Site Sponsors
Page 4 of 5 FirstFirst ... 2 3 4 5 LastLast
Results 151 to 200 of 249

Thread: E-5 better in low light than D7000? Yes, maybe.

  1. #151
    Senior Member
    Join Date
    Dec 2010
    Posts
    2,118
    Post Thanks / Like

    Re: E-5 better in low light than D7000? Yes, maybe.

    Quote Originally Posted by ptomsu View Post
    Another two in C1Pro, again standard settings .....

    Again E5, 2/35-100, ISO 3200
    You can see banding in the first shot and these are size reduced to the XGA class size. Most cameras will look great at these sizes and banding is a PITA to get rid of.

    But if your target delivery medium is this kind of image size for the web, seems ok (except for the banding of the first).

    - Raist

  2. #152
    Senior Member
    Join Date
    Dec 2010
    Posts
    2,118
    Post Thanks / Like

    Re: E-5 better in low light than D7000? Yes, maybe.

    Quote Originally Posted by ptomsu View Post
    E5 with 35 year old 3.5/50, both ISO 1600, no adjustments in post processing!
    These look much better though the burned highlights suggest you had over exposure in a lot of the shot, a good light or somewhat strong light. Plus of course the resize, though Olympus is very good at keeping overall tone up.

    - Raist

  3. #153
    Subscriber Member Jorgen Udvang's Avatar
    Join Date
    Nov 2007
    Location
    Pratamnak
    Posts
    9,346
    Post Thanks / Like
    Images
    2157

    Re: E-5 better in low light than D7000? Yes, maybe.

    Raist,
    I don't think many by the E-5 and/or the 14-35 on a whim. They are tools that do certain things extremely well. Very sharp photos at low ISO is obviously the main strength, and that in combination with enormous reach with moderately sized telephoto lenses. My two main alternatives (if I change from Nikon, which is tempting for me at the moment) are this combo or Sony A850 with Zeiss 24-70mm f/2.8. The Sony obviously has more resolution, but not nearly as much as the pixel count indicates, and that combo is somewhat cheaper.

    However, if I add a 300mm eqv. prime, the whole equation is turned upside down. The Sony 300mm f/2.8 (and equivalent lenses from Canon and Nikon) is 2-3 times as expensive as the Zuiko 150mm f/2.0 (in Thailand it's three times as expensive), and now we're talking differences of $3-4,000. The same kind of differences can be seen for other telephoto lenses where the reach is similar.

    For people like me, who use legacy lenses, Olympus is an ideal solution, and the combination 4/3 plus m4/3 seems to work rather well. Being a GH1 user already, I have a backup camera that will work well for most of what I do.

    Although I agree that the future is all Pen, I also believe that future pro cameras from Olympus will include solutions that also cater for the current range of top pro glass. I don't think Olympus will invest in another round of that kind of lenses, and dumping what was probably a considerable investment wouldn't be smart. Not for Olympus and not for the customers. Actually, I wouldn't be surprised if we see a hybrid solution similar to what Sony has developed for their SLT cameras, but with m4/3 lens mount and possibly a tailor-made adapter for 4/3 lenses.

    Time will show, but whatever happens, a couple of E-5s plus lenses bought today will bring lots of photographic joy for the owner for many years to come.

  4. #154
    Workshop Member ptomsu's Avatar
    Join Date
    Nov 2007
    Location
    Austria, close to Vienna
    Posts
    3,872
    Post Thanks / Like

    Re: E-5 better in low light than D7000? Yes, maybe.

    Quote Originally Posted by raist3d View Post
    That is not true, the E-5 doesn't outperform the A900 in low light. Using modern raw conveyers like LR 3.3 gives at best the same kind of 100% view performance, but since the A900 has so much more megapixels than the E-5, for a valid comparison you resize down and the A900 wins that hands down. That is- if the A900 doesn't perform a bit better at high ISO to begin with which could very well be the case. Moreover, the A900 has definitively more DR.

    - Raist
    Raist

    not sure how you evaluate cameras and how you compare. I know what I am comparing as I had A900, D700, E3 and now E5. Plus I know pretty well how a D3X performs, as a good friend of mine is shooting one.

    From what I have seen in my own experience, the E5 is equal or even outperforming a A900 and a D3X at lower ISO - say up to 800 for sure, if you use a pro grade lens on it. Then you see very brutally what all these AA filters are doing and how they are destroying what higher MP count brings. As soon as you move above ISO1600 I would say there is no clear winner, then all of these cameras - A900, D3X and E5 - become somehow tricky to get good results.

    But what I have seen from my E5 at ISO1600 and even 3200 are results I am completely satisfied with. I cannot say that from the D700, which without any arguing is dealt as one of the best low light DSLRs. But no noise comes always from lot of processing, which partially is already done inside the camera, even to the RAW outputs. Which takes lot of details away. But for me I rather like all details and do post processing of noise as I like it.

    Finally - if I want really high resolution - NOTHING and none of these DSLRs tops my H3D39. But of course only till ISO400 or better even ISO200.

    So it is obvious what the E5 brings to me: the most flexible and high quality DSLR I could buy today, delivering results which are not second to much higher MP counts in FF and APSC, of course in combination wit good lenses. If I need more MP I take MFD, if I really would need highest ISO day in and day out I would go for a D3S, or maybe the next incarnation of the camera like a D4S. But as I do not need this I am perfectly happy with my E5 as what it is!

  5. #155
    Workshop Member ptomsu's Avatar
    Join Date
    Nov 2007
    Location
    Austria, close to Vienna
    Posts
    3,872
    Post Thanks / Like

    Re: E-5 better in low light than D7000? Yes, maybe.

    Quote Originally Posted by raist3d View Post
    The future is all pen, not 4/3rds. Something to keep in mind when buying the 14-35 though if they honor their promise to support it in the prospec body it should be ok.

    What I would warn yo about the 14-35 is this: Yes it is very sharp, and nice. Great contrast. It's excellent *but* in moderate to low light it hunts and takes forever to AF. Now if you are a guy that does MF this doesn't affect you at all. But if you rely on AF, watch out. Even on the E-5, has been reported by quite many.

    - Raist
    Maybe I am missin something, but I do not recognize any banding

    So maybe there is none, or I am banding blind, which again is good for me as I the can shoot much different cameras without any headache obviously

  6. #156
    Subscriber Member Jorgen Udvang's Avatar
    Join Date
    Nov 2007
    Location
    Pratamnak
    Posts
    9,346
    Post Thanks / Like
    Images
    2157

    Re: E-5 better in low light than D7000? Yes, maybe.

    Quote Originally Posted by ptomsu View Post
    Raist

    not sure how you evaluate cameras and how you compare.
    I have a list of wants that includes a tack sharp standard zoom, the ability to use legacy lenses, good viewfinder, IBIS, good weather sealing, articulated LCD... do you see where this is leading?

    The fact that I like the output from Olympus cameras helps a lot too of course, as does the detail rendering of the E-5 at low ISO.

  7. #157
    Member bcf's Avatar
    Join Date
    Jan 2008
    Location
    Lyon, France
    Posts
    125
    Post Thanks / Like

    Re: E-5 better in low light than D7000? Yes, maybe.

    Quote Originally Posted by Jorgen Udvang View Post
    This one surprises me. Any noise treatment?
    Do you mean in good or in bad?

    For me, in bad. I do see banding, and a lot of noise. I like Olympus cameras, but lower noise at ISO 1600 or 3200 is important to me, and a D700 would have fared much better for this type of shot.

    In a "reportage" type situation, the freedom brought by the ability to shoot at "autoISO" between say ISO 200 and 1600, with a minimal shutter speed, and essentially a noise-free image (or with some easily dealt with noise) cannot be overemphasized.

    I still fail to see the point of all this: by getting an E-5 and f/2 lenses, you will get a camera and lenses just as big as heavy as a D700, but limited to ISO 800-1600 at most. I myself like my D700 but find it too big - why would I get an E-5 which is just as big and has less capabilities?
    -- Bernard

  8. #158
    Workshop Member ptomsu's Avatar
    Join Date
    Nov 2007
    Location
    Austria, close to Vienna
    Posts
    3,872
    Post Thanks / Like

    Re: E-5 better in low light than D7000? Yes, maybe.

    Quote Originally Posted by bcf View Post
    Do you mean in good or in bad?

    For me, in bad. I do see banding, and a lot of noise. I like Olympus cameras, but lower noise at ISO 1600 or 3200 is important to me, and a D700 would have fared much better for this type of shot.

    In a "reportage" type situation, the freedom brought by the ability to shoot at "autoISO" between say ISO 200 and 1600, with a minimal shutter speed, and essentially a noise-free image (or with some easily dealt with noise) cannot be overemphasized.

