Reading mode: Light Dark

5D mk III: DR hysteria

Go to last post
etto72

etto72

Forum Enthusiast
Messages
410
Reaction score
82
Location
NY, US
There are lots of folks in this forum that are saying that the DR of the 5D III has similar values has the model that replaces ..... creating lots disappointing feelings

I would like everyone to make a simple consideration.....

Looking at the last 4 Canon sensor generation

-1Ds mk II, 5D,

-1Ds mk III,1D mk III,5D mk II, 40D

-1D mk IV, 7D

-1DX, 5D mk III

Lets look at the pixel pitch and at the equivalent DR values for each generation:

5D Pixel size 8.2 = 9.2 DR

1DS mk III Pixel size 6.4 = 11.3
5D mk II Pixel size 6.4 = 11.1
40D Pixel size 5.7 = 11

7D Pixel size 4.3 = 9.8
1D mk IV Pixel size 5.7 = 12

1DX Pixel size 6.9 = ?
5D mk III Pixel size 6.25 = ?

As you can see each generation gave an increase in DR except with the 7D wich has an extremely smaal 4.3 pixel pitch size

Last generation sensor gave a 12 EV with 5.7 pixel size

The mk III has 6.25 pixel size

Even if Canon didn't progress in DR with the newest sensor engineering...

..... I have really hard time to believe that with the 5d mk III with 6.25 they can't get same or better DR results with a sensor that has 5.7 !!!!

Time will tell....

Thanks
Ettore
 
I really don't trust all of these DR measurements. (not just from you but from everyone)

I mean, the 40D has more DR than the 5Dc? That's a joke. 5Dc dynamic range blew the 40D out of the water and was probably the primary reason for me to go full frame back in the day.

I will get a 5D3 and I will take actual photos with it and I will see what the DR is. The 5D2 didn't have much real world improvement over the 5D in my opinion. But again, that's not what the DXOmark number show.
 
agree that dr number comparisons appear suspect

and what people actually mean is better IQ....they just call it dr because they reckon you can measure that... wrongly!

so 5d3 having same dr as 5d2 means to most people that you won't see any IQ change worth having.... time will tell.
 
expro said:
agree that dr number comparisons appear suspect

and what people actually mean is better IQ....they just call it dr because they reckon you can measure that... wrongly!
Click to expand...
IQ is a general quality, a term whose definition is comprised of several elements: color depth, dynamic range as measured in EV steps, and amount of noise as ISO increases in stops.

So if DR is just one element of IQ, how do you figure what people really mean when they say DR is really IQ?
expro said:
so 5d3 having same dr as 5d2 means to most people that you won't see any IQ change worth having.... time will tell.
Click to expand...
But people are already seeing an overall IQ improvement over the 5D 2. High ISO seems to have less noise. But one area where the 5D Mk II's fairly thin DR is in video. There is very little exposure latitude, and you really need to nail it, almost like shooting reversal film as opposed to color negative. I know many videographers, and photographers, for that matter, who are hoping for an improvement in this area.
 
Randplaty said:
I really don't trust all of these DR measurements. (not just from you but from everyone)

I mean, the 40D has more DR than the 5Dc? That's a joke. 5Dc dynamic range blew the 40D out of the water and was probably the primary reason for me to go full frame back in the day.
Click to expand...
...what DR is. People misuse and misinterpret the measure just like they misuse the term "bokeh" to mean how blurred the background is, or people thinking how changing the ISO changes the exposure, etc., etc., etc.

That's the difficulty with technical discussions here on DPR -- people using terms without actually understanding what they mean. Of course, what else would one expect? Still...
Randplaty said:
I will get a 5D3 and I will take actual photos with it and I will see what the DR is. The 5D2 didn't have much real world improvement over the 5D in my opinion. But again, that's not what the DXOmark number show.
Click to expand...
...as I said...
 
your numbers are off, some are way off.

