that their DSLR may have a mirror-up position than when used with like-view is results in essentially a mirror-less camera.
That mirror flippin’ around will definitely cause jerkiness, especially at slow shutter speeds when firmness is needed most.... case in point for mirrorless I suppose.
Nope, there are plenty of solutions for that - including the fact that most DSLR today can do live view - with more mass and thus more dampening of the shutter movement (if even that is still applicable) than the mirrorless toys do... The depicted 7D has an electronic first shutter curtain usable in live view without drawbacks (unlike for example the Olympus mirrorless which drop their quality because the electronic first shutter curtain drops A/D conversion depth)... So on a DSLR you get the best of both worlds!
DSLRs use the same design as the 35mm celluloid film cameras of days gone by. The inertia of the flipping mirror that retracts just before the shutter exposes the sensor is scary. The moving shutter wobble doesn't look convincing neither. The whole thing looks odd – fine photographic apparatus using mechanical parts that would be more at home inside a car engine.
That inertia is tiny, mirror is very light and thin - that's why it looks quite dramatic, but in reality energies involved are minuscule. If you want to look for a real problem - a combination of large shutter with small body causes notably higher vibrations that easily propagate to the lens if camera got a short flange distance. Look: A7 mirrorless cameras. There's plenty of discussions about that issue, but I guess everyone came to realize that it's just something that can't be fixed unless we get full electronic shutters without the issues current electronic first curtain got.
Yeah in slow motion the mirror flip looks dramatic but if you look closely the mirror is completely still by the time the shutter starts to move.
It's annoying when people try to discredit mirror less by calling them toys. Unless you has a full blow career in photography you are a hobbies and any camera equipment you have are just toys for your hobby. Get over yourself and your equipment. No matter how good you are I'm sure I could find better photographers online using mirror less cameras.
@Boky: nope, actually people have been reporting "softness" shooting with mirrorless cameras with large sensors like the a7R while using slow shutter speeds. In fact, it is not softness but vibration of the shutter mechanism causing a mess and since the camera body is smaller than DSLR, it can't act as cushion for the vibration.
"By Karl Gnter Wnsch The depicted 7D has an electronic first shutter curtain usable in live view without drawbacks (unlike for example the Olympus mirrorless which drop their quality because the electronic first shutter curtain drops A/D conversion depth)..."
This is incorrect/misinformation--Olympus does not drop bit depth with electronic shutter. Some Panasonic cameras do, but not Olympus.
I stand corrected, it's the Panasonic which drop A/D conversion depth - the Olympus I meant doesn't even have a proper electronic first shutter, it still uses the mechanical shutter and increases shutter lag to have the camera dissipate that shutter shock to then employ the electronic first shutter curtain - which doesn't really solve the problem... http://www.dpreview.com/articles/4134393686/olympus-updates-om-d-e-m1-with-electronic-first-curtain-anti-shock
The main difference is that vertical shutter allows for faster shutter speeds. Cloth shutter like OM-1 or K1000 was limeted to 1/1000, vertical shutter even back in that day could go to 1/4000, although mass cameras like that (FA and FM2) showed up may be a decade later.
Good explanation of how the shutter works and what 'rolling shutter' is. I guess my question is, does the shutter somehow reset after each shot so the curtains would always be traveling from top to bottom??? I didn't see anything like that in the video. Just the single top to bottom travel.
Yes, the shutter does reset afterwards. If I'm not mistaken, the single-digit Canons have a separate drive to do this, while the three-digit models reset the shutter together with the mirror. That's why the mirror cycles on the 600D (for example) every time you take a shot in live view mode.
@D200 - NO, there are micro-sized Loompa-Doompas that resets the shutter after each shot. . . . I am kidding. My smart-a$$ reply, of course the shutter blades resets it self.
What cam do you have that is shutter-less ? It may not have a mirror, but it has a shutter !!
Now when electronic shutter sensors come about (global shutter) then you will NOT have any moving objects except for your finger - no not in your nose, on the 'shutter' button !! ;)
So, maybe LOL in the 'mirror' , is that your reflection ??
Paraphrasing Clint Howard' character in The Waterboy - "The good Lord chose not to bless me with a fully functioning AF, but you're an inspriation to all of us who do not care to learn how to use mirror lockup!"
@coyote@suave Lol lol lol u guys funny! Ok mechanical shutters, ok? Good catch boys, good catch. I can see a finger alright and it's pointing in someones general direction just about now...
