Mel Gross
Offline
• • •
Upload & Sell: Off
|
Lotusm50 wrote:
I'm not so sure the original 5D comes up "wanting' relative to the Nikon D3/D700. While the Nikon goes higher, at ISO the 5D is capable of it seems to be able to produce nearly equivalent results as the Nikons -- at least from what I have seen. I agree, however, that the 5D MkII should be even better -- at least from what I have seen to date.
I was not suggesting that downsizing the image file "reduces" the noise, but rather as you suggest it become less apparent, due to be smaller. There have been several comments about the a900 noise performance being good and competitive, if you look at the file downsized to the same size as the Nikon. My comment was made with those comments in mind.
...Show more →
What I've read so far about the noise from the A900, is that it's about the same as that from the 1Ds (the only other comparable camera actually out right now). The hands on reviews have said that it's a bit noisier at 400 and above. Given the high level of noise reduction that Sony feels necessary, the smearing of fine detail that both reviews I've read have noticed, show that there is a problem with the ability of the chip itself when it comes to noise.
It seems that Sony biased the chip more towards resolution than total IQ, and is hoping that noise reduction will take care of the rest.
Sony is not a long time camera manufacturer in this still photo business. I think they are treating it much more like a consumer business unit than a professional business unit. From my own experience with Sony's professional ENG business, I can see the difference in the marketing. The A900 leads me to believe that Sony really thinks the the semi pros who buy $3,000 D-SLR's will simply look at the pixel count. We know that for most, this is not true.
The 5D, however is being bought in large numbers by the same pros who have bought 1Ds's, and even 1D's, as back-ups, if for nothing else. And it enjoys a large market in the wedding and event photohraphy market, as well as elsewhere.
Sony can't claim to have an existing market in any of those areas, nor can it claim a high end pro market which will want a smaller, lighter, less expensive body as a back-up, as Canon (and Nikon) have for their medium price FF cameras.
Sony really needed to hit a home run with this, and it surely didn't. Maybe a double.
I'm sure it will lure a few away from Canon and Nikon, because there are always those for whom the grass is greener, etc. But, it will do more to keep most people buying into the systems they already have.
Overall, I'm just not impressed with the thing.
Getting back to the D3/D700, I have my own findings. Using a friends D700, I did some comparisons with my old 5D, which has better noise control, and a larger dynamic range than either the 1D, or 1Ds.
What I find is that up to 200, the cameras are about equal, except when I tried to bring detail out of shadow areas. The Nikon did better there, which shouldn't be surprising.
At 400, it was a bit of Canon, and a bit of Nikon, except, again, when going into shadows.
At 800, the Nikon did pull ahead, and you could see it, no question. While I like the Canon "grain", as I have CR set to the standard 25% chroma, 0% luminance noise reduction, the Nikon simply has less. Shadow correction now becomes impossible for the Canon, but surprisingly good, still, for the Nikon.
1250 (a speed I like a lot with the 5D for indoor and portrait) is noisier, but still pleasing on my 5D. I've read a description of it as velvetly, which, if the noise reduction is as above, is true, but again, Nikon is simply quieter. By a lot.
1600 is more of the same, but the 5D is now, running out of steam. The Nikon still looks quite good, and even at 3200 is pretty nice. Shadows do begin to start acting up at 1600, but are not terrible until 3200 when opening them up.
I haven't been posting images, because I don't find posted images to be helpful. What I find, and this is my own belief, of course, that what really matters is what shows on the print. so, the printers used will have their affect as well.
I have tried to print from some samples seen on the web, but the results are often ambiguous.
Since I really don't care about the file itself nearly as much as the printed results, I don't post examples.
Often, I find that differences are more muted when printed out.
This post wandered much further than I first intended.
Sorry about that. I just hope it gave something useful to the thread.
|
