|
Forums >
Photography Talk >
Canon 20D Under-exposure
Have 20D and always shoot raw in rgb. My images are usually a little under-exposed whether i use the in-camera meter or my sekonic. Has anyone experienced this with their 20d? Sep 02 05 10:44 pm Link Every digital sensor is different, just as every batch of film is different. If you find that it's underexposing a little, compensate. If it's a lot, check with Canon to see if it's really a problem. Sep 02 05 11:23 pm Link I use a sekonic and set the iso to 64 when I'm shooting at 100. It would be best to do a one-time test of your camera/meter combo. Take a series of seven shots at 1/3 stop increments. It'd be best if your subject had on a white top (so you can watch for blow-out). Pick the image that comes just close to blowing out and compensate accordingly from then on. Paul Sep 03 05 12:50 am Link Ditto on what Paul said -- testing is key -- I believe even more so with digital then film. We get trapped into thinking the LCD tells the story  Sep 03 05 05:57 am Link Check the histogram, adjust your exposure accordingly. Sep 03 05 07:20 am Link Thank you all for your input. Sep 03 05 10:34 am Link I have a 20D also, but mine seems to like to overexpose a little. I mostly just leave it adjusted down a fraction of a stop. Some of it depends on the dynamics of what you're shooting. I find shooting a little under is safer. It's a lot easier to fix (minor) underexposures than blown-out whites. Sep 06 05 07:43 am Link I've found that the two Canon digital cameras that I have (the 20D and the Digital Rebel) underexpose at about 2/3 of a stop. My in studio testing confrimed this also. I adjust for this a couple of ways. I either set a new default with Camera Raw at plus .67 for the exposure for images not shoot with this adjustment or I will do an in camera EV adjustment of 2/3 stop while shooting. In studio my Sekonic light meter is adjusted with a -.4 as my testing indicated that this was the needed adjustment. Sep 06 05 08:02 am Link What mode are you shooting? If you have a fast enough lens and you are checking your exposure correctly, all should be well.......I ask what mode because shooting manual is the only way you have control...... Sep 06 05 05:45 pm Link You always need to calibrate your light meter to your camera. This goes for film as well. I have the Sekonic 558 and I know on my 20D its a 1/3 of a stop off, my 1D is dead on, and my film camera (MF Mamiya), all depends on which slide film I'm using. Once you figure it out make a note card and keep it in your bag. With digital its an easy fix for under exposure. Slide film needs to be spot on, very little latitude. Why slide???? Because several magazines still insist on it, which in my opinion sucks.........Digital is still my fav. Sep 06 05 06:06 pm Link This may sound weird, but why do digital shooters even need a light meter? Sep 06 05 09:15 pm Link Because the histogram displays data for the entire scene, not just the subject. And the LCD isn't accurate enough. If your camera has a highlight blow-out indicator, you could keep jacking up the exposure until you see some blow-out but that's too much like work. It's much simpler to meter the light and know your exposure will be on the money. Paul Sep 06 05 10:50 pm Link Jay Farrell wrote: They don't. Sep 07 05 12:22 am Link Personal Observation and Opinion. Your 20D may be different, and your milage may vary... and I'd like to hear if others have had similar experiences with the 20D's histograms. First, I don't intend to bash the 20D. It is a great piece of equipment, and offers features and capabilities I never dreamed of when I picked up my first SLR a few decades ago. But, as with every piece of equipment, the more it can do, the more it seems it should do. Thus, the push to expand the technology and improve it goes on and on. That said... My 20D likes to overexpose. If I use the internal meter and meter off a gray card the result is between 1/3 and 2/3 stop overexposed (in reality, probably 1/2 stop). That might not sound like much, but it can be a real problem in high contrast situations. In my case the problem appears to be in the camera's meter, not the sensor, as in the studio I set the 20D to the exposure my Polaris indicates and the resulting exposures are right-on. As to the blow-out indication on *MY* 20D, if an image is overexposed to where the blow-out indication appears in the LCD, the image is probably so blown out it is worthless. Which brings me to a particular overexposure problem I have with the 20D which I never had when shooting film (Canon A1). Shooting in the late afternoon orange glow. I find that the histogram is only a guide, as it too has its drawbacks. The histogram displayed on the LCD appears to be the *average* of the red, green, blue values, and that works well when you are shooting properly white balanced average subject material and the subject doesn't have any very bright, very vivid colors. However, what if you are trying to take advantage of that wonderful orange-red light before sunset, and shooting a model with the white balance set for daylight to get that orange glow? You meter with your external meter (incident) and set the 20D accordingly. The exposure meter in the 20D agrees with the external meter reading. The histogram you see on the LCD looks great, and the images in the LCD display loop pretty good too! You are happy! Unfortunately, when you get back to the computer and view the images, they look sick! That can't be! The histogram looked great! You take the memory card out of the reader, put it back in the camera, and sure enough, the histogram looks great! What the? So, you go back to the computer and THIS time you look at the separate red, green, and blue histograms for the image. Expletive deleted. The blue channel looks dark, the green channel looks normal, and the red channel looks very blown out. Average the three together and you see that great