    I still fail to see the point of all this: by getting an E-5 and f/2 lenses, you will get a camera and lenses just as big as heavy as a D700, but limited to ISO 800-1600 at most. I myself like my D700 but find it too big - why would I get an E-5 which is just as big and has less capabilities?
    I think you just do net get it, which is perfectly fine! I did not get it long time

    If you like the clinical output from a D700, which draws everything like under a smoothing plastic skin at higher ISO then it is ok! Maybe for your type of shooting this is essential. For mine it is not.

    Second as I already stated, I cannot see banding, but maybe i am blind

    Third, why not just be happy with the D700 and continue? I for myself decided to sell this camera and I can tell you one thing for sure: I never was happier than now since I am shooting an E5 with any other DSLR.

    Does it mean the E5 is perfect? For sure NOT! I could give you a list with at least 10 items to be improved, first and foremost the sensor be replaced by the GH2 sensor. But for me the E5 has reached a level of perfection which makes it an ideal tool. Which was not the case BTW with the E3.

    PS: BTW that smoothing plastic skin is also present at lower ISO level with the D700, whereas with the E5 I have first time the feeling that I can see CLEAR!

  9. #159
    Senior Member
    Join Date
    Dec 2010
    Posts
    2,118
    Post Thanks / Like

    Re: E-5 better in low light than D7000? Yes, maybe.

    Quote Originally Posted by Jorgen Udvang View Post
    Raist,
    I don't think many by the E-5 and/or the 14-35 on a whim. They are tools that do certain things extremely well. Very sharp photos at low ISO is obviously the main strength, and that in combination with enormous reach with moderately sized telephoto lenses. My two main alternatives (if I change from Nikon, which is tempting for me at the moment) are this combo or Sony A850 with Zeiss 24-70mm f/2.8. The Sony obviously has more resolution, but not nearly as much as the pixel count indicates, and that combo is somewhat cheaper.
    I am not sure what you are trying to say. That's part of what I said. The 14-35 is a fantastic lens. I certainly didn't buy one on a whim. I tested it myself. It is a fantastic lens. Super sharp and great except for the moderate-low to low light AF issue (it does lock focus but takes forever). Many already with E-3, E-30 and even E-5 have corroborated the issue is still there and I have seen it with my own eyes. If you are a MF guy, you have nothing to worry about. If you rely on AF in low light, then beware. I wasn't commenting on your alternatives, but on this lens specifically.

    However, if I add a 300mm eqv. prime, the whole equation is turned upside down. The Sony 300mm f/2.8 (and equivalent lenses from Canon and Nikon) is 2-3 times as expensive as the Zuiko 150mm f/2.0 (in Thailand it's three times as expensive), and now we're talking differences of $3-4,000. The same kind of differences can be seen for other telephoto lenses where the reach is similar.
    I find that the normal ranges overall, other than the specific example, comes out still very pricey for Olympus, but certainly each individual needs to look at their cost for their needs.

    For people like me, who use legacy lenses, Olympus is an ideal solution, and the combination 4/3 plus m4/3 seems to work rather well. Being a GH1 user already, I have a backup camera that will work well for most of what I do.

    Although I agree that the future is all Pen, I also believe that future pro cameras from Olympus will include solutions that also cater for the current range of top pro glass.
    That's sort of what I am expecting. Otherwise the backlash will be huge.

    I don't think Olympus will invest in another round of that kind of lenses, and dumping what was probably a considerable investment wouldn't be smart.
    It's a sunk cost. You can't take into account what you spent on these lenses. I actually think new micro four thirds pro spec lenses will come out and eventually the 4/3rd lenses discontinued in production though still supported.
    That's fine by me. I say that too because there's no point in going small and carry big lenses.

    Not for Olympus and not for the customers. Actually, I wouldn't be surprised if we see a hybrid solution similar to what Sony has developed for their SLT cameras, but with m4/3 lens mount and possibly a tailor-made adapter for 4/3 lenses.
    That's certainly possible.

    Time will show, but whatever happens, a couple of E-5s plus lenses bought today will bring lots of photographic joy for the owner for many years to come.
    Sure.

    - Raist

  10. #160
    Senior Member
    Join Date
    Dec 2010
    Posts
    2,118
    Post Thanks / Like

    Re: E-5 better in low light than D7000? Yes, maybe.

    [self nuked post].
    Last edited by raist3d; 10th March 2011 at 09:57.

  11. #161
    Senior Member
    Join Date
    Dec 2010
    Posts
    2,118
    Post Thanks / Like

    Re: E-5 better in low light than D7000? Yes, maybe.

    Quote Originally Posted by ptomsu View Post
    Maybe I am missin something, but I do not recognize any banding
    Do you authorize me to download your photo so I can point it out? [del]

    So maybe there is none, or I am banding blind, which again is good for me as I the can shoot much different cameras without any headache obviously
    [del]

    - Raist
    Last edited by raist3d; 10th March 2011 at 09:58.

  12. #162
    Senior Member
    Join Date
    Dec 2010
    Posts
    2,118
    Post Thanks / Like

    Re: E-5 better in low light than D7000? Yes, maybe.

    Quote Originally Posted by Jorgen Udvang View Post
    I have a list of wants that includes a tack sharp standard zoom, the ability to use legacy lenses, good viewfinder, IBIS, good weather sealing, articulated LCD... do you see where this is leading?

    The fact that I like the output from Olympus cameras helps a lot too of course, as does the detail rendering of the E-5 at low ISO.
    Go for it man. Sounds like you know the pros and cons and you have a good match. Really, go for it and congrats when you get it.

    - Raist

  13. #163
    Super Duper
    Senior Member
    Godfrey's Avatar
    Join Date
    Jan 2009
    Location
    Near San Jose, California
    Posts
    7,931
    Post Thanks / Like

    Re: E-5 better in low light than D7000? Yes, maybe.

    Quote Originally Posted by ptomsu View Post
    ... {more E-5 joy} ...
    Sheesh, this thread is still rambling on? All of you: get a life, make some photographs.

    Sounds like you're enjoying the E-5, Peter. Good luck with it! I remain delighted with it, it is perhaps the most satisfying camera I've owned since I sold my Nikon F3/T ... which is still one of the few finest cameras of all time. IMO, of course. ]'-)

    Perhaps Jorgen will join our small cadre of Olympus E-5 users soon.

    ps: I like the Nikkor-H 85/1.8 so much that I just ordered another one. The BGN one I got from KEH has perfect glass but needs a CLA badly: the focusing helicoid is rough and it's got many signs of hard use. They had an EXC graded example as well for another hundred bux; I decided it was worth it as a CLA for this one will cost me $150 and won't clean up the cosmetics at all. A friend of mine already wants to buy this one too, so I'm not out any money.

    It's a great lens on the FourThirds bodies, fits the E-5 perfectly, and a fantastic focal length and speed.

    enjoy!



  14. #164
    Senior Member
    Join Date
    Dec 2010
    Posts
    2,118
    Post Thanks / Like

    Re: E-5 better in low light than D7000? Yes, maybe.

    Looks like you missed this:

    http://raist3d.typepad.com/photos/sm...tfolio_sample/

    - Raist

  15. #165
    Workshop Member ptomsu's Avatar
    Join Date
    Nov 2007
    Location
    Austria, close to Vienna
    Posts
    3,872
    Post Thanks / Like

    Re: E-5 better in low light than D7000? Yes, maybe.

    @raist3d

    well, I see that you are starting number of responses and attacks, which I am definitely not willing to go into. I have absolutely no intention to defend an Olympus camera against any other camera, nor to prove that my standards and decision criteria are good or bad or whatever. All of this would go just in a pure waste of time - not only wasting my time but also yours.

    Enjoy your K5, make great photos and please do no longer care about what I am doing and thinking it is good. We are not compatible - sorry

  16. #166
    Senior Member
    Join Date
    Dec 2010
    Posts
    2,118
    Post Thanks / Like

    Re: E-5 better in low light than D7000? Yes, maybe.

    @ptomsu - you make some claims but not seeing the backing up. Two people pointed out to you the banding in your ISO 3200 shot, but you can't see it and reply with a "cute" thing which seems rather odd. I offered if you authorized it to download your shot and mark it, but never heard back on that. You say you don't have any intentions to defend a particular camera yet you jump with your alleged observations doing exactly that. Not sure what to make of that.

    As for responses, I thought that was what normally happens in a thread or conversation. Otherwise it would be monologues.

    I know I do photography, which is why I ask often when others talk about their experience and observations to share what they do. Very commonly all of a sudden upon asking for a portfolio or gallery showing their experience, they vanish, for some reason.