You Toronto guys! Don't get me into the last Provincial election and as to hockey what can I say!! The answer to your question is that the DSLR with its mirror plus focussing module lets pro photographers (and keen amateurs) track birds, planes, football players etc. better than any other camera design.
It's clear to me that the mechanical parts of a camera are with their days numbered. In a few years the cameras will be fully electronic, and will take hundreds or thousands of high-resolution pictures per second.
" thousands of high-resolution pictures per second" That implies extremely short exposure time. You'll need some mighty powerful light to achieve that.
In fact, the sunlight is strong enough for very short exposures. For example, according to the "Sunny 16" rule, for a F2.8 lens and ISO1000, an exposure time of 1/32,000 seconds is enough for outdoor photography in bright sunlight. Modern sensors capture images virtually free of noise at ISO1000, so 32,000 fps is perfectly possible in principle, provided the electronic circuits are fast enough to process and store the huge mass of data.
Frank_BR - So how do you capture images in rainy, foggy, and cloudy days? What about dawn, dusk, and indoors. Of the scenario you described, are you satisfied with the "virtually free of noise" images you obtained. It seems to me that the article you read in Wiki is from a wishful thinker.
ISO 1000 images are chock full of noise in the shadows -- because light itself is noisy (photon shot noise: the statistical variation in photon emission rate). I'm working on a new type of camera sensor that can sustain the equivalent of better than 1000 FPS, but it does that by having a different integration time for each pixel, so each pixel integrates photons as long as is needed to get a specified accuracy and darker pixels update much slower than bright ones. The real trick is that you're trying to image the scene, not the light used to sample it, so you need enough photons to build a scene model and can then temporally interpolate between samples at the pixel level.
Conventional high-speed camera sensors need very bright lighting to get past photon shot noise within a 1/10000s integration time. In fact, I'd be hesitant to expose the 7D's sensor to the level of lighting their 10K FPS camera needed... but they aren't too careful with that 7D. ;-)
Frank - Perhaps one day but currently all cameras have plenty of mechanical parts. What about IBIS in some mirrorless designs (including my Sony a72). Why is a mechanical mechanism chosen to perform this function rather than a software solution?
Mirrorless cameras use mechanical shutters and go through a sequence of mechanical aperture and shutter control to take an image.
The mirror isn't all bad! It enables you to view the image for as long as you like without powering up the sensor, resulting in great battery life and lower camera temperatures. The mirror protects the sensor from excessive light so the lens aperture can be fully open until you press the shutter button.
I will say that a DSLR isn't the ideal camera to take thousands of hi-res pictures per second but it is a mirrorless camera in Live View mode.
"What about IBIS in some mirrorless designs (including my Sony a72). "
Good question, Camley! I believe that all optical/mechanical image stabilization systems, such as IBIS, will be replaced in the near future by purely electronic IS systems, which could function more or less as follows: long exposure times are divided into very short intervals, say 1ms or less, and the multiple images captured in each interval are stacked to form the final image. Even without assistance of an optical/mechanical IS system, individual image captured in very short intervals are sharp. So the final image would be sharp, too. In fact, this technique has been used in rather incipient form in many cameras to capture night shots. The electronic circuits aren't yet sufficiently powerful for general photographic captures, but this technological limitation should be removed in a few years.
In the not so distant future of perhaps 50 years, we may not even need to carry those bulky camera and lenses. Super nano-technology may even allow implants like the Borgs (Star Trek Next Generation) obviating the need for wearable camera. Such advance technology may not even need visible light to capture images. I am sure I would be just ashes by then.
For now and in the next 5 years, I am settling for my current Nikon DX and FX DSLRs and complement of lenses. All my DSLRs allow me to capture night scenes without electronic or electrical powered lights. I capture indoor scenes illuminated by candle lights. I capture the stars on a moonless night.
I marvel at the digital imaging technology and would not waste my time defending one technology from another, much less, one brand from another nor different models from the same manufacturer.
I hope you also enjoy your cameras as well, if not more, than I enjoy mine.
Nice video! I think most DSLR users are not aware of the moving slit, which is a cloth curtain that moves left-right in many old SLRs (and why we still call them curtains). The temporal artifacting, nearly identical to electronic rolling shutter, is also something I think many people don't know about. It's also worth mentioning that there is a very different artifacting for leaf shutters used in compact cameras: the iris goes from open to closed and then open again, effectively varying the aperture diameter through the exposure interval (which actually can improve bokeh, but causes strange temporal artifacts within the bokeh for fast-moving objects).
