maharzan said:
I don't use auto focus. I am using AI Servo but all settings are default. I just tested few shots of an object and every time I hear that beep (half pressed), I take the pictures. I see that not all images are sharp. perhaps like 2 out of 5.
This doesn't make sense to me. How can it be that you, "don't use autofocus," but are using AI Servo, which is an autofocus mode? Also, in AI Servo mode there is no focus confirmation, so when you half press the button there should not be a beep.
Ok, looking back at a previous post:
maharzan said:
Who ever thinks, a 1800$ camera will have flaws??
Any piece of equipment can have flaws. Sometimes there's a defect, other times the flaw is with the design, or a lack of proper understanding of the capabilities of the device.
maharzan said:
I was never professional so, all I did was manual focus, adjust aperture and shutter speed to I could get the meter to 0 (recommended exposure?).
Honestly, that sounds like how many pros shoot - full manual exposure and manual focus. They do it successfully because they've been shooting like that for a long time (I'm certainly not a pro, but I started shooting with film SLRs in the days before autofocus was even available - automatic aperture control was the latest and greatest innovation). But shooting like that takes skill and an understanding of the limitations of your gear (more below). The automatic and semi-automatic modes do make it easier for a novice/amateur to get better images.
maharzan said:
I always got soft image too and I blamed myself all this time as it was very difficult to focus. I mean, even if you feel its focused in the LCD or view finder, when you look into the computer, I always saw photos were out of focus.
The resolution of your computer display is much higher than the camera's LCD. The on-camera review is a bad place to judge sharpness, as only significantly misfocused images will be obvious. Note that there's a substantial difference between actively focusing (either through the viewfinder or using Live View) and reviewing an image. Actively focusing (racking back and forth and centering on the best focal plane) can be quite accurate, especially in live view. But, depending on your lens, that process might not be very accurate using the viewfinder.
maharzan said:
Recently, I bought 35mm 1.4L Canon and videos / photos were soft with this lens too.
This is one of those limitations I mentioned. Old film cameras had focusing screens with a split screen and microprism collar to aid in manual focusing. More importantly, current cameras are geared for consumers using consumer lenses and autofocus. One problem with slow lenses (e.g. kit zoom lenses with f/5.6 max apertures) is that the viewfinder is very dim with those narrow apertures. So, camera makers use focusing screens that are microetched to result in a brighter viewfinder. That sounds good, but the consequence is that the viewfinder can only show the depth of field of ~f/2.8. So, with your 35mm f/1.4L, the real DoF is much thinner than what you're seeing in the viewfinder (i.e. something can look perfectly in focus in the viewfinder, but actually be out of focus). Live View does actually show the true DoF for fast lenses, which is why it's better to use Live View for manual focusing. For some cameras, users can change the focusing screen (for my 5DII, I have the EG-S focusing screen, which shows the true DoF of fast lenses, but then slow lenses are dark). The 7D does not have a user-replaceable focusing screen (at least, that's Canon's position, although there are 3rd party screens available, great care is needed to avoid damaging the transmissive LCD).
Bottom line, with the 7D and a fast lens, meaning a wider max aperture than f/2.8, you're better off using a properly microadjusted autofocus, or manual focus with Live View and 10x magnification, than trying to manually focus through the viewfinder.
maharzan said:
At least when I shoot my MBP's keyboard, when I focus on say Y key, the focus is actually on the N key. I even did MA but then I just reset it to 0. I took several pictures outside and they seem good. I will still need to experiment more and practice focusing.
Here's another area of not understanding exactly how the system works. You look in the viewfinder, put the AF point rectangle over what you want to focus on (the Y key), half-press the shutter button to AF, then take the picture. The problem is that the actual AF point on the AF sensor is larger than the little box that represents it in the viewfinder.
You know what feature you're trying to focus on, but the camera does not. In your example, the actual AF sensor might be covering the from the F6 key down to the N key, and from the R key to the I key horizontally. The camera is going to lock onto an area of high contrast anywhere in it's sensitive area, but in your test the whole area has highly contrasting lines (black keys with a silver surround, at least on my MBP), so the AF systems doesn't know where to lock focus.
I posted some examples of this on
another forum thread, where the issue was a fast lens (135mm f/2L) lens supposedly focusing inconsistently, but in reality the lens and AF system were doing their job correctly, but the 'test' was not set up correctly (just like yours wasn't).
Hope some of that helps...