    *shrugs*

    I'll enjoy shooting with what I have which are multiple brands. My photography goes beyond the concept of a brand.

    - Raist

  17. #167
    Workshop Member ptomsu's Avatar
    Join Date
    Nov 2007
    Location
    Austria, close to Vienna
    Posts
    3,872
    Post Thanks / Like

    Re: E-5 better in low light than D7000? Yes, maybe.

    Quote Originally Posted by raist3d View Post
    @ptomsu - you make some claims but not seeing the backing up. Two people pointed out to you the banding in your ISO 3200 shot, but you can't see it and reply with a "cute" thing which seems rather odd. I offered if you authorized it to download your shot and mark it, but never heard back on that. You say you don't have any intentions to defend a particular camera yet you jump with your alleged observations doing exactly that. Not sure what to make of that.

    As for responses, I thought that was what normally happens in a thread or conversation. Otherwise it would be monologues.

    I know I do photography, which is why I ask often when others talk about their experience and observations to share what they do. Very commonly all of a sudden upon asking for a portfolio or gallery showing their experience, they vanish, for some reason.

    *shrugs*

    I'll enjoy shooting with what I have which are multiple brands. My photography goes beyond the concept of a brand.

    - Raist
    Let me know an email address and I will send you the RAW file.

    Peter

    PS: I am also brand agnostic, having (unfortunately) owned almost all of them - I think I am missing Fuji and yes, Sigma

  18. #168
    Member
    Join Date
    Aug 2009
    Location
    Warsaw, Poland
    Posts
    188
    Post Thanks / Like

    Re: E-5 better in low light than D7000? Yes, maybe.

    Quote Originally Posted by raist3d View Post
    The future is all pen, not 4/3rds. Something to keep in mind when buying the 14-35 though if they honor their promise to support it in the prospec body it should be ok.

    What I would warn yo about the 14-35 is this: Yes it is very sharp, and nice. Great contrast. It's excellent *but* in moderate to low light it hunts and takes forever to AF. Now if you are a guy that does MF this doesn't affect you at all. But if you rely on AF, watch out. Even on the E-5, has been reported by quite many.

    - Raist
    And you surly shoot lot E5+14-35?
    Cause I do, and have not noticed "taking forever to AF".
    Give your body/lens combo for a check, they must be faulty.

  19. #169
    Senior Member
    Join Date
    Dec 2010
    Posts
    2,118
    Post Thanks / Like

    Re: E-5 better in low light than D7000? Yes, maybe.

    Quote Originally Posted by nugat View Post
    And you surly shoot lot E5+14-35?
    Cause I do, and have not noticed "taking forever to AF".
    Give your body/lens combo for a check, they must be faulty.
    There was a thread in which several E-5 owners mentioned that the issue still happened though it seemed a little better than say E-3 or E-30. The issue doesn't happen in good light. It's in low light/or moderate low and I can only imagine (this part I haven't tried), tungsten since tungsten is just traditionally nasty or fluorescent. The lens does lock focus, but it basically goes to AF fast, then as it approximates the point of focus starts doing this "little steps". It easily takes more than a second.

    Given this issue has been reported with E-5, have seen it myself with 420, E-3, E-30 and Olympus has not addressed it with the previous cameras I certainly do believe the E-5 still has it with it.

    Again, low light.

    - Raist

  20. #170
    Senior Member
    Join Date
    Dec 2010
    Posts
    2,118
    Post Thanks / Like

    Re: E-5 better in low light than D7000? Yes, maybe.

    Quote Originally Posted by ptomsu View Post
    Let me know an email address and I will send you the RAW file.
    Peter, sent you a message. Let me know if you want posted it here or privately.

    I'll be busy tomorrow but should find time at night, thanks again.

    - Ricardo

    PS: I think Nik's whatever it is noise reduction program they do, supports banding removal. I also believe since then there's another one that does too (forgot if it was something DXo did or someone else).

  21. #171
    Senior Member Riley's Avatar
    Join Date
    Dec 2007
    Location
    Adelaide Hills South Australia
    Posts
    441
    Post Thanks / Like

    Re: E-5 better in low light than D7000? Yes, maybe.

    Quote Originally Posted by raist3d View Post
    That is not true, the E-5 doesn't outperform the A900 in low light. Using modern raw conveyers like LR 3.3 gives at best the same kind of 100% view performance, but since the A900 has so much more megapixels than the E-5, for a valid comparison you resize down and the A900 wins that hands down. That is- if the A900 doesn't perform a bit better at high ISO to begin with which could very well be the case. Moreover, the A900 has definitively more DR.

    - Raist
    thyre very close as noise reading per ISO, at print sizes A900 has a fraction more than a stop on E5 (not enough to call 1.3 stops).
    Like E5 A900 isnt a high ISO machine, and E5 would give it a run for its money in jpeg at any ISO and probably has faster zooms available to it,
    so I think Godfrey might just be right,..... thats the way it is

  22. #172
    Workshop Member ptomsu's Avatar
    Join Date
    Nov 2007
    Location
    Austria, close to Vienna
    Posts
    3,872
    Post Thanks / Like

    Re: E-5 better in low light than D7000? Yes, maybe.

    Quote Originally Posted by raist3d View Post
    Peter, sent you a message. Let me know if you want posted it here or privately.

    I'll be busy tomorrow but should find time at night, thanks again.

    - Ricardo

    PS: I think Nik's whatever it is noise reduction program they do, supports banding removal. I also believe since then there's another one that does too (forgot if it was something DXo did or someone else).
    Ricardo,

    sent you the image! Not sure if it is useful as it could be because it is a LR3.3 DNG conversion.

    Feel free to post here!

    If you need more stuff, just let me know

    Peter

  23. #173
    Senior Member
    Join Date
    Dec 2010
    Posts
    2,118
    Post Thanks / Like

    Re: E-5 better in low light than D7000? Yes, maybe.

    Quote Originally Posted by Riley View Post
    thyre very close as noise reading per ISO, at print sizes A900 has a fraction more than a stop on E5 (not enough to call 1.3 stops).
    Like E5 A900 isnt a high ISO machine, and E5 would give it a run for its money in jpeg at any ISO and probably has faster zooms available to it,
    so I think Godfrey might just be right,..... thats the way it is
    I can't agree to that. The megapixel difference is too big to not account for it. And they don't suffer from banding, and do have more dynamic range/better shadows. The early raw conversions of the A900 didn't look very good originally, but the ones after the RAW converters got tuned for it showed that the A900 wasn't as noisy as it was first thought.

    Something like LR 3.3 actually makes it look pretty decent actually and you have tons of data to use (compared to the E-5). It's thanks to that extra data that while you will have to "burn" a lot of it at higher ISO you can still come out ahead.

    The best: download raws from both and see your yourself (said in general).

    And a note on zooms: not all lenses are or have to be zooms. There are primes too. Just because someone likes to use zooms shouldn't in any shape or form those who shoot with primes (and viceversa).

    - Raist

  24. #174
    Senior Member
    Join Date
    Dec 2010
    Posts
    2,118
    Post Thanks / Like

    I used the one you posted....

    Quote Originally Posted by ptomsu View Post
    Ricardo,

    sent you the image! Not sure if it is useful as it could be because it is a LR3.3 DNG conversion.

    Feel free to post here!

    If you need more stuff, just let me know

    Peter
    Ill check the raw later. I highlighted some of the bands. There are more (see another one at the very bottom of the shot all the way across).



    Here's another example in a thread- look at the bird shot (not linking directly so you can see the original post).

    http://forums.dpreview.com/forums/re...ssage=37298036

    There is more to this issue that simply getting it. If you need to do any kind of post where you need to brighten the RAW image, the image falls apart super fast. This behavior was documented int he E-3 review at Luminous Landscape:

    http://www.luminous-landscape.com/re...3-Second.shtml

    Look a the section "A more serious issue:"

    "An E-3 frame shot at 1600ASA.
    Banding is manifesting itself quite obviously in the flat areas of the frame
    and there is absolutely no latitude for pushing the exposure in post production."

    ----

    There are things you can do to minimize the impact:

    - Avoid high iso shots with areas of dark and bright
    - Avoid tungsten light
    - Try your darndest to get the exposure right- exact. This is a rule that all cameras benefit from obviously but here you have to really get it. Don't think about under exposing and having to develop up.
    - Try out software that does noise reduction and supports banding noise reduction.

    Nik's Define 2 does this:

    http://www.niksoftware.com/company/u...uncement.shtml

    I have tried it the first time they came out with my e-330 and it did help a bit, though varied by banding situation. But this is "version 2" of that process, so it should work hopefully pretty good for many situations.

