Go Back   Steve's Digicams Forums > Digicam Help > General Discussion

Reply
 
Thread Tools Search this Thread
Old Jan 12, 2008, 7:36 PM   #1
Senior Member
 
Join Date: May 2006
Posts: 181
Default

How do they compare? I'm interested in the dynamic range that each of the above can render at once. For eyes, the amount of range that can be seen at one pupil setting and for cameras, with same aperture and shutter speed.

When I look up, I can see the writing PHILLIPS, the difference in brightness from the end of lamp to center,the shadow the fixture cast on the ceiling as well as the pattern of the ceiling

With my Canon S3 IS (1/2.5" sensor) set to minimum compression JPG,EV = 0 at asterisk, the entire lamp becomes saturated (i.e. 255,255,255 or whitewashed).


If I set it to EV=0 on the center of the lamp, I can read PHILLIPS and see the difference between the lamp end brightness and the center, but most of the ceiling is pure black (0,0,0).


How well could a camera like Canon EOS 30D render this?
What about ordinary film or positive film?
How about a professional DSLR?
itsme000 is offline   Reply With Quote
Sponsored Links
Old Jan 12, 2008, 9:08 PM   #2
Senior Member
 
Join Date: Jun 2005
Posts: 804
Default

Here's some interesting stuff that I found on the subject:

"The human eye can actually perceive a greater dynamic range than is ordinarily possible with a camera. If we were to consider situations where our pupil opens and closes for varying light, our eyes can see over a range of nearly 24 f-stops.

On the other hand, for accurate comparisons with a single photo (at constant aperture, shutter and ISO), we can only consider the instantaneous dynamic range (where our pupil opening is unchanged). This would be similar to looking at one region within a scene, letting our eyes adjust, and not looking anywhere else. For this scenario there is much disagreement, because our eye's sensitivity and dynamic range actually change depending on brightness and contrast. Most estimate anywhere from 10-14 f-stops. The problem with these numbers is that our eyes are extremely adaptable. For situations of extreme low-light star viewing (where our eyes have adjusted to use rod cells for night vision), our eyes approach even higher instantaneous dynamic ranges."

And from another source:

"...However, most Digital Single Lens Reflex cameras as of Fall 2006 use a 10 to 14-bit A/D converter, which translates into a dynamic range of 10 to 14 stops. Even then this dynamic range is limited by noise levels, meaning that shadow areas in the image where detail was captured may exhibit so much digital noise as to be visually unattractive and best rendered as pure black.

Sensors are far better than film when it comes to dynamic range, being able to capture a much wider dynamic range than film, on average. However, this pales in comparison to the human eye which is not only able to see details in a scene containing a contrast range of nearly 24 stops, but also able to instantly change its contrast-perception ability in order to see, alternatively, details in highlights and in shadow areas without the conscious awareness that we are doing so."

Here's a good chart from Clarkvision:



Again from Clarkvision:

"There seems to be an urban legend that says digital cameras have less dynamic range than film. The legend is wrong. The above plot (Grant's note: Not the chart shown above. I didn't include the referenced chart.) shows the comparison of a DSLR with print and slide film. The slide film records only about 5 photographic stops of information (a stop is a factor of 2, so 5 stops is 32). The print film shows about 7 stops of information. The digital camera shows at least 10 stops of information (this test was limited to 10 stops). Other tests show the Canon 1D Mark II camera has about 11.6 stops of information (a range of 3100 in intensity). Other DSLR cameras, like the Canon 10D have around 11 stops. Point and shoot digital cameras, somewhat less."

This might not quite address all the questions, but it's a good start!

Grant

granthagen is offline   Reply With Quote
Old Jan 13, 2008, 6:48 PM   #3
Senior Member
 
BillDrew's Avatar
 
Join Date: Jun 2002
Location: Hay River Township, WI
Posts: 2,512
Default

granthagen wrote:
Quote:
Here's some interesting stuff that I found on the subject:

...

"...However, most Digital Single Lens Reflex cameras as of Fall 2006 use a 10 to 14-bit A/D converter, which translates into a dynamic range of 10 to 14 stops. Even then this dynamic range is limited by noise levels, meaning that shadow areas in the image where detail was captured may exhibit so much digital noise as to be visually unattractive and best rendered as pure black.

...
Hard to tell if that follows the same misconception I used to have that the bit depth of the A/D converter has anything to do with the dynamic range. The flaw in that "logic" that there is some relationship can be seen easily by analogy with a stairway. If a set of stairs has a total height of 10 feet, that total height remains the same regardless of whether the stairs have 14 steps or 10 steps or any other number.

I have a problem with his 12 A/D limit lines for the same reason: I see no reason why a 15 stop dynamic range could not be coded into a 12 bit conversion. Each bit would represent something like 1.2 stops. No reason I can think of to assume that a doubling of the value represnts exactly one stop.

