So, being largely a novice in the world of Reflex cameras, I was talking with a friend who mentioned this issue. While he is definitely more expert than I am, none of us has any accurate data, and we are curious.
Is it (still?) true that DigitalSLR are slower that analog SLR in taking pictures, precisely since the time yuo push the shutter button to the time the image is captured? In other words, do you really capture what you see when you push the shutter button, or at least are you as fast as old analogs?

Usually reviews I've read, dpreview ones in particular, measure the time needed to take the first shot in terms of OFF-Shot. In other words, they include the time needed to power the camera on.

I can understand there could be a delay there compared to analog reflex (?), but I can also accept the solution of keeping the camera in Stand-by and pay the price in terms of battery life.

But, other than that - assuming current cameras are still affected by this - is there any other soruce of slowness?

I would tend to say no, and this is my reasoning. As you can see though, I am missing some data.
1. Disabling AF, I suppose that pushing the shutter button in a dDLR is like pushing the button on a SLR: shutter opens and light poors in.
2. Light reaches the sensor, which I would guess to be much faster than the film in capturing light and retaining it to replicate how exposure changes in time.
3. Processor needs to take data collected by the sensor, produce a file out of it, and store it in RAM first, memory card then. Am I missing something?
If not, I am assuming that delay can only be generated here. And the fact that the Canon 7D, for example, uses two image processors instead of one, makes me think that there is a need for speed there, especially, I suppose, with larger nr of bits needing to be processed when using high resolution sensors.

So, if I am correct in this simple thinking, are current cameras (image processors) perfectly comparable to old analogs? More realistically, are they all, or just some of them?

What do you guys think / know about this?

I don't know about all brands, or all models.

But for the top-of-the-line film and digital cameras (professional boat anchors) from Canon, the 1V (film) and 1D (digital), the 1D is quicker.

AF cameras are slower than non-AF cameras, film or digital. Turning AF off closes the gap, though.

Cheers

I am, once again, not an expert, but I cannot imagine how film couldn't be faster. You're opening the shutter for the same amount of time (assuming the same film/sensor speed), but the film records that data directly due to the photons affecting the chemicals on the film, whereas the sensor has to do the same thing, plus pass that information on to a bunch of electronics, which in turn do more processing.

However, there are really at least two questions here. The one I just answered is "how long does it take the camera to record an image?". The next is "would the image I captured be the same between these two cameras?". I think the answer there is "yes". I don't see any great reason why there would be a significant amount of difference between the time you press the shutter button and the shutter actually opens, and I've already pointed out that the shutter is open for the same amount of time. Since neither camera can record images at any time when the shutter isn't open, that would seem to indicate that they'd be taking the same photo.

(Actually, there might be an electronic delay on the digital camera, as it might have to tell the sensor and related electronics to get ready, while the film is always ready. That said, I imagine that mechanical linkages would make that delay irrelevant.)

Then there's another question, which is "how much longer until I can take another photo?". I'd imagine that the digital could come out ahead here, as it doesn't have to wait for the physical movement of a piece of film. There are probably other factors that are more relevant, though.

Yes, my question was precisely: "would the image I captured be the same", the variable being TIME and not QUALITY, to be accurate.
In other words, will I be capturing the same moment in time?

And, you have a great point there in answering this question. As a silly novice, I was forgetting a quite obvious piece of information: the shutter >>stays open<< for time X for both cameras, and that's decided by the operator.
My thinking was wrongfully based on the inconscious assumption that the shutter would close "when done", which make no sense in a camera.

So, yes, the only parameter left, as far as I can tell now, is whether the shutter-release button in dSLR does more than mechanically opening the shutter, or not. If not, there must be no difference.

Thank you!

Press the button, the camera takes the picture.
The time between those two events is called "shutter lag".

An old-fashioned film camera used mechanical linkages from that button to the actual shutter. But for the past 25 years, nearly all SLRs have had electronic linkages, film or digital.

Shutter lag is comprised of AF delay, AE delay, processing delay, and shutter actuation delay. These are all the same between film and digital, or at least should be all the same.

So film and digital ought to be pretty much identical here, on similar platforms.

