sk66
Advertiser
- Messages
- 8,677
- Name
- Steven
- Edit My Images
- Yes
I would agree with ISO performance (data collection) but not so much with "detail" (resolution).
(related but different)
Full Frame v Crop - Real World Opinions
- Thread starter taxboy
- Start date
(related but different)
RedRobin
Dances With Dogs
- Messages
- 9,314
- Name
- Robin
- Edit My Images
- Yes
Although both this discussion and the differences described are very interesting, what matters most is what type of subjects you shoot and consequently which lenses you use.
Full-frame is 'better' for some subjects and Crop-sensor is 'better' for others - It's simply Horses-for-Courses.
The beauty of a DSLR camera system such as Canon and Nikon offers, is that we have choices to buy what suits our individual needs best. The rest is interesting but it's somewhat academic.
Personally, I only take images of wildlife with DSLR and only have telephoto lenses. I use Canon and therefore I have Canon's current flagship crop-sensor body - The 7D Mark II. The reason I prefer crop-sensor (apart from cost) is both for reach and because the majority of wildlife shots get cropped it doesn't matter what's at the edges. In fact, I think that there are usually more megapix in the central areas of a crop-sensor than in a full-frame. Swings vs Roundabouts
I don't care how my images compare with others, I only care about whether my images satisfy me. I'm an amateur and not a professional in competition.
Full-frame is 'better' for some subjects and Crop-sensor is 'better' for others - It's simply Horses-for-Courses.
The beauty of a DSLR camera system such as Canon and Nikon offers, is that we have choices to buy what suits our individual needs best. The rest is interesting but it's somewhat academic.
Personally, I only take images of wildlife with DSLR and only have telephoto lenses. I use Canon and therefore I have Canon's current flagship crop-sensor body - The 7D Mark II. The reason I prefer crop-sensor (apart from cost) is both for reach and because the majority of wildlife shots get cropped it doesn't matter what's at the edges. In fact, I think that there are usually more megapix in the central areas of a crop-sensor than in a full-frame. Swings vs Roundabouts
I don't care how my images compare with others, I only care about whether my images satisfy me. I'm an amateur and not a professional in competition.
- Messages
- 1,078
- Name
- Matt
- Edit My Images
- Yes
One taken with a d810
The other I think was a D5300.
I know Kinda apples and oranges.
I went from a D7k and was worried until the first time I tried to get the milky way and could see it clearly on the rear screen of the D810. Then I knew I'd made the right choice.
At some point I'll try the same shot with my D7k with a 14-24 and then again with the D810 and a 14-24.
Like I said not scientific and results WILL vary but for night photography FF wins.
Going even further maybe saying V High ISO performance would win. But at the moment you only get that on FF.
That same shot I've taken with the D810 would be even more OMG with a D4s. Less MP's but more ISO performance.
EDIT: Just to add My e-Penis defo got bigger when I open then D810 box.....Just saying
Last edited:
sk66
Advertiser
- Messages
- 8,677
- Name
- Steven
- Edit My Images
- Yes
"Detail" can be light collected or it can be "resolution" (MP's) and they counter each other; both are pixel size related.
If you do take the same shot w/ a D7000 and a D810, at a pixel level they will look about the same (i.e. same shot/same lens/D810 cropped). But when viewed at the same size the compression of the D810 image will cause it to look cleaner with better color. But the D810 "resolution" (pixel detail) will also be compressed to ~D7000 equivalence.
In all cases the larger sensor wins in overall IQ. But if you take the smaller sensor image and compress it (print smaller than 1:1) it will also benefit in noise reduction/color improvement. Possibly to the point that the difference is negligible/insignificant for the intended use. The D7000 image would be compressed slightly for an A3/250dpi print; more so for a 200dpi print (and very significantly for 2048x web display).
woof woof
I like a nice Chianti
- Messages
- 39,671
- Name
- Alan
- Edit My Images
- No
I went through this when I first bought into MFT. I compared my GF1 and then G1 to my Canon 20D and 5D and after processing each shot for best effect and comparing on screen and after spending lots of dosh printing up to and including A3 I concluded...
- At low to medium ISO's you have to look very closely to see any differences.
- At the highest ISO's and when looking closely (do a big print and look very close or look at very high magnification on screen) and FF wins.
- This is a gear head geek thing. Normal people don't care.
I gave up comparing low to medium ISO shots when trying to show someone two shots, one taken with my G1 and the other taken with my 5D. When shown the first shot which just happened to be the G1 shot the person I'd roped in excitedly exclaimed "I want that picture!" I decided then and there it was a non issue and I stopped comparing shots... until I got my A7 and GX7
After using them both for some time my views haven't changed. The A7 is better than the 5D and the GX7 is better than the G1 and the A7 is better than the GX7 but only at the highest ISO's or when looking very very closely.