BTW, a high-speed video looking at the other end of the camera is interesting too: the lens in a DSLR typically has to stop down to the desired aperture as the mirror is being lifted. In fact, in some cameras that's what's still in progress in the short interval between the mirror being up and the shutter curtain starting the exposure.
Leaf shutters can be anywhere, but they are usually near the aperture (rarely, even double as the aperture) to keep the artifacts to the bokeh issues I mentioned as opposed to vignetting.
There is a mirrorless without a mechanical shutter; the Nikon J1. Others have the option to use an electronic shutter, including the Sony A7S, Panasonic GH4 and Nikon V1, although it has limitations due to the relatively slow transit time of the "back curtain" which is limited by the sensor read rate.
The fully electronic shutter is one of those holy grails of camera engineering, especially if it's global. If/when it can be perfected without compromising IQ (as all CMOS global shutters do), it will have huge advantages for both mirrorless and DSLRs, although only the former can ever be truly silent (if you discount liveview modes, which are essentially just a DSLR in mirrorless mode).
Enjoyed the video, well aware of the content before watching, but found it entertaining. I appreciate the the time spent putting this together and perhaps for some of the newer members of the imaging community it offered good insight into workings of a DSLR.
I am hoping that we lose the "R" of DSLR i.e. mirror. Technology is converging on this solution. I also suspect that the disrupter in the DSL market isn't Sony as much as it is Samsung. IQ debates aside, their first real salvo demonstrated that software is effectively free so why cripple a camera with poor software features?
CMOS sensors are the rule of the day, for now and favor rolling shutter. I suspect the next real breakthrough (50 mega pixel canon is a catch up not break through) is a global shutter in a body that is under $4,000 us at 18-24 MP -
Global shutter at these resolutions at video frame rates pretty much leaves ergonomics to the differentiation of vid/stills bodies
The reason Canon and Nikon have dominated the professional SLRand DSLR market is their ability to manufacture high speed shutters and mirrors. In contrast, companies like Minolta/Sony, Pentax and Olympus simply could not and cannot equal Canon and Nikon in this area.
That's a lovely bunch of BS. Before Sony chose to move forward towards where cameras are going with HQ EVFs and the SLT that makes them possible now before coming advances in sensor technology for electronic shutters and focus. The Sony A700 had the same capability as the 7D 1/8000th of second. And even now the A7M2 can rattle off a measured upto 14 FPS (rated at 12 FPS) in a camera that sells for half the price of the 7D2. And it has a Sony sensor. So Sorry.. there are many nice things about Canon and Nikon, from brand power, to lens collection to some body features that make them popular fast flappy mirrors and shutters are not the differentiating feature. And at some point both of those are going away as technology improves. So one would hope Canon and Nikon have more than that going for them as Sony is the one leading the technology that removes a 6o year old mechanical design from 21st century cameras.
Actually, focal-plane shutters are commonly made by companies like Copal (e.g., http://www.nidec-copal.com/02/02.html ) and purchased as modules by Canon, Nikon, Sony, etc.: they're often the exact same mechanisms across many camera brands.
You know Pentax's mid range DSLRs have 1/6000 shutter speeds, where Canon and Nikon both have 1/4000 right?
Also the shutter manufacture is done by completely different 3rd party companies as stated by ProfHankD. In fact copal and some of the other shutter makers have been making shutters for many decades.
So why have the Sony A900 and A99 all had faster shutters than (every version of) the Canon EOS 5D? The top end Sonys can sync with any flash or strobe at 1/250s, use wide apertures in daylight at 1/8000s, and use up the shutter life at 5 fps, whereas the 5D has rarely delivered on its advertised sync speeds, which are much lower to begin with. If you want speeds like this with a Canon mount, you need to fork over many thousands of dollars more for a Canon 1D...or pay even less for a Sony A7 + mount adapter.
Apparently you don't know that minolta holds the record for the fastest shutter speed. 1/12 000 on the 9xi. I don't think Canon and Nikon ever matched that.
This excellent presentation shows well that mechanical shutters are old rudiments which will gradually be replaced with modern mirrorless applications.
Of course primitive shutter mechanism creates fancy effects to the photos, but I still think that Photoshop is better tool for these :)
Most mirrorless cameras, especially the ones with APS-C or larger sensors, are still using mechanical shutters. Just ask all the Sony a7R users about their shutter shock issue.
