I knew I needed more time to reach any conclusions and still do. I have been experimenting with ACR vs DPP. Softness issues are related to noise reduction in DPP. When converting with ACR, noise is less well controlled but detail is much better. I think I would rather do my own noise reduction in NN and retain detail. With no noise reduction, the 50D files converted in ACR are showing as much detail, or maybe a little more (in some cases) than the 1D2N. Noise levels are about the same or maybe slightly better than 1D2N, which is pretty good considering the increase in mpx.. I also think the NO NR noise levels are better than the D300. I'm much more encouraged today than last night....Show more →
I collected some 50 images from the web... Some converted by DPP and others by ACR/LR. I came to the same conclusion as you. The DPP converted images are much smoother but at the expense of detail. The ACR/LR converted images show a bit more noise but show a great deal more detail. Maybe you get the best possible IQ if you convert by ACR/LR and apply noise reduction in NN indeed...
DPP doesn't work for me. So IF I decide to buy the 50D, I will convert RAW images through ACR/LR only.
Questions for you: if you use ACR/LR for RAW conversions, what ISO value is still usable in your opinion...
a) if you use ACR/LR default settings only (+25 color noise)?
b) if you use the noise removal tool from ACR/LR?
c) if you convert by ACR/LR and apply NN?
I don't shoot beyond ISO1600 myself. Based on this, would you recommend the 50D to me?
Daan B wrote:
Questions for you: if you use ACR/LR for RAW conversions, what ISO value is still usable in your opinion...
a) if you use ACR/LR default settings only (+25 color noise)?
b) if you use the noise removal tool from ACR/LR?
c) if you convert by ACR/LR and apply NN?
I don't shoot beyond ISO1600 myself. Based on this, would you recommend the 50D to me?
As I've said before, I hate to reach conclusions without spending more time with the camera. Last night, I was seriously considering selling the camera but this morning, I'm much more encouraged having tried ACR vs DPP. Detail is much much better with ACR conversions and I need to go back to DPP as see what I can do by making some adjustments there. I think ISO 1600 is useable especially if you compare by normalizing identical crops to the same size.
I shoot mostly BIF's (or at least that's my main interest) and I've been very anxious for this camera. I thought the D300 would be the perfect BIF camera but it hasn't worked out for me primarily due to lenses. I'm a pilot by profession and travel all over the world but I can't take the big telephoto lenses easily so I'm limited to the smaller/long lenses like the 400/5.6 in Canon or the 300/4 in Nikon. I've alway enjoyed much better results from the 400/5.6 or 100-400L than I get from the AF-S 300/4 with a TC. The Nikon 80-400 would be great if the focus was a little faster. The D300 is a great camera though and there are many things I prefer with that body. IQ is not one of them, especially in the telephoto range.
I hate to recommend based on my limited time with the camera but so far it's looking good to me. I'll continue to add any meaningful comparisons I find. So far, I'm impressed with the auto WB (I like the out of camera colors better than the 1D2N), also with the AF (seems quite fast even in low light). Hope this helps
csd2020 wrote:
As I've said before, I hate to reach conclusions without spending more time with the camera. Last night, I was seriously considering selling the camera but this morning, I'm much more encouraged having tried ACR vs DPP. Detail is much much better with ACR conversions and I need to go back to DPP as see what I can do by making some adjustments there. I think ISO 1600 is useable especially if you compare by normalizing identical crops to the same size.
I shoot mostly BIF's (or at least that's my main interest) and I've been very anxious for this camera. I thought the D300 would be the perfect BIF camera but it hasn't worked out for me primarily due to lenses. I'm a pilot by profession and travel all over the world but I can't take the big telephoto lenses easily so I'm limited to the smaller/long lenses like the 400/5.6 in Canon or the 300/4 in Nikon. I've alway enjoyed much better results from the 400/5.6 or 100-400L than I get from the AF-S 300/4 with a TC. The Nikon 80-400 would be great if the focus was a little faster. The D300 is a great camera though and there are many things I prefer with that body. IQ is not one of them, especially in the telephoto range.
