snubbr.com

Best dynamic range point and shoot
I have a Canon G7 and a Fuji F31. I am coming back from a trip in Asia and lots of my pictures have overexposure or highlight clipping problems even with -2/3 compensation. I have to PP most of them to adjust the exposure. Is there in the market cameras who deal better with exposure than others ?Thanks..

Comments (29)

I don't know about the Fuji, but you can get a little bit more range from the Canon if you set your MyColors to Neutral..

I had posted these in a thread but it got lost with the crash. Here are three images from my A710 IS that shows the effect of MyColors on the same exact scene. On the Neutral settings the highlights aren't blown even though the exposure is the same. (NOTE: the Vivid shot shutter speed is faster due to my movement.).

This series demonstrates that some highlight clipping occurs as a result of post-processing, and not because of over exposure..

MyColors Vivid.

Image control:Zoom outZoom 100%Zoom inExpand AllOpen in new window.

MyColors Off.

Image control:Zoom outZoom 100%Zoom inExpand AllOpen in new window.

MyColors Neutral.

Image control:Zoom outZoom 100%Zoom inExpand AllOpen in new window..

Comment #1

That's a good test and thanks for posting, but from the histogramshape it seems that Neutral is also clipping, it's just moved the clippingpoint down to about 97% of the histogram scale? Would you agree?.

Off does seem to provide more highlight data than Vivid, but betweenOff and Neutral, I think I would prefer Off since the range is less compressed..

Graystar wrote:.

MyColors Vivid.

Image control:Zoom outZoom 100%Zoom inExpand AllOpen in new window.

MyColors Off.

Image control:Zoom outZoom 100%Zoom inExpand AllOpen in new window.

MyColors Neutral.

Image control:Zoom outZoom 100%Zoom inExpand AllOpen in new window.

For a compact with better DR, one with RAW would be preferablebecause there is often hilight detail to salvage from the channelsthat clip later and also more shadow detail that the JPEG throwsaway..

Just my two oereErik from Sweden.

Image control:Zoom outZoom 100%Zoom inExpand AllOpen in new window..

Comment #2

Neutral didnt clip the highlights. Otherwise, the screen would be blinking like the other two..

All three test shots clip the shadows somewhat because the scene had a dynamic range that exceeds the cameras dynamic range. It doesnt matter what the settings are or if youre shooting RAW...if the scene exceeds the DR of the camera then youre going to have clipping at one end or the other..

On P&S RAW Ive found that RAW doesnt really get you any more than Neutral does. The reason RAW can get you more highlight information in the first place is due to the exact reason that Ive described...the camera is boosting the colors and causing highlight clipping. Once you eliminate the boost you get nearly the same results as RAW. In my testing my A710 IS provided about 7 stops of DR in RAW, and about 6 2/3 stops from a low-contrast JPEG. With MyColors off the camera gets about 6 stops of DR..

The Canon G9 should have slightly better DR. But its still a far cry from what a DSLR can do...

Comment #3

Graystar wrote:.

Neutral didnt clip the highlights. Otherwise, the screen would beblinking like the other two..

It shifted the clipping point down to a level where the screen didn'tsignal it with blinking. But the histogram still looks clipped, it's just thatit's lost the highest end of the range for no purpose..

Just my two oereErik from Sweden.

Image control:Zoom outZoom 100%Zoom inExpand AllOpen in new window..

Comment #4

Ehrik wrote:.

Graystar wrote:.

Neutral didnt clip the highlights. Otherwise, the screen would beblinking like the other two..

It shifted the clipping point down to a level where the screen didn'tsignal it with blinking. But the histogram still looks clipped, it'sjust thatit's lost the highest end of the range for no purpose..

Do you realize how utterly ridiculous that sounds?.

I've checked this extensively. The pixels in question are not clipped...

Comment #5

All compacts suffer from poor dynamic range, can't have your cake and eat it too!.

The best sensors in terms of DR (compacts and DSLR's) are fujifilms Super CCD's, in addition to the Fuji cameras, these appear to be used on some Olympus compacts as well, allthougth not officially acknowledged..