    - Raist

    PS: Thank you Peter for allowing me to use your image.

  25. #175
    Senior Member Riley's Avatar
    Join Date
    Dec 2007
    Location
    Adelaide Hills South Australia
    Posts
    441
    Post Thanks / Like

    Re: E-5 better in low light than D7000? Yes, maybe.

    Quote Originally Posted by raist3d View Post
    I can't agree to that. The megapixel difference is too big to not account for it.
    lets keep this short
    well you are an advocate of DxO, go take a look
    you are wrong

    Quote Originally Posted by raist3d
    And they don't suffer from banding, and do have more dynamic range/better shadows. The early raw conversions of the A900 didn't look very good originally, but the ones after the RAW converters got tuned for it showed that the A900 wasn't as noisy as it was first thought.

    Something like LR 3.3 actually makes it look pretty decent actually and you have tons of data to use (compared to the E-5). It's thanks to that extra data that while you will have to "burn" a lot of it at higher ISO you can still come out ahead.

    The best: download raws from both and see your yourself (said in general).

    And a note on zooms: not all lenses are or have to be zooms. There are primes too. Just because someone likes to use zooms shouldn't in any shape or form those who shoot with primes (and viceversa).

    - Raist
    zooms offer more utility than primes
    fast zooms offer more utility than slow zooms
    whats not to like....

  26. #176
    Senior Member
    Join Date
    Dec 2010
    Posts
    2,118
    Post Thanks / Like

    Re: E-5 better in low light than D7000? Yes, maybe.

    Quote Originally Posted by Riley View Post
    lets keep this short
    well you are an advocate of DxO, go take a look
    you are wrong
    (edited)
    I can't quite agree to that- dxo says:

    E-5 DR = 10.5 ev. Sony A900 = 12.3 EV
    ISO score (low iso) = 519, Sony A900 = 1431 (that's almost 3 times the score of the e-5)
    Color bits = 21.6 bits , Sony = 23.7 bits

    (edited) If there is something wrong, by all means point it out.

    The S/NR graph favors the A900 even without taking into account print (i.e. the extra resolution) and separates more as ISO goes up.

    (edit)as I encourage everyone to download A900 raws, E-5 raws and convert and see for themselves.

    zooms offer more utility than primes
    fast zooms offer more utility than slow zooms
    whats not to like....
    Size and price for marginal improvements? Zooms offering more utility than primes depends. Primes can be faster (F1.4) and far smaller at a fraction of the price with comparable quality. I am not going to make "the case for primes" because this case has been long made since film photography. That doesn't even take into account the issue of seeing, which zooms can let you slide oh so easy without remembering the issues of perspective. Faster zooms don't mean anything when the other bodies outclass the ones you put those in ISO performance by a significant amount.

    There's pros and cons.
    Last edited by raist3d; 10th March 2011 at 10:02.

  27. #177
    Senior Member Riley's Avatar
    Join Date
    Dec 2007
    Location
    Adelaide Hills South Australia
    Posts
    441
    Post Thanks / Like

    Re: E-5 better in low light than D7000? Yes, maybe.

    Quote Originally Posted by raist3d View Post
    I don't think so- I mean, dxo says:

    E-5 DR = 10.5 ev. Sony A900 = 12.3 EV
    ISO score (low iso) = 519, Sony A900 = 1431 (that's almost 3 times the score of the e-5)
    Color bits = 21.6 bits , Sony = 23.7 bits

    What did I miss here? Where am I wrong?
    well Im a bit used to the way you backpedal from dpr, this is what I said

    Quote Originally Posted by Riley
    "thyre very close as noise reading per ISO, at print sizes A900 has a fraction more than a stop on E5 (not enough to call 1.3 stops).
    Like E5 A900 isnt a high ISO machine, and E5 would give it a run for its money in jpeg at any ISO and probably has faster zooms available to it,
    so I think Godfrey might just be right,..... thats the way it is"
    see any mention of DR there?

    Quote Originally Posted by raist3d
    The S/NR graph favors the A900 even without taking into account print (i.e. the extra resolution) and separates more as ISO goes up.
    Quote Originally Posted by Riley
    at print sizes A900 has a fraction more than a stop on E5 (not enough to call 1.3 stops)

  28. #178
    Senior Member
    Join Date
    Dec 2010
    Posts
    2,118
    Post Thanks / Like

    (!!!!!) Re: E-5 better in low light than D7000? Yes, maybe.

    Quote Originally Posted by Riley View Post
    well Im a bit used to the way you backpedal from dpr, this is what I said
    see any mention of DR there?
    I am sorry but I am not back pedaling at all. Ignore the DR, you certainly mentioned ISO. Then you said I was wrong because Dxo which I go by proved me wrong. So I mentioned everything I saw at Dxo which does include ISO performance btw and I couldn't see where I was wrong. I asked you what I missed on that and you replied with "I am used to the way you back pedal from dpr."

    Nobody mentioned dpr. (edit) I don't understand. (edit) The dxo ISO score and the ISO S/N ratio chart talk about all these things.

    (edit) I am going to focus on the banding that was the last big thing being talked about anyway.


    - Raist
    Last edited by raist3d; 10th March 2011 at 10:04.

  29. #179
    Senior Member Riley's Avatar
    Join Date
    Dec 2007
    Location
    Adelaide Hills South Australia
    Posts
    441
    Post Thanks / Like

    Re: E-5 better in low light than D7000? Yes, maybe.

    I said
    "thyre very close as noise reading per ISO, at print sizes A900 has a fraction more than a stop on E5 (not enough to call 1.3 stops)"

    to which you said
    "I can't agree to that. The megapixel difference is too big to not account for it."

    now you know what I said was vindicated by DxO and I was right all along

  30. #180
    Senior Member
    Join Date
    Dec 2010
    Posts
    2,118
    Post Thanks / Like

    Re: E-5 better in low light than D7000? Yes, maybe.

    Quote Originally Posted by Riley View Post
    I said
    "thyre very close as noise reading per ISO, at print sizes A900 has a fraction more than a stop on E5 (not enough to call 1.3 stops)"

    to which you said
    "I can't agree to that. The megapixel difference is too big to not account for it."

    now you know what I said was vindicated by DxO and I was right all along
    No, (edit) I do not believe that's not what Dxo says but the opposite. The claim was - that DXo supports the claim presented and I posted what DXo talks about in the comparison; don't see where does DXo supports that claim. (edit)

    I asked for supporting evidence for that claim and doesn't look like it's been produced (edit).

    If you could please, point out where exactly Dxo proves the claim presented. Having an ISO score at DXo 3 times the one of the E-5, I don't understand how is so (edit).

    - Raist
    Last edited by raist3d; 10th March 2011 at 10:07.

  31. #181
    Senior Member Riley's Avatar
    Join Date
    Dec 2007
    Location
    Adelaide Hills South Australia
    Posts
    441
    Post Thanks / Like

    Re: E-5 better in low light than D7000? Yes, maybe.

    I said
    "thyre very close as noise reading per ISO, at print sizes A900 has a fraction more than a stop on E5 (not enough to call 1.3 stops)"

    this grab is of the screen size. taking a sample SN at 400 ISO
    E5 31.4db
    A900 32.1db
    the difference is 0.7db, 0.231 of a stop difference, thats consistent to ISO3200



    this grab is of the print size, taking a sample SN at 400 ISO
    E5 33.4db
    A900 36.9db
    the difference is 3.5db, 1.165 stops difference, thats consistent to ISO3200



    now enough of this nonsense I do not think it fair of you to blindly disagree at every opportunity and try to wear people down. Im busy and dont have time for further questioning of results that are right on the site for anyone to investigate.

    IMO this matter is closed

  32. #182
    Senior Member
    Join Date
    Dec 2010
    Posts
    2,118
    Post Thanks / Like

    Re: E-5 better in low light than D7000? Yes, maybe.

    Hmmm I am not convinced the matter isn't closed, I am not sure the graph was read accurately- (edit)

    #1. First graph- mentioning this graph at this point doesn't provide additional data. This case was already covered and mentioned- doesn't take into account extra A900 information for a resize down.

    #2. (edit)
    If we look at the graph on print section, I believe we can see that at the lower ISO's the distance is akin to what you mentioned earlier until we hit measured true ISO 800 on the graph. The difference does not seem consistent to me. At that point they just start growing apart more and more.

    e-5 is about 28db and A900 is about 33 DB. 33 (A900) -28(E-5) = 5 / 3 = 1 stop and 2/3rds of EV.

