High Resolution Or Low Light Performance?

[Batdude]

Guys, there is no way the 40MP sensor is going to do miracles when it comes to high ISO performance. Is just physics. Is the XH2S doing miracles at 26MP? No, so why are some already guessing that the 40MP is going to have "improved" high ISO performance?

So in reality, who needs what? Do you need a lot of resolution to make prints of the size of the empire state building or for cropping birds heavily, or high ISO performance? You can't get both.

It looks like the trend is resolution, and this seems to keep going and going. I'm looking at the images from the 26MP sensor and darn it man, I don't need more than that. I actually need a sensor with top notch low light performance. I don't need more than 20MP that's the truth.

I wonder what Fuji is going to do with the XT5 and so forth but it looks like the camera industry has brain washed many making them think they "must have" that high resolution sensor. So the question to me is will Fuji ever make a low light top of the line low light performer? What's up with that?

And don't tell me there is software for that. I have no desire to add more extra steps and time wasted to my existing workflow which is in fact already time consuming. Some things are just starting not to make any sense sorry.

 

[a_c_skinner]

You've moved on from things you are disappointed with to things you anticipate you will be disappointed with!

I'm sure you are correct but frankly performance at high ISOs is so good few people will be pressing low light performance to the point where this is going to be troublesome.

 

[Jerry-astro]

Guys, there is no way the 40MP sensor is going to do miracles when it comes to high ISO performance. Is just physics. Is the XH2S doing miracles at 26MP? No, so why are some already guessing that the 40MP is going to have "improved" high ISO performance?

So in reality, who needs what? Do you need a lot of resolution to make prints of the size of the empire state building or for cropping birds heavily, or high ISO performance? You can't get both.

It looks like the trend is resolution, and this seems to keep going and going. I'm looking at the images from the 26MP sensor and darn it man, I don't need more than that. I actually need a sensor with top notch low light performance. I don't need more than 20MP that's the truth.

I wonder what Fuji is going to do with the XT5 and so forth but it looks like the camera industry has brain washed many making them think they "must have" that high resolution sensor. So the question to me is will Fuji ever make a low light top of the line low light performer? What's up with that?

And don't tell me there is software for that. I have no desire to add more extra steps and time wasted to my existing workflow which is in fact already time consuming. Some things are just starting not to make any sense sorry.

Right. So given your knowledge, insight, and expertise in sensor development, please explain why you think that Fujifilm's engineers are leaving IQ, DR, etc. on the table. What technology improvements that you are aware of are not being utilized? The latest technology (X-H2S) is all about speed and improved AF, tracking, etc. If that's not enough and you want to squeeze major IQ improvements out of your sensor, then you'd better get serious about moving to FF or MF.

So, it stands to reason that recent improvements are more about resolution. You might not need it, but I'm guessing that others will enjoy taking advantage of the ability to print big with higher detail definition and also to have a greater ability to crop while still making decent sized prints. If that doesn't float your boat, then you'd probably better start looking at going FF or MF, as I suspect that huge gains in IQ for crop format sensors are not in the offing.

Bottom line: If you feel that other vendors offering cameras with crop format sensors are accomplishing that, then please, by all means, share your data along with some examples. I'm more than happy to be proven wrong here. Otherwise, you have a choice to make. If you truly feel limited by crop format sensors, then you have little choice but moving to FF or MF. Complaining here won't make it happen.

--
Jerry-Astro
Fuji Forum co-Mod

 

[Erik Baumgartner]

There’s no reason the 40MP sensor shouldn’t be able to manage better high ISO performance if they keep the readout speed reasonable. Yes, a 40MP sensor will have higher per pixel noise, but there is no reason it can’t match or improve upon the current sensors at at the same viewing size (as going from 16MP to 24/26MP did). With greater pixel density it should be able to significantly reduce moire issues and X-Trans demosaicing/interpolation errors for cleaner detail (both luminance and color) which should improve image quality with any lens. If they can manage to keep the noise at X-Trans III levels, I’m all for it.

