skimmel Posted September 11, 2009 Share #1 Posted September 11, 2009 Advertisement (gone after registration) OK, this is a pretty quick test of increasing ISO with M9 done yesterday (firmware 1.002 -- I think I got the number right). They were taken in rapid succession with same lens, aperture and just increasing ISO. Shot in RAW, developed with Lightroom defaults -- no noise reduction. The reason I did this was to see if there was an incremental difference between, say, 640 and 800. It has been noted on some cameras (and I noticed it on my old Canons) that the difference in noise was apparent at full stops but at 1/3 stops, the noise often just jumped up to the next level (e.g., 1000 looked pretty much like 1600). Not sure this is going to come out well on the web, but on full screen computer, there is clearly a difference with each 1/3 ISO. I tested this by putting up pairs without looking at ISO and I could, 100% of the time, pick the higher ISO. To me this is very good news -- it means I can bump from, say, 800 to 1000 and not get a large jump in noise (to, say the 1600 level). Below are crops -- I'll post the full picture next. Welcome, dear visitor! As registered member you'd see an image here… Simply register for free here – We are always happy to welcome new members! Link to post Share on other sites Simply register for free here – We are always happy to welcome new members! ' data-webShareUrl='https://www.l-camera-forum.com/topic/96274-increasing-iso-test-with-m9/?do=findComment&comment=1028575'>More sharing options...
Advertisement Posted September 11, 2009 Posted September 11, 2009 Hi skimmel, Take a look here Increasing ISO Test with M9. I'm sure you'll find what you were looking for!
skimmel Posted September 11, 2009 Author Share #2 Posted September 11, 2009 OK. Here's the full picture. Welcome, dear visitor! As registered member you'd see an image here… Simply register for free here – We are always happy to welcome new members! Link to post Share on other sites Simply register for free here – We are always happy to welcome new members! ' data-webShareUrl='https://www.l-camera-forum.com/topic/96274-increasing-iso-test-with-m9/?do=findComment&comment=1028576'>More sharing options...
barjohn Posted September 11, 2009 Share #3 Posted September 11, 2009 For those of you that think the M9 is clean, here is a link to and ISO 4000 shot taken with the lowley E-P1. Compare detail and noise levels. Flickr Photo Download: Self Iso 4000_3 Link to post Share on other sites More sharing options...
skimmel Posted September 11, 2009 Author Share #4 Posted September 11, 2009 I was quite pleased that noise seemed to increase incrementally as ISO went up by small steps. I was worried that they'd put in 1/3 stops but that the noise wouldn't be much different between 1000 and 1250. My take from a very quick test of M8 v M9 was about 1 stop improvement. I'll be very curious to see what Sean Reid finds. Link to post Share on other sites More sharing options...
Guest roey Posted September 11, 2009 Share #5 Posted September 11, 2009 I was wondering about just this question (the inbetween ISO steps) and it is good to hear that Leica got it right. Thanks for sharing! Link to post Share on other sites More sharing options...
skimmel Posted September 11, 2009 Author Share #6 Posted September 11, 2009 I was wondering about just this question (the inbetween ISO steps) and it is good to hear that Leica got it right. Thanks for sharing! Your welcome. It's the first time I tried to post comparative pics and I really wanted to provide info for a question I was wondering about as well. I know how thorny these comparisons can get on forums. So, I really appreciate your "thanks." Link to post Share on other sites More sharing options...
chris_tribble Posted September 11, 2009 Share #7 Posted September 11, 2009 Advertisement (gone after registration) Skimmel - your experience accords with mine - one stop difference. For me this is really good. Although I use 2500 from time to time on the 5D2, it's almost exclusively when I'm working in low light (concert / theatre) with long telephoto (300 2.8) - not the conditions in which I'd be using an M camera. Where I do work with the M, clean 1250 is what I really need. This covers night / street, documentary work and the rest extremely well, with higher ISO available which I'd treat as B&W in post anyway - I don't find colour that meaningful when the light really drops apart from for fine art effects... Link to post Share on other sites More sharing options...
tomasis7 Posted September 11, 2009 Share #8 Posted September 11, 2009 For those of you that think the M9 is clean, here is a link to and ISO 4000 shot taken with the lowley E-P1. Compare detail and noise levels. Flickr Photo Download: Self Iso 4000_3 very awful.. see noise at shoulders it looks overprocessed Link to post Share on other sites More sharing options...
barjohn Posted September 11, 2009 Share #9 Posted September 11, 2009 very awful.. see noise at shoulders it looks overprocessed Then show me an M8 or M9 image with as much detail and as clean shot at ISO 4000. IT should be easy for you given the much larger sensor and much larger photo receptors. Link to post Share on other sites More sharing options...