    By true measured ISO 3200, the Db difference is at a about 26 db (A900) - 19 DB (E-5) = 7. This is 2 full stops and 1/3rd of a stop. Ditto for further. That number as we see further down matches the DXo's ISO score expectations. (edit)

    (edit) Let's take a look at this:

    http://dxomark.com/index.php/en/Lear...-Scores#Sports

    I cite:

    Sports & action photography: Low-Light ISO

    .... An SNR value of 30dB reflects an excellent image quality. Low-Light ISO is then the highest ISO setting for the camera such that the SNR reaches this 30dB value while keeping a good dynamic range of 9 EVs and a color depth of 18bits.
    As cameras improve, the Low-Light ISO will continuously increase, making this scale open.
    A difference of Low-Light ISO of 25% represents 1/3 EV and is slightly noticeable.
    The iso score of the E-5 is 519 and the A900 score is 1431. That from the get go is saying that to keep the certain image quality, all you can go on the E-5 is up to 519 while on A900 is 1431 ISO.

    If you look at the % of difference: (1431-519)/519 = 175.7%(!). According to their explanation, every 25% is 1/3rd of a stop 175.7%/25% = 7.028 1/3rd's of a stop.

    That's basically 2 full stops and 1/3rd of a stop of an ISO gap. Which is kinda like.. exactly the same number and difference that is on that graph above.

    So on ISO's lower than ISO 800 which isn't really high iso, the noise is still 1 stop to 1 stop and a 3rd in favor of the A900. And at ISO 1600/3200 is on the order of 1 stop and 2/3rds, 2 stops or higher. Sounds about right with my own examinations of the raw files.

    - Raist

    PS: And I don't "blindly try to disagree at every opportunity and try to wear people down" A discussion happens and it's only natural to try and provide evidence when people differ. (edit)

    (edit)
    PSS: This doesn't cover the banding issue the E-5 has at ISO 3200 and up, which can sometimes even start as low as ISO 1600. Haven't seen the same banding from pushed A900 files though it's possible that there could be some on some rare occasion, but haven't seen it.
    Last edited by raist3d; 10th March 2011 at 10:13.

  33. #183
    Subscriber Member Jorgen Udvang's Avatar
    Join Date
    Nov 2007
    Location
    Pratamnak
    Posts
    9,346
    Post Thanks / Like
    Images
    2157

    Re: E-5 better in low light than D7000? Yes, maybe.

    This thread may be interesting reading:

    http://forums.dpreview.com/forums/re...ssage=37928329

    The photos are apparently done mostly with the 12-60 giving apertures from 3.4 to 4.0.

  34. #184
    Senior Member Riley's Avatar
    Join Date
    Dec 2007
    Location
    Adelaide Hills South Australia
    Posts
    441
    Post Thanks / Like

    Re: E-5 better in low light than D7000? Yes, maybe.

    ok Ive got some time between shoots, and maybe some a little later so lets look at this now

    Quote Originally Posted by raist3d View Post
    No, the matter isn't closed, you misread the graph, but we can close the case soon enough- let's take this one step at a time:

    #1. First graph- completely meaningless. That's what I said when I referred to that the A900 has so much more data that you can do more with it bringing the effective ISO higher (hence that Dxo print button). This "100% graph" is meaningless in this discussion. I already covered that case. Moving on...

    #2.
    If you look carefully at the graph on print (which is as I said, what matters, since I mentioned that the A900 has much more data to do something with),
    yes but to do what with?
    For instance we can see form data we can gain from numerous sources that while A900 has very deep extinction resolution the termed 'absolute resolution' isnt particularly different, so that doesnt necessarily translate into anything special other than especially robust files that wont fall apart as easily. Likewise more data doesnt necessarily mean less noise although outside of DxO we just dont have a numeric to operate from

    Quote Originally Posted by raist3d
    you will see that at the lower ISO's the distance is akin to what you are say until you hit measured true ISO 800 on the graph. No the difference is "not consistent" as you say and it's pretty obvious. At that point they just start growing apart more and more.
    on reflection i can see that they do, however i still think ISO 400 to be a respectable snapshot since your snapshot is just the extreme end of the scale. Its only a difference of where one takes ones snapshot

    Quote Originally Posted by raist3d
    e-5 is about 28db and A900 is about 33 DB. 33 (A900) -28(E-5) = 5 / 3 = 1 stop and 2/3rds of EV.

    By true measured ISO 3200, the Db difference is at a about 26 db (A900) - 19 DB (E-5) = 7. This is 2 full stops and 1/3rd of a stop. Ditto for further. Keep that little number in mind because it gets interesting how DXo doesn't contradict itself (see below).
    ok this is my problem with your observations right here, you are going to have to explain to me what you mean by 'true measured ISO' and I want you to consider this.

    Lots of folks get pretty ratty about perceived differences in ISO from observations of DxO data. What they need to remember is DxO record data in a completely different way to everyone else. Where manufacturers need to uphold to the standards of ISO definition given to them by ISO SOS and ISO REI. DxO measurements are good for DxO, not especially useful to photographers (same it seems is true of their DR measurements)

    the bottom line is, if you shoot a jpeg from a camera its ISO will be pretty close to right b/se it matches the tone curve the manufacturer set conditional to ISO measurement. Since DxO for whatever reason use their own curve that measurement becomes displaced. We could assume this is why they maintain a reflection of the manufacturers setting on their charts.

    Thats the way I interpret it, and that isnt an unfair observation nor is it 'erroneous' as you put to me later

    Quote Originally Posted by raist3d
    Of course your erroneous conclusion seems to put at a conflict the graph and DXo's own own ISO score which is easily explained here:
    'erroneous conclusion' which is just inflammatory BS that I suggest you stay away from in this place.

  35. #185
    Senior Member
    Join Date
    Dec 2010
    Posts
    2,118
    Post Thanks / Like

    Re: E-5 better in low light than D7000? Yes, maybe.

    Quote Originally Posted by Riley View Post
    yes but to do what with?
    You have more information to begin with which allows you to resample down and keep valid information/detail.

    For instance we can see form data we can gain from numerous sources that while A900 has very deep extinction resolution the termed 'absolute resolution' isnt particularly different, so that doesnt necessarily translate into anything special other than especially robust files that wont fall apart as easily. Likewise more data doesnt necessarily mean less noise although outside of DxO we just dont have a numeric to operate from
    Just because the A900 has a very "deep extinction resolution" doesn't mean that it doesn't have also more real world extra resolution over the e-5. It's also better at higher ISO to begin anyway, but we were addressing the particular claim made that it was only about only 1 stop + 1/3rd in what you could do in the end regardless of the extra data/resize. (edit)

    on reflection i can see that they do, however i still think ISO 400 to be a respectable snapshot since your snapshot is just the extreme end of the scale. Its only a difference of where one takes ones snapshot
    Well two things:

    (edit)
    #1. A claim in general terms. The fact that it only goes to ISO 400 (or 800), proves that claim wrong.

    #2. This whole thread context is high ISO. And it is also the discussion we were having all along in this thread as far as how the cameras perform in lower light situations over the other. The Sony still does better anyway, but putting the bar at ISO 400 doesn't mean much. I mean, from my view at least, I already agreed the E-5 does fantastic at low iso. Even the A900 still doing better here won't matter much though you sure get much better shadows, though that touches more on DR however related they may be.

    ok this is my problem with your observations right here, you are going to have to explain to me what you mean by 'true measured ISO' and I want you to consider this.
    (edit)
    I believe DXo themselves does an excellent job of explaining that themselves.

    http://dxomark.com/index.php/Our-pub...-make-it-clear

    Lots of folks get pretty ratty about perceived differences in ISO from observations of DxO data. What they need to remember is DxO record data in a completely different way to everyone else. Where manufacturers need to uphold to the standards of ISO definition given to them by ISO SOS and ISO REI. DxO measurements are good for DxO, not especially useful to photographers (same it seems is true of their DR measurements)
    (edit)
    Actually I don't believe that is true at all. That graph quoted from Dxo themselves it shows how sensitive each camera will be to light sensor wise, plus how the extra resolution gives you more data for a higher ISO when printed. That should be pretty straight forward for photographers and if not the graph, the DXo ISO score.

    And if the graph is examined on the ISO's, Olympus ISO 3200 is less than Sony's ISO 3200 even. So while someone is shooting at ISO 3200 on the e-5, they only need to put their A900 at 2800 (around so) for an equivalent sensitivity. Knowing this is important because it's going to affect that shutter speed in low light, having to do ISO higher than 3200 on the Olympus then.

    the bottom line is, if you shoot a jpeg from a camera its ISO will be pretty close to right b/se it matches the tone curve the manufacturer set conditional to ISO measurement. Since DxO for whatever reason use their own curve that measurement becomes displaced. We could assume this is why they maintain a reflection of the manufacturers setting on their charts.
    They do that so you know what the effective raw ISO of your camera is when you set it to the different numbers. I can corroborate what they are saying in the ISO sensitivity graph, with my 620. They described the graph appropriately and what I found matches completely- that ISO 100 = ISO 200 under exposed. Shooting a JPEG doesn't reveal the sensor potential as JPEGS engines vary wildly.