Except that every 36th 37th frame (or less), there's a 1+ minute delay before a film camera can take the next shot. Digital cameras only hit this delay once every several-hundred (or several thousand) frames.

The next factor is shot-to-shot delay. Digital wins here, hands down: no mechanical delay while the film is wound to the next frame. The result is that digital cameras boast FPS (Frames per second) way in excess of similar film cameras.

And don't even get me started on shot-delay when changing ISO (film speed). At best, it's perhaps 5-10 seconds using film, when one switches from one camera body to another. At worst, we're talking minutes to change films. With digital, the delay can range from zero ("auto-ISO") to perhaps 5 seconds of fumbling with controls.

Cheers

There should be a moody index as well. I had a couple of minolta aslr cameras that were very fast, when they were happy. But if they thought all the settings were not perfect, they would just refuse to take the picture. I hated those cameras...

All my digitals are this way, they get pissy and refuse to take pictures if all is not perfect. I would love a camera that just takes the shot when I push the shutter button, correct settings or not.

I don't know too many ways to set up my D300 to prevent it from taking a shot when the shutter button is pressed. One sort-of those way is setting into self-timer mode - but it will still take the shot - after a delay.

Canon DSLRs (or at least the ones I have used), when in Program, Aperture or Shutter modes will refuse to take the shot when the shutter speed would have to exceed the max speed. With auto focus on they will also refuse to take the shot if they have failed to get a focus lock. They will also refuse to take the shot in many modes if the flash is raised and the shutter speed needed would exceed the sync speed.

I'm guessing the Nikons and other DSLRs behave similarly ?

But of course is that is a problem you'd just switch to manual exposure and turn off auto focus.

Yes, and that "+" can be very long, I suppose, if you're in a harry and drop your camera, or the film, or you mess it all up somehow.

Thank you all, guys, this all makes very much sense and answered my questions.

I noticed that too, but was not sure if I was doing something else wrong. It is a bit annoying, I have to say. Again, I guess that one gets used to that. I too have only used Canon DSLR, though, so I don't know how Nikons behave.

Annoying ? Why ?

The whole point of the Program, Aperture and Shutter modes is that exposure is automatic. When you select those modes you are saying "please get the exposure 'right' for me".

Say you have the aperture set to say f/2, the ISO set to 400 and it is a nice sunny day. That could mean that the shutter speed is 1/8,000 of second or some equally high number.

The camera could either just go ahead and take the shot at 1/4000, resulting in an over exposed shot or it can not take the shot and indicate why it can't (i.e. flash the shutter speed on the display). I for one would prefer it to alert me to the problem rather than just slavishly taking the shot.

If you want to take a shot regardless of how it is exposed, there is always full manual mode.

A third choice would be to allow you to choose what it should do in that situation via an option. Some DSLRs may well do that, I've only ever used a couple of Canon models for any length of time, so wouldn't know if any did.

Don't get me wrong, I am not saying it is the Camera's fault.

The behavior I would have consider "intuitive", as a novice, is to offer me an option: wait for the camera to complete all automatic adjustments when the button is half pressed and push the button down only after the OK beep is heard, or not wait for the beep and go all teh way down with the button before the beep, taking a "wrong" picture.

Having a different expectation, I felt frustrated when, in "wrong" exposure/aperture conditions, I could just not push the button and "see what happened", which was my approach to begin with. Of course, I resorted to manual mode then.

Sorry, I suppose I wasn't taking into account purposely setting up the camera to fail, such as not making any shutter adjustment in shutter priority or aperture adjustment in aperture priority, trying to exceed the limitations of the software. In full program it should be able to make the necessary adjustments to always take a shot. Though to be sure, one can also use auto-ISO which makes even more automatic in the vein of a P&S.



This is an option for Nikons in at least the pro-sumer range, such as the D300. I'm pretty confident it's available at least in the D70/D90 class machines too.



Hmm, this one I can't recall, but it too might be influenced by some configuration setting as well.

I think that once you're used to the limitations of the camera and more so the relationship of sensitivity, shutter and aperture (your calculated exposure), these situations will be very very rare.

Why not both? I can't imagine an instance where a bad photo would be worse than no photo.