histogram that you saw on the LCD on the back of the 20D. Maybe if the 20D had the ability to show not only the "average" histogram, but also the histograms of the individual red, green, and blue channels you could put a lot of weight in what the histogram(s) say. My solution (for color-biased situations) has been to either deliberately underexpose by an experimentally derived amount, or connect a laptop to the 20D, take a test exposure, and look at the individual histograms to make sure nothing is blowing out. Note that I find that if I meter with an external meter under color-biased situations and set the 20D manually according to the external meter's reading, I'll still end up with an over-exposed channel (red in the case of shooting just before sunset). Another example. I'm shooting stage productions where they are using the new LED light banks. They can create any color and any intensity and any color balance they want with the light banks. Beautiful vivid greens, reds, blues - colors so saturated you can hardly believe it - until you try to photograph them. And as a limitation, I can't go up on stage to meter, so I use my experience to temper what the in-camera meter and histogram say. If I can get into the rehearsal, I bring my laptop and shoot test exposures and look at the individual red, green, blue histograms, record what exposure values to use for each section of the program, and during the live production shoot to the values I determined during the rehearsal. In the old days with my Canon A1 I'd bracket exposures, and eat up a LOT of slide film (and sometimes miss the best shot) in the process. Today the 20D and laptop combination works much better. Bits are a LOT cheaper than film! While I love my old A1 and it has given me a couple of decades of great service, I love and appreciate my 20D as the wonderful machine it is. Your thoughts? Sep 07 05 12:45 am Link I think it varies from camera to camera...I find that mine does underexpose a bit if I use the in-camera meter, but I typically shoot manual, use a handheld meter, and check my histogram often. As for whether or not you need a light meter - I wouldn't say you "need" one, but I love mine! Especially for studio work. :-) Michelle Sep 07 05 02:15 pm Link I think we do need a light metter. I totally agreed with John Korb. The light metter and the build in metter in the camera always give me diff result. The LCD and a computer screen got the same problems. I think we have to play around and learn how to look on the LCD. The best is doing in RAW if possible, because Raw can go up/down 2 F stop. Regards Nick p.s I really not happy with my pictures although i have the top digital camera (1DS Mark II ) in the world. That's suck Sep 09 05 05:27 pm Link John Korb wrote: This doesn't make sense because earlier you said that your internal meter overexposed. John Korb wrote: That makes some sense if you're positive of the above results. I don't have a 20D so don't have any history with them. Sep 09 05 07:15 pm Link I use a canon 20d and a canon 10d and a Minolta, it's right on everytime I use it. just wish I used it more:) Sep 10 05 08:41 am Link I always check my histogram... true - it shows values for the *scene* BUT if you know what your scene should look like (according to the result you want of course) then you can look at the histogram to see if it matches what you expect, if you have any blow-outs or if it's underexposed. I also use my histogram to verify my light meter readings...using a gray card or expodisc. I *always* check my histogram, at the beginning of a shoot and anytime conditions change considerably. I also shoot RAW, which is great for that added flexibility post-processing. However, I find that my images are *much* nicer if exposure is right SOOC, rather than tweaked post-process. ;-) Sep 10 05 04:10 pm Link kmwphoto wrote: Well, for one,..it's better to have a slight under exposure than a slight over exposure when shooting digital.. Sep 10 05 04:31 pm Link Paul Ferrara wrote: John Korb wrote: This doesn't make sense because earlier you said that your internal meter overexposed. Hi Paul, I'm probably not explaining this that well, but let me give it another try. Sep 10 05 07:08 pm Link John Korb wrote: Maybe. Sep 10 05 08:19 pm Link ok...sorry I could not read ALL the post due to time, but here are some tips (not sure if it was mentioned... If you are judging under exposure on your Monitor..make sure your monitor is calibrated correctly. If you are Judging it on your LCD Vs your Monitor...remember one thing, your LCD its just a preview to see compositions etc. it is not acurate, as a matter of fact, must have their LCD brightness right in the middle..that is just the brightness of the LCD and it does not affect youe file...said that if you set the LCD all the way to the left it may be a bit closer to the actual recorder file, on my 10D it still lighter on the LCD than the actual file, so I learn to compensate as long as the info does not reflect blown outs. Check those two things and test them again and let us know if that resolves your problem. As someone else mentioned...the Histogram is an overall guide..not exact..if you have alot of black in the pic it may show the histogram to be a bit underexposed..but when you see the file look at the subject and make your decision,,,that is what it counts and don't be fulled by gadjedts, use your instincts and test it out, even if you have to use another computer to compare results. Juancho Sep 10 05 08:41 pm Link Jay Farrell wrote: Mine Broke in 1999.. Sep 10 05 08:50 pm Link Hugh Jorgen wrote: Mine broke in 2003, but repaired it this month. Jay is right, using a lightmeter is better than messing around all the time with the exposure, or go on a rescue mission in photoshop later. Something that seems to have become a standard nfor some!!!! Sep 11 05 12:48 pm Link |
But I've never seen a situation where an incident meter got fooled. 