    Thats the way I interpret it, and that isnt an unfair observation nor is it 'erroneous' as you put to me later
    No, (edit) the actual sensor sensitivities vary and you get the slice on the graph that you get. The A900 does better than the iSO difference you said.

    (edit)

    'erroneous conclusion' which is just inflammatory BS that I suggest you stay away from in this place.
    I suggest you look at the language you used yourself when you started to reply. A claim was made that I believe mis represented Dxo. To me it looks pretty factual. (edit) Pointing this out shouldn't be inflammatory.

    (edit)

    - Raist
    Last edited by raist3d; 10th March 2011 at 10:18.

  36. #186
    Senior Member
    Join Date
    Dec 2010
    Posts
    2,118
    Post Thanks / Like

    Re: E-5 better in low light than D7000? Yes, maybe.

    Quote Originally Posted by Jorgen Udvang View Post
    This thread may be interesting reading:

    http://forums.dpreview.com/forums/re...ssage=37928329

    The photos are apparently done mostly with the 12-60 giving apertures from 3.4 to 4.0.
    There's some banding in some of the shots, others have a chunk of noise, but they still hold the overall color tone which I think is something Olympus does very very well particularly with the sensor they are dealing with.

    Some of the shots, like the three guys with the white dressed suit guy in the middle are 1/400- that's somewhat strong theater light. Everything where the light hits will show up pretty decent with reasonable detail, it's the mid tones and the lower light areas that go muddy.

    While I applaud him for posting 1600ish shots, it's still quite the resize down. I am not quite as convinced as he is that the 620 couldn't get closer to what he did, and I say this from my relative experience where I shot a bar concert at super low iso in really bad light. Certainly not JPEG though.

    - Raist

  37. #187
    Senior Member Riley's Avatar
    Join Date
    Dec 2007
    Location
    Adelaide Hills South Australia
    Posts
    441
    Post Thanks / Like

    Re: E-5 better in low light than D7000? Yes, maybe.

    Quote Originally Posted by raist3d View Post
    I don't have to do that. DXo does an excellent job of explaining that themselves.

    http://dxomark.com/index.php/Our-pub...-make-it-clear
    - Raist
    thats not good enough

    I want you to tell me why you think there is an ISO and a 'true measured ISO'. You seem to be operating on the impression one of them is wrong, I want to hear in your own words why you think that is

  38. #188
    Senior Member
    Join Date
    Dec 2010
    Posts
    2,118
    Post Thanks / Like

    Re: E-5 better in low light than D7000? Yes, maybe.

    Quote Originally Posted by Riley View Post
    thats not good enough

    I want you to tell me why you think there is an ISO and a 'true measured ISO'. You seem to be operating on the impression one of them is wrong, I want to hear in your own words why you think that is
    Why is what they explain not good enough? They are even going by an ISO standard, and setting a baseline for comparison in sensor sensitivity. As far as the graph goes I don't even have to touch that. This is about the more straightforward concept that the camera sensors vary in their sensitivity at the different rated manufacturer ISO.

    As a quick example- the E-3 gets the short end of the stick in highlight range comparison but that's vs an e-30/e-5 with the whole DR tones shifted up by one stop. You can say ISO 200 on the E-5 is a true ISO with less shadow DR, or a pushed ISO 100 to preserve highlights that could have more shadow DR if it wasn't so.

    In any case, I do think DXo explains this rather well and this doesn't take away from the measured sensor performance, in which the A900 clearly has better effective ISO performance than the E-5 in low light/high iso.

    - Raist

  39. #189
    Subscriber Member Jorgen Udvang's Avatar
    Join Date
    Nov 2007
    Location
    Pratamnak
    Posts
    9,346
    Post Thanks / Like
    Images
    2157

    Re: E-5 better in low light than D7000? Yes, maybe.

    Raist,
    One of those shots (the woman talking on the phone) is at ISO 6400 at 137mm and f/3.4. If we assume that he needs that much DOF, with an FF camera, he would have to use 274mm, f/6.8 and ISO 25,600. It would be interesting to see a photo shot at those specs that improves more than just a fraction over the photo taken with the E-5.

    I think 1600px is a relevant size. Larger than that isn't usable for web, and for print at anything up to A3, the photo will look very similar to this, particularly with commercial offset printing, where the grid hides a lot of the noise (and details) anyway.

  40. #190
    Senior Member Riley's Avatar
    Join Date
    Dec 2007
    Location
    Adelaide Hills South Australia
    Posts
    441
    Post Thanks / Like

    Re: E-5 better in low light than D7000? Yes, maybe.

    2nd opportunity and you ditched, since you are unable to answer I will tell you.
    The primary locations for ISO sensitivity are legitimately placed to fit with DxO's own curves for their own purposes.

    The markings treated as secondary relate to the locations for the manufacturers selected curves that comply with ISO SOS and ISO REI, DxO's curve selection which is assumed (but not stated) to be the same for every sensor they test does not comply with ISO SOS and REI. DxO then attach what they see as the ISO in comparison to their own, that doesnt make them right, moreover for photography purposes its not especially useful. Whereas the alternate used to comply with the standard will be within specification

    This is why review sites stopped testing ISO accuracy some time ago, compliance to the standard meant the test was pointless and simply a waste of time.

    is there any evidence of this ?
    heres the low down on the standards from Andy Westlake in responce to this message from a user confused about the conflicting data DxO present
    http://forums.dpreview.com/forums/re...ssage=32963533
    Andy's answer
    Quote Originally Posted by Andy Westlake
    http://forums.dpreview.com/forums/re...q=dxo+iso&qf=m
    The difference here is pretty straightforward - the issue is that there are several definitions and methods of measuring ISO, and the one DxO uses is different from everyone else's.

    The relevant standard, ISO 12232:2006 has five possible definitions; the ones used by camera manufacturers are about image brightness (called standard output specification and recommended exposure index, SOS and REI). In the simplest fashion, according to SOS if you point the camera at a grey card, meter and shoot it should be rendered as middle-grey (50% luminosity). This makes sense - it's the way most photographers think about exposure. REI extends this to take into account the fact that pattern metering of high contrast scenes becomes a matter of interpretation rather than 'right' or 'wrong'.

    DxO's method is based on something different - highlight clipping. This is useful for comparing raw sensor performance but doesn't reflect how cameras will render the image. Expose cameras from various manufacturers based on DxO's measured ISO and they'll all clip whites at the same point, but give images of different brightness.

    The difference between DxO's measured ISO and the manufacturer's recommended ISO gives an indication of what we call the highlight dynamic range of the camera. Ironically, the greater the difference (i.e. the lower DxO's ISO relative to the indicated), the larger the highlight dynamic range, resulting in smoother rendition and rolloff of highlight detail.

    So it's not all one big con after all.
    Critical to these ideas is that of scene brightness and resulting exposure, DxO measure from the highlight clipping point, the standard considers the middle (50%) grey for its measure of exposure (and yes these impacts subvert the DR data too). The differences are commonly conversed as those between jpeg and RAW, this however isnt true either. For the more salient points from Andy Westlake again.

    http://forums.dpreview.com/forums/re...ssage=36602488
    Quote Originally Posted by Andy Westlake
    This is the crucial point - different tests will bring different results. What then becomes important is to understand what those results mean.

    The CIPA standard for ISO, as used by Japanese camera manufacturers and which is essentially incorporated into ISO 12232:2006, only considers camera JPEGs. The moment you switch to a third-party raw converter, you're not technically talking about ISO at all, at least within the terms of this standard.

    The fact that the CIPA standard simply doesn't relate to the raw file meant DxOMark had to use a different definition, that's within ISO 12232 but not used by the camera manufacturers. It's based on the white point of the file, which is why it considers ISO 100 and 200 the same on Olympus cameras (both clip to white at the same point in the image). It doesn't consider the intended image brightness, i.e. the level in the raw file the camera's metering assigns to middle grey, at all. This is why it considers 100 and 200 to be identical in terms of noise and DR, and within its own terms it's absolutely correct.

    When you convert a raw, it's instructive to compare the result to the camera's JPEG. If the converted file is lighter or darker, then the raw converter's default settings misrate the ISO relative to the manufacturer's intent. But don't worry, it's almost certainly got an 'Exposure' slider, so you can change the brightness at will. This slider is in fact an ISO control - it changes the image brightness for a given shutter speed/aperture combination. ISO is a very flexible concept indeed when you're using RAW.