This film shows the absolute oldfascionality of the mechanical shutter. No need to design new DSLRs using this technology from the past. DSLRs are from the past, already now. Let's start to learn to forget their existence and stop to give some support to them in buying these museum monuments.
And still this old fashioned tech works leaps and bounds better than the mirrorless which for all the money in the world can't provide a decent viewfinder image and can't autofocus on moving subjects even if you put the subjects in a freezer at 0°K...
I find the evf image actually better most of the time. I assume I'm not the only one. Off course this has also to do with the size of the image seen (the Olympus Vf4 evf image size is much larger than the one in the Canon 6D, for example). And I would not want to live without the ability to view the image in the viewfinder after taken the shot (the only way to really judge an image).
@datiswous: "And I would not want to live without the ability to view the image in the viewfinder after taken the shot (the only way to really judge an image)." You do know that the quality of the EVF doesn't allow this to happen in any meaningful way!
What about when the light gets low? +1 for EVF over an optical viewfinder, even when the camera has a fast lens. There are very good EVFs now, they are only going to get better as resolution and refresh rates improve. There is no blackout and they can clearly show me a view that is hard to see with the naked eye in dark situations. Flapping mirror technology and its amazing damping systems is very impressive but, sadly, rapidly losing its relevance.
@Karl Gnter Wnsch It actually does! Much better to see your photo in the EVF than on the screen. I personally get the idea you never tried a high quality EVF..
So, did you enjoy the slow mo presentation, or not? I also enjoy driving my mom's '77 Triumph Spitfire, but I don't take the opportunity to evaluate how primitive it is compared to my Lexus.
I have tried the latest generation and they are still absolute garbage when compared to an OVF... They can only show about 6EV dynamic range - so when your scene has a decent 10 EV you will neither see the shadows nor the highlights before taking the shot and neither will you do so afterwards. So you'll never know what rubbish you ended up with during composition and thus ruined the shot and can't redo it because you wont see that before you open the image on a decent screen - the same applies to the LCD screen (worse if it's an OLED, they are horrendous). With an OVF you would have seen the problem before you took the shot and had had a chance to react (if you are knowledgable about photography) and not just the EVF junkie who has no clue about anything not shown in the electronic crutch...
Karl, Sorry, but your comments indicate a very biased and close-minded understanding of the current capabilities of EVF's. If you don't care for using an EVF, that's fine -- not everyone does; but to make such ill informed, arbitrary statements doesn't reflect much objectivity. I've used optical view finders for about 25 years of my 30 years of serious photography, and I can assure you that current EVF's offer enough advantage to me that I don't want to revert to an OVF -- although I can still switch back and forth without any problem.
Karl, Your comment about the DR advantage of an OVF is also silly. An EVF will allow you to view and correct for any blown highlights or blocked shadow details in scene in real time -- before you do a shot. This is absolutely not true of an OVF -- it will portray the scene exactly the same, regardless of exposure settings.
@daddyo: False. on an OVF you can see the full DR your scene has, on the EVF you don't even get the part of the DR that your sensor can capture - and worse of all you can't evaluate the shadows or highlights for junk that's hidden there because the EVF completely fails to show that. If you are so dumb as to correct for blown highlights or blocked shadows on the bases of the EVF you are overcompensating by 2-3 stops on either side of the DR - because the shadows block up 2-3 stops in the EVF before even a lowly JPEG would and the same applies to the blown highlights.
EVF's show the full DR of what the camera is capable of. If I'm incorrect, can you give a good source of information about this topic?
Off course with your eyes through an ovf, you can actually see more DR than the camera is capable of. If this is a good thing or not, is a matter of taste.
"EVF's show the full DR of what the camera is capable of. If I'm incorrect, can you give a good source of information about this topic?" A good LCD can only display 8 EV at most - in video less. An EVF without localized backlight illumination can never ever manage more than 6EV. That's it. If you want to evaluate a scene and determine which part to capture you need all information, not just a small part. Just for example try to do a portarait against backlight when using fill flash - total failure with the EVF, with the OVF no problem to make up the expression on the face of your subject.
Karl, You need to know what you are talking about, before you refer to someone as being "so dumb".
The histogram, or 'blinkies' available in real time using an EVF are a measure of the exposure from the camera's meter -- not a reading of what the EVF is capable of displaying. For your edification, here is a quote from the Nikon website touting use of their D5100 LCD to accurately check exposure:
"But you should check the histogram when a scene's lighting is especially tricky; when there are areas of deep shadow and bright light in the same scene; and when you're going to take a series of images in the same setting and want to be sure your exposure is right on target. A glance at the histogram will tell you if parts of your photo are over- or underexposed. Overexposure means lack of detail in the highlights; underexposure, loss of detail in the shadows."
