Camera talk...

[raja manuel]

PPI has no significance unless you are printing. You can set the same file to 72 PPI or 300 PPI and the resolution of the file will be the same. Only printing size will change. I am surprised that the Raw file has any PPI setting as it doesn't make any sense until the file is demosaiced. Did the Raw data have a PPI setting or was it the embedded JPG in the Raw file that had the PPI setting?

What is the -5 and what is its relevance to focus lock?

 

[nac]

PPI has no significance unless you are printing. You can set the same file to 72 PPI or 300 PPI and the resolution of the file will be the same. Only printing size will change. I am surprised that the Raw file has any PPI setting as it doesn't make any sense until the file is demosaiced. Did the Raw data have a PPI setting or was it the embedded JPG in the Raw file that had the PPI setting?
What is the -5 and what is its relevance to focus lock?

When RAW files are opened in photoshop, it shows PPI/DPI.
Are you sure? When I change dpi, image gets pixelated. I will try and see if there is any impact on the size.

Jpeg files from my camera SX130, is 180ppi/dpi.

"-5" is exposure value. You know reviewer say that "autofocus doesn't hunt and lock quickly even in negative exposure".

- - - Updated - - -

Okay, I read and understood little better.
Image loses it's sharpness because I was resampling which explains "pixelating"

So why does manufacturer picks different value. D800/D5500/D90 jpegs are 300dpi, SX130 jpeg is 180dpi, 1DX Mark 2 jpeg is 72ppi

 

[raja manuel]

The print point is 1/72 of an inch so a monitor with a 72 PPI resolution would display at exact print size, which is the standard that Apple used in the early Mac days (lifting straight from Wikipedia here). I don't know about the other DPIs. For a compact camera it might have something to do with the kind of printer that the photo could typically be expected to be printed on (pure speculation on my part).

I still don't understand the -5 bit. EV represents the f-stop/shutter speed combination and unless you're shooting with a Sony this setting shouldn't have any impact on focus lock. Additionally, under what circumstances are the reviews talking of negative exposure? If I spot meter off a black surface and set -5 exposure compensation I may actually end up with correct exposure, while spot metering off a white surface and setting -5 exposure compensation will result in severe underexposure - but again, this will not affect focus lock. Do they mean that the scene light is very low and therefore the focus point is unable to lock? I don't see how you can set that in camera, though.

 

[nac]

The print point is 1/72 of an inch so a monitor with a 72 PPI resolution would display at exact print size, which is the standard that Apple used in the early Mac days (lifting straight from Wikipedia here). I don't know about the other DPIs. For a compact camera it might have something to do with the kind of printer that the photo could typically be expected to be printed on (pure speculation on my part).

I still don't understand the -5 bit. EV represents the f-stop/shutter speed combination and unless you're shooting with a Sony this setting shouldn't have any impact on focus lock. Additionally, under what circumstances are the reviews talking of negative exposure? If I spot meter off a black surface and set -5 exposure compensation I may actually end up with correct exposure, while spot metering off a white surface and setting -5 exposure compensation will result in severe underexposure - but again, this will not affect focus lock. Do they mean that the scene light is very low and therefore the focus point is unable to lock? I don't see how you can set that in camera, though.

As of now, I don't have a clue why manufacturer choose to have different dpi. As you already mentioned, dpi is irrelevant until we print. So we leave there for now.

Camera tend to struggle focusing in low light, when the reviewer say that high end cameras are capable of quickly focusing even in low light or negative exposure without hunting, my understanding was it can lock focus whether ambient light is low or photographer dial down to -3 or -5. Now you're raising a doubt and I don't have a DSLR to check. Let me google if I can find the reviews and post the link here. We will discuss about whether dialing down do have any impact on focus lock.

- - - Updated - - -

These may not be the reviews I read, but reviewers are talking about low light AF performance.
Canon EOS-1D X Mark II Review

Another 1D X Mark II improvement is the 1 stop better low light AF capability, now rated to EV -3 in One Shot mode. Being able to focus in 1/2 as much light as with the 1D X should not go un-noticed by those shooting in light-starved venues.

CANON 1DX mark II: SETUP, NOISE PERFORMANCE AND REVIEW

AF working range down to EV -3 (vs -2 in 1DX) = better low light performance.