The flipside is that Fuji force you into some pretty rougth trade offs in terms of the rest of the body...

Comment #6

You know, if you were a little less condescending I probably wouldn't have bothered.

Graystar wrote:.

On P&S RAW Ive found that RAW doesnt really get you any more thanNeutral does. The reason RAW can get you more highlight informationin the first place is due to the exact reason that Ivedescribed...the camera is boosting the colors and causing highlightclipping..

In RAW, tis more usual for settings like colour choices are rendered non effectivethat in it's essence is one reason why we have RAW.

Once you eliminate the boost you get nearly the sameresults as RAW. In my testing my A710 IS provided about 7 stops ofDR in RAW, and about 6 2/3 stops from a low-contrast JPEG. WithMyColors off the camera gets about 6 stops of DR..

The Canon G9 should have slightly better DR. But its still a farcry from what a DSLR can do..

Canon G9.

Http://www.digitalcamerainfo.com/...era-Review-15723/TestingPerformance.htm.

"The G9 has decent dynamic range at ISO 80 and 100, but at higher sensitivities it falls off quickly (see the graph above). At ISO 800 and 1600 dynamic range is severely limited. Note that the graph indicates the best possible dynamic range with the camera, and these values my not be achievable in normal shooting conditions. The G9s dynamic range performance is worse than average for 2007 cameras, though better than the Canon PowerShot G7 and Canon PowerShot S5 IS, and identical to the Nikon Coolpix P5000.".

Fuji F30 (very similar to F31).

Http://www.digitalcamerainfo.com/...al-Camera-Review/TestingPerformance.htm.

"The Fujifilm FinePix F30 turned in an excellent performance. We've noted a handful of compact cameras lately with very high scores in the Low Quality measure at ISO 100, but they tend to be unimpressive at High Quality, and their scores look worse at higher ISOs. The F30 turns this phenomenon on it's head: it's 10.3 EV score at ISO Low Quality is good, but not outstanding, while it's High Quality scores match up well with any camera short of a good DSLR.".

And the chartsCanon G9.

Image control:Zoom outZoom 100%Zoom inExpand AllOpen in new window.

Fuji F30.

Image control:Zoom outZoom 100%Zoom inExpand AllOpen in new window.

In jpeg only cameras where you have such control, I would try more muted colours, lower contrast, lower sharpness settings. This both rounds off the curve for better rolloff, pushing clipping farther out, and lessens noise effects which broadens the DR range.Riley.

I'd give my right arm to be ambidextrous..

Comment #7

Interesting comparision, but the graphs have different layout even though they seem to be from the same site. Do you know if the G9 graph represent high quality or low quality DR (as found on the F30-graph), or maybe an average? And is the same test method used for both cameras?..

Comment #8

Svein Eriksen wrote:.

Interesting comparision, but the graphs have different layout eventhough they seem to be from the same site. Do you know if the G9graph represent high quality or low quality DR (as found on theF30-graph), or maybe an average? And is the same test method used forboth cameras?.

Use the links provided, they go to the pages were the charts are.

Yes the charts are a different format, over time ive seen 3 formats their I think.

I cant speak for them about their test procedures, but the differences they quote are so stark, I think it would be difficult to misinterpret themRiley.

I'd give my right arm to be ambidextrous..

Comment #9

Rriley wrote:.

You know, if you were a little less condescending I probably wouldn'thave bothered.

You probably shouldnt have. Did you read all the posts?? The OPs concern was highlight clipping. I gave him the benefit of my testing and experience in explaining whats happening in the Canon camera. But rather than accept my results, or ask for actual images for evaluation, or perform some comparative testing, the OP dismissed the results based on nothing but the tiny histograms from a 115,000-pixel LCD, and instead provides his own nonsensical explanation of whats going on...that the camera is clipping but not showing that its clipping. Its a ridiculous explanation..

In RAW, tis more usual for settings like colour choices are renderednon effective.

