Page 1 of 2 12 LastLast
Results 1 to 25 of 43

Thread: ISO Bracketing with E-5: Too good to be true?

  1. #1
    Join Date
    Sep 2011
    Location
    Naperville, IL 60563
    Posts
    66
    Thanks
    0
    Thanked 0 Times in 0 Posts
    Feedback Score
    0

    Default ISO Bracketing with E-5: Too good to be true?

    Hey all!

    Just came back from an awesome Joel Grimes workshop in Michigan this weekend. He showed some amazing techniques including the ability to process 32bit HDR files from a 3 frame bracket capture (using the new Cam Ranger that works with Nikons and Canons). Being the odd man out, I navigated the menu on my E-5 to see if I could even do ISO Bracketing, and apparently I can. The undocumented caveat is that you _have_ to set your ISO to a minimum of 200 because you get 3 exposures for one shutter cycle- it goes up to 1 stop under to 1 stop over (other options are .3 and .7 EV, but no idea how the ISO value would actually turn out when looking at the aftermath).

    Now, the "too good to be true" part. You actually get 3 exposures for every shutter cycle... how is this even possible? Is this considered true bracketing? Why don't you have to go through 3 shutter cycles like you would have to in AE bracketing?

    I essentially got 3 exposures at ISO 200, 100, and 400. The crazy thing is that because the trio was produced from just one shutter cycle, there is essentially no ghosting or alignment issues. I'm going to email Olympus and see what they have to say. This could be the world's best HDR portrait secret that no other camera system can provide today.

  2. #2
    Join Date
    Jul 2008
    Location
    NH, USA
    Posts
    2,592
    Thanks
    57
    Thanked 30 Times in 26 Posts
    Feedback Score
    13 (100%)

    Default Re: ISO Bracketing with E-5: Too good to be true?

    Do you mean that for each of the 3 ISOs (200, 100, and 400) you get three images and that WITHIN each ISO each image is varied with a different exposure (via a change to the shutter speed or aperture)?

    Or do you mean that you get 3 images within each ISO, all with the same exposure? If they are all the same, I don't see how that is very useful.
    Rich
    Olympus E-M10; Panasonic GM5
    m4/3 lenses: Oly 75-300; Oly 14-42 f3.5-5.6 II R; Oly 17 f1.8; Oly 40-150 f4.0-5.6 R; Oly WCON-P01 adapter; Rokinon f7.5 fisheye; Sigma 19 f2.8; Pan 20 f1.7; Pan 12-35 f2.8; Pan 12-32

  3. #3
    Join Date
    Jul 2008
    Posts
    839
    Thanks
    0
    Thanked 1 Time in 1 Post
    Feedback Score
    3 (100%)

    Default Re: ISO Bracketing with E-5: Too good to be true?

    I tried it and it seems to work as advertised. It is three images with the shutter and aperature locked so it would not be the same exposure.

  4. #4
    Join Date
    Jul 2008
    Posts
    839
    Thanks
    0
    Thanked 1 Time in 1 Post
    Feedback Score
    3 (100%)

    Default Re: ISO Bracketing with E-5: Too good to be true?

    There are actually multiple threads out on DPReview about this topic going back to at least 2008. The big question is how much different is this then just importing the image into an image editor and creating additional images with +/- exposure adjustments applied.

  5. #5
    Join Date
    Sep 2011
    Location
    Naperville, IL 60563
    Posts
    66
    Thanks
    0
    Thanked 0 Times in 0 Posts
    Feedback Score
    0

    Default Re: ISO Bracketing with E-5: Too good to be true?

    @RAH- one shutter cycle, 3 images, all varied ISO. It is EXTREMELY useful because you get details across a larger spectrum with no "ghosting" because it was just 1 shutter cycle. The intent is to use Photoshop CC's (CS6 also works) new PhotoMerge algorithm to let you work in 32bit mode. Because it's still in RAW, detail recovery should be insane. However, at the workshop, Joel was using a Canon 5dMark 3 and needed 3 shutter cycles to do ISO bracketing (it was cool because he set it up via the CamRanger where you can control your camera settings and look at previews via your iPad or iPhone). He had to account for ghosting and alignment since he was photographing a model and not a stationary, inanimate object. The ability to pull back details from shadows and highlights on a 32bit image doesn't even compare to a 16bit image. But his results were dramatically better but there was just a slight hint of blur. If there is a way I can get this method to work, then this will put the E-5 in a completely different class compared to any other camera out there. I'll upload some test shots later.

  6. #6
    Join Date
    Feb 2008
    Location
    Traverse City, MI
    Posts
    1,025
    Thanks
    16
    Thanked 17 Times in 9 Posts
    Feedback Score
    8 (100%)
    Real Name
    Clint

    Default Re: ISO Bracketing with E-5: Too good to be true?