Canon EOS-1D X Mark II vs EOS-1D X: 12 things you need to know - Techradar India

Talking of centre AF points, the EOS-1D X Mark II's is sensitive down to EV -3, versus the EOS-1D X's EV -2, making it more effective for low-light photography.

- - - Updated - - -

Few weeks back I mentioned about 1D X for wedding photography. Around that time I googled about it and read this article. Here photographer talks about his experience using 1D X Mark 2. He is talking about low light AF not dialing down EC

Initial thoughts on shooting weddings with the EOS-1D X Mark II - Canon Professional Networ

If I’m completely honest, I wasn’t quite prepared for the autofocus on this camera. It is very sophisticated and customisable and so I left it in default mode – one shot – and used the centre focus point (as I do on my 5D-series). At these settings, the low light accuracy was quite superb. It didn’t hunt or lock on to higher contrast elements in the background even in those situations, where the older cameras always need a little bit of help to focus. It was fast. Accurate. Amazing.

- - - Updated - - -

Couple of discussions about ppi

Spoiler

As Todd Gardiner says, they don't -- but the cameras have to produce a regular image file, and many of these file formats have a field for the DPI. So the cameras just plug in a default value like 72, 180 or 300 for the files they produce. Adobe Camera RAW arbitrarily assigns a DPI value for RAW files it processes -- you can change it.

Historically (the 80s and 90s) DPI was very important when you created a digital image by scanning something physical, like a slide or a print. A higher DPI meant a higher resolution image -- there was a huge difference in the amount of detail and information in a 300 DPI scan versus a 72 DPI scan (the "rule of thumb" DPI of monitors was 72, and a laser printer was 300 in case you were wondering where those values come from)

Thus DPI became a "shorthand" for image quality -- which is somewhat misleading, but it was such a useful shorthand that it stuck. Editors and pre-press production people would ask for a "high resolution file", meaning a high DPI value. They knew from bitter experience that if they got a 72 DPI scan (or other low resolution), the image would be useless for printing.

Fast forward to digital image capture, where none of this makes any sense because there is no scan of a physical intermediary like a slide any more. There's just what comes off the camera sensor, and DPI has no meaning in this context. The amount of useful data in a digital photo straight out of the camera is determined by the dimensions and pixel arrangement of the sensor. You can set the DPI value of the resulting image file to be 72 or 3,000 and the image still has the same amount of information in it -- DPI doesn't matter.

The idea that the DPI value of the file determines image quality persists, however, long after it has outlived its usefulness for digitally captured images. I sell stock photos and still get requests for "high resolution files" from book and magazine production people. I send them the largest file I have without upsizing (typically matching the original pixel dimensions of the camera sensor), with the DPI value set to 300. I have a Photoshop Action that does this for me. The power of DPI is so strong that some people still get upset if DPI isn't >= 300, even though I could set DPI to 1 and it would still be the same image data containing the same amount of information.

DPI is an often misunderstood subject. The short answer is that the dpi you find within the exif data of the jpg to be irrelevant. It is just a number for an application to use as a baseline to interpret the pixels at what physical dimensions it should be rendered.

The DPI doesn't change the fact that the 700D produces 18 MP images. The pixels aren't changed, the representation is changed. If 72dpi is set and you load it up in photoshop, it will happily print it way too large on your paper. If set to 300 dpi it might fit a 40x30 cm sheet of paper.

The 700D resolution is: 5184 x 2912 pixels.

DPI = inches / number of pixels

72 = inches / 5184

inches = 5184 / 72 = 72"

inches = 2912 / 72 = 40"

72" x 40". That's the physical size you'd need to print the image.

You do not need to shoot raw, merely to understand what dpi actually represents.

So if you set dpi to 300, the size will be

dpi = inches / pixels

300 = inches / 5184

inches = 5184 / 300 = 17"

inches = 2912 / 300 = 9"

The size you'll end up with physically is 17" x 9". If you print it smaller, all it will do is give you a smaller yet sharper image, the number of pixels will not change, it's still the same 5184 x 2912 pixel image.

It's like drawing a grid on a deflated balloon. Inflate the balloon the number of squares in the grid doesn't change, but the physical size of the grid will increase. The lines may become more blurry though.