And this is from Nikon bragging about the 'Live Histogram' available on their LCD. Do you suppose Nikon is also "so dumb"?
False the histogram and the blinkies don't have anything to do with the metering, they are an expression of the final image as they are derived from the JPEG as per settings in the camera. And that only applies to the histogram visible after the shot has been taken - if you messed that one up you can redo it - which you often can't do. And yes, both Canon and Nikon are so stupid...
Karl is right, and raises several very valid points. That dismissal from mirrorless boys is laughable.
I used EVFs for over 3 years and pretty much the only thing I'm missing in OVF is live histogram, but I still will always pick OVF over EVF, precisely because issues raised by Karl and a fact that, well, viewing world on a tiny TV really breaks the connection with a scene and I always, even after 3 years of running with EVFs in my primary camera, got distracted by EVF. Image is different in EVFs, you always have to get back into "I'm operating the camera now" mode, while with a DSLR it's much more transparent, you are always there with a scene, no tiny TVs standing in your way.
The shutter is always moving down while the exposure is happening. You have an opening shutter blade, which moves down first, exposing the sensor (or film), followed by a closing blade, which covers it up, stopping the exposure after the amount of time corresponding to the set shutter speed. When they reach the bottom the camera resets the shutter by moving both blades back to the top. When it does this, they are both together so they don't let light through. This was important in the days of film, which would be double exposed by any gap between the blades, but doesn't matter with digital as the sensor is not recording while the shutter is resetting.
I hope that explains the last chapter in the journey of a focal plane shutter.
And why this camera wasn't mirrorless one? Mirrorless is the way ahead. There would be no need for videos like this if all cameras were mirrorless. There are many cameras with wi-fi and selfie functions. Why don't they film those?
What the F*&^ are you talking about? How does a camera having wi-fi or selfie function relate to this video in any way? The video has absolutely nothing to do with mirrorless vs mechanical shutter.
There is no such thing as mirrorless vs mechanical shutter. Most mirrorless today HAVE a mechanical shutter. What they don't have is the bouncy mirror.
It's kind of sad to see my comment has achieved it's purpose of exposing disturbed souls. I didn't expect to see as many as 18 of them. People who are almost fanatical about the gear and without sense of irony, humour involved in photography? Nah! I don't think so.
Bouncy mirror is not really a problem, it's very light, energies involved are tiny, and everything is very well dampened. DSLRs got plenty of space for that. Unlike mirrorless cameras where you have no space at all, so vibrations from mechanical shutter get pronounced all over the body and lens. And if you still have a problem with DSLRs and their vibrations - just use damn mirror lockup feature. Learn your camera.
Long live mechanical shutters! The maximum flash synchronization speed is the fastest speed the shutter moves WITHOUT have to resort to the moving slit to achieve the set speed. It used to be about 1/60 for horizontal travel cloth shutters and 1/125 for vertical travel metal shutters. The latter are faster because the distance top to bottom is less than across the horizontal axis. Since the digital SLRs use smaller crop formats and mechanical technology has improved we can get sync speeds up to about 1/250 ... but the flashes today can fire stroboscopically achieving sync speeds as fast as the shutter allows, just at reduced power, a limitation of the flash.
1/250 flash sync has been available well before the advent of digital cameras. Ones that I've used Nikon FM2, FE2 and FA (1982-83 first manufacture dates), and later Nikon F4. I'm sure there would have been some Canon models as well :)
Strobostropic flash allow you to use a higher shutter speed per image area, but there is still a 1/250th second or longer time difference between the top and bottom part of your image, creating some occasionnal distorstion. That's occasionnaly annoying ;)
Given the fact that that technology works and is robust (look at how beat up that 7D is) - why should it be going away? It provides things that no mirrorless junk could ever provide - like a viewfinder image.
For me the EVF are a pure eye sore - you can't see your subject properly (dynamic range of the mini displays is woefully inadequate), it lags, the viewfinder blackout time for the best is 5 times as long as on a modern DSLR, it disturbs the light adaptation of the viewfinder eye resulting in headaces and the inability to see in dark environments after its use for minutes at a time endanger you. EVF are a crutch for those that can't photograph and a bad one at that because with an EVF you'll never learn.