I can, for me at least. When you have forgotten you have it on too high and ISO, are quickly taking a shot and not bothering to review the shot. I'd far rather have the camera clearly let me know by not taking the shot so I can fix the problem.

My camera, if the exposure is outside the camera's limits to set correctly, just changes the information in the viewfinder to red to let me know there's a problem, but it nonetheless takes the picture. This seems to me to be the best compromise.

tanstaafl.

Nikon, at least, lets you set a variety of different autofocus policies with regard to what happens when you want to shoot at the precise moment when the image is out of focus. I've got mine set such that it always takes the picture immediately and autofocus is completely disconnected from the shoot button. Instead, I use the "AF-ON" button the back to pick my focus, then reframe the shot. (Exactly like I used to do with manual focus cameras and that goofy split-screen focusing gizmo.)

Yes, this is best. I tend to take pictures of folks doing things, so I don't always have time to set the camera up properly. I would rather have a crappy pict than the memory of a defiant camera.

I miss a lot of shots because of my cameras these days. That never happened with the AE1...

Maybe I should look at your camera, it seems a good all around choice.

You are assuming light meters are always correct. They are not, in fact reflected light meters are very easily fooled and tricked. For instance, what if you were trying to take a picture of something in a snowy scene with a ISO of 400 and aperture of f/2. It could well be that the correct shutter speed to render the snow white would be less than the camera's maximum shutter speed, however the light meter would think the correct shutter speed is 1/8000 (which few if any cameras can do) because it tries to render the snowy scene a neutral 18% gray. Blind reliance on light meters is a very bad way to take pictures- you can almost always get a better picture if you carefull spot meter the scene or use an incident meter. No one does this anymore in the age of digital because they can assume "I can just fix it later"- but they actually can't because of digital's low dynamic range..

I'm not, hence my quotes around the word 'right' in the phrase "please get the exposure 'right' for me".

But then you asked the camera to prevent you from taking the shot if the exposure is not "right". So the camera should trust it's light meter more than the user?

I suppose you could design the camera to force you to take the shot in manual mode- but that's not exactly elegant behavior...

No I didn't.

I said that if when in Aperture priority the camera's light meter decided that the shutter speed needed to be higher than the max shutter speed the camera is capable of then I wanted it to make me very aware of the fact (but not taking the shot) rather than just taking a shot at the max shutter speed.

If I want to second guess the light meter because of challenging lighting conditions or dynamic range, I'll be in full manual mode.

IMO, that's the only way to manual focus and without it, manual focus is almost useless. So I never ever manual focus with my Nikon D300.

Only way? I kind of like what my camera does. If I am using manual focus, as soon as I move the focus ring, the center of the image in the viewfinder is magnified eight times. This allows very precise focusing on whatever element of the picture is most important. When I stop moving the focus ring, the viewfinder reverts to normal view.

FWIW, I always use spot metering and (when in auto-focus mode) spot focusing. Another trick I nearly always use is to gauge exposure by what the viewfinder shows. When the shutter release is pressed halfway, the viewfinder image darkens (or lightens, as the case may be) to show what the actual exposure will look like (something a DSLR with optical viewfinder can't do) and locks the exposure. If that image in the viewfinder is not what I want, I move the spot focus/exposure point to a lighter or darker portion of the image and lock that exposure. I might do this three or four times (it takes less than a second per try) until the preview exposure looks like what I want, then press the shutter release all the way to actually take the picture. The metered exposure gets me in the ballpark, but I make the final adjustments by eye. Depending on what operational mode I am using (shutter preferred, aperture preferred, or program auto) the camera adjusts to compensate for how I want the picture to look. Sometimes I'll set the camera to do three-shot bursts, one under-, one normal-, and one over-exposed (according to what the light meter is telling it) in combination with the half-shutter-release compensation described above.

tanstaafl.

You can't magnify an optical image in the same way though. If your eyes are the least bit off, even if only temporary because some dust blew in them, it's a complete PITA to focus manually without the prism on the focus screen - though my old Minolta actually had this feature on a specific lens, go figure (I have no idea how that was accomplished).

FYI, that's called bracketing.