Yes I know. That was the point of my posts...that the camera isnt really overexposing. Its the in-camera post-processing that causing the highlight clipping. If you review the test shots youll see that MyColors Off and MyColors Neutral are shot at the same speed, aperture, and ISO, yet one is clipped and the other isnt. The Neutral setting gets you very close to what you get in RAW. Its not exact because there is still post-processing applied.



Did you even read any of what was posted other than my last remark?..

Comment #10

I did follow the links before responding, and didn't see any info on the pages. Didn't search the site for explanation, but thought maybe you had checked it before posting...

Comment #11

Henri74 wrote:.

I have a Canon G7 and a Fuji F31. I am coming back from a trip inAsia and lots of my pictures have overexposure or highlight clippingproblems even with -2/3 compensation. I have to PP most of them toadjust the exposure. Is there in the market cameras who deal betterwith exposure than others ?Thanks.

Just shoot manualy observing the LCD....Helps a lot. You can see the effect of your aperture/shutter changes. I nearly always shoot this way if I have time....:)Alek..

Comment #12

Svein Eriksen wrote:.

I did follow the links before responding, and didn't see any info onthe pages. Didn't search the site for explanation, but thought maybeyou had checked it before posting..

Svein, maybe you followed the wrong links or something, but the info is there, and I just followed them to the same place. They appear to use IMATEST for there results so the reviews are by the same measurement system.

Canon G9.

Http://www.digitalcamerainfo.com/...era-Review-15723/TestingPerformance.htmFuji F30.

Http://www.digitalcamerainfo.com/...al-Camera-Review/TestingPerformance.htm.

Riley.

I'd give my right arm to be ambidextrous..

Comment #13

Yes, they use Imatest (1.6.16 for F30 and 2.3.17 for G9), but they don't say if the G9 curve is for high or low quality. If you assume it's low quality, then the DR of F30 is obviously better. If you assume it's high quality then G9 and F30 is very close. Btw, you'd expect the 6MP Fuji sensor to be better because of the bigger pixel size, I'd just like to know exactly what you're comparing...

Comment #14

Graystar wrote:.

Rriley wrote:.

You know, if you were a little less condescending I probably wouldn'thave bothered.

You probably shouldnt have. Did you read all the posts?? The OPsconcern was highlight clipping. I gave him the benefit of my testingand experience in explaining whats happening in the Canon camera.But rather than accept my results, or ask for actual images forevaluation, or perform some comparative testing, the OP dismissed theresults based on nothing but the tiny histograms from a 115,000-pixelLCD, and instead provides his own nonsensical explanation of whatsgoing on...that the camera is clipping but not showing that itsclipping. Its a ridiculous explanation..

In RAW, tis more usual for settings like colour choices are renderednon effective.

Yes I know. That was the point of my posts...that the camera isntreally overexposing. Its the in-camera post-processing that causingthe highlight clipping. If you review the test shots youll see thatMyColors Off and MyColors Neutral are shot at the same speed,aperture, and ISO, yet one is clipped and the other isnt. TheNeutral setting gets you very close to what you get in RAW. Its notexact because there is still post-processing applied.



Did you even read any of what was posted other than my last remark?.

Sure I did, and his polite reply.

You can be saying the essence of what is right (which you are) but just come off sounding like you are somehow a tad too obnoxious (which you do) mostly because Erik probably misunderstood your convoluted long winded reply (which it is) english not being his first language..

Which despite english being your first language, you misunderstood in his reply anyway, because you seem unaware that RAW is irrelevant in an F31 (which it is) because it doest have a RAW file system, and he is therefore dealing with jpgs on the Fuji at least..

You might try this text, I'm told it can be very helpfulhttp://www.amazon.com/How-Win-Friends-Influence-People/dp/0671723650.

@ Erik.

You might try plug in -1/3 ev which is a standard for exposure on F31, and keep away from auto mode when the situation looks as though the lighting is tricky or dim. In this way you keep it off AUTO iso, which might be responsible for a loss of DR..

Riley.

I'd give my right arm to be ambidextrous..

