MikeRichards wrote:
So, I am going to put up a couple of shots in a minute but I went out and shot two images, exactly the same shot, nothing changed but the file format. The jpg was MUCH cleaner than the RAW file. I wonder if that is something to do with LR or what but I was fairly happy with the results from the jpg.
I'd chalk that up to Canon's NR algorithms removing the noise more than anything else?
ausmr wrote:
For clarification 1Ds3 cannot do ("natural") ISO 3200.
The max is 1600. You can extend it to 3200, which I never do.
You can push 7D to 12800, which is substantially more.
And I haven't yet tried this in the field.
-mirek
none of the cameras, aside from perhaps the 7D can really do natural ISO 3200 and that doesn't mean the 7D won't be a lot noisier at ISO3200 than a D700/5D2/etc.
MikeRichards wrote:
Okay, here is what I mean. Canon 7D, 50L, f1.2, 1/2000, ISO 320.
Thanks for posting.
You are not imagining things. This is really noisy
The thing is, yours are not isolated samples.
All the samples that I've looked at are either noisy or show NR artifacts (digital dandruff).
It’s going to take some time and multiple formal reviews before people start accepting that the 7D is noisier even than the 50D.
Don’t know what Canon was thinking when they decided to put this sensor in an expensive almost pro body .
Hopefully this is not their new standard for ISO performance on 1.6x cameras.
And here I thought it was my imagination but at ISO 800 and up it seems a lot noisier than my prior bodies. Another photographer friend of mine keeps trying to tell me it's because it's more sensors and 18 mp. If I down sample it it'd be fine. I shouldn't have to do that I'm thinking or what's the point of having the megapixels in the first place?
But right now I'm thinking that the 7D is going to prove useless for concert shooting.
(My copy seems to not focus sharp with any lens wide open. especially on my longer lenses.) I'm trying to love this camera but I'm wondering if I just blew close to 2 grand on a camera that does nice portraits.. you know anything f8 or better if the lens isn't too long. but so far has been hit or miss with anything high ISO or moving subjects or god forbid less than full sunlight or flash.
Jack M wrote:
Can anyone that owns the camera comment on the battery life?
I don't want the grip if I don't need it.
I've taken 693 test shots so far (started at 95% charge) with a lot of chimping and some video (with review), battery life is still 45% as per the camera.
I moved to this camera to stay away from the grip... :-)
All this talk of noise - the examples all look pretty good to me. Instead of noise, I will refer to it as absense of plasticky tones. It looks like very fine film grain to me, and it will improve the sense of detail in your prints. I noticed this when I switched from 5D/1D3 to 1Ds3. At 100% it looks noisier, but in print it just sparkles ;-)
Just on a hunch, I called a couple of Best Buy stores in the Columbus area during lunch today. I had called them previously during the week to no avail. To my very pleasant surprise, one of the stores (Polaris) had received two camera kits in stock only this morning, and had already sold one. I immediately told them to hold it for me, and I bolted over to the store. I just couldn't wait any longer, as I am scheduled to be out of town for a week on business, and didn't want my pre-ordered shipment to stay in UPS' hands.
I'll try to put the 7D through various paces even while away on business, and hopefully I'll be able to provide some useful observations between now and when I return on the 19th.
To help dispel the noise issue, do you know of an easy to use formula that works out equivalent viewing distances. For example, to be able to say "If you view a 18.1 MP sensor at 100% on your monitor at 2 feet, it is equivalent to viewing a X by Y print viewed at 2 feet."
I suppose monitor size/resolution is a component, but I just don't think some people understand the fact that viewing high resolution images at the pixel is equivalent to shoving your nose up against a print.