My advice to anyone lusting after a bigger sensor is to think about printing or viewing size and distance and ISO setting. When pushing things to extremes I think that bigger is probably better but when shooting at low to medium ISO's and printing or displaying at relatively "normal" sizes and distances (and I mean real normal person sizes and real normal person distances and not gear head geek sizes and distances) I think that mostly it doesn't matter and that other things will possibly matter more... like focus accuracy and speed and handling etc.
If anyone is genuinely and truly interested and not trolling or looking to flame I can provide A7 and GX7 raws for comparison. The points of focus and composition will be slightly different as they're real world shots. I could also supply test shots if it's a genuine request.
Mostly though I think that all this is pretty much a waste of time. If shooting at low to medium ISO's and doing so for best effect (I always ETTR if at all possible) and if processing for best effect in many instances anything from MFT and upwards should be ok. IMVHO it's only when pushing the absolute limits of ISO and print size or viewing magnification that significant image quality differences show up in the kit I've owned.
sk66
Advertiser
- Messages
- 8,677
- Name
- Steven
- Edit My Images
- Yes
yes/no...
The biggest "error" I see in the comparison is that pixel size is not really the main determiner of SNR; sensor size is. If you have a given sensor size with fewer/larger pixels it will have better SNR per pixel. And a sensor of the same size with more/smaller pixels will have lower SNR per pixel. But the total SNR (light/photon shot noise) collected will be about the same.
The main difference will be in read/amplification/conversion noise which will be system dependent (not sensor dependent). In general, the smaller pixels will have a little more noise. If you go stupid small with the pixels they will have a lot more noise relative to the signal collected. (the added noise is relatively constant/increases whereas the collected signal drops)
Last edited:
- Messages
- 1,078
- Name
- Matt
- Edit My Images
- Yes
When I mentioned detail I was really referring to the amount of stars the D810 captures at its "best" settings against the crop at its "best" settings. Both say, 30sec @14mm (ignoring crop factor) and both with the highest ISO poss without massive noise.
- Messages
- 2,531
- Name
- Paul
- Edit My Images
- Yes
Precisely - which is exactly why a "scientific" debate on this subject as to which is 'technically' better is largely pointless.
sk66
Advertiser
- Messages
- 8,677
- Name
- Steven
- Edit My Images
- Yes
I figured as much, particularly when you said this:
quote edited... ISO performance could be very similar/the same due to total SNR collected at anything other than higher ISO's (same viewing size).
sk66
Advertiser
- Messages
- 8,677
- Name
- Steven
- Edit My Images
- Yes
Ah, so different exposure settings for the same image...
At the same ISO, where the smaller pixels take a big hit is in color depth. The "measurements" don't seem significantly different (i.e. DXO) unless you understand what a "bit" is. Bit depth is 2 to the bit power (i.e. 2 to the 21st power for 21 bit) and even less than 1 bit differences result in significantly different sensitivities.
The other thing that makes the D810 significantly more noisy is that it is using digital gain at ISO 3200 and above where the D4s is not using digital gain at all. But those are camera specifics and not really sensor/pixel related.
Last edited:
- Messages
- 1,078
- Name
- Matt
- Edit My Images
- Yes
Comparing what a crop can do and what a FF can do in the same scenario using the equipment to the best of IT'S abilities.
I don't see the point in setting the cameras up with the same settings and comparing. Thats like driving a Fiat uno and a Ferrari at 70mph and arguing which is faster.
I'd love to play with a D4s at night. ooh er missus
odd jim
Flimsiest Lambresta
- Messages
- 9,209
- Name
- Jim
- Edit My Images
- Yes
But you can then see which will then get to 100 mph quickest.
The answer is obvious, much like this thread.
That said, the differences between a crop and FF sensor aren't quite as pronounced as a Fiat Uno and Ferrari, so maybe that's not the best comparison!
Last edited:
- Messages
- 872
- Edit My Images
- No
The DSLR's were used with the same lens : 17-40 and 135 F2. The full frame camera (6D) is visibly better than the crop camera (20D) in print.
odd jim
Flimsiest Lambresta
- Messages
- 9,209
- Name
- Jim
- Edit My Images
- Yes
A 6d v 20d isn't a fair comparison. There's a 7 year development difference between the two which rules out any meaningful comparison.
6d and 70d, or 7d2 (as per my post) is a more level and realistic comparison development / tech wise.
Last edited:
sk66
Advertiser
- Messages
- 8,677
- Name
- Steven
- Edit My Images
- Yes
Comparing the same settings is comparing the benefit/differences for what you currently do and the way you do it.
Comparing different settings is comparing the benefit/differences that might allow you to do something "better" or you can't currently do.
Both are valid questions/approaches, just depends on your needs/goals.
FWIW, with the smaller pixel/higher MP sensors I find I have to use ~ 1 stop more SS which means a 1 stop ISO penalty. So, in comparing cameras like my D810/D4 for "what I can do" capabilities regarding ISO/SS I would set the D810 one stop higher on ISO.
But if I shot static subjects using a very sturdy tripod, then maybe not.