Originally posted by mattyjm Thanks for your replys every one. Some of that technical stuff has just gone over my head!!!!!
I'd recommend doing some research into the topic of exposure and metering. Bryan Peterson's "Understanding Exposure" is the usual recommendation here. I learned a lot from the National Geographic Field Guide to Photography. Really, any decent book on photography (that focuses on SLR's as opposed to P&S cameras) should explain these concepts. You really need to get a solid of understanding of these concepts to get the most effective use out of any SLR.
Quote: I am not convinced that you can call this normal. I have used Canon digital Slr's and have not had this problem.
I'd challenge you to take pictures of the same scene at the same time with both cameras to see if the differences are as big as you are remembering. I'm guessing you'll find they aren't: Canon cameras designers surely read the same books on exposure and metering than Pentax camera designers read. If Canon produced that first picture much brighter, it would be blowing out highlight left and right, and violating ANSI standards for average brightness in the processes.
Quote: I generally take my pictures on center weighted metering as I did with that sample one I posted.
OK, so in that case, it wasn't the camera attempting to preserve highlights: it was the camera trying to make the scene look a little darker than an 18% gray card, as camera meters have been doing for decades. When shooting a backlit scene like this - one in which the background is quite bright but your subject is much darker - it is and always has been necessary to use exposure compensation to force the camera to overexpose the highlights in order to produce the exposure you want in the shadows.
Quote: I find I also have the same problem with nigh photos. I have attached more samples.
The nigh pic shutter speed was 30 sec.
30 seconds is the longest shutter speed available; that's probably why it was chosen. And assuming the increasing aperture or ISO was not an option for the camera (tha would depend on your camera settings and lens used), then that may be as bright as the camera is capable of rendering the scene with that lens and using those settings.
Although again, those bright lights are already clipping - in terms of *average* brightness, I'm not sure that's really so far off. If you want the rest of the scene to look good when including light sources in a picture, you need meter a part of the scene that doesn't include the light sources.
EDIT: actually, I checked the EXIF, and the shutter speed was actually 1/20, not 1/30. So the camera *could* have chosen to expose half a stop more. Meaning it really is just a matter of the bright light source within the scene dominating the average. Metering off the ground would have helped, but then you'd have run into the 30 second limit. So you'd have had to have tried a larger aperture (you were at f/27 - that's crazy small!) or a higher ISO or both. Another thing to note is that it is true that in very dark conditions, the metering of any camera is not too accurate. The lower limits different from camera to camera, and metering off the ground at night does likely exceed the lower limits of what the meter cn deal with effective. Which doesn't mean you can't take pictures - just that you might need to use some trial and error.
As for the other scene, that's also a classic demonstration of what happens when you meter a white or nearly white object - the camera tries to render it as 18% gray or a little darker. Exposure compensation has always been the traditional way of dealing with that.
Again, any book on exposure will explain all of this.