 

[raja manuel]

OK, this is where it gets interesting (or confusing, depending on which way you interpret it). EV is Exposure Value, and is a way of referring to exposure settings without referring to the specific exposure setting i.e., with the same ISO setting an exposure of f/2.8 at 1/100 of a second will have the same EV as an exposure of f/5.6 at 1/25 of a second (reciprocity). So if you use Av mode at f/5.6 and set exposure compensation of -3 EV, you are asking the camera to underexpose by 3 stops based on the meter reading. Since you've locked the aperture the camera will change ISO and shutter speed. If you've also locked ISO then the camera will only vary the shutter speed. This is NOT what we are talking about in this discussion.

There is another way in which EV is used, as a measure of scene lighting on a fixed scale. Here, if f/1.0 1 sec ISO 100 gives you a middle/balanced exposure in a photo of a middle grey card, it means that the scene light is EV 0. If it is one stop over, it means the scene light is EV 1, etc. It is this meaning that reviewers are using (or should be using if they know what they are talking about) when they speak of autofocus point sensitivity. If they say that the camera's autofocus point can focus at -3 EV, it means that the camera can lock focus in light so low that at ISO 100 f/2.8 you will need a 64 second exposure (calculating reciprocity from f/1.0 1 sec ISO 100) to get a middle exposure of a grey card. Note that this use of EV has nothing to do with camera settings. It is only used as a point of reference to describe scene lighting irrespective of whether there is a camera present to photograph the scene or not, and therefore does not affect the autofocus point in any way.

The PPI settings discussions illustrate the problem we are facing nowadays: a lot of the information we are getting on the internet is outdated or wrong or both, but such information spreads easily and many people insist it is correct while the people who give the correct explanation are often shouted down because they are in the minority. Just look at the number of forums where someone says their pictures are never correctly exposed and they receive a chorus of 'use the spot meter in the camera' or even worse 'get a light meter'. Completely useless advice now that we've moved past the film age (and not very useful in the film age either), but its uselessness is matched only by its popularity.

 

[nac]

Okay, it is my understanding which was wrong about EV reference then.

In that case, should I comment that 1 DX M2's AF was not as good as they claim (or not as much I expected it to be) when I tried. It was past sunset around 6:45. Focal length 200mm, Single point AF, Single shot mode, Center AF point, Av priority @ f/2.8, Auto ISO (@ISO-2000), Ev -5, camera picked 1/200th sec

If they say that the camera's autofocus point can focus at -3 EV, it means that the camera can lock focus in light so low that at ISO 100 f/2.8 you will need a 64 second exposure (calculating reciprocity from f/1.0 1 sec ISO 100) to get a middle exposure of a grey card.

By the how it is 64 sec? f/1.0 to f/2.8 is just 3 stop, right? Shouldn't this be 8sec?

- - - Updated - - -

[strike]If f/1.0 1 sec ISO 100 is Ev 0, the setting I posted gives EV + 0.5. Is that right? Just manually calculated, could be wrong.[/strike]

Okay, strike that out. It's confusing.

1/200th to 1 sec is 7.33 stop -ive
ISO-2000 to ISO-100 is 4.33 stop +ive
f/2.8 to f/1.0 is 3 stop -ive
So this means I was 6 stop down?

 

[raja manuel]

In that case, should I comment that 1 DX M2's AF was not as good as they claim (or not as much I expected it to be) when I tried. It was past sunset around 6:45. Focal length 200mm, Single point AF, Single shot mode, Center AF point, Av priority @ f/2.8, Auto ISO (@ISO-2000), Ev -5, camera picked 1/200th sec

In what way was it lacking? But remember, this gets even more convoluted. Focus point sensitivity is not the same as focus point accuracy, and no matter how sensitive the focus point is or whether it is cross or dual cross, it still needs contrast to lock focus. All this technology doesn't remove the need to know what to focus on.

By the how it is 64 sec? f/1.0 to f/2.8 is just 3 stop, right? Shouldn't this be 8sec?

Yes, but remember the -3 EV. That's another factor of 8. If it was an f/1.0 lens then it would be 8 seconds but we're compounding a change in aperture with a change in EV.

So this means I was 6 stop down?

Down from what? Unless you were photographing a middle grey card you have no point of reference. What was the metering mode?

