I've seen examples of "upgraded" higher MP cameras with downgraded IQ and low light performance. The Fujifilm Finepix F30-to-F50 "upgrade" is one.
Marcus Watts wrote:
Of course the 5D2 will be improved. Anyone who knows technology would understand that and only someone desperate to have a cry would doubt it.
I also find it interesting all the talk about about wether the lenses are up to the task. Given that there have been no complaints in the last year from the 1Ds 3 i don't see why. Let's also remember that 21 megapixels is not even double the resolution of 6 megapixels.
I wouldn't take a chance with that. Just because it fits doesn't mean that it's designed be be used with the 5D II. We will know soon enough. If Canon introduces a completely different set of focussing screens for the Mk II, I will go with the Mk II version to play it safe. After all, they are not expensive .... I believe around $35 each.
ulrikft wrote:
According to dpreview, the old screens will fit the new 5d.
Got to play with the camera today at Photokina. Of course time for hands on was limited but i did test out a few things. First thing i noticed was the high res rear LCD, amazing detail and color. It was the first time seeing a hi-res LCD on a DSLR so for me it was a huge improvement. Also since i'm not familiar with the newer body menus, the menu layout was clean and easy to manage. The Canon rep says customizable menu lists(?)
Of course I had to get my hands dirty with the video...so I hit the Live View button and then clicked SET and started my video capture. Toyed around with some manual focusing as well as if you have the Live View AF box displayed you can use the AF On button to select your AF point (not one of the 9 AF points but any spot on the screen) all while recording.
Apparently details were still a bit scarce as far as video info goes per the Canon rep at the show. I'm not sure how it works on good video setups but i asked about Aperture settings if they were fixed with the lens or could they be set, sadly i didn't quite understand the guy so i really didn't get an answer.
Again, not familiar with the previous 5D autofocus but toying with the MkII i set the cam AF and tested it in both dark and bright "trade show" lighting and the camera focused quick and precise. Since I'm used to using AI Servo i set the cam to that and did some quick movements around the booth from far to close range subjects and the camera was able to track the target without hesitation and locked on without a problem.
I wish i had more 5D knowledge to better differentiate between the old and the new but i don't!
With what I experienced today in my first hands on with the camera I will be making my purchase next spring after i return from the Mid-East on a "vacation"
Mel Gross wrote:
Accurate, but a real pain to do. If a few people think that the 5D mkII auto focus is slow, then that method for manual is hibernating.
For macro and other tripod use it sometimes works out fine, but not for normal handheld shooting.
well it's a bit slower.
i have used it for handheld stuff though including wildlife in africa and nght shoots over LA though and you can't beat the precision of it.
yeah it's a HW change but one that would not require much change in AF logic coding
Do you think that represents a small effort? Changing hardware and software--even for what you think is "not much change" is an immense project in a complex device.
Changing tooling requires redesigning the assembly devices, numerous tests of the redesigned tooling, recoding the assembly computers, and retraining workers--and that has to be integrated into the assembly process of everything you didn't change. All that is just to build apparatus to build the new AF system.
The new AF code itself has to be designed, unit tested, system tested, and regression tested.
The new hardware has to be designed, unit tested, and system tested.
Then you put them into an actual device toether for further system testing and regression testing.
All of this requires hundreds of hours by expensive workers, and that's even before you do usability testing--before you even get to the "beta" stage.
There is no "easy" change. It's always expensive and time consuming.
WilliamG wrote:
Make sure you have 2GB. Yes, you can theoretically edit 21MP images with 1GB, heck even 512MB, but slowness will ensue. Just use 2GB+ and you'll be dandy.
I have played with a couple of 1Ds MkIII RAW files on my PC (Intel Core 2 Duo E6550 with 3Gb RAM).
Converting to TIFF or JPEG took about 12 s per file. I guess it's fine with me.
It also took about 6 to 7 seconds to display a 100% crop of a RAW file in DPP. This is a bit annoying because I like to check all pics for sharpness.
All this focus on autofocus is misplaced. In my own personal photography, 9 of my top 10 best photographs that I have taken were taken with manual focus cameras. This even though in the last 5 years I have taken ten times as many photographs using auto focus cameras as during the previous 25 years using manual focus cameras. It is something besides autofocus that brought out the very best in my photographic skills for those 9 of 10 best shots. That factor (I'm still not certain what it was or is) is what I am still looking for in a DSLR. I think that part of that factor is in the camera IQ - I used to shoot Kodachrome and print to giant ciba-chrome enlargements. There was something magic in the way you could peer into the depths of those slides and prints. I think that it is also in part the way manual focused cameras and viewfinder designed forced your brain to really think about the view through the view finder. The 5D is the closest I've gotten to the photographic experience of shooting with the F1n on Kodachrome. Hopefully the 5DII will take me one step closer. It is not the autofocus that is going to make that difference.
Mark Bishop wrote:
Pertinent point, but consider this.
