Lessons

Lesson Info

Sensor Q&A

So, we'll start with some of these ISO questions. One is how well does the noise reduction setting within a camera work? Should it only be set within the camera, or done in editing? This is one of those things that I talk more about in my fast start classes for the particular cameras. That's why I don't have it in here, but I will address that. I normally turn off in-camera noise reduction. I turn it off completely in my camera. The reason is because when I shoot a photo at, let's say right when it's twilight and the light's changing and I have a really dark scene in front of me and I do a 15 second exposure. I shoot a photo for 15 seconds, I stand there and wait around for 15 seconds, and then if I have noise reduction turned on the camera then processes that image for the next 15 seconds. And I can't do anything for another 15 seconds, and this sounds like I'm being extraordinarily impatient but when you're out working in the cold and the light's changing quickly that 15 seconds ca...

n mean everything to you. So, the choice is, I can let the camera process for 15 seconds and give me a result that's pretty good, that's an improvement off the original image. Or, I can take that original 15 second image and take it in my software that I use with Lightroom or Photoshop or whatever else, and I can customize the fix on that image to exactly the needs of that image rather than having a generic fix done in camera. So you can do it in camera and it makes it easy and fast and you don't have to worry about dealing with it later, but I would recommend for the more serious night-time shooter to turn that feature off and just work with it afterwards. Because, if you're able to work with it on a larger screen that has more controls, you'll be able to do a better job than what your camera can in and of itself. Thank you, go ahead. So, I found like a cool little Easter egg on the 5D3 by changing it from 100 to 50, but now I've been hearing that 100 is the optimal setting for that camera. So, I'm wondering for whatever reason, if you had the tools to say, maybe use, instead of going to 50, keep it at a hundred and use like a neutral density filter or do it in software, just for you, do you have a preference as to what you think about how you solve whatever issue? A number of cameras have this low setting, and the low setting has a different penalty than the high setting. The high settings, you get more noise. On the low setting, you don't get more noise, you don't get less noise. What happens is, the dynamic range of light, from low light to bright light, slightly compresses and you're able to get less dynamic range. So whether you should use that feature or not has to do a little bit with what you are photographing. There's been many times where I have been photographing a waterfall and I thought to myself I'm in ISO I'm at half a second, and I would really like to get down to one second. The solution is to set ISO 50, but I'm going to compromise the dynamic range. Then I would say, wait a minute, this scene doesn't have that great a dynamic range. This scene will fit within the dynamic range so I can use 50 and it's not going to be a problem. So, in most of those evenly-lit situations, that extra low setting is perfectly fine to set and you're unlikely to notice any difference at all. In theory, you could add a filter but then that adds kind of another thing in front of the lens and then you have to compare the two to see which one is better quality and worse quality. So I will use that low setting on whatever camera I have if I look at the scene and it's not a really contrast-y high contrast scene. So it can be judicially used with virtually no penalty at all. Alright, great, thanks for that question. So, you've talked a lot about the different ISO's and sensor sizes, can you sum up for people again, this is Alicia Young S, how does the sensor size affect noise at different ISO's. Different sensor sizes at different ISO's is starting to throw too many wrenches into this problem because it's very easy if we can just line things up and say everything is a 24 megapixel sensor how does it change from different size sensors and the bigger the pixels the less noise you're going to have. So, when you start throwing in the pixel size well, then you can also complicate that by asking about the age of the camera. Because what happens, is that, let's take the Sony a7S II, it's a 12 megapixel camera and it is arguably the best low light camera on the market today, it's a full frame camera with 12 megapixels. Let's go back 10 years in time to the Cannon 5D, the original 5D. That too, is a full frame camera with 12 megapixels. It is blown out of the water by the new Sony camera. Same number of pixels, same size sensor the processing is the difference. They have much newer technology that's able to process the information, look at that same original information and come up with a better result. Also, the sensor itself is better at absorbing light. The pixels themselves are more sensitive and able to pick up that light better. So, we have three different factors which are going to impact the camera: the size of the sensor, the number of the pixels, and how new or old the camera is, how good the processing and quality of the pixels and the whole image scene system is in the camera. It's something that people do a lot of pixel peeping. That the words, pixel peeping. I'll throw a shout out to our buddies over at DP Review, they do a great testing if you want to get in and check out what is one camera compared to another camera. DP Review, and you can dial in different ISO's different cameras, and you can compare away until your heart's content. And you can judge which camera is best in that regard. John, I love your classes because I learn something new every time. I didn't know that pixel was made up of picture element. That's like the most basic thing Kayla! I know, did you guy's know? I didn't know that. Anyhow, more on ISO's. This is from Robert Dawson and has a number of