    As it happens, Lightroom generally aims to match the camera's metering intent as closely as possible, but this can depend on the version you're using. We frequently see early support that doesn't match the jpegs at all, which is then fixed in later versions.

    The take home message should be that the raw converter is a crucial part of the imaging chain, and itself influences the effective ISO. So it's incorrect to say manufactuers are mis-rating their ISOs in this case.
    There is a far higher population of users with RAW converters adjusted to manufacturer specs rather than straight RAW converters with no front ends such as dcraw, by and large these follow those same curves as the curves set by the manufacturer which is not to suggest that cant be changed but it is the default position for RAW. Indeed these converters are these days looked upon to be the most favourable comparison, giving the better noise performance and colour gamut than front endless dcraw.

    Since you never answered me twice Im rightly given to assume your position is that the manufacturer position on ISO is the lesser to DxO, which is quite wrong. DxO data fits its own position on ISO and no-one elses, certainly not complying with the manufacturers (all of them) standards. Yes DxO shows data that agrees with your position, the problem all along was you chose the wrong data set.
    Last edited by Riley; 9th March 2011 at 22:51. Reason: additions

  41. #191
    Workshop Member ptomsu's Avatar
    Join Date
    Nov 2007
    Location
    Austria, close to Vienna
    Posts
    3,872
    Post Thanks / Like

    Re: E-5 better in low light than D7000? Yes, maybe.

    Raist,

    sure I can also see some banding. But maybe acceptance level is different for different people, as I did actually not even recognize it before you showed me and even knowing it is there I am not too scared, as for my eyes it is very weak. Having said that it would be definitely better if it was not there at all!

    I will try Nik SW at see what it does.

    PS: Not to play down banding, but sometimes printers - even high end printers - introduce more "banding" or however it is called there. So maybe on screen it looks really worse, while when printed it becomes no longer relevant.

  42. #192
    Member
    Join Date
    Aug 2009
    Location
    Warsaw, Poland
    Posts
    188
    Post Thanks / Like

    Re: E-5 better in low light than D7000? Yes, maybe.

    Peter,
    Judging from the experience of the DPreview Olympus forum, now you will never free yourself from the Oly hunters. Your pictures will be magnified 600% and dissected into subpixel accuracy. You will see things you have never seen before in your digital files. You will be proven wrong on all accounts and forced to return your gear and cancel orders. You will regret the day you heard of Olympus.
    You have been warned.

  43. #193
    Senior Member
    Join Date
    Dec 2010
    Posts
    2,118
    Post Thanks / Like

    Re: E-5 better in low light than D7000? Yes, maybe.

    Quote Originally Posted by Jorgen Udvang View Post
    Raist,
    One of those shots (the woman talking on the phone) is at ISO 6400 at 137mm and f/3.4. If we assume that he needs that much DOF, with an FF camera, he would have to use 274mm, f/6.8 and ISO 25,600. It would be interesting to see a photo shot at those specs that improves more than just a fraction over the photo taken with the E-5.
    The photo of the woman has banding, something the author himself acknowledged. Also it doesn't have as much details.

    I think 1600px is a relevant size. Larger than that isn't usable for web, and for print at anything up to A3, the photo will look very similar to this, particularly with commercial offset printing, where the grid hides a lot of the noise (and details) anyway.
    I agreed 1600 is a relevant size. All I am saying is that even then it's still a big resize (12 MP to 2 MP). I applaud the OP for posting that and not the typical SVGA/XGA size (800x600 or 1024x768 respectively). However, to mention a recent example, an A900 will definitively show much more detail at this resolution and this ISO coming from ISO 6400. Dpreview has an ISO 6400 shot from the A900 that is done in lower light than this, and resize it to this size and see how much detail there is.

    Now one more thing to remember- the scenario you mentioned is where we want the same DOF so we close by two stops. What if we have enough DOF and we care about getting a higher quality shot at a lower ISO? Both scenarios are valid and real.

    So in one case, we get better anyway than an E-5 from say an A900/A850, and in the other we get much much better. If we compare vs the non FF latest APS-C competition, we also get better by ISO or DOF case.

    I want to make clear though, that shot is usable for some scenarios, so I am not saying it's useless, and in particular for the web, Olympus is good at preserving their color profile in a resize down. But FF or APS-C competitors would certainly do better at that ISO.

    Something else to consider in scenarios that the Zuiko 14-35 is brought up - at that point buying an E-5 + that lens in the USA puts you at $4,000 USD (bhphoto) vs $3,600 of an A850 + Carl Zeis F2.8 24-70 high end lens. The A850 combo is actually $400 USD cheaper...

    The A850 performs like the A900 DR/ISO wise.

    - Raist

  44. #194
    Senior Member
    Join Date
    Dec 2010
    Posts
    2,118
    Post Thanks / Like

    Re: E-5 better in low light than D7000? Yes, maybe.

    Quote Originally Posted by ptomsu View Post
    Raist,

    sure I can also see some banding. But maybe acceptance level is different for different people, as I did actually not even recognize it before you showed me and even knowing it is there I am not too scared, as for my eyes it is very weak. Having said that it would be definitely better if it was not there at all!
    Well different people/different needs/different sensitivities/etc. I think the Nik will help quite a bit in that case. The main PITA of banding per se is that it can show up when you least want it and if you need to push up a shot, or push the shadow up, it can show the bands very fast. What I don't understand is why it seems the E-30 is better at this from what I am seeing. Weird.

    I will try Nik SW at see what it does.

    PS: Not to play down banding, but sometimes printers - even high end printers - introduce more "banding" or however it is called there. So maybe on screen it looks really worse, while when printed it becomes no longer relevant.
    Oh, I can't agree with that :-) If you are paying for a high end printer and gives you banding, it's time to switch! :-) Printer bands are different from these too.

    - Raist

  45. #195
    Senior Member
    Join Date
    Dec 2010
    Posts
    2,118
    Post Thanks / Like

    Hunters says you?

    Quote Originally Posted by nugat View Post
    Peter,
    Judging from the experience of the DPreview Olympus forum, now you will never free yourself from the Oly hunters. Your pictures will be magnified 600% and dissected into subpixel accuracy. You will see things you have never seen before in your digital files. You will be proven wrong on all accounts and forced to return your gear and cancel orders. You will regret the day you heard of Olympus.
    You have been warned.
    Nobody except you or Riley have mentioned preview here. The whole thread is dissecting other cameras. If you haven't noticed I continuously post photographs from several cameras, not just talk and not walk. And don't be silly, I have recommended the E-5 to a few people. Where were you when I was singing praise of the e-330 and e-420?

    Looks like you were referring to other type of hunters of which you form part of.

    - Raist

  46. #196
    Subscriber Member Jorgen Udvang's Avatar
    Join Date
    Nov 2007
    Location
    Pratamnak
    Posts
    9,346
    Post Thanks / Like
    Images
    2157

    Re: E-5 better in low light than D7000? Yes, maybe.

    Quote Originally Posted by raist3d View Post
    Something else to consider in scenarios that the Zuiko 14-35 is brought up - at that point buying an E-5 + that lens in the USA puts you at $4,000 USD (bhphoto) vs $3,600 of an A850 + Carl Zeis F2.8 24-70 high end lens. The A850 combo is actually $400 USD cheaper...

    The A850 performs like the A900 DR/ISO wise.

    - Raist
    If the A850 is better at ISO 6400 than the E-5 at ISO 3200, the A850 will be the better option. That is, until you need a fast 300mm eqv. Sony doesn't even have a 300mm f/4, so you are stuck with the $6,000+ 300mm f/2.8 against the $2,000 Zuiko 150mm f/2.0. You can of course buy the 70-400mm, but I assume that's beyond f/5.0 at 300mm, and since the A850 maxes out at ISO6400, that alternative isn't really relevant.

    Just to mention it:
    The K-5 would of course be an interesting alternative. Unfortunately, the 16-50mm f/2.8 isn't nearly as sharp as either the Zuiko or the Zeiss. If I was going to use only primes (and there was a reliable distributor for Pentax in this country), that would probably have been the best low light alternative.

  47. #197
    Senior Member
    Join Date
    Dec 2010
    Posts
    2,118
    Post Thanks / Like

    Re: E-5 better in low light than D7000? Yes, maybe.