Yes. Do other cameras have that burst mode where it takes the three shots in less than one second with the bracketed exposures? I can adjust the degree of over/under exposure in 1/3 f-stop increments up to +/- two stops. In other words, as much as two stops under and over, or as little as one-third stop under and over.

tanstaafl.

Yes, many do, especially from the film era.

Bracketing, yes, but not necessarily the burst part that Doug refers to. In the (D)SLR world, bracketing and "burst" (or continuous mode) are generally separate features. If you choose, you can use them together, giving what he describes, but you could also choose to shoot your bracketed frames one at a time

That's called a split-image rangefinder. A lot of former full-image Rangefinder users (read: mostly Leica users) wax poetic about them and mourn that no decent digital cameras have them today. They are kind of nice when you get used to them- especially full image rangefinders like the Leica's of old.

Yes, if one is expecting the old film camera method of showing focus.

My 40D, like many digital SLRs, will confirm focus by lighting up the "focus dots" that are in-focus during manual focusing. Those are every bit as easy to see as the old style split-prism view, and the camera doesn't need to be centered on the target either.

Cheers

You can even get adapters that will let you use old manual focus lenses on DSLR bodies, some of these adapters will actually light up the AF Confirm dot when they are in focus. Using old lenses like this can be a good antidote to bored-with-all-your-lenses syndrom. For instance you can pick up an old Pentax or Nikon 50mm f/1.4 lenses for $30 or so. A 50mm Canon 1.4 costs about 10 times that. Of course you don't get autofocus and have to use stop-down metering but it works.

Except of course Leica's own top end digtial cameras.

Caon also has the AF-ON button. I am not sure, yet, that it can ben used as you describe as well. Definitely, it will perform the same function as the half-press shutter release (AF and AE).

Right like the M9 I should have said digital cameras that mere mortals can afford....

You can buy replacement focus screens for Nikon and Canon bodies with the split-image in various configurations.

The one I was already familiar with is Katz Eye:
http://www.katzeyeoptics.com/

But there are others:

http://www.focusingscreen.com/
http://brightscreen.com/enter.html (their HTML web page looks like complete shit, I shudder at the thought of seeing their Flash version)
http://haodascreen.com/othercameras.aspx

The benefit of these is that you don't have to walk around like a pretentious Leica user.

When photographers talk about a "faster" camera or lens, they are often talking about the light-gathering capability. A "fast" camera is one that can use a faster shutter speed, given the same environment, than a "slower" camera. This has to do with the maximum aperture size of the lens and the speed of the film (or sensor). In digital cameras, which I don't know much about, I understand that "noise" is an issue in low light conditions, which would be another form of "slowness", since you'd need a longer exposure to deal with this. The sensor may function like relatively slow-speed film in digital cameras. I know most digital cameras have ISO settings up to 1600, but I also understand this comes with quite a lot of noise in the image. For usable quality, I can imagine that a digital camera is "slower". Then again, film cameras also give extraordinarily higher quality, especially with slow film or negatives larger than 35mm, so comparing like-to-like image quality is a fool's endeavor anyhow.

Are you sure he wasn't talking about exposure time and not time between successive shots?

I don't know of any film that can be pushed to 16000+ ISO with any level of quality. I don't even know of any film body that allows even close to that to be configured.

Digital sensors will produce noise at all sensitivities, the least amount at the sensitivity they "default" with. But current low-noise and high-ISO capable sensors, along with the software controlling them, are producing some amazing results that you can't really expect from film.

Yeah. It's techno-weenies (like us!) that chat mostly about "sensor noise" in DSLRs. In pocket digicams, with teensy sensors, sure, this is a real issue. But not in DSLRs.

Film suffers from coarse grain at high ISOs, which is vaguely similar to digital sensor noise.

My 40D, at ISO 1600, is way, way, WAY better than ISO 1600 colour film. Oh, wait.. there's no such thing as ISO 1600 colour film..

And the better DSLRs go much, much higher than ISO 1600.

No, he is not talking about fast exposure times.

Conversation is still ongoing and at present it evolved as as follows:



Questions bein asked now are:

1. Is 0-1 time interval the same between analog and digital? I suppose the answer is "yes", meaning that it depends on the electronics between the shutter release button and the shutter itself, and has nothing to do with the camera being digital rather than analog.

