FX (Crop) VS DX

[Timppa]

Hello,

I know the differences between FX and DX, I am also aware that when you use a FX camera in DX mode, you are cropping the image, resulting in less pixels.

But here is my question;
a FX sensor is known for better high ISO performance. According to what I found on the internet, it can result up to 3 stops of light (ISO 100 on DX = ISO 800 on FX in terms of amount of grain --> this formula is definitely not correct and also depends on type of sensor and many other things, but it's a rough comparison).
When cropping a FX down to a DX, does this benefit remain?

Would it be better to use a 46mp FX camera, cropped down to DX mode (resulting an image around 20mp), or would it be better to use a 20mp DX camera? Will I be able to use the FX camera at an ISO roughly 3 stops higher than the DX and remain the same amount of grain?
Or will the results be exactly the same?

Thanks in advance!

I am just wondering in general if it would be better to get 2 cameras (a DX and FX around 20-24mp) or just 1 high mp fx camera.
I know there are many other factors in this decision, but I would just like an answer on the above question.

 

[Braineack]

Would it be better to use a 46mp FX camera, cropped down to DX mode (resulting an image around 20mp), or would it be better to use a 20mp DX camera? Will I be able to use the FX camera at an ISO roughly 3 stops higher than the DX and remain the same amount of grain?
Or will the results be exactly the same?

Quite possibly on all accounts. The D850 is a better imager than the D500 -- but the pixel pitch is almost the same size.


4.35 microns vs. 4.22 microns, whereas my d610 is 5.97 microns

However, if you look at the DXO mark results, somehow Nikon is able to keep the D850 on par with the D610 in terms of DR, color sensetivity, tonal range, and SNR. That's what makes it somewhat magical -- like shooting between my D800 and D610. Even though the pixel size is much smaller in the 36MP sensor, I achieve identical results in terms of image quality, but with the added bonus of extra resolution and image detail. I've posted these results here once.

 

[Ysarex]

Hello,

I know the differences between FX and DX, I am also aware that when you use a FX camera in DX mode, you are cropping the image, resulting in less pixels.

But here is my question;
a FX sensor is known for better high ISO performance. According to what I found on the internet, it can result up to 3 stops of light (ISO 100 on DX = ISO 800 on FX in terms of amount of grain --> this formula is definitely not correct and also depends on type of sensor and many other things, but it's a rough comparison).
When cropping a FX down to a DX, does this benefit remain?

Basically no. The fundamental reason an FX sensor provides better low-light performance is it's bigger. If you crop it so that it's no longer bigger then you're giving up the reason for the advantage.

Because other factors (different sensor architectures, etc.) play secondary roles you can't just apply that evenly to any two FX/DX cameras. The best low-light performance crop camera these days will be very close to a medium performance FF camera. The best way to think about it is this: Crop an FX camera to DX size and the camera's low-light performance will be downgraded accordingly.

Joe

 

[Braineack]

Basically no. The fundamental reason an FX sensor provides better low-light performance is it's bigger. If you crop it so that it's no longer bigger then you're giving up the reason for the advantage.

You're stuck in the past, and this is simply flawed logic.

The best way to think about it is this: Crop an FX camera to DX size and the camera's low-light performance will be downgraded accordingly.

Again, this is not a good rule-of-thumb anymore. Techonology has caught up to 1995.

 

[Ysarex]

Basically no. The fundamental reason an FX sensor provides better low-light performance is it's bigger. If you crop it so that it's no longer bigger then you're giving up the reason for the advantage.

You're stuck in the past, and this is simply flawed logic.

Then the FX advantage comes from what?

Joe

 

[Braineack]

4.35 microns ( expeed 5 ) vs. 4.22 microns ( expeed 5 ) vs. 5.97 microns ( expeed 3 ). Image processors > pixel pitch.

 

[Ysarex]

4.35 microns ( expeed 5 ) vs. 4.22 microns ( expeed 5 ) vs. 5.97 microns ( expeed 3 ). Image processors > pixel pitch.

View attachment 190050

View attachment 190052

View attachment 190053

View attachment 190054

I said other factors like sensor architecture play a role. There's five years of tech advance between the D610 and the D850. You can't compare the two and control other variables. Compare a D850DX with a D500 (one year apart) -- here's Bill Claff's DR comparison chart:

Photographic Dynamic Range versus ISO Setting

And the D500 DR performance is slightly better.

The D850's performance drops when you switch it from FX to DX. What causes that drop?

Joe

 

[RVT1K]

Basically no. The fundamental reason an FX sensor provides better low-light performance is it's bigger. If you crop it so that it's no longer bigger then you're giving up the reason for the advantage.

You're stuck in the past, and this is simply flawed logic.

Then the FX advantage comes from what?

Joe

It has more to due with the size of the individual elements on the sensor, not the size of the sensor itself.
The larger sensor has, typically, allowed larger individual elements a.k.a. pixels.

This comes up a lot and in reality it is an electronics/semiconductor question more than it is a camera question.

 

[Ysarex]

Basically no. The fundamental reason an FX sensor provides better low-light performance is it's bigger. If you crop it so that it's no longer bigger then you're giving up the reason for the advantage.

You're stuck in the past, and this is simply flawed logic.

Then the FX advantage comes from what?

Joe

It has more to due with the size of the individual elements on the sensor, not the size of the sensor itself.
The larger sensor has, typically, allowed larger individual elements a.k.a. pixels.

This comes up a lot and in reality it is an electronics/semiconductor question more than it is a camera question.

No, it has more to do with the size of the sensor and not the size of the pixels/sensels on the sensor:
The effect of pixel size on noise
The effect of pixel size on noise

Joe

 

[Braineack]

The D850's performance drops when you switch it from FX to DX. What causes that drop?

Joe

you've zoomed in. which brings us back to the OP's question.

 

[Ysarex]

The D850's performance drops when you switch it from FX to DX. What causes that drop?

Joe

you've zoomed in. which brings us back to the OP's question.

Zoomed in causes the drop in performance? In other words smaller sensor. Why is smaller sensor less performance?

Joe

 

[Braineack]

no. sensor sizes dont change.

 

[Ysarex]

no. sensor sizes dont change.

When you throw the switch in the D850 from FX to DX you use less of the sensor -- that = smaller. No question or argument the performance drops. What causes the performance drop? Zooming in causes the performance drop? How is that any different than using a smaller sensor?

Joe

 

[Braineack]

persception.

Let's frame it this way:

Is there a differnece between using DX mode, or taking a full RAW image and cropping it to 20MP in post?

 

[Ysarex]

persception.

Let's frame it this way:

Is there a differnece between using DX mode, or taking a full RAW image and cropping it to 20MP in post?

So if you crop in post the performance drop that happens when you switch FX to DX doesn't occur?

Joe