I couldn't disagree more with your last sentence. I think if you want to learn photographing, you are way better of with an evf. More information on the screen, better preview of the image-outcome, you can better see the result afterwards in the evf.
Btw. I disagree with the opinion of the original poster. I think there's technology-wise nothing wrong with using a dslr. It performs better or worse depending on interest and subject.
Lag time one most modern cameras is below 10ms. Which is about twice as fast as the fastests human eye refresh rate.
> the viewfinder blackout time for the best is 5 times as long as on a modern DSLR
That simply is untrue. The black out is only a problem with long exposure. Because that's the time when sensor cannot be used for EVF.
But if you are into a long exposure, things like Live Composite (see your long exposure developing on your LCD/EVF as it develops) are simply impossible with OVF.
> you can't see your subject properly (dynamic range of the mini displays is woefully inadequate)
Nonsense. Some cameras have blinkies, which can show you the highlights/shadows. Something you can never have in OVF.
> EVF are a crutch
Your previous comments could be attributed to lack of information, but that's just silly. The whole digital photography is a poor crutch for those who failed to learn to photograph with the film.
I'd disagree on the lag, I had a lot of difficulties shooting macro shots with a GH3 as even though I clicked the second the flying insect appeared it was usually half out of the frame in the photo. Switching to a DSLR (and not one noted for high speed) I got almost all with the insect where I wanted it to be.
The other EVF issues are power consumption (I take 2-3 batteries for my mirrorless cameras per one for my DSLR), low DR (the EPSON LCDs that are widely used run about 240:1) or bad colour accuracy (usually with the OLED ones, which do have better DR to make up for it). The big pluses are more on-image symbology (although you can do that with an OVF and a optical combined LCD, e.g. X100t) and in-VF image review (which I do miss when using an OVF camera).
I use both types and don't see either as a problem, but if I'm at an event I'd much rather be watching it through a big FF optical finder than on a baby TV.
Dr_Jon Agree mostly, although the X-T1 demonstrates a larger Viewfinder than I think any 'FF' DSLR. It's actually the reason that I ditched my Nikons - the EVF is that good.
Yes, it sucks batteries like nothing else, but batteries are cheap and easy to carry so I'm good.
"Lag time one most modern cameras is below 10ms. Which is about twice as fast as the fastests human eye refresh rate." - ONLY IN PERFECT LIGHTING CONDITIONS!
In real life lag can go below 1/24 of a second as sensor needs to be exposed for a time, and all that smudging you see in your EVF while in low light is because exposure time can go below 1/24s.
"Nonsense. Some cameras have blinkies, which can show you the highlights/shadows." - FYI: These are useless cause they're made on 8 bit output for EVF, not the full output you get in RAWs.
I've read how that works, but interesting to see. I didn't realize that the shutter seems to move at about the same speed, just with larger or smaller openings. I wish he had shown under 1/200 so you could see why max flash sync speed exists.
It's not only the 7D mirror box; actually, there are many, many things that can be filmed @ 10,000 fps to be better undestood. Nikon FF mirror boxes, bullets ripping thru apples, H-bomb deflagrations...
And, you know, mirror box can't film, actually. One also needs sensor, controls, CPU, battery, memory, etc, basically full camera in order to film something.
What's most informative here is the vast amount of mirror bounce and shutter-curtain bounce. If you ever needed a demonstration of why mirror lock-up is essential when shooting on a tripod and you want to achieve the best-possible results, this is your proof.
Mirror bounce isn't a problem here. It's vibration transmitted to the camera body that's important & it shows non of that. In fact, the reverse is true. The mirror bouncing about shows that most of the shock is being absorbed & dissipated by the mirror, rather than being transmitted to the camera body.
When I saw that mirror bounce and all the other complex mechanism on this video, I thought the designers of mirrorless systems got their motivation from such observations. The mirrow is out, now it is time for taking the shutter out!
I asked this question on the YouTube page, but maybe I'll have better luck here:
I'm sure I'm making an obvious mistake here, but it's killing me :-)
At 4:20 he explains how the top of the picture is "older" than the bottom, due to the way the shutter leaves work. Then at 4:28 he further explains that this is in fact reversed due to the way lenses invert the image, i.e. the bottom of the picture will be "older".
So why is the shadow of the cork, which is lower in the frame and therefore should be "older", actually younger (i.e. it happens later in time than the image of the cork itself, which has only just left the bottle?
