Thats not a fair example as it's clearly a stitched image - if you wanted to shoot that in 1 frame any DSLR is the wrong tool and 22mp+ digital MF is the way to go. And if you regularly stitch like this then any DSLR will do the job nicley.
> Thats not a fair example as it's clearly a stitched image
...which makes my point, that wedding photography does "make massive demands on resolution" at pretty much every wedding. I would love to be able to take this shot in one exposure because it would save me a lot of effort, but it is 12000 pixels wide and my budget doesn't stretch to an MF back, let alone something with 12000 pixels per line. I try to avoid this kind of picture, but in this case the bride specifically asked me to do shots of the whole family. What can I do? They are paying the bill. :(
I think the photographer chooses his tool and makes the client want the quality he can provide, at least wedding photographers were the first to shoot 3mp cameras. speaking for small business only
Point taken but if you regularly shoot large groups like this at weddings (or any event for that matter) and require the kind of resolution you have indicated in the crop your only options are to stitch, use 6x7+ film or 22mp+ digital - even a future M9 aint going to cut it so why wait for something that will still not do the job anyway. Additionaly stitching prevents distortion of faces at the edge which you would have got if you did it in 1 shot from the same position.
Dave, I would really like you to tell us if you were comparing 0.6 of the Canon frame to the whole M8 frame by not up-rezzing the M8 frame. I think it is really important because people read this forum and they might get the impression from your comparison that the M8 resolution is 1.7 times as good as it is. It is an easy thing to overlook, but the oversight could be very very misleading to people who have an idea of 1Ds2 quality and who are considering the M8.
Are you still there Dave?
(more apologies to Clarke and Kubrick)
Why 0.6 and 1.7 ?
You need to count the linear pixels to get an idea of the added resolution (which is a linear measure).
1DsII has 1.27x the M8 resolution, or M8 has 0.79x the 1DsII resolution.
I agree though that while per-pixel sharpness is higher in the M8 (CCD and lack of AA filter), you need two identical frames to judge both systems.
Dave's comparison really shows only that if you crop a 1DsII image about 30% more than a M8 one, the M8 will be slightly better... even if shot through a window...
Yes - thanks for correcting me Marco - I should have used the ratios of the baselengths not the areas. It is still a significant misrepresentation though, equivalent to downrezzing the 1ds2 image to 10 megapixels before assessing it, and it shows that when judging resolution we should be careful not simply to judge 100% crops.
In fact it calls into question any comparison between 1Ds2, 5D, DMR, M8, D2xs, and D200, and of course, others, UNTIL the lower megapixel cameras have been interpolated into a number of pixels that matches the highest volume sensor in the group. I think a great deal of online discussion has taken place in the past on this forum, and others, without properly making this adjustment. Consequently there have been quite a few bogus claims about resolution equivalence between some higher and lower megapixel devices.
brainiac wrote:
BTW - Dave, I don't want to bite your head off at all, and I am grateful to you for sharing your images, but the reason I would like to see the full frames of the doorway shots
I am currently 300 miles away from where the files are, so won't be able to post these for a few days. These are 100% crops, and not resized. The 1Ds2 is marginally larger that the M8 crop because of the very reasons you are questioning them..... some frames were approx the same FOV, and some were larger/smaller. I found the ones nearest for the shots I posted, but they are not perfectly the same.
I understand why you are questioning the crops 'as posted' and have no problems with that at all. Human knowledge only moves forward by first questioning, and then proving or dis-proving .
As I have said, I don't see the M8 replacing the 5D for the things that the 5D is good at. Why would it need to? Pick the best tool for the job, on a job by job basis
When I get back, I will find out the files, and maybe post several variations as required.....
> When I get back, I will find out the files, and maybe post several variations as required...
That's very kind Dave - thanks. I am sorry to have been so insistent, but there has been a lot of hype about the M8 and I really want to get to the bottom of what it can do in an objective and truthful manner. Despite the fact that this is FM's very own 'Hands-on M8' thread, that's proving pretty hard at the moment.
DaveEP wrote:
I am currently 300 miles away from where the files are, so won't be able to post these for a few days. These are 100% crops, and not resized. The 1Ds2 is marginally larger that the M8 crop because of the very reasons you are questioning them..... some frames were approx the same FOV, and some were larger/smaller. I found the ones nearest for the shots I posted, but they are not perfectly the same.
I understand why you are questioning the crops 'as posted' and have no problems with that at all. Human knowledge only moves forward by first questioning, and then proving or dis-proving .
As I have said, I don't see the M8 replacing the 5D for the things that the 5D is good at. Why would it need to? Pick the best tool for the job, on a job by job basis
When I get back, I will find out the files, and maybe post several variations as required........Show more →
If the 1Ds2 shows as bad as the first example my advice is to send it to Canon repair...The M8 should go back also to "get in line" for the big Leica fix.
Good luck.
The Reb
> Does anybody on this thread actually have their hands on a M8?
Dave says he's going to come back with actual pictures.
I am very surprised that no more comprehensive and credible comparison between the performance of the M8 and 5D has emerged on this thread.
It is a shame that various trusted reviewers allowed themselves to be gagged by Leica too.
If the camera gave worse results than a 400D, then it wouldn't be surprising that objective analysis was not being encouraged by M8 users who have spent so much, but a number of people have said they are thrilled by the camera's performance. Let's hope there will be a few more who are willing to share their results and allow them to be assessed in an unsentimental way.
If there aren't, then I think we have to ask if it's a conspiracy of silence, with the obvious implication that people are unhappy with the camera's output.
Or it could be that in my quest for some actual data on this subject, I have just bored everyone to death.
The 100% crop (means 1-1 pixels no?) on my screen (96 P/in) at 2.7" (and its not quite that) is less than 300 pixels and it is a crop that is less than 1/20 of the total width. so 5500-6000 pixels
that what a 22Mp camera has.
. 
?