votes. He's been told by other photographers that it is better to shoot at native ISO's for example 100, 200, 400, 800, as I will get sharper images. The problem is that when you use one third stops, 125, 160, 250 etc these are calculated from the camera by the closest native ISO and you loose quality. Is that true? Oh, we're digging deep folks, aren't we? In general, it's not something to worry about, but there is some truth to this. Because we don't know exactly what the manufacturers are doing behind the scenes, it's quite possible when you go from 100 to 200, then you decide 160, what the camera is actually doing is, it's setting it at 100 or and then by software, it just darkens it to look like 160. But, unless you do some incredibly accurate testing at all the different ISO's will you know the difference. So, the ultimate solution is to do your homework assignment, learning project number 3 or something and check out your own camera, because some people have found that some cameras have better ISO's that don't make perfectly logical sense. For instance, 160 might be better on a camera than 100, but you are gonna have to really dig deep to see this information. If you were to print this on a 16x20 photo and walk down the street and ask people to see the difference, they're probably not going to see the difference. In fact, I would guarantee they're not going to see the difference, but it's the type of thing where if you blow it up and take a magnifying glass and, ah yeah, I can see this one a little bit better than that one, you'll see a difference. So this is where we get into the constant arguments, this is the source of so many arguments in photography, not this particular topic, but the concept between this is technically better, and that really can't be argued with, but for all visual practical purposes, it makes no difference. So you'll get somebody that argues on the technical side of the thing being accurate and somebody else, "Well listen in real world this is how it works and it doesn't matter." And that is the source of so many of our problems. And a relief if we're on the side of, it doesn't really matter, right? It does, but it's so small, I have to give them credit because I do like people who are very technically accurate in what they are doing, but in a moment we have to step back and we say, okay this car has 459 horsepower and this one has 460, and yes it has more, but is it going to make any difference driving down to the store? Alright, well you just mentioned printing, and several folks were asking about whether or not sensor size affects print quality, and how quality degrades as print size increases. Hmmm, well printing is a whole nother world unto itself. I don't know if Creative Life ha-- I mean, they should have like a week long master print class, cause that, there's a whole lot of things that go into that. One key number I will give you is 300. 300 dots per inch is usually what the human eye can pull out. Now in many cases some people kinda say, well, it's only 200 and some, well I say it's 360. Here we are arguing again. So you want to be able to print about 300 dots per inch. You can start doing the math out as to how many pixels you need. If you want to make, lets keep this very simple, 10 inch print, a 10 inch print, you would need 3,000 pixels by 3,000 pixels. Now you can start looking at how many pixels you have on your sensor as to how big you can print. Now with printing, you can actually push things quite a bit, to 200 pixels for instance, per inch and still get good quality results. So, it isn't directly related to sensor size. It is indirectly related to sensor size, because generally bigger sensors have more pixels. On average, if you look at full frame cameras right now, they're averaging probably in the range of around 30 megapixels, whereas crop frame are averaging more around 20 megapixels. And so in that general sense the larger frame, the larger sensor is giving you more pixels, which is giving you more resolution to print larger prints, and so having more pixels helps out directly for making prints. Now, what type of prints, what type lighting conditions, well then that goes back to where we were going though the whole last section in the sensors. So, it gets to be a little bit of a complicated answer and I don't know if I got anywhere close to it. What difference is there in image quality between a 16 megapixel image shot on a full frame camera and a 16 megapixel image shot on a crop frame camera? I know we talked about that, but could you re-sumarize for people? Okay, I hate to say this, but there was a slide for that. I think I compared a 24 crop frame and a 24 full frame and what happened is that ISO 100, there was no easily discernible difference. If you get a big enough microscope, you'll find a difference. But you couldn't see really any difference at ISO 100, but when you crank the ISO up to, I believe we went up to 25,600, you started to see more noise in the smaller size sensor. So, if you have a lot of light to work with, if you said, all I do is photograph in the middle of bright sunny days, you don't need a big sensor in order to get good results, you can do that with any small sensor, because you're getting nice, lots of light into your sensor. So in their example, the smaller sensor is going to get more noise, and what's going to happen is if you were to test these things out, you're gonna reach a break point where you're gonna say 100 it looks totally fine, 200 good, 400 good, 800 good, 16 hold on, I'm starting to see a difference. At some point is where you're gonna start seeing the difference. And that's where it just helps to know what type of photography you do, you know if your kids are playing basketball down at the local gymnasium and it's terribly dark down there, and you shoot with your small frame camera, or your small sensor camera, excuse me, and you shoot at ISO 6400 and you're getting these really noisy, grainy shots, that's probably your sensor that needs to be larger in that particular situation.