5d1 = 11.1 stop DR. (30D had 10.8)
http://www.dxomark.com/index.php/Cameras/Camera-Sensor-Database/Canon/EOS-5D

40d arriving 2 years later had 11.3
http://www.dxomark.com/index.php/Cameras/Camera-Sensor-Database/Canon/EOS-40D

5d2 arrive another year later had 11.9

http://www.dxomark.com/index.php/Cameras/Camera-Sensor-Database/Canon/EOS-5D-Mark-II

that same year 50D had 11.4

7D arrived another year later had 11.7 and 1D4 had 12.

what you can see is a a steady but very slow progress. In aps-c, it is averages about 0.2 stop per generation of sensor. for larger sensor it is about 0.1 stop per year.

But we now know that DR has little to do with pxiel size, it is more to do with pixel quality and the quality of electronics, the higher grade the less contamination the higher DR. This is why 7D test shows slightly more DR than 60D, despite having (more or less)the same sensor and one year older.
 
correction to the post above, it should read average of 0.3 stop improvement per generation for APS-C.

but then again if we compare 50D to 60d, the number is back to average 0.2 stop per generation.
 
there are many ways to calculate DRs and with the same DR cameras may perform differently so it's really complicated and confusing.

if anyone wants a "one phrase answer" I think the SNR-print at DxOMark should be the nearest one, though there are issues like RAW cooking in some cameras (like Nikon, Sony, and Pentax). after all the SNR is the most important thing that when people talk about IQ, usually they mean SNR.
 
Dynamic range is determined by full well capacity and read out noise not other things mentioned.

Full well capacity is a result of pixel size and voltage.
Readout noise is from the sensor.

So if smaller pixels generated better DR that means readout noise was improved.

Readout noise is something you often see in CCDs that is improved from one generation to another.

So if mark iii DR is the same it as Mark ii it means readout noise is slightly less as pixel size is slightly less so full well capacity (how much charge the pixel can hold before it is saturated) must be slightly less.

So if high ISO performance in RAW is improved a lot over ii whilst DR is the same or lower as posted in another post (it may not be accurate) then that implies Canon is no longer pure RAW but applying smoothing to RAWs like Nikon used to (may still do).

Greg.
 
I also may add these links to similar RAW photos from Mark iii and D800.

Hard to tell exactly from the D800 as the image is so small but it would appear at ISO6400 images are similar.

Mark iii seems good up to about ISO10,000 then starts to degrade and show fixed pattern noise at 12800.

I don't have a mark ii but perhap others who do can comment about whether they think their mark ii could match these mark ii RAWs?

Greg.
 
Great Bustard said:
Randplaty said:
I really don't trust all of these DR measurements. (not just from you but from everyone)

I mean, the 40D has more DR than the 5Dc? That's a joke. 5Dc dynamic range blew the 40D out of the water and was probably the primary reason for me to go full frame back in the day.
Click to expand...
...what DR is. People misuse and misinterpret the measure just like they misuse the term "bokeh" to mean how blurred the background is, or people thinking how changing the ISO changes the exposure, etc., etc., etc.

That's the difficulty with technical discussions here on DPR -- people using terms without actually understanding what they mean. Of course, what else would one expect? Still...
Great Bustard said:
I will get a 5D3 and I will take actual photos with it and I will see what the DR is. The 5D2 didn't have much real world improvement over the 5D in my opinion. But again, that's not what the DXOmark number show.
Click to expand...
...as I said...
Click to expand...
Okay enlighten us then and explain to me why the 40D cannot handle backlit photos whereas the 5Dc could. I'm guessing that has nothing to do with DR?
 
Astrophotographer 10 said:
So if high ISO performance in RAW is improved a lot over ii whilst DR is the same or lower as posted in another post (it may not be accurate) then that implies Canon is no longer pure RAW but applying smoothing to RAWs
Click to expand...
Not necessarily, depending on how you define high ISO performance. High ISO performance could also be improved by increased quantum efficiency, which has nothing to do with DR.
 
etto72 said:
Those values are from DP.......
which are certainly more real life like than DXO!!!
Click to expand...
No, dp is far less real life than dxo, because dp uses ooc jpg to determine dr, their figure is next to useless. Dxo measures dr of raw files
 
Randplaty said:
Great Bustard said:
Randplaty said:
I really don't trust all of these DR measurements. (not just from you but from everyone)

I mean, the 40D has more DR than the 5Dc? That's a joke. 5Dc dynamic range blew the 40D out of the water and was probably the primary reason for me to go full frame back in the day.
Click to expand...
...what DR is. People misuse and misinterpret the measure just like they misuse the term "bokeh" to mean how blurred the background is, or people thinking how changing the ISO changes the exposure, etc., etc., etc.