@Pecolpan - false application of Occam's razor. The true result is simply that the sensor is read from the top of the frame which happens to be at the bottom of the physical sensor.
Thanks Karl, that makes sense. In the video, his hand actions at 4:36 got me thinking the other way around, leading to the seemingly-unlikely possibility that Pecolpan described.
@Karl: What Pecolpan wrote is not an application of Occam's razor, but a short version of Sherlock Holmes' motto "When you have eliminated the impossible, whatever remains, however improbable, must be the truth".
So should I use Occam's razor to shave upwards or downwards? And does the same apply if I use my own razor? I suppose one should shave the slowest growing hair first and then progress to the fastest growing. In that way, at some time, all the hair should be at the same length.
@Karl Gnter Wnsch - do you know why in video the frame is read from the top of the frame while in still photografy it is exposed form the bottom of the frame? It's certainly a design feature but an intriguing one.
@Pecolpan: It because stills use a shutter curtain to expose and it's been the design with mechanical shutters to travel vertically down the sensor vertically for many decades. This allows super fast shutter speeds.
Older cameras (usually rangefinders) had horizontally traveling shutter curtains which inhibited the speed.
With video the exposure is electronic. Honestly, it could be programmed to go either way, but I guess the engineers decided to go from bottom to top because of the lenses image inversion.
@JDThomas: Thanks. That's more or less what I was guessing.
One small detail I do not agree with you: the shutters traveling vertically down has very little to do with how fast the shutter travels across the sensor (I am thinking about the time it takes the curtain to travel the sensor hight, not the exposure time as this one related to the delay between the first and second curtain).
The shutter is spring loaded and the force it applies on the curtain is much larger than the force due to gravity (and that's why the camara works well upside down). An example: if the shutter of a full frame camera takes 1/500 s to cross the sensor when shooting down (so the sensor is horizontal and gravity does not add to the movement) with the aid of gravity with the sensor vertical it would take about 1/514 s.
@Pecoplan: You're misunderstanding what I'm saying. The reason why camera manufacturers started using vertical traveling shutters is that the distance from up to down is shorter than the distance from left to right.
Older cameras had horizontal shutters meaning the shutter curtains had to travel across 36mm per shutter cycle. A vertical traveling shutter only has to cover 24mm of real estate.
That's why older cameras typically have shutter speeds from 1/500-1/1000 and newer cameras from 1/4000-1/8000.
It has nothing to do with gravity. It has to do with mechanical and physical limitations.
I've always found it fascinating when watching these kinds of videos. They kind of show how much mankind has progressed in different areas (mechanics, technology..). So much that we don't pay too much attention to these stuff anymore in our everyday lives.
What boggles me, is that someone originally developed this tech without aid of computer. Heck, someone invented computers without the aid of a computer.
If the big one hits, those that are left are going to have to start all over and re-live another 200 year industrial revolution just to get back to being able to complain about that "ancient" Canon technology. (if they can figure out how to feed ourselves without modern technology)
Computers were actually invented _with_ the aid of computers. Computer used to be a profession: rows upon rows of mostly women sat in offices and computed numbers all day. Sometimes multiple times over, in parallel, for verification. Talk about a mind numbing job.
Canon has released firmware version 2.0.0 for the EOS 7D. The update, originally announced in June, is the most comprehensive we can think of - significantly expanding the capabilities of the 7D. Changes include improved continuous-shooting buffer depth (to 25 Raws, up from the original 15), customizable Auto ISO, control of audio recording level for video and the ability to re-process Raws and rate images in-camera. The update also enables the use of the GP-E2 GPS module.
Canon has updated the firmware for its EOS-1D Mark IV and EOS 5D Mark II digital SLRs. Firmware versions 1.1.0 and 2.0.9 respectively both improve read/write speeds with UDMA 7-compatible CF cards, and also correct minor bugs and menu language errors. The firmware will be available for download from Canon's regional websites.
Canon has posted a firmware update for its EOS Rebel T3i / 600D digital SLR. Firmware v1.0.1 clarifies the on-screen feature guide description of the digital zoom function in movie mode, and fixes a bug relating to computer connectivity via USB. The firmware is available for immediate download from Canon's website.
Canon has posted a firmware update for its EOS 7D digital SLR. Firmware v1.2.5 fixes bugs relating to stills and video recording with specific CF cards and image stabilization with specific lenses in custom setting mode. The latest version also improves read/write speeds when using UDMA 7 compatible CF cards. The firmware is available for immediate download from Canon's website.
