Question to those who moved from crop sensor to full frame camera-low light Perf

[shadowlands]

Just do it. It's worth it. I'm in love with FX already, and I've only had it for 4 days...

I hear ya!

Noise looks roughly the same on my D700 at ISO 6400 as my D300 does at ISO 1600.

 

[MSnowy]

I had the opportunity to shoot a D3S today. The speed was pretty sweet tFlickr

So for now it will be the D7100. Hopefully next month will bring a new DX king...

Wow those flight shots are nice. I see they were shot at iso 2800 which let you shot at 1/2000 f 7.1

 

[djacobox372]

I just ran a test in a semi-dark room.
using the same 50mm f/1.8 AF-D lens

My nikon d7000
f/4
ISO 100
for proper exposure --> 30 seconds

My Nikon d600
f/4
ISO 100
for proper exposure --> 15 second

pics looks about the same, except the longer exposure is a bit more blurry becz I wasn't expecting to hold it still for so long.

This doesn't make any sense how in the heck can the pics look the same when one is a stop more exposed.

 

[ruifo]

This doesn't make any sense how in the heck can the pics look the same when one is a stop more exposed.

The sensors have different size... That's why.
The full frame sensor collects more light than the crop sensor, and needed 1 stop less light to get the same exposure.

 

[480sparky]

The sensors have different size... That's why.
The full frame sensor collects more light than the crop sensor, and needed 1 stop less light to get the same exposure.

Huh?

You mean full-frame users need to learn the Sunny 11 rule?

 

[ruifo]

No.
Sunny 16 for FF, as always.

 

[480sparky]

No.
Sunny 16 for FF, as always.

Then your statement about FF sensors collecting more light and need one less stop of exposure doesn't make sense.

 

[ruifo]

No.
Sunny 16 for FF, as always.

Then your statement about FF sensors collecting more light and need one less stop of exposure doesn't make sense.

Let's put it the other way to make it easier. Reversed thinking.

The FF 35mm is a long lasting size, since the film era. It always needed this specific light. Like the f/16 sunny rule. That size is today reproduced by the FF sensors.

With the popularization of croped sensors, with the digital era, the smallest sensors needed more light to reproduce the exact same scene of the full frame big brother, that all were used to with the 35mm film/sensor. So it is the crop sensor who needs more light, not the full frame brother.

The "sunny 11" you played with means one more stop of light to the full frame, not one stop less. One less would make it "sunny 22", and in a old school 35mm film, that would be the "snow/sand 22", if I may say so. Who needs more light is the croped smallest sensor, not the big FF one, that continues with th "sunny 16" as always.

And the sunny 16 thing is just a starting point, no more.

Got it?

But I have to admit that many people have different opnion about that debate, and I may be wrong. That was debated a lot right after the release of the first digital croped sensors. Not sure, but that's my feeling and experience so far. I'm happy to be convinced of the opposite.

 

[coastalconn]

I had the opportunity to shoot a D3S today. The speed was pretty sweet tFlickr

So for now it will be the D7100. Hopefully next month will bring a new DX king...

Wow those flight shots are nice. I see they were shot at iso 2800 which let you shot at 1/2000 f 7.1

Thanks Michael! I shoot in Manual with Auto-ISO so I picked my SS and Aperture. I was trying to get the ISO up there so I could see real life samples...

 

[Derrel]

The sensors have different size... That's why.
The full frame sensor collects more light than the crop sensor, and needed 1 stop less light to get the same exposure.

Huh?

You mean full-frame users need to learn the Sunny 11 rule?

Companies often "fudge" the ISO ratings a bit, to earn betetr scores on DxO mark and other test sites, since NOISE is what the sites usually measure and report on, so if say, ISO 6400 is actually ISO 5,000, it makes the noise stats skewed a bit in favor of the camera. There's also some differences in how the data off the sensor is processed, and how the black point is set, and also maybe what type of curve has been applied to the raw data. The slightest change in the way the raw data is output can make the image look "brighter", by quite a lot. The SUnny 16 or Sunny 11 rule ONLY applies if the nominal ISO is in 100% accord with the ACTUAL ISO level, and looking at DxO mark, one can see--that is not usually the case at the higher ISO values. Also, not all "raw" files are truly "raw"...the black point can be set, and it makes the images look different. I've also seen a couple article mentioning that sensors do not respond quite linearly above f/2.8, and that wide-wide,wide apertures on d-slrs seem not to give the FULL, reciprocal exposure one woukd expect, and f/4 is just one stop below f/2.8, so even if the loss is say 1/3 stop due to wide aperture, or the way the sensor collects the light, and the ISO maybe being fudged by a bit, or the way the NEWER, BTETR FX sensor with much highee quantum efficiency handles light in longer exposures, YES, it's esily possible that in the real world, an 80-year old film "rule" no longer is always accurate in actual,practical application.