I hate to recommend based on my limited time with the camera but so far it's looking good to me. I'll continue to add any meaningful comparisons I find. So far, I'm impressed with the auto WB (I like the out of camera colors better than the 1D2N), also with the AF (seems quite fast even in low light). Hope this helps ...Show more →
Thanks for sharing your impressions so far... Since I have an interest in BIF as well, I am curious to know what you think about the 50D's AI Servo tracking... From the looks of it, I think ISO 1600 will be usable, especially in good light. I don't think ISO 3200 will be very usable. At least not without doing some heavy PP + noise removal.
BTW One thing I noticed about your samples, is that the 1D2n crops are much sharper than the 50D's (when equalized). Maybe your 400 5.6L must be calibrated to your 50D? Or maybe the 400 5.6L can't keep up with the 50D's sensor? Were these crops focused manually?
thw2 wrote:
Whatever max ISO you felt comfortable with in the 40D/450D, you'll experience the same with the 50D.
I have never used any Canon crop cam before, only FF (5D + 1Ds3). But I get your point... the 40D and 50D are probably very similair in noise performance, especially when converting RAW through ACR/LR.
Daan B wrote:
BTW One thing I noticed about your samples, is that the 1D2n crops are much sharper than the 50D's (when equalized). Maybe your 400 5.6L must be calibrated to your 50D? Or maybe the 400 5.6L can't keep up with the 50D's sensor? Were these crops focused manually?
What an astute observation. I was getting discouraged by 400/5.6 comps. I did some micro-adjustments and found +10 the sweet spot with the 400/5.6. I have re-posted those comparisons if you care to check. I uprezzed the 1D2N for comparison (held up surprisingly well).
csd2020 wrote:
What an astute observation. I was getting discouraged by 400/5.6 comps. I did some micro-adjustments and found +10 the sweet spot with the 400/5.6. I have re-posted those comparisons if you care to check. I uprezzed the 1D2N for comparison (held up surprisingly well).
Could be possible for you to take the exact same shot at ISO 3200 with RAW and sRAW1, and compare the noise (of course reducing the resolution of the RAW file in PS to the same level of the sRAW1)?
I want to know if pixel binning at the sensor level has an increased effect on noise reduction compared to a pure resolution reduction in PS.
One more picture, this time ISO 1600, 2.8, 1/60, Canon 16-35/2.8L MkII
I'm using sRAW (7Mb), and converting to jpg using DPP, with default values, except:
Sharpening : 4
Luminance Noise : 4 (default)
Chrominance Noise : 0 (I like it that way, so I keep detail)
Here you have an example of big dynamic range, from very bright inside the store, and much more darker outside.
csd2020 wrote:
What an astute observation. I was getting discouraged by 400/5.6 comps. I did some micro-adjustments and found +10 the sweet spot with the 400/5.6. I have re-posted those comparisons if you care to check. I uprezzed the 1D2N for comparison (held up surprisingly well).
ebe5000 wrote:
One more picture, this time ISO 1600, 2.8, 1/60, Canon 16-35/2.8L MkII
I'm using sRAW (7Mb), and converting to jpg using DPP, with default values, except:
Sharpening : 4
Luminance Noise : 4 (default)
Chrominance Noise : 0 (I like it that way, so I keep detail)
Here you have an example of big dynamic range, from very bright inside the store, and much more darker outside.
Go pull the fruit shot from dpr, apply a curve, and run it through noise reduction. You end up with a very usable iso6400 image. I don't think you could get the same thing from a 40d with the same amount of effort.
Corndog is right! If you believe the crap you read on DPR or that website then you will make an uninformed decision about the 50D. Most of the morons on DPR would rather bitch about something they know very little about, like photography, while convincing themselves that their an expert.
Most of the shots used as evidenced in that website were poorly exposed and even the best high ISO noise cameras will show noise when not exposed properly.
Absolutely worthless as far as credible information.