Comment #15

Graystar wrote:.

Ehrik wrote:.

Graystar wrote:.

Neutral didnt clip the highlights. Otherwise, the screen would beblinking like the other two..

It shifted the clipping point down to a level where the screen didn'tsignal it with blinking. But the histogram still looks clipped, it'sjust thatit's lost the highest end of the range for no purpose..

Do you realize how utterly ridiculous that sounds?.

Since you didn't offer the jpegs, I had to go by the histograms, and from whatit looks like to me, the Neutral histogram goes from a peak (at about 97%along the x axis) straight down to zero. Such an abrupt cutoff seems unlikely;a roll-off would have been expected..

Image control:Zoom outZoom 100%Zoom inExpand AllOpen in new window.

So the choices are that (i) the light was distributed in this unnatural lookingway in real life, or (ii) that the camera chose to shift the clipped valuesdown a bit from 255, and left the highest values unpopulated. Which wouldexplain the histogram look and the lack of blinking highlights..

Having seen camera manufacturers doing weird, or if you like,ridiculous, things before, I found (ii) to be slightly more plausible..

I've checked this extensively. The pixels in question are not clipped..

If (ii) is correct, then the pixels would not have to be (255,255,255) to beclipped..

Just my two oereErik from Sweden.

Image control:Zoom outZoom 100%Zoom inExpand AllOpen in new window..

Comment #16

Svein Eriksen wrote:.

Yes, they use Imatest (1.6.16 for F30 and 2.3.17 for G9), but theydon't say if the G9 curve is for high or low quality. If you assumeit's low quality, then the DR of F30 is obviously better. If youassume it's high quality then G9 and F30 is very close. Btw, you'dexpect the 6MP Fuji sensor to be better because of the bigger pixelsize, I'd just like to know exactly what you're comparing..

Well I'm glad you found it anyways.

I cant speak for how they make their reviews, which as you have noted might change over time anyway..

What we do know is that super ccd's seem to have higher DR than most by there architecture, and that stuffing a sensor full of pixels will reduce DR, which according to them is true of the Canon G9 and as we know from elsewhere the newer versions of these Fuji cameras..

Riley.

I'd give my right arm to be ambidextrous..

Comment #17

Rriley wrote:.

Which despite english being your first language, you misunderstood inhis reply anyway, because you seem unaware that RAW is irrelevant inan F31.

My first post started with:"I don't know about the Fuji,".

Everything I wrote was in reference to the Canon only. Also, my first post was entire about JPEG in-camera processing and never mentions RAW. I only addressed RAW after the OP brought it up..

It's responses like this that again question whether you actually read all the words that were written before attacking me...

Comment #18

Ehrik wrote:.

Along the x axis) straight down to zero. Such an abrupt cutoff seemsunlikely;a roll-off would have been expected..

Maybe what we have here is a slight misunderstanding of histogram dynamics..

If you were to take a properly exposed image of a perfect gray card, you will get a single tall line in the middle of your histogram. In similar fashion, digital exposure targets, with a white, gray, and black sections, give three sharp peaks in the histogram. Whether the histogram "rolls-off" depends entirely on the scene being photographed..

In the case of my test image, the room was dark with the only light up against a white wall. This created a scene with a stark contrast between light and dark (as indicated by the histogram.) In this type of scene, a "rolling-off" to the right would not be expected. What you'd expect is a "rolling off" to the left, where the bright center of the image causes a peak and darkens as you move away from the center. That's exactly what the histogram shows..

I've checked this extensively. The pixels in question are not clipped..

If (ii) is correct, then the pixels would not have to be(255,255,255) to beclipped..

That is correct, and they are not..

Hope that helps clarity a bit...

Comment #19

This won't work with your G7 but most Canon P&S using the Digic II chip can be "coaxed" into capturing a raw image with the CHDK firmware. You will probably be able to squeeze out a little more DR with the raw file. Below is the link. There have been extensive discussions of this firmware on the Canon Talk forum but with the recent server problems I don't know if they will all be accessible..