BillDrew is offline   Reply With Quote
Old Jan 13, 2008, 9:29 PM   #4
Senior Member
 
VTphotog's Avatar
 
Join Date: Mar 2005
Location: Extreme Northeastern Vermont, USA
Posts: 4,214
Default

Bill is quite correct. The A/D converter has nothing to do with the dynamic range of the sensor, which can be determined by measuring the saturation voltage vs. the dark voltage of the pixels themselves. This is an absolute measurement and doesn't indicate useful DR. Info I have on hand regarding an older 5MP 2/3" sensor gives about a 5-stop DR, which is pretty common for digicams. DSLRs do better, with some measurements I have seen on another site showing about 10-11 stops when processed from RAW.

DR of an outdoor scene on a sunny day, is IIRC, about 10 stops. This agrees with what I have found myself from shooting bracketed exposures with digicam and DSLR. Where it gets complicated, is in the translation from the camera to a .tiff or .jpeg file, and displaying on a monitor or printing.

brian
VTphotog is offline   Reply With Quote
Old Jan 13, 2008, 10:04 PM   #5
Senior Member
 
Join Date: Sep 2003
Location: Dallas, Texas USA
Posts: 6,483
Default

It's funny how stuff like this gets tested to death today now that it's so easy to do with digital technology, which seems to have attracted as manytechno geeks as it haspeople interested in simply takingwell composed and exposed images, evenwith a little detail lost to one extreme or another.The sad part is, how much credibilitymost give the geeks.

In going back to images shot with 35mm equipment I see manypictures where highlights are blow or detail was lost to shadows. This "phenomenon" didn't start with the first digital image. Having complete control and more access to control adjustments, even with the smaller sensored digicams, has proven much better than the use of larger 35mm film, especially in the case of all those one-time use camerasI used to see so many carrying around which were, and arejust plain awful,and leaving the final image quality to the lab wheremost have little tono say in how the print turns out.

Dynamic range, to a large extent, is fine in most digicams if one would just learn how to use the most basic exposure controls, such as exposure compensation andmetering options and actually use them when it makes sense to. Thisimage, shot with a Panasonic TZ3,is a typical example..



When I first framed this, the brighterside of the tombstonewas completely blown using multi-segment metering, so I change to spot andmetered off that side of the tombstone with some exposurecompensation, which completely revealed allthe detail. The background has "blown" detail in the sky. Does this ruin the photo? To me, no.

Regarding the OP question regarding the posted light bulb imagesto an actual image where ceiling light bulbs would be included in the field of view, how important in relationship to all the other parts of the sceneis it to actually reveal every possibledetail in the bulb or mounting equipment? I don't think there's any question our eyes will reveal much more dynamic range than any film or digitalmedia, unless you go to extra-ordinary lengths to do it digitally, which make it, in my opinion, more trouble for mostthan it's worth in more cases than not.

Greg Chappell is offline   Reply With Quote
Old Jan 14, 2008, 1:55 AM   #6
Super Moderator
 
peripatetic's Avatar
 
Join Date: Nov 2004
Posts: 3,599
Default

One area where film does still seem to hold an advantage is B&W negative film, which many people claim can give up to 15 stops of DR. Although I have used (even occasionally still use) B&W film, I have never owned my own darkroom, so am really a beginner in that regard.

From my experience it is certainly true that P&S have a much smaller DR than DSLR. And also that my 5D has a sometimes noticable advantage over my old 20D, in "usable" DR. There were often shadow areas in my 20D photos where the noise was quite intrusive, the 5D by contrast, even if it's not actually capturing a great deal of extra range, renders the shadows with less noise making the images better.

I agree that Norman Clark's site is an excellent resource for understanding all this stuff. Additionally the various Astro-photography websites delve into these matters in great detail.

A few other things to mention...

Firstly I think it is important to understand how best to expose your digital images - for a good starting point about "Expose to the right" ETTR.

http://www.luminous-landscape.com/tu...se-right.shtml

Which leads onto the second point about why the bit-depth might be important for optimising DR in real-world images from digital. In general it is certainly true that the best analogy is to say that the DR is the height of the staircase, and the bit depth relates to the size of the stairs. However combined with the information in the article above it seems to me that we get to a point where the maximum practical DR encodable does depend somewhat on the bit depth. 8-bit encoding (like JPG) provides a maximum of 8 stops - why? Well because images are encoded such that each stop (working from the right) takes up half the remaining values. So for the darkest stop of an 8-bit encoding you only have 2 values to encode that whole stop. Assuming your sensor is capturing 10 stops, how would you encode those bottom two stops? If you look at the DR analysis on dpreview you find that pretty much all the modern DSLRs max out at 8 stops of DR in JPEG mode.

Given 12-bit encoding and a sensor DR of 11 stops, if you ETTR, then you are able to discard the bottom stop with only 2 levels (2^1 bits), and use the next stop with its 4 levels (2^2) to encode that information. With 14 bit encoding you would be able to discard the most information poor 3 stops at the bottom, leaving your worst stop of DR with 16 levels of information (2^4).