That's the difficulty with technical discussions here on DPR -- people using terms without actually understanding what they mean. Of course, what else would one expect? Still...
Randplaty said:
I will get a 5D3 and I will take actual photos with it and I will see what the DR is. The 5D2 didn't have much real world improvement over the 5D in my opinion. But again, that's not what the DXOmark number show.
Click to expand...
...as I said...
Click to expand...
Okay enlighten us then and explain to me why the 40D cannot handle backlit photos whereas the 5Dc could. I'm guessing that has nothing to do with DR?
Click to expand...
Give me an example of a photo from the same scene with the 40D and 5Dc shot with the same exposure that the 5Dc handled and the 40D did not.
 
You are wrong
DP major both jpeg and raw (with ACR in manual mode)
in same case raw will give one extra stop in other more

beside that they show real test recovering of high contrast scene !

This is were DXO......in my opinion fails
DR is not just number!
ultimitsu said:
No, dp is far less real life than dxo, because dp uses ooc jpg to determine dr, their figure is next to useless. Dxo measures dr of raw files
Click to expand...
 
etto72 said:
You are wrong
DP major both jpeg and raw (with ACR in manual mode)
in same case raw will give one extra stop in other more

beside that they show real test recovering of high contrast scene !

This is were DXO......in my opinion fails
Click to expand...
DPR measures the DR of a processed photo, which is a function of the tone curve applied. DxOMark measures the DR of the unprocessed photo, which represents the range of tone curves that can be applied to the photo without loss of detail (that is, between the noise floor and completely blown).
 
etto72 said:
You are wrong
DP major both jpeg and raw (with ACR in manual mode)
Click to expand...
No, dpr did not do proper raw dr measurements. That is why your 5d1 numbers are so far off the mark.
etto72 said:
in same case raw will give one extra stop in other more
Click to expand...
Because IOC jpeg is not a consistent indicator to begin with.
etto72 said:
beside that they show real test recovering of high contrast scene !
Click to expand...
What makes you think Dxo doesn't? U just don't get to see their scene.
etto72 said:
DR is not just number!
Click to expand...
Whatever this means...

Bottom line is - you arrived at a wrong conclusion because it was based on wrong Numbers, you then made a thread complaining that such conclusion makes no sense. You are being told why it doesn't make sense but you insist on relying on these blatant erroneous numbers, so you can stick to your incorrect conclusion and keep complaining ranting. I am not even sure if this is healthy.
 
Randplaty said:
Randplaty said:
Randplaty said:
I mean, the 40D has more DR than the 5Dc? That's a joke. 5Dc dynamic range blew the 40D out of the water and was probably the primary reason for me to go full frame back in the day.
Click to expand...
Click to expand...
Okay enlighten us then and explain to me why the 40D cannot handle backlit photos whereas the 5Dc could. I'm guessing that has nothing to do with DR?
Click to expand...
Dr has nothing to do with full frame or not, people who bought full frame for dr were obviously mistaken, current dr king is an apsc sensor, and the tiny g11 sensor has better dr than a lot of the m43 sensors..

But having that said 5d1 and 40d had almost identical dr, so this particular example would get nowhere.
 
You must log in or register to reply here.

Keyboard shortcuts

f
Forum
m
My threads
Mobile site
About
Editorial content
Cameras & Lenses
Community
All content, design, and layout are Copyright © 1998–2025 Digital Photography Review All Rights Reserved.
Reproduction in whole or part in any form or medium without specific written permission is prohibited.
When you use DPReview links to buy products, the site may earn a commission.
©GPS Media - Guides, Products, Services.
Back
Top