Ugly Hedgehog® - Photography Forum
DX v. FX - Should Crop Factor Be Applied to Aperture/ISO?
If you want to reply, then register here. Registration is free and your account is created instantly, so you can post right away.
May 22, 2014 22:03:09   #
gessman Loc: Colorado
 
PetaPixel and Tony Northrup addressed these questions in a video and the url was posted here on uhh on Wednesday of this week but was put in the links section so I want to see if I can move it the main discussion so perhaps it will get more attention.

I feel this is important information that can be of benefit to many. It could even influence some buying decisions. The video is a bit long but worth watching in its entirety. I will not be involved in discussions but if any of you feel you have better information, please feel free to present your opinion with some substantiating reasoning.

Hopefully it won't get moved again. I will not put the link in here so it will be moved again to the links section so you will have to pull it together with a period where I have put a space in the url after petapixel and copy and paste it into a browser rather than make it a clickable url:

petapixel com/2014/05/18/are-camera-manufacturers-misleading-us-by-not-calculating-the-sensor-size-into-specs/

| Reply
May 22, 2014 22:07:14   #
Racmanaz Loc: In my bedroom why?
 
gessman wrote:
PetaPixel and Tony Northrup addressed these questions in a video and the url was posted here on uhh on Wednesday of this week but was put in the links section so I want to see if I can move it the main discussion so perhaps it will get more attention.

I feel this is important information that can be of benefit to many. It could even influence some buying decisions. The video is a bit long but worth watching in its entirety. I will not be involved in discussions but if any of you feel you have better information, please feel free to present your opinion with some substantiating reasoning.

Hopefully it won't get moved again. I will not put the link in here so it will be moved again to the links section so you will have to pull it together with a period where I have put spaces in the url after petapixel and copy and paste it into a browser rather than make it a clickable url:

petapixel com/2014/05/18/are-camera-manufacturers-misleading-us-by-not-calculating-the-sensor-size-into-specs/
PetaPixel and Tony Northrup addressed these questi... (show quote)


Actually this was posted on here 3 days ago :)

| Reply
May 22, 2014 22:08:58   #
gessman Loc: Colorado
 
Racmanaz wrote:
Actually this was posted on here 3 days ago :)


I knew it was recent and was thinking it was two days ago but I'll give you that day. I've been busy and time flies... :thumbup:

| Reply
May 22, 2014 22:44:22   #
Mr PC Loc: Austin, TX
 
It's gotten a lot of reaction at Tony's site, some violently negative. I don't think the naysayers were quite up to speed on the points he was trying to make. I found it enlightening, I'm sure anybody can quibble with something in a 37 minute video.

| Reply
May 22, 2014 22:52:11   #
amehta Loc: Boston
 
I agree that it is a video worth watching. Tony Northrup raises some very interesting points, and he backs it up with some valid math. I'm still not entirely convinced about it, so I'll ponder some more.

In a nutshell, let's say that we have an APS-C2 sensor which has a 1.4x crop factor. He says that, in addition to applying the crop factor to focal length, we need to also add one stop to the aperture (f/4 --> f/5.6) and one stop to ISO (ISO 200 --> 400). The aperture is the focal length divided by the lens opening, so that's his reasoning behind aperture. The ISO is related to the sensitivity to light, and he says that we should consider the total light, which depends on the area of the sensor, or the crop factor squared. In this case, that doubles the ISO, a one stop increase. This keeps the exposure triangle at the same exposure value.

I guess the first thing is that you can still use the actual focal length, aperture, and ISO values, those are real. The Sony RX-100 has a 10.4-37.1mm f/1.8-4.9 lens with a base ISO 125. All of that is true.

The issue becomes when we "translated" things to 35mm equivalents to use values which mean approximately the same thing across sensor sizes. For example, with the Sony 1"/Nikon CX sensor (13.2 x 8.8mm, 2.7x crop factor), it is very useful to say that camera has a 28-100mm focal length (35mm equivalent), because most of us know that 28mm is a wide angle and 100mm is a medium telephoto. Now the f/1.8-4.9 lens is fast at the wide end, so I could think that I would get the same performance as having a 28mm f/1.8 lens on a full frame camera. By performance, of course he means depth of field, except depth of field depends on sensor size (for everyone except imagemeister ;-) ). In this case, he says the DoF is comparable to about a f/5.6 lens, 2.7x the f/1.8 that it states.