Canon has issued a service notice for the EOS 5D Mark II and EOS 7D offering a paid-for option to replace the cameras' mode dial with a redesiged lockable version. Similar to the one seen the company's latest EOS 60D DSLR, the mode dial features a centre button which has to be pressed to allow a change of exposure mode. This option will be available from December 6 and for a cost of $110 in the US. The European service charge is to be confirmed.
The Nikon Z6 may not offer the incredible resolution of its sibling, the Z7, but its 24MP resolution is more than enough for most people, and the money saved can buy a lot of glass. Find out what's new and notable about the Z6 in our First Impressions Review.
Many cameras today include built-in image stabilization systems, but when it comes to video that's still no substitute for a proper camera stabilization rig. The Ronin-S aims to solve that problem for DSLR and mirrorless camera users, and we think DJI has delivered on that promise.
The SiOnyx Aurora is a compact camera designed to shoot stills and video in color under low light conditions, so we put it to the test under the northern lights and against a Nikon D5. It may not be a replacement for a DSLR, but it can complement one well for some uses.
At its core, the Scanza is an easy-to-use multi-format film scanner. It offers a quick and easy way to scan your film negatives and slides into JPEGs, but costs a lot more than similar products without a Kodak label.
If you're looking for a high-quality camera, you don't need to spend a ton of cash, nor do you need to buy the latest and greatest new product on the market. In our latest buying guide we've selected some cameras that while they're a bit older, still offer a lot of bang for the buck.
A question frequently asked on the Internet is 'what's the best DSLR?' In this buying guide we've answered that question – but also whether it's the right question to be asking in the first place.
What's the best camera for under $500? These entry level cameras should be easy to use, offer good image quality and easily connect with a smartphone for sharing. In this buying guide we've rounded up all the current interchangeable lens cameras costing less than $500 and recommended the best.
Whether you've grown tired of what came with your DSLR, or want to start photographing different subjects, a new lens is probably in order. We've selected our favorite lenses for Sony mirrorlses cameras in several categories to make your decisions easier.
Whether you've grown tired of what came with your DSLR, or want to start photographing different subjects, a new lens is probably in order. We've selected our favorite lenses for Canon DSLRs in several categories to make your decisions easier.
On their latest models the two smartphone manufacturers have replaced the dreaded display notch by a design that features a circular hole for the front camera in the display.
With the latest version, Adobe Camera now lets you import Raw files from the newest iPhones, Pixel devices, the Samsung Galaxy Note 9 and Nikon Z6 among others.
The Nikon Z6 may not offer the incredible resolution of its sibling, the Z7, but its 24MP resolution is more than enough for most people, and the money saved can buy a lot of glass. Find out what's new and notable about the Z6 in our First Impressions Review.
The Sigma 56mm F1.4 combines a sensible sub-$500 price tag and excellent performance, providing a portrait-friendly 85mm equiv. view on Sony's APS-C mirrorless cameras.
The 250mm F4 is Fujifilm's longest lens for its medium-format system. It's equivalent to about 200mm on a GFX camera, and we put it to work on some portraits as well as some scenes around Seattle's waterfront – take a look.
A new patent from Canon lays out the schematics for a speedbooster-style adapter for mounting Canon EF lenses onto EOS M cameras, but with a variable baffle to reduce the risk of flare.
The Jackson Hole Travel and Tourism Board has started a campaign asking visitors to stop geotagging their specific locations when visiting Wyoming's national parks.
Film simulation app Filmborn has been updated with new presets, features, and overall improved support on Apple's latest mobile operating system and devices.
We've been shooting with the LX100 II both in and out of the studio, as part of our ongoing review. We're pretty impressed, so far, with the revised JPEG color and addition of a touchscreen both noticeable improvements.
Skylum's Luminar 3 arrives on December 18 with the long-awaited ability to manage your photo library. However, it won't be a full DAM (digital asset manager); the company plans to roll out features throughout 2019 and won't charge for updates from Luminar 2018 during that time.
Nikon's IPTC Preset Manager, a tool for creating predetermined sets of metadata, has received an update. Version 1.1.0 no longer uses Microsoft Silverlight, sheds the network connection requirement, adds extended language support, updates support for Windows 8.1 and Windows 10, and ends support for Windows Vista and Windows XP.
Camera app developer Hipstamatic says it has found a way to use the depth data generated by the iPhone X to improve the way its TinType app works out which areas of a picture to render out of focus.
Comments