 

[forest dream]

There’s no reason the 40MP sensor shouldn’t be able to manage better high ISO performance if they keep the readout speed reasonable. Yes, a 40MP sensor will have higher per pixel noise, but there is no reason it can’t match or improve upon the current sensors at at the same viewing size (as going from 16MP to 24/26MP did).

Both X-H2 and OM-1 is BSI-CMOS. Are X-H2 similar or less DR than MFT camera?

• OM System OM-1 20MP Stacked BSI-CMOS MFT sensor, pixel size=3.36µm
• Fujifilm X-H2 40MP BSI-CMOS APS-C sensor, pixel size=3.03µm

 

[John Gellings]

If you don’t need to print more than 20mp sizes… say 13x19” at 100% / 300dpi, then a 40mp sensor will allow you to print at smaller percentages at 300dpi making your high iso look better.

 

[forest dream]

If you don’t need to print more than 20mp sizes… say 13x19” at 100% / 300dpi, then a 40mp sensor will allow you to print at smaller percentages at 300dpi making your high iso look better.

Agree that "print at smaller percentages at 300dpi making your high iso look better".

When print at 300dpi, which sensor (26MP / 40MP) produce better IQ for high ISO? Or both sensor produce similar IQ at 300dpi?

• High resolution allow more crop later
• High resolution produce bigger RAW file.
• High resolution usually has lower e-shutter fps.

 

[Erik Baumgartner]

There’s no reason the 40MP sensor shouldn’t be able to manage better high ISO performance if they keep the readout speed reasonable. Yes, a 40MP sensor will have higher per pixel noise, but there is no reason it can’t match or improve upon the current sensors at at the same viewing size (as going from 16MP to 24/26MP did).

Both X-H2 and OM-1 is BSI-CMOS. Are X-H2 similar or less DR than MFT camera?

• OM System OM-1 20MP Stacked BSI-CMOS MFT sensor, pixel size=3.36µm
• Fujifilm X-H2 40MP BSI-CMOS APS-C sensor, pixel size=3.03µm

Overall dynamic range and noise has to do with sensor size, not pixel size. The significantly larger APS-C sensor will always come out on top in this regard.

 

[Typoz]

Both X-H2 and OM-1 is BSI-CMOS. Are X-H2 similar or less DR than MFT camera?

• OM System OM-1 20MP Stacked BSI-CMOS MFT sensor, pixel size=3.36µm
• Fujifilm X-H2 40MP BSI-CMOS APS-C sensor, pixel size=3.03µm

Pixel size has no relation to dynamic range. The only thing that matters is sensor size and sensor tech (well depth, BSI, and so on). That myth has been debunked so many times already... For instance compare the Sony A7S III (12MP) to A7R IV (62MP) and A7 III (24MP): the A7S III with huge pixels actually has worse dynamic range... https://www.photonstophotos.net/Charts/PDR.htm#Sony ILCE-7M3,Sony ILCE-7RM4,Sony ILCE-7SM3

Pixel size also does not/barely impact low-light performance, especially with BSI sensors. (

, https://www.dpreview.com/articles/5559775087/choosing-a-camera-should-i-worry-about-pixel-size))

The only drawbacks of higher resolution are the increased file size and the increased readout time, leading to rolling shutter when using electronic shutter (no effect when using mechanical shutter).

I would love to have a 100 MP sensor. Lots of cropping potential if the lenses are sharp enough (in the center at least).

Guys, there is no way the 40MP sensor is going to do miracles when it comes to high ISO performance. Is just physics. Is the XH2S doing miracles at 26MP? No, so why are some already guessing that the 40MP is going to have "improved" high ISO performance?

No sensor is going to improve low light performance in the future. It's just the lack of photons that is the limit now. That's the physics reason (not the pixel size...).

So in reality, who needs what? Do you need a lot of resolution to make prints of the size of the empire state building or for cropping birds heavily, or high ISO performance? You can't get both.