He also translates the ISO, which we use to talk about how noisy the image is. With the RX-100 at ISO 100, his claim is that the noise is comparable to the noise in a similar full frame sensor at ISO 800.

Since many would say that full frame has the characteristics of better high ISO performance and shallower depth of field compared to crop sensors, what he says is consistent, and he quantifies it fairly simply.

| Reply
May 22, 2014 23:08:18   #
Bmac Loc: Long Island, NY
 
Thanks for the link Gessman, I enjoyed the video. 8-)

| Reply
May 23, 2014 01:21:26   #
Erik_H Loc: Denham Springs, Louisiana
 
gessman wrote:
PetaPixel and Tony Northrup addressed these questions in a video and the url was posted here on uhh on Wednesday of this week but was put in the links section so I want to see if I can move it the main discussion so perhaps it will get more attention.

I feel this is important information that can be of benefit to many. It could even influence some buying decisions. The video is a bit long but worth watching in its entirety. I will not be involved in discussions but if any of you feel you have better information, please feel free to present your opinion with some substantiating reasoning.

Hopefully it won't get moved again. I will not put the link in here so it will be moved again to the links section so you will have to pull it together with a period where I have put a space in the url after petapixel and copy and paste it into a browser rather than make it a clickable url:

petapixel com/2014/05/18/are-camera-manufacturers-misleading-us-by-not-calculating-the-sensor-size-into-specs/
PetaPixel and Tony Northrup addressed these questi... (show quote)


Very interesting. However, and please correct me if I'm wrong, but when he was talking about the manufacturers being deceptive about the speed of their bridge camera lenses. I don't agree that they are. If they are claiming for example, to have a 150mm lens with a 35mm equivalent field of view of 300mm, that's fine. But f/1.8 is just that, f/1.8. It doesn't matter what lens you are talking about because it's a factor of the actual focal length divided by the aperture. Both are physical things, focal length and aperture are FIXED. You would only do the equivalence on the field of view .

| Reply
May 23, 2014 05:48:05   #
selmslie Loc: Fernandina Beach, FL, USA
 
gessman wrote:
PetaPixel and Tony Northrup addressed these questions in a video ...

Do yourself a favor and skip the video. Northrup has taken a previous mistaken assertion about the relationship between crop factor and aperture and doubled down by applying the same faulty logic to ISO. Despite all of his half-truths and pseudo-scientific explanations, he simply does not know what he is talking about.

A crop sensor is just that. It crops the image projected by the lens. This is obvious to anyone who has a camera with an FX/DX switch. The only thing that happens when you flip the switch is that you crop the image on the sensor. It has no effect on ISO, aperture, the lens-to-sensor distance or anything else. You get exactly the same result if you take an FX image and crop it later in post processing.

However, what you do with the resulting image after cropping it does matter. When you enlarge it you will change a number of things such as the apparent depth of field, the visibility of noise, camera shake, lens aberrations, subject movement, etc. These are all geometric consequences with which we should all be familiar.

| Reply
May 23, 2014 06:59:10   #
amehta Loc: Boston
 
I think this whole issue is about language, not optics.

For example, the Panasonic FZ200 with a 4.5-108mm f/2.8 lens is pretty impressive. But he Panasonic web site says, "WorldÂ’s First 25-600mm Lens with Full-Range F2.8 Aperture", implying that it is a "25-600mm f/2.8" as B&H puts it. That is the lie.

If we never talked about crop factor, then none of this would matter. We would talk about the focal length of a lens, and the field of view with a particular sensor. We would talk about the maximum aperture of a lens, and the depth of field we could get with a particular sensor. And we would talk about the ISO of a sensor and the noise level. But after a while, that becomes a lot more trouble than it's worth, and we would take short cuts, ending up someplace with as much confusion and inaccuracy as we have now.

| Reply
If you want to reply, then register here. Registration is free and your account is created instantly, so you can post right away.
UglyHedgehog.com - Forum
Copyright 2011-2019 Ugly Hedgehog, Inc.