2. Is 0-1 shorter in completely mechanical cameras, where the shutter is linked to the shutter release button via leverages?
I woudl say that if good electronics is used, it can be just the same as a mechanical one.

3. Is there a 1-2 time interval at all, in digital cameras? That is, does the sensor reacr immediately as it is hit by photons, or is there some lag time there?
I would guess it does react fast enough not to introduce any delay perceivable by humans compared to the chemical reaction occurring on a film, but it is just a guess.

This is just, pure intellectual curiosity.

Originally the question asked, that generated this thread, was actually different: if I want to take a picture of a specific moment in time and push the button of wto cameras identical in all but one being digital and the other analog, will they take the picture of the same "moment" in time? This already found an answer: yes. They will obviously take the picture of the "moment" in time defined by the exposure time interval.

For all we know there is a 1-2 interval in analogue as well, who is to say that film starts being exposed the instant light falls on it.

You could of course test this by rigging up two cameras to take the same shot at the same time of say a scrolling wave form on an oscilloscope (or some sort of hires digital timer on a screen/display, a second hand on a clock isn't going to work due to parallax issues). But I'll bet any difference (either way) is so small that it will be lost in the noise in the data.

That is, when the shutter opens, right? I mean, we ARE assuming light travels at infinite speed, here, am I correct? Taking into account light speed seems to be a bit too much accuracy.

1-2 was referring to the moments immediately AFTER lights hits the surface of sensor/film, in case - just in case - sensor does not respond immediately with an electric signal or, similarly, chemical reaction on a film does not actually start.

I think we are agreeing, that just as the sensor takes some finite time to start reacting to the light, the film stock almost certainly has some similar delay before it starts reacting.

The issue is: does the sensor need to be turned on or otherwise activated in some way? This might include turning on or activating whatever electronics are accepting the signal behind the scenes. This is almost certainly an infinitesimal amount of time, but film simply does not have this issue. Chemicals react to photons as soon as the photons arrive. Yeah, it may take a few femtoseconds for something to start happening, but delaying the incidence of light would also delay that delay. On the other hand, if the data bus behind the sensor simply isn't on for that period of time, for example, the sensor is essentially unreactive and the photons that hit it during that time were irrelevant.

It's not an issue at all Bitt. Well, I suppose if you want to split hairs you could do all kinds of analysis. I don't think you'd be able to measure the speed of activating the sensor in any meaningful way though.

The bottom line is that digital is simply a lot faster than film. Especially if you're talking about anything but the highest-end F6 (the most advance film body ever created).

Since the time the shutter opens, light hits both film and sensor at the same time, of course.

My thinking regarding Bitt's observation is that the sensor is already ON even when in darkness, before the shutter opens. It would just make no sense not to have it on and ready to accept light.When photons hit the sensor, I would agree with Bruno that the consequent electron-flow (signal) generated by the sensor itself takes place "immediatly", that is in a time which is not significant. But, are we sure? I am just guessing here.
But, I also think that chemical reaction on film is slower than sensor in anyway.
Again, just guessing. I'd be interested in finding some data.

The only importance of the "time" it takes to cause the reaction/sensing is whether it causes a delay or otherwise impacts the time it takes to get to the next shot. With a digital SLR you can take another shot immediately, with film, you either have to manually advance to the next frame or wait for an automatic motor winder to advance the frame for you. Consumer automatic 35mm cameras were insanely slow. And even the highest end motor winders and extra battery packs (for cameras with built-in motors, were considerably slower than the quick DSLRs of today. Some of them are of course able to beat some consumer DSLRS of today, but dollar for dollar, DSLR frame capture is just all-around so much faster.

Of course there is. For example.

However, since you weren't aware that ISO1600 film exists, it seems unlikely that you know for certain that the 40D is WAY better at that speed. Admittedly, 1600-speed film is very grainy. It is also possible to "push" film a couple of stops faster than its rated, and make up for it with longer development time, and some people used to use 800 speed or 400 speed film that way.