It also helps of course when you don't manage to perfectly ETTR and end up "discarding" some of the information space of the encoding.

In theory therefore the new 14-bit cameras should provide images with much better shadow detail. I haven't seen a proper review of this on the internet anywhere, but I would be very interested if anyone has a link.

Photoshop provides many options for merging multiple exposures into a single High DR image (HDR). By taking multiple exposures of the same scene using exposure bracketing you can merge them so that you are using the right-hand side of the histogram and discarding the information-poor left hand side to get greater shadow detail. Such images really can look quite spectacular.

And finally of course, not every scene in the real world actually has the range of light such that it exceeds the 10-11 stops that digital sensors can capture. So extending DR isn't always necessary.
peripatetic is offline   Reply With Quote
Old Jan 14, 2008, 4:51 AM   #7
Senior Member
 
BillDrew's Avatar
 
Join Date: Jun 2002
Location: Hay River Township, WI
Posts: 2,512
Default

peripatetic wrote:
Quote:
...

Which leads onto the second point about why the bit-depth might be important for optimising DR in real-world images from digital. In general it is certainly true that the best analogy is to say that the DR is the height of the staircase, and the bit depth relates to the size of the stairs. However combined with the information in the article above it seems to me that we get to a point where the maximum practical DR encodable does depend somewhat on the bit depth. 8-bit encoding (like JPG) provides a maximum of 8 stops - why? Well because images are encoded such that each stop (working from the right) takes up half the remaining values. So for the darkest stop of an 8-bit encoding you only have 2 values to encode that whole stop. Assuming your sensor is capturing 10 stops, how would you encode those bottom two stops? ...
Good point: there can be a gain from increased bit depth. Much less than simply doubling for each added bit though.

There can be very real improvements resulting from increased bit depth that have nothing to do with dynamic range, in particular when doing various adjustments. Especially when "pushing" the ISO or correcting underexposure: more bit depth will decrease posterization. I did not mean to imply that increased bit depth was purely a marketing gimmic like digital zoom, only that it has little to do with dynamic range.

BillDrew is offline   Reply With Quote
Old Jan 14, 2008, 9:37 AM   #8
Administrator
 
Join Date: Jun 2003
Location: Savannah, GA (USA)
Posts: 22,378
Default

Keep an eye on the new Sony models like the Alpha 700 (DSLR-A700) for improvements in Dynamic Range compared to previous models. It looks like Sony has been working very hard in this area.

If you look through some of DIWA Labs' newer camera tests using DxO Analyzer, the new Sony DSLR-A700 has better dynamic range compared to any of the cameras they've tested, besting the Canon EOS-40D, Nikon D300, Nikon D3, Olympus E-3, Fujifilm S5 Pro and more.

Here's a chart from it. It's best DR comes in at ISO 200.

http://www.diwa-labs.com/photoalbum/...&id=191649

To see their test results, click on a camera model, then click on "Test results for camera performance" when you get to choices on the next screen.

http://www.diwa-labs.com/wip4/test_result.epl

Dynamic Range from this new Sony is impressive compared to other cameras.

Note that I'm biased (since I now shoot with a Sony DSLR-A700). :-)

Phil Askey over a dpreview.com also commented on it in his A700 review conclusion as having Excellent Dynamic range with more highlight range than they are used to seeing. For example, his tests in the A700 review show it measured at 2/3 stop better than the Nikon D200.

I'm waiting for Dave Etchells' Imatest results, too. I suspect he's probably retesting a lot of stuff since his preview, since Sony has upgraded the firmware to this camera since his initial test shots. He usually measures DR shooting both RAW and JPEG with cameras he reviews. So, I'd keep an eye out for his tests to be finished in the full reviews of newer models over at imaging-resource.com for Imatest test results of Dynamic Range.

JimC is offline   Reply With Quote
Old Jan 24, 2008, 7:59 AM   #9
Member
 
Join Date: Jan 2006
Posts: 96
Default

Hi
I thing you are wrong. AD convertor works with usually from 0-5 V and we can say dinamic range is from 0 to 5 V, but question is how many steps are between. 8 bit AD makes 256 steps, 12 bit AD makes 4096 steps. Sensor can be more or less linear, or has more or less noise, or more or less distortion etc what can reduce advantage of 12 or more bit convertor.
My question is can we see more than 8 bit from monitor or from paper?
Thanks
emes is offline   Reply With Quote
 
Reply


Thread Tools Search this Thread
Search this Thread:

Advanced Search

Posting Rules
You may not post new threads
You may not post replies
You may not post attachments
You may not edit your posts

BB code is On
Smilies are On
[IMG] code is On
HTML code is Off
Trackbacks are Off
Pingbacks are Off
Refbacks are Off



All times are GMT -5. The time now is 12:23 AM.