Its the same AF that is still, to this day, in the 1V. A 'pro camera' no less.
In fact many consider the 1V the pinnacle of film cameras in general, I havent shot the Nikon so I cant say, but I know no one moans that the 1V has poor focus.
And yet, it isnt as good as the 40D in most situations.
That could be due to film photographers printing out their shots and viewing them at a reasonable distance as opposed to digital photographs that view their shots at 100 percent, 12 to 24 inches away on a monitor.
RDKirk wrote:
Do you think that represents a small effort? Changing hardware and software--even for what you think is "not much change" is an immense project in a complex device.
Changing tooling requires redesigning the assembly devices, numerous tests of the redesigned tooling, recoding the assembly computers, and retraining workers--and that has to be integrated into the assembly process of everything you didn't change. All that is just to build apparatus to build the new AF system.
The new AF code itself has to be designed, unit tested, system tested, and regression tested.
The new hardware has to be designed, unit tested, and system tested.
Then you put them into an actual device toether for further system testing and regression testing.
All of this requires hundreds of hours by expensive workers, and that's even before you do usability testing--before you even get to the "beta" stage.
There is no "easy" change. It's always expensive and time consuming.
listen over three years switching 1 single extra bar to f/2.8 (when they did it and more 30d to 40d in half the time) is not a huge deal!
sure it involves some work, but come on!
and the software to control it would not be that different.
now someone said something about slower lenses perhaps not working as well with dual f/2.8 1/3 DOF sensors and whether that is the case or not I don't know.
anwyay, i didn't want to talk about the AF anymore or complain about this and this is not so much a complaint now that they didn't do it but just saying come on that would hardly have been the end of all engineering tasks for canon!
the way you are talking they might as well have given up and stopp ed with the D30 since everything is too hard and expensive and time consuming to improve.
the way you are talking they might as well have given up and stopp ed with the D30 since everything is too hard and expensive and time consuming to improve.
The point is that every change costs a considerable amount of money, and in a profit enterprise, every expense is justifiable only if there is a sufficient promise of greater profit--that's a marketing decision, not an engineering decision.
Harry T wrote:
i asked about Aperture settings if they were fixed with the lens or could they be set, sadly i didn't quite understand the guy so i really didn't get an answer.
According to Laforet, you can set a fixed aperture with a kludge. Something about pointing the camera at something bright, or covering the lens, and then locking it down. He said that he didn't have an instruction manual, so there might be a better way. But it can be done.
For those that are griping about the 5d focus, here's something I found interesting today after rereading Rob Galbraith's 1d3 August update. Maybe it will make you rethink wanting the latest 45 point focus in the cam, and why Canon maybe decided not to use the 1d focus...
"Rounding out the list of static focus troubles we encountered before the firmware updates is a problem with the Mark III series autofocusing on low or no contrast stationary subjects. This would sometimes translate to the camera misfocusing without ever finding the correct focus, or not autofocusing at all. During a full day of shooting at a local conference centre, in mostly lousy light, the camera struggled or failed to focus perhaps a dozen times. To make it through the event, it became necessary to swap the EOS-1D Mark III for an EOS 5D........................
After a few minutes of struggling with the camera, and with the light going away, engaging Live View mode and then manually focusing saved the day. While we might have chalked the autofocus problem up to light that was simply too tough for any camera's AF, a 5D mated to the same lens accurately autofocused the same scene without incident during scouting a few evenings earlier, plus a Nikon D300 attached to an AF-S VR Nikkor 300mm f/2.8G IF-ED, set up alongside the EOS-1Ds Mark III at the actual shoot, was able to autofocus and get the distance right on the first try."
bobbytan wrote:
The grid screen is nothing more than just that - it's not meant to assist you in manual focussing.
I concur about the grid screen. I bought it for use primarily for use with the TS lenses. It also helps me in handheld shooting since I apparently 'tilt' a bit LOL. It makes no difference in focusing AFAICT but I like it and its not meant for 'field replacement'
mchahn wrote:
Ouch. So I guess you just pick the one you like the most and keep it in.
You definitely do not want to be changing screens in the field. You need a clean, well lit, dust free environment. Some place you do not mind having the lens off for a few minutes. Also, you want to have your blower handy to get clean the dust off the focusing screen and SI plate.
The more blog posts and comments Canon sees from people requesting this, the more likely they will pay attention and make it happen.
I also have a bunch of 5D MKII related links on the post (audio interview, sample images and videos, articles, etc). If you have a link to add, please leave a comment there and I'll add it.
With regard to stills, I have a 3-4 year old Alienware single-core 3.0 GHZ machine with 2GB of RAM, and when I have 4x5 film Creo scanned, I end up with individual TIFFs between 250 - 400MB.
I can edit this stuff no problem, and have done so now for a couple years. But, the limit of what I do typically is level correction and dust spotting. If you are using a lot of layers, it will make a difference.
1080p video? Yeah...you're going to want a dual-core box with loads of RAM.