You can have both. Because pixel size does not impact high ISO performance...

 

[Morris0]

You've moved on from things you are disappointed with to things you anticipate you will be disappointed with!

I'm sure you are correct but frankly performance at high ISOs is so good few people will be pressing low light performance to the point where this is going to be troublesome.

He is simply not as entertaining with out Robin.

Morris

 

[Morris0]

If you don’t need to print more than 20mp sizes… say 13x19” at 100% / 300dpi, then a 40mp sensor will allow you to print at smaller percentages at 300dpi making your high iso look better.

Some people crop :-}

Morris

 

[DJF55]

There’s no reason the 40MP sensor shouldn’t be able to manage better high ISO performance if they keep the readout speed reasonable. Yes, a 40MP sensor will have higher per pixel noise, but there is no reason it can’t match or improve upon the current sensors at at the same viewing size (as going from 16MP to 24/26MP did). With greater pixel density it should be able to significantly reduce moire issues and X-Trans demosaicing/interpolation errors for cleaner detail (both luminance and color) which should improve image quality with any lens. If they can manage to keep the noise at X-Trans III levels, I’m all for it.

My understanding is that keeping the readout speed reasonable is not today's flavor. I have read most posts on this forum regarding the X-H2s / X-H2, and I see no reason at this stage to upgrade my X-H1, for my stills photography. To tempt me to upgrade to a new model down the line, besides what I written elsewhere on this forum, my criteria would include exactly that: reasonable readout for better ISO performance (incl. min. ISO of 100 or even 64) in comparison to the X-H1. Besides of course better DR. Thanks to Fuji, my X-H1 will last me many years.

 

[John Gellings]

If you don’t need to print more than 20mp sizes… say 13x19” at 100% / 300dpi, then a 40mp sensor will allow you to print at smaller percentages at 300dpi making your high iso look better.

Agree that "print at smaller percentages at 300dpi making your high iso look better".

When print at 300dpi, which sensor (26MP / 40MP) produce better IQ for high ISO?

Well, that depends on the sensor right? But if you are printing both at 300dpi and at 100%, the 26mp sensor most likely will do better. Also, the 40mp print will be bigger.

Or both sensor produce similar IQ at 300dpi?

The idea is this... if we print a 26mp print at 300dpi at 100% and a 40mp print at the same size, you are printing the 40mp file at like 70% of max res. That will hide some noise. So basically, if you do not print need to print your high ISO photos at 300dpi and 100%, you can have great high ISO and high resolution with a high MP sensor. People tend not to think of that.

• High resolution allow more crop later
• High resolution produce bigger RAW file.
• High resolution usually has lower e-shutter fps.

Ok...

--
https://www.johngellings.com
Instagram = @johngellings0

 

[John Gellings]

If you don’t need to print more than 20mp sizes… say 13x19” at 100% / 300dpi, then a 40mp sensor will allow you to print at smaller percentages at 300dpi making your high iso look better.

Some people crop :-}

But that is not what we are talking about...

 

[Morris0]

If you don’t need to print more than 20mp sizes… say 13x19” at 100% / 300dpi, then a 40mp sensor will allow you to print at smaller percentages at 300dpi making your high iso look better.

Some people crop :-}

But that is not what we are talking about...

Oh, crops are not susceptible to noise? It's the same discussion!

Morris

 

[Batdude]

There’s no reason the 40MP sensor shouldn’t be able to manage better high ISO performance if they keep the readout speed reasonable. Yes, a 40MP sensor will have higher per pixel noise, but there is no reason it can’t match or improve upon the current sensors at at the same viewing size (as going from 16MP to 24/26MP did).

Both X-H2 and OM-1 is BSI-CMOS. Are X-H2 similar or less DR than MFT camera?