I have pulled out a couple of wheel-hubs-deep-stuck 250-horsepower V-8 pickup trucks down in the creek-side gravel with a 1949 John Deer Model M tractor with a 100 cubic inch 2-cylinder gas engine which produced a mere 18 drawbar horsepower when it was NEW...and this was 30 years later when it was tired... Statistics don't mean a lot sometimes...

1949 Model M http://johnnypopper.com/jdgifs/mseylar.jpg

 

[480sparky]

Let's put it the other way to make it easier. Reversed thinking.

........So it is the crop sensor who needs more light, not the full frame brother.
.......
Got it?........

Nope.

I cannot fathom why you think the size of the sensor has anything to do with exposure.

So do cell phones with their itty-bitty pencil - lead sized sensors need a radically different exposure than a medium format Hasselblad?

 

[Derrel]

Sparky, I too have seen this "size of the sensor" determines the needed exposure level line of thinking. It seems very odd that people think this. I think it's much more likely that the way the various sensors and readout electronics and the way the differing camera makers and different models have been actually programmed to output their sensor data is "cooking" the data somewhat. Even the slightest tweak to the tone curve can make the image appear "brighter", and if there's any fudging of the real sensitivity level and the stated level, that adds into the mix.

 

[ruifo]

Interesting points, Derrel.
I never of that, this way.

 

[Tailgunner]

You may have seen my thread last week, because I was teetering on the edge of buying a D800. The general consensus for how I shoot and what I shoot(birds) was there was not a huge advantage for me to go FF since I'm always cropping. My biggest problem is all the comparisons show the comparisons with different focal lengths. Shooting at 600mm is 600mm so I would always be cropping the D800 so much that the noise difference wouldn't be huge. For the most part with careful noise reduction, I can still do OK at ISO 6400 and still get use-able 8x10 prints or maybe slightly bigger. This was what I could produce at 6400 on the D7100... But it isn't always so good, you really have to nail exposure...
;Typical George 7_15 by krisinct- Thanks for 2! Million + views!, on Flickr

VS this on a D600 at 6400.. which I didn't even have enough DOF to get the whole head in focus..
lJuvi Red Tail Hawk 5_29 4 by krisinct- Thanks for 2! Million + views!, on Flickr

Nice work, I knew if could fully test the D7100 it would be you.

Anyhow, I really hate to suggest the D810 and it not work out. That is a lot of coin for a body but I'm starting to think it maybe a good fit. I mean cropping at 1.2x still gives you 18mp and from what I hear about the new Expeed 4 processor, it's supposed to be really quick...originally designed for the D4. Then may install a grip and bump the frame rate up to 6-7 fps. It could be a good deal. I would still probably rent one to see how it works first.

 

[coastalconn]

You may have seen my thread last week, because I was teetering on the edge of buying a D800. The general consensus for how I shoot and what I shoot(birds) was there was not a huge advantage for me to go FF since I'm always cropping. My biggest problem is all the comparisons show the comparisons with different focal lengths. Shooting at 600mm is 600mm so I would always be cropping the D800 so much that the noise difference wouldn't be huge. For the most part with careful noise reduction, I can still do OK at ISO 6400 and still get use-able 8x10 prints or maybe slightly bigger.

Nice work, I knew if could fully test the D7100 it would be you.

Anyhow, I really hate to suggest the D810 and it not work out. That is a lot of coin for a body but I'm starting to think it maybe a good fit. I mean cropping at 1.2x still gives you 18mp and from what I hear about the new Expeed 4 processor, it's supposed to be really quick...originally designed for the D4. Then may install a grip and bump the frame rate up to 6-7 fps. It could be a good deal. I would still probably rent one to see how it works first.

I think the D810 would be an OK fit, but I can't justify the price. I do have a show opening tomorrow and running for 2 months at a nature center, maybe if I sell a ton of stuff, who knows, lol. I'm still holding my breath for September to see what gets announced... Being a dedicated birder really calls for a certain camera and Nikon has just refused to produce it so far. But after shooting the D3S, I am more impressed by the D7100 considering the price to performance ratio, lol...