View Full Version : R-D1 noisy at ISO 200?
Got my new R-D1 and CV 35/1.7 Ultron. Took it out (cloudy day) for a few tests. Below is a crop from one picture at ISO 200, f4, 1/125s. Am I crazy or is this a lot of noise for that setting? The second picture is the same thing after Neat Image processing, but I don't think I should be needing anything like that. My Canon 10D certainly doesn't. Any clues?
BTW, converted with Bibble Lite, vanilla settings (no sharpening/noise reduction/etc). All pictures look like that.
My Canon 10D certainly doesn't.
I'm not an RD-1 expert, but I understand your surprise. The Canon 10D has some of the cleanest ISO 100 and 200 photos I've ever seen, RAW or JPEG from the camera.
The sample you posted is too small to really tell, though.
They're both 250x250 crops. What I do is to put each on one tab on my browser and switch back and forth between the tabs. The difference is quite clear then.
I found the R-D1 very clean up to ISO 400, quite usable at 800 and moderate at 1600 but I could clean it up pretty well. However, compared to the M8 it was a level below at 400 and maybe two levels below at 800 and three at 1600 but the M8 at 2500 (3200 actual) needs a lot of cleanup.
What is impressive to me (so far) is the amount of noise present at the lowest possible ISO setting. It just seems really too much, and I surely can not qualify it as 'very clean'. I certainly did not expect to have to use noise reduction software at the lowest ISO setting. As it stands, the same size picture (the 10D is also 6MP) shows much more detail on my 10D than on my R-D1. I'll keep on testing for a few more days.
One could say I'm spoiled by the 10D. Perhaps. But I'd never imagine seeing what I'm seeing from the reviews I read before taking the plunge. I'm hoping somebody can explain this finding... Images from my 10D with either my 35/2 or 50/1.8 look much, much better at this stage.
PS: I must say that the R-D1 noise seems much "nicer" than the 10D noise. When the 10D is noisy (say ISO 3200), the noise is a "bad" one, not related at all to film noise. Nevertheless, at ISO 100 (or 200) it just has none, and that's exactly what I expected from the R-D1.
Are you doing any exposure adjustment in Bibble? Brightening an underexposed image can increase noise. Also, have you tried any other image processing program for a comparison?
Remarkably clean at ISO 200.
Some beautifully smooth 11x17" prints will attest to that.
I think you need to check your workflow. Nobody has ever complained about noise from the R-D1 at ISO 200 or 400. In reality noise at 800 and 1600 is not a problem either - not as clean as some at that level, but certainly not unusable. Check the results using Epson Photo Raw. If you are having to lift exposure significantly in the raw converter this will most certainly add noise, but at 200 or 400 even that should not be problematic.
Make sure you're not underexposing. I often dial in +2/3 stop exposure compensation which tames any noise issues.
The meter tends to underexposure a bit.
I've used a lot of digital cameras, and the R-D1 has as clean ISO 200 performance as any of them I've tried.
Noise is fine at ISO 200. Used to own a Canon 10D, and personally prefer the R-D1 files.
I dont see noise in the firtst picture, but the "neated" version looks sharper and more saturated. did you apply any unsharp masking or other sharpening?
I agree with Didier, if you overlay the pictures and flick between them, the second shows sharpening artifacts and slightly higher saturation.
I can't see any noise at all on my screen.
Of course, we're all different, but there's nothing there that would worry me.
Regarding the noise, it is most visible on the first picture on the green border of the book. Maybe it is more visible on an LCD, but to me it is clearly visible. Even the white part of the cover has noise. If I look for it, I'll find noise everywhere without having to try hard. Not so with the 10D. The second picture shows sharpening artifacts because Neat Image sharpened the picture... But I did not touch the saturation at all.
Regarding exposure, I don't think it is the problem in that case. I can see the noise without touching the curves at all. But I noticed that compared to the 10D it underexposes by about 2 stops. Since the 10D usually overexposes, I'll try shooting in +2/3EV and see what happens.
Now, the converter might be the problem. I haven't tried anything but Bibble (have only a Linux box, saving to buy a Mac), and given how unanimous seems to be the opinion that at 200 there is no noise, I'll have to borrow a computer from someone around me and try the Epson software.
vBulletin® v3.6.8, Copyright ©2000-2013, Jelsoft Enterprises Ltd.