• OM System OM-1 20MP Stacked BSI-CMOS MFT sensor, pixel size=3.36µm
• Fujifilm X-H2 40MP BSI-CMOS APS-C sensor, pixel size=3.03µm

Overall dynamic range and noise has to do with sensor size, not pixel size. The significantly larger APS-C sensor will always come out on top in this regard.

Please forgive my question, when you say "the significantly larger APS-C sensor" do you mean the (more resolution)?

 

[Erik Baumgartner]

There’s no reason the 40MP sensor shouldn’t be able to manage better high ISO performance if they keep the readout speed reasonable. Yes, a 40MP sensor will have higher per pixel noise, but there is no reason it can’t match or improve upon the current sensors at at the same viewing size (as going from 16MP to 24/26MP did).

Both X-H2 and OM-1 is BSI-CMOS. Are X-H2 similar or less DR than MFT camera?

• OM System OM-1 20MP Stacked BSI-CMOS MFT sensor, pixel size=3.36µm
• Fujifilm X-H2 40MP BSI-CMOS APS-C sensor, pixel size=3.03µm

Overall dynamic range and noise has to do with sensor size, not pixel size. The significantly larger APS-C sensor will always come out on top in this regard.

Please forgive my question, when you say "the significantly larger APS-C sensor" do you mean the (more resolution)?

No, the physical size of the sensor. Dynamic range and noise (given similar tech) is primarily determined by sensor size, not pixel size/resolution.

 

[John Gellings]

If you don’t need to print more than 20mp sizes… say 13x19” at 100% / 300dpi, then a 40mp sensor will allow you to print at smaller percentages at 300dpi making your high iso look better.

Some people crop :-}

But that is not what we are talking about...

Oh, crops are not susceptible to noise? It's the same discussion!

Morris

Of course they are, but it wasn’t what I was talking about.

 

[sluggy_warrior]

And don't tell me there is software for that. I have no desire to add more extra steps and time wasted to my existing workflow which is in fact already time consuming. Some things are just starting not to make any sense sorry.

Do you want them to cram a whole power-hungry GPU or TPU with real-time constraint into the camera to address noise? Or would you prefer shifting that heavy task of denoising to your computer and leave your camera doing its best at shooting?

Although, given they all ready have AI/neural-network processor implemented for object detection, I suspect it won't take long for them to add a neural-network denoiser that utilizes the same processor. It can be trained specifically for that particular sensor, so the power/memory consumption can be minimized. Obviously, will consumers swallow the additional cost for better NR?

Anyway, everything is a compromise, there's no perfect solution. Adjusting your workflow would be the easiest option. For example, I can edit 2,000+ RAF files in two days (without applying denoise or sharpening), then let the time-consuming CPU/GPU-heavy export process (with neural-network denoiser and RL-deblur) runs overnight. Everything is ready the next morning when I wake up. There's no dead/waiting time to me.

 

[forest dream]

And don't tell me there is software for that. I have no desire to add more extra steps and time wasted to my existing workflow which is in fact already time consuming. Some things are just starting not to make any sense sorry.

Do you want them to cram a whole power-hungry GPU or TPU with real-time constraint into the camera to address noise? Or would you prefer shifting that heavy task of denoising to your computer and leave your camera doing its best at shooting?

Although, given they all ready have AI/neural-network processor implemented for object detection, I suspect it won't take long for them to add a neural-network denoiser that utilizes the same processor. It can be trained specifically for that particular sensor, so the power/memory consumption can be minimized. Obviously, will consumers swallow the additional cost for better NR?

If camera built-in DeepPrime like denoising engine (few seconds via hardware acceleration from FPGA processor), native support demosaiced and denoised DNG raw/JPEG, allow direct print Film Simulation photo to Instax Mini/Wide Link printer....

Consumer allow only demosaiced and denoised SOOC JPEG. Consumer can review & adjust (via front/rear command dials) photo when photography, if not satisfied denoised result, can retake again. What You See Is What You Get.

Are it considered improve IQ for SOOC JPEG (Film Simulation)?

Are it considered improve workflow for RAW processing?