Rick Posted October 7, 2009 Share #10 Posted October 7, 2009 barjohn, it doesn't look so good to me either, but it is hard to tell with the processing you have done. To really be fair you need to post the picture as is, without any processing. If, you would like me to compare noise and detail for you, could you post the color version with no changes? Thank you. Link to post Share on other sites More sharing options...
ulrikft Posted October 7, 2009 Share #11 Posted October 7, 2009 Then show me an M8 or M9 image with as much detail and as clean shot at ISO 4000. IT should be easy for you given the much larger sensor and much larger photo receptors. I can shoot them side by side to avoid differences. That shot looks weirdly clean (and noisy) at the same time, got exif-link on it? or full res image? Link to post Share on other sites More sharing options...
sean_reid Posted October 7, 2009 Share #12 Posted October 7, 2009 I was quite pleased that noise seemed to increase incrementally as ISO went up by small steps. I was worried that they'd put in 1/3 stops but that the noise wouldn't be much different between 1000 and 1250. My take from a very quick test of M8 v M9 was about 1 stop improvement. I'll be very curious to see what Sean Reid finds. About which aspect? You know the ISO noise tests are already in the review, right? Cheers, Sean Link to post Share on other sites More sharing options...
ulrikft Posted October 7, 2009 Share #13 Posted October 7, 2009 Iso 6400 with the EP1 Flickr Photo Download: Olympus E-P1 ISO 6400 Iso 3200 with the EP1: Flickr Photo Download: Olympus E-P1 ISO 3200 Great performance from such a small sensor, but.. I somehow feel that your image is cleaner at 4000, than the 3200 one is at 3200. So either the small size is fooling me, or you have done lots of NR, or something funky is about Link to post Share on other sites More sharing options...
Rick Posted October 7, 2009 Share #14 Posted October 7, 2009 ulrikft, thanks for the post of the image. That is more what I would expect the image from the EP1 to look like at that high iso. I hope barjohn will post the unprocessed image but, I'm not holding my breath. I suspected that the image he posted as an example of the superiority of the EP1 was pretty well processed, converted to B&W to hide the chroma and otherwise doctored up. Link to post Share on other sites More sharing options...
thrice Posted October 8, 2009 Share #15 Posted October 8, 2009 For those of you that think the M9 is clean, here is a link to and ISO 4000 shot taken with the lowley E-P1. Compare detail and noise levels. Flickr Photo Download: Self Iso 4000_3 Oh damn, the M9 sure sucks compared to that. Must be awesome not to feel tempted to buy an M9 now... If you have a pre-order you should definitely cancel it Link to post Share on other sites More sharing options...
lars_bergquist Posted October 8, 2009 Share #16 Posted October 8, 2009 Here's a thought. Suppose that you tested the M9 and your EP-1 side by side, ISO for ISO, but on a resolution test chart? Would the supposedly 'cleaner' image of the EP-1 resolve as much detail as the M9 does? You must understand that the Leica and the Japanese philosophies are very different. The Japanese way is to apply lots of noise suppression in the camera, to make the camera file as 'noiseless' as possible. This assumes that users are allergic to noise but do not want to do any noise suppression in PP -- or any PP, for that matter. Strong noise suppression does also suppress fine detail, however, lowering the resolution, just as an AA filter does. The Leica way is to capture and preserve as much detail as possible in the camera file. What else do we have those premium grade lenses for? So Leica accept more noise as the price to pay for detail. Please note that I am speaking of detail at high local frequencies of 40--80 lp/mm. Higher levels of noise suppression are left to the user's choice (definitely not the Japanese way). So you should make sure to compare apples with apples, preferably by making the test outlined above. The old man from the Age of the Siemens Star Link to post Share on other sites More sharing options...
250swb Posted October 8, 2009 Share #17 Posted October 8, 2009 If I can interject there is no background noise reducion in EP-1 RAW files, and I would expect to see similar results as those posted above from my EP-1. But I think it is difficult to compare directly. The EP-1 has a very weak AA filter, but its pixel density is higher than the M9, meaning its already high resolution (higher than a Nikon D3 using a common test lens) is confusing how detail and noise is assesed. That initial high resolution will start to degrade earlier in like for like tests as the image size gets bigger than its native dimensions. And of course the dynamic range isn't as high as an M9, meaning you need to expose very carefully (yes, don't we all!) for noise not to get excessive with the EP-1. So I won't be cancelling my order for an M9, even though the m4/3 sensors in the Pana G1 and GF1, and Olympus EP-1 are very impressive performers. Link to post Share on other sites More sharing options...
Recommended Posts
Archived
This topic is now archived and is closed to further replies.