    Quote Originally Posted by Jorgen Udvang View Post
    If the A850 is better at ISO 6400 than the E-5 at ISO 3200, the A850 will be the better option. That is, until you need a fast 300mm eqv. Sony doesn't even have a 300mm f/4, so you are stuck with the $6,000+ 300mm f/2.8 against the $2,000 Zuiko 150mm f/2.0. You can of course buy the 70-400mm, but I assume that's beyond f/5.0 at 300mm, and since the A850 maxes out at ISO6400, that alternative isn't really relevant.
    That only depends on the lenses you need. As I pointed out, a typical common zoom range + the A850 is cheaper. Moreover, as we examined in this thread, the A850 has a 2 stop+ advantage at high ISO over the E-5. This means for an equivalent type of shot you can even push it up (i..e shot ISO 6400 under exposed). The E-5 at ISO 3200 is has banding somewhat often. You may want to compare vs ISO 1600.

    We can make contrived comparisons all day. For example, there's an F1.8 135mm Carl Zeiss for $1,479. Olympus doesn't even make a Zuiko faster than F2.0, you have to rely on Panasonic-Leica or Sigma on a native 4/3rds mount. The lens I quoted is a pretty standard very flexible range zoom- the range of the 14-35. If you need 300mm then you spend less with the Zuiko (but still no better ISO performance). If you need a more standard zoom range, you come ahead with the Carl Zeiss I mentioned.

    But yeah, if you need that zoom, there's a big jump from E-5 to A850 in price *in that particular situation*.

    Just to mention it:
    The K-5 would of course be an interesting alternative. Unfortunately, the 16-50mm f/2.8 isn't nearly as sharp as either the Zuiko or the Zeiss. If I was going to use only primes (and there was a reliable distributor for Pentax in this country), that would probably have been the best low light alternative.
    The 16-50 F2.8 is actually pretty good. Combine the part of the body in the equation and it certainly will look sharp at high ISO. And as you say, primes galore that are good. But yeah, I am not suggesting you don't pick an E-5 or such. If you need that 300mm range and you only see Oly products around, sure, that makes a lot of sense. That's a bit different from equating the high ISO of the E_5 to the other options as I have seen in some posts (said in general).

    - Raist

    PS: When are you getting your E-5? It's an honest question, it seems like it's a good match for what you want.

  48. #198
    Subscriber Member Jorgen Udvang's Avatar
    Join Date
    Nov 2007
    Location
    Pratamnak
    Posts
    9,346
    Post Thanks / Like
    Images
    2157

    Re: E-5 better in low light than D7000? Yes, maybe.

    It's all a question of individual needs of course. The choice for me is between Olympus and Sony. The telephoto need is real, since I shoot motor-sports and sometimes golf. For golf, I even need longer, but a TC or even a good old 500mm f/8 reflex can solve that (golfers move slower than cars). What makes Olympus stand out when it comes to the fast 300mm is a combination of optical quality, weight, size and price, and I've never doubted that 4/3 would be the best solution for that.

    What I needed to know was if an E-5 can also solve other parts of my photography, and solve the annoying standard zoom challenge. This discussion, and lots of going trough photos and doing all kinds of research indicates that it can. Again, Sony is the strongest competitor, and if I can afford two kits (question of priorities and time) I might go for both. Time will show.

    A design assignment forces me to buy a new Mac as soon as possible, so I may have to postpone the E-5 a month or two. I might buy the 14-35 first to check out the potential, using it on the GH1 or even buying an E-620 for that. At the moment, I don't have a sharp standard zoom, so that part is rather urgent.

  49. #199
    Senior Member
    Join Date
    Dec 2010
    Posts
    2,118
    Post Thanks / Like

    Re: E-5 better in low light than D7000? Yes, maybe.

    Quote Originally Posted by Jorgen Udvang View Post
    It's all a question of individual needs of course. The choice for me is between Olympus and Sony. The telephoto need is real, since I shoot motor-sports and sometimes golf. For golf, I even need longer, but a TC or even a good old 500mm f/8 reflex can solve that (golfers move slower than cars). What makes Olympus stand out when it comes to the fast 300mm is a combination of optical quality, weight, size and price, and I've never doubted that 4/3 would be the best solution for that.
    Well if that's the case, that you never doubted 4/3rs would be the best solution for that and that's what you need, why a choice at all? i.e. why is Sony even in the running? Sounds like you just just get the E-5, like now and call it done :-)

    What I needed to know was if an E-5 can also solve other parts of my photography, and solve the annoying standard zoom challenge. This discussion, and lots of going trough photos and doing all kinds of research indicates that it can. Again, Sony is the strongest competitor, and if I can afford two kits (question of priorities and time) I might go for both. Time will show.
    I don't know how specific are your needs and if you have a particular photograph business, but it sounds to me that if the E-5 has the ISO quality you are going to be shooting at, I wouldn't even look at the Sony. I mean personally to me this sounds like an either or brand thing because in both you will spend a significant amount of cash, and sounds like both can more or less meet your needs with a few cons the other may do a bit better.

    The E-5 is very well built. So is the A900/850. To me personally what attracts me in this comparison the most is the ergonomics and OVF. The OVF of the A900 is simply glorious and the ergonomics/interface on the A900 are to me much better thought out than the E-5. The E-5 got the E-3's which in my opinion, has several issues. This is nothing against Olympus- Olympus had it right with the E-1 but dropped the ball with the E-3.

    While I can understand some of their design goals- like changing all camera settings at once, removing "lock" buttons for dials by having you press and hold a button + turn a dial, how they implemented that on the E-3/E-5 is a bit puzzling to me in particular the three buttons on the upper right.

    Now this is my opinion, I am not trying to dispute or convince you or anything. If you like the E-5 ergonomics as is you are set. I personally made my peace with the interface, so to me it's annoying but ok. I guess I can deal. The A900's ergonomics though are a dream. Shows that the Minolta team brought good know how.

    A design assignment forces me to buy a new Mac as soon as possible, so I may have to postpone the E-5 a month or two. I might buy the 14-35 first to check out the potential, using it on the GH1 or even buying an E-620 for that. At the moment, I don't have a sharp standard zoom, so that part is rather urgent.
    I can tell you the 14-35 is outstanding as a lens. The only issue it has is that moderate to low light AF speed. But if you do MF or daylight shooting, it's a non issue. The lens itself is really really really good.

    - Raist

  50. #200
    Workshop Member ptomsu's Avatar
    Join Date
    Nov 2007
    Location
    Austria, close to Vienna
    Posts
    3,872
    Post Thanks / Like

    Re: E-5 better in low light than D7000? Yes, maybe.

    Quote Originally Posted by Jorgen Udvang View Post
    It's all a question of individual needs of course. The choice for me is between Olympus and Sony. The telephoto need is real, since I shoot motor-sports and sometimes golf. For golf, I even need longer, but a TC or even a good old 500mm f/8 reflex can solve that (golfers move slower than cars). What makes Olympus stand out when it comes to the fast 300mm is a combination of optical quality, weight, size and price, and I've never doubted that 4/3 would be the best solution for that.

    What I needed to know was if an E-5 can also solve other parts of my photography, and solve the annoying standard zoom challenge. This discussion, and lots of going trough photos and doing all kinds of research indicates that it can. Again, Sony is the strongest competitor, and if I can afford two kits (question of priorities and time) I might go for both. Time will show.

    A design assignment forces me to buy a new Mac as soon as possible, so I may have to postpone the E-5 a month or two. I might buy the 14-35 first to check out the potential, using it on the GH1 or even buying an E-620 for that. At the moment, I don't have a sharp standard zoom, so that part is rather urgent.
    All these are valuable considerations of course!

    Fr me the E5 has something else as that high ISO junkie capabilities, which are less important to me, for me it reproduces very similar images as Leica M8, even 9 and DMR. This is partially because of the processing and partially because of the lenses, if you use high grade glass. This is a combination which I cannot achieve with any other DSLR currently (well the S2 for sure, but that is arguably a very different area). So I prefer the E5 over Nikon, Pentax and even Sony. Canon is out of the game for me anyway - do not like their colors and rendering!

    A Sony could definitely be an option, especially if I needed a very high ISO beast or if I needed high MP count in a DSLR. I am sure the next incarnation of their FF A900 will be a stellar performer, I also hope the do with much weaker AA filter - SIGH! So I would buy a Sony in a minute if I had the need for these options and of course with some of the Zeiss glass (although it is not same level for me as it was during Contax times for example). And this high res FF beast would also for sure be much cheaper than any CanNikon FF top of the range DSLR.

    So there are multiple additional option, but I am kind of pretty sure I will nicely survive with the E5 the next years to come. Especially with the 14-35 and 35-100 in my bag

Tags for this Thread

Posting Permissions

  • You may not post new threads
  • You may not post replies
  • You may not post attachments
  • You may not edit your posts
  •