Kerkula.

Http://chdk.wikia.com/wiki/Main_Page..

Comment #20

Graystar wrote:.

Ehrik wrote:.

Along the x axis) straight down to zero. Such an abrupt cutoff seemsunlikely;a roll-off would have been expected..

Maybe what we have here is a slight misunderstanding of histogramdynamics..

Or of the physics of light....

If you were to take a properly exposed image of a perfect gray card,you will get a single tall line in the middle of your histogram..

Even if the card were perfect, light isn't, it has a random nature, so eachpixel will not be struck by an identical number of photons, instead some willget a bit more than average and some a bit less. And the histogram will bea bell shaped spread around the average (actually a bit distorted by thetone curve). It's called shot noise, google it!.

Even if you never heard of that before, you are still assuming that yourwall has perfect uniform reflectance which gives a perfect gradient towardsthe light source. Which is quite a stretch, IMHO..

In similar fashion, digital exposure targets, with a white, gray, andblack sections, give three sharp peaks in the histogram..

If you by "sharp" still mean "single tall line", then, no, by reasons givenabove (with a possible exception from the black peak which gets itsspread from read noise which may be too little to register in the histogramresolution)..

Whether thehistogram "rolls-off" depends entirely on the scene beingphotographed..

In the case of my test image, the room was dark with the only lightup against a white wall. This created a scene with a stark contrastbetween light and dark (as indicated by the histogram.) In this typeof scene, a "rolling-off" to the right would not be expected..

If you had used a sufficiently fast shutter speed to avoid clipping I thinkyou would have found that the peak at the bright end would have someroll-off to the right. I guess you could still try that if you want to..

Just my two oereErik from Sweden.

Image control:Zoom outZoom 100%Zoom inExpand AllOpen in new window..

Comment #21

.. any P&S F I L M camera using any N E G A T I V E film would give you far greater DR then any digianything p&s past the king of all DSLRs..

Henri74 wrote:.

I have a Canon G7 and a Fuji F31. I am coming back from a trip inAsia and lots of my pictures have overexposure or highlight clippingproblems even with -2/3 compensation. I have to PP most of them toadjust the exposure. Is there in the market cameras who deal betterwith exposure than others ?Thanks..

Comment #22

MrRoger wrote:.

All compacts suffer from poor dynamic range, can't have your cake andeat it too!.

The best sensors in terms of DR (compacts and DSLR's) are fujifilmsSuper CCD's, in addition to the Fuji cameras, these appear to be usedon some Olympus compacts as well, allthougth not officiallyacknowledged..

The flipside is that Fuji force you into some pretty rougth tradeoffs in terms of the rest of the body..

What makes you think so (Olympus using Fuji's CCD's)?.

My Flickr:http://www.flickr.com/photos/royiavital..

Comment #23

Royi Avital wrote:.

What makes you think so (Olympus using Fuji's CCD's)?.

I have a C-770. Although the sensor manufacturer is anonymous, the documentation describes the dual size pixel structure typical of Fuji's super CCD's. Also, the performance, given the cameras age and the fact that it has a 1/2.5" sensor i.e. half the size of many of the better compacts...

Comment #24

Your Fuji is at the top of the heap and pretty close to some DSLRs. Check out this link -.

Http://www.digitalcamerainfo.com/...al-Camera-Review/TestingPerformance.htm.

Also, search for the G9 tests and you will probably be shocked at how it compares. Dare I say it - too many pixels in a smallish sensor!.

Cheers..

Comment #25

Your eloquence is only exceeded by your intelligence on the matter. Seems you've been providing incorrect information about digianything's latitude..

But just so that in the future you can better articulate the truth here's a resource for you - direct from the proverbial horses mouth. Canon itself states in their White Paper - available at http://www.robgalbraith.com/bins/content_page.asp?cid=7-7897-8537. In page 7 the King of All DSLRs only matches the latitude of slide film. Also that no other none full frame sensor even matches that..

Don't blame the messenger, the head sensor maker said so itself. .