 

[nac]

In what way was it lacking? But remember, this gets even more convoluted. Focus point sensitivity is not the same as focus point accuracy, and no matter how sensitive the focus point is or whether it is cross or dual cross, it still needs contrast to lock focus. All this technology doesn't remove the need to know what to focus on.

Down from what? Unless you were photographing a middle grey card you have no point of reference. What was the metering mode?

Both speed and accuracy

Down from 0 Ev. It was spot metering.

 

[raja manuel]

Unless you were photographing a grey card, the 0 EV of the meter also has no significance because it also has no point of reference. Aside from that, Spot metering + Av mode will give poor exposure unless you are dialing exposure compensation to target a particular shade.

 

[nac]

There is too much emphasis on grey card. Do people use it widely in this digital era?

 

[raja manuel]

Where is there too much emphasis on grey card? I would say it, or something equivalent is more required in the digital era as it is now much easier to use.

 

[nac]

Digital cameras comes with their own metering system and they do have their own algorithm to figure out the "right" exposure. Since we have the luxury of reviewing photos right after taking photos, we can dial up/down exposure and re-shoot if we need. So I think we can work without 18% grey card.

 

[raja manuel]

Except that they don't. DSLR's TTL metering is based on the middle grey card. And reviewing photos from a camera screen which can be set to different levels of brightness, and who's brightness is perceived differently by the human eye in different levels of ambient light, is useless unless you're really experienced and have first practised with a grey card, and are used to interpreting the histogram.

 

[nac]

I don't know how many use grey card in their everyday work. And I don't see how practical it is to use a grey card on every work.

 

[raja manuel]

We seem to be shifting goalposts a lot here. It isn't practical to use a DSLR, or a L lens, or a flash gun, or a soft box, yet it is what you do if you want quality. As for how many use a grey card, I suspect that number is greater than you think. I know of at least one internet group where this is a highly recommended method, and there are photographers who go beyond grey card and even use a full blown colour target for every shoot. The fact that many photographers don't use it doesn't prove anything. Most photographers I've met, both online and in-person, don't know how metering works. They only know to brag about how they always use manual mode without a clue about what they are doing.

And even amongst those who don't use a grey card, there are many who use a light meter. And a large number of them don't know that the light meter has to be calibrated to each body + lens combination using (what a surprise) a grey card or equivalent.

 

[nac]

I have been in an assumption that with digital cameras, need of grey card/light meter is pretty much gone. One way or other, photographers learned to get correct exposure without grey card/light meter.

As for how many use a grey card, I suspect that number is greater than you think.

 

[CRACING]

Hello Friends,

I'm looking for used 50mm lens for my Nikon D3200. AF or AF D is fine for me as I do manual focusing.

Used Nikon AF 50mm f/1.8 (Not D) is listed in ebay India for 3.2k [Strike]which is pretty decent price I think [/strike] but it has issues with auto focusing (Not a problem in D3200) and slight fungus.

Link: Nikon Nikkor F Mount 50mm F1.8 Manual Focus Fast Prime Lens | eBa

The price for the same was 3.8k a week ago and now the seller has reduced to 3.2k.
Do you guys think its worth to buy this lens or should I look for better piece?
Do you know where can I find used 50mm lens? I have been checking jjmehta for long but there are only f/1.4.

50mm lens is tempting for me, even though I have 35mm lens but I don't like to spend much for it.

Thanks...
Best Regards

 

[nac]

3.2k for a non D lens and in this condition, isn't pricey?

 

[sujoyp]

I think its pricy...a slightly damaged lens with fungus should attract lesser money....for 3.2 you should get AFD easily. I have my non D 50mm. I bought it for 2.5k 5 yrs back.

 

[CRACING]

3.2k for a non D lens and in this condition, isn't pricey?

I think its pricy...a slightly damaged lens with fungus should attract lesser money....for 3.2 you should get AFD easily. I have my non D 50mm. I bought it for 2.5k 5 yrs back.

Hmm, I always thought lenses have more value then other stuffs but you guys are right, 3.2k is looks pricey. I will send message to seller and ask if he is willing to reduce.

Meanwhile if you find such lens anywhere else, please let me know.

Thanks...
Best Regards