I don't take many pictures of fast-moving subjects, so I mostly used ISO25 or ISO50 film when I was doing a lot of photography. While we're on the subject, digital may be nearly as good as a 35mm film camera for image quality, but film is available in 6x7cm or 4x5in and the results from those negatives can be incredible. Digital is cheaper, and "good enough", but I still use my film cameras for stuff that I know is going to be enlarged -- typically my 6x7 Crown Graphics.

There's currently no easily attainable digital substitute for very large format film, that's for sure.

I've pushed both B&W and colour neg a few stops when I was in school and developing it and the prints myself.

As an extreme example, let's look at ISO 6400, which is still a modest capability of the new pro Nikon bodies.

Well, there are "medium format" digital cameras. Hasselblad and Mamiya, possibly amongst others, make them. Of course, at that point, you're talking about spending as much for a camera as for a car.

Why would that be, in your opinion?
Leaving aside the comparison with film, I would argue that if there's a time needed by the sensor to react (1-2 in my little diagram above), which has no impact on the final result of the shot you're taking (shot 1), that same delay would simply transfer to the following shot (shot 2) and in that case, too, it would have no impact on the final result of shot 2.

In any case I still think reasonable to assume that time to react for a sensor is irrelevant and possibly significantly less relevant that that of the chemical reaction of a film, assuming both sensor and film are hit by photons at the same time.

I've tried this trypod in a shop:

http://www.zipshottripod.com

I was impressed on how light it is and how quickly it opens. The idea is extremely simple and very effective, I think. Too bad it only holds 1.3 Kg. Several body+lens combination in the market will be too heavy, including mine (1.7Kg). I would have probably bought it, otherwise. Anybody owns one?

That tripod appears to be way too flimsy to do what a tripod is meant for: holding the camera dead steady.

The more conventional heavy Manfrotto tripod I have here, has quick-release snaps on the legs, and opens out in about ten seconds. Fold up time is similar, and the legs adjust in length to handle irregular terrain (hillsides, for example..).

Cheers

Yes, Manfrotto are great, this one can't compare. It is more supposed to be the option for when you want to stay light.
You can't even adjust its height.


But, although it looks flimsy, after trying it and holding it I did change my mind in that. I looks as some very cheap product, but it is actually very well designed and deliberately simple. Not as steady and strong as a Manfrotto, it is incredibly steady nontheless! Before I read the specs, I thought it could sustain even something huge. I have been pushing on it, trying to bend it, and it felt great in my hands.
But again, specs tell me its limit is nonetheless quite low.

I would bet Ilford Delta 3200 in 120 (Medium Format) would beat your DSLR. Color? Who needs it.

Heh. Okay, so there are a few fast films still left in inventories around the planet.

But the better comparison for that 120 format stuff would be a 120 format digital medium format sensor.. and there, the digital wins hands down again on the noise/grain issue. Or so say the experts.

Anyone still shooting film, should run out and buy a case or two of the stuff while it's still in stock. And put it in your freezer. What's left on the market now isn't going to be manufactured for much longer. Or switch to digital, and save a ton of money!

Cheers

I do agre they're some amazing cameras, but...

Super expensive, and compared to their film counterparts, lower resolution. Their sensors are also not the same size as the medium format film frame, are they? What are they up to now, 40MP? Even Canon is already at 24 with a FF "35mm" sensor.

60 megapixels, full 645 medium format.

I'm curious when Kodak, Fuji, and Ilford will finally give up on manufacturing film. Certainly, less popular films are rapidly going away (e.g., Kodachrome). With demand contracting, it wouldn't surprise me at all if somebody like Kodak or Fuji just packs up and stops making film, period, in the next few years.

I totally agree with Mark: if you care about shooting film, stock up now.

Will film have a minor resurgence like vinyl records are apparently having currently?

Kodak has actually released two new films in the last year or so. It may be they plan on consolidating there film line-ups to save on manufacturing costs, but whatever the reason you can still buy film, and I can even get film for my 5x7 inch view camera (an uncommon size view camera). The future of color film is uncertain though as it is not easy to make and requires knowledge and equipment that are quickly disappearing. B&W film will probably be around for quite a while, however. People have even been successful creating it in there basement- some people are even going back to coating glass plates just for the unique look of it. It's definetly an interesting time in Photography.