Sigezar wrote:.

Go get lost or something http://www.apug.org/forums/home.php..

Comment #26

LesDMess wrote:.

Seems you've been providing incorrect information aboutdigianything's latitude..

Show me where, brainiac..

Oh wait, you can't, seem you've been saying black people are inferior and eat bananas. Here's an unrelated link http://www.albinoblacksheep.com/flash/youare..

Comment #27

Sigezar wrote:.

Show me where, brainiac..

Yeah, I must seem like a brainiac to you so let me help you out further..

In my previous post I posted the following:.

LesDMess wrote:.

But just so that in the future you can better articulate the truthhere's a resource for you - direct from the proverbial horses mouth.Canon itself states in their White Paper - available athttp://www.robgalbraith.com/bins/content_page.asp?cid=7-7897-8537. Inpage 7 the King of All DSLRs only matches the latitude of slide film.Also that no other none full frame sensor even matches that..

1. Do you see the link that is of a different color above in my reply? Click on that..

2. Then when you get there, there is a canon white paper PDF, click on that too..

3. If it doesn't automatically come up, perhaps you don't have a PDF viewer installed. You will have to go tohttp://www.adobe.com and search for the adobe reader. Note it is a free download so please resist any credit card questions..

Then try again at step 1 above. If it doesn't work this time, then I would recommend a local community college for beginning internet coursses. Good luck and have fun. ..

Comment #28

You would be surprised how may P&S cams have good DR. Apart from Fuji F30, look at the Sony W100. Their figures are almost the same as these DSLRs - Olympus E330, Pentax K100, Nikon D40..

It's a simple rule but true in most cases that a sensor that is good in low light has good DR..

Refer to this site for tests on the cameras I have mentioned -.

Http://www.digitalcamerainfo.com/...al-Camera-Review/TestingPerformance.htm.

On the point of clipped highlights, it's not simply a case of limited DR. Some manufacturers deliberately set the contrast a little high (as well as over-sharpen, over-saturate) to 'enhance' the perceived image quality of their cameras..

The DR of images from my S6500 (from RAW) have to be seen to be believed although, like many P&S when left in JPEG mode it tends to clip highlights more because the tone curve is set by default for high contrast..

Setting the S6500 to RAW not only allows me to adjust the contrast to the correct level (as I like it) but to bring detail out of the shadows that you just cannot reveal when post processing the contrasty JPEG..

Cheers..

Comment #29

Click Here to View All...

Sponsored Amazon Deals:

1. Get big savings on Amazon warehouse deals.
2. Save up to 70% on Amazon Products.


This question was taken from a support group/message board and re-posted here so others can learn from it.

 

Categories: Home | Diet & Weight Management | Vitamins & Supplements | Herbs & Cleansing |

Sexual Health | Medifast Support | Nutrisystem Support | Medifast Questions |

Web Hosting | Web Hosts | Website Hosting | Hosting |

Web Hosting | GoDaddy | Digital Cameras | Best WebHosts |

Web Hosting FAQ | Web Hosts FAQ | Hosting FAQ | Hosting Group |

Hosting Questions | Camera Tips | Best Cameras To Buy | Best Cameras This Year |

Camera Q-A | Digital Cameras Q-A | Camera Forum | Nov 2010 - Cameras |

Oct 2010 - Cameras | Oct 2010 - DSLRs | Oct 2010 - Camera Tips | Sep 2010 - Cameras |

Sep 2010 - DSLRS | Sep 2010 - Camera Tips | Aug 2010 - Cameras | Aug 2010 - DSLR Tips |

Aug 2010 - Camera Tips | July 2010 - Cameras | July 2010 - Nikon Cameras | July 2010 - Canon Cameras |

July 2010 - Pentax Cameras | Medifast Recipes | Medifast Recipes Tips | Medifast Recipes Strategies |

Medifast Recipes Experiences | Medifast Recipes Group | Medifast Recipes Forum | Medifast Support Strategies |

Medifast Support Experiences |

 

(C) Copyright 2010 All rights reserved.