    Gonna need to see a step by step with examples :-)
    Clint
    Rockin' it Mirrorless Style
    Olympus EM-1, E-PM2, LX-7, S110
    7.5, 12-32, 12-40, 14-42, 20, 25, 45, 75
    Asahi 35 2, SMC Super Tak 135 2.5, Super Tak 200 4
    HLD-7, FL-600r, FL-36r, FL-36, Cactus V5 Duo (x2)

  7. #7
    Join Date
    Feb 2008
    Location
    Traverse City, MI
    Posts
    1,025
    Thanks
    16
    Thanked 17 Times in 9 Posts
    Feedback Score
    8 (100%)
    Real Name
    Clint

    Default Re: ISO Bracketing with E-5: Too good to be true?

    Interesting...I just checked my EPM2...one thing to keep in mind with it, is that the iso has to be set to 400 for +1 compensation, since the lowest ISO is 200. Neat trick...I'll have to test this out in real life...sure beats doing the burst mode and counting 3 -5 times.
    Clint
    Rockin' it Mirrorless Style
    Olympus EM-1, E-PM2, LX-7, S110
    7.5, 12-32, 12-40, 14-42, 20, 25, 45, 75
    Asahi 35 2, SMC Super Tak 135 2.5, Super Tak 200 4
    HLD-7, FL-600r, FL-36r, FL-36, Cactus V5 Duo (x2)

  8. #8
    Join Date
    Mar 2007
    Posts
    1,652
    Thanks
    0
    Thanked 0 Times in 0 Posts
    Feedback Score
    0

    Default Re: ISO Bracketing with E-5: Too good to be true?

    I use ISO bracket all the time
    I dont think I would get the same results in the same time using some version of RAW processing
    the downside to it is that dark areas at ISO400 (100, 200, 400 ISO) come in quite noisy
    the new Sony sensor should make short work of same

    looking forward to it

    despite this pita of an image uploader I will persevere with some obligatory samples






    Riley

    Olympus User, Pro Photographer since 2003

  9. #9
    Join Date
    Jul 2008
    Posts
    839
    Thanks
    0
    Thanked 1 Time in 1 Post
    Feedback Score
    3 (100%)

    Default Re: ISO Bracketing with E-5: Too good to be true?

    The big downside would seem to be the limitation of only 3 images with a maximum of 1EV step in each direction. On the E-P5 specs it has this for the HDR function.

    bracketing for HDR postprocess 3 or 5 frames in 2.0/3.0EV steps selectable, 7 frames in 2.0EV steps selectable.
    *This function does not execute HDR processing.
    *HDR function can be programmed to the Fn-button.

  10. #10
    Join Date
    Sep 2011
    Location
    Naperville, IL 60563
    Posts
    66
    Thanks
    0
    Thanked 0 Times in 0 Posts
    Feedback Score
    0

    Default Re: ISO Bracketing with E-5: Too good to be true?

    @saburns- Having 3 exposures with 1 stop difference is plenty for getting details in the shadows and the highlights across the set. Doing HDR _in-camera_ totally defeats the purpose of what I am referring to since the goal is to process a 32bit file in camera-raw.
    @Riley- nice examples! perhaps this method would fare best doing high key compositions, although you can still do noise-reduction to the raw files prior to processing via LR or Bridge.

    Here is a test example of the ISO bracket trio as well as the HDR composite prior to doing any additional processing. In other words, only working in Adobe Camera Raw after running HDR Photomerge in Photoshop CC:

    ISO 200:
    Name:  ISOsplash-test-101.jpg
Views: 181
Size:  172.2 KB
    ISO 100:
    Name:  ISOsplash-test-102.jpg
Views: 181
Size:  130.7 KB
    ISO 400:
    Name:  ISOsplash-test-103.jpg
Views: 197
Size:  171.7 KB
    HDR Photoshop CC Photo-Merge-to-Camera Raw
    Name:  ISOsplash-test-104.jpg
Views: 448
Size:  153.4 KB

    again, the key benefit here for portrait work and living subjects is no ghosting- 1 shutter cycle, 3 exposures with bracketed ISO values. If it were for landscapes and even products, I would most likely use the typical AE bracketing.

  11. #11
    Join Date
    Feb 2008
    Location
    Traverse City, MI
    Posts
    1,025
    Thanks
    16
    Thanked 17 Times in 9 Posts
    Feedback Score
    8 (100%)
    Real Name
    Clint

    Default Re: ISO Bracketing with E-5: Too good to be true?

    Unless it were really windy, then I could see this coming in handy...I wonder how it would have worked for fireworks...
    Clint
    Rockin' it Mirrorless Style
    Olympus EM-1, E-PM2, LX-7, S110
    7.5, 12-32, 12-40, 14-42, 20, 25, 45, 75
    Asahi 35 2, SMC Super Tak 135 2.5, Super Tak 200 4
    HLD-7, FL-600r, FL-36r, FL-36, Cactus V5 Duo (x2)

  12. #12
    Join Date
    Oct 2009
    Posts
    659
    Thanks
    1
    Thanked 1 Time in 1 Post
    Feedback Score
    0

    Default Re: ISO Bracketing with E-5: Too good to be true?

    Thanks for sharing this E-5 gem.

    I never used it since I didn't see the point of ISO bracketing thinking it would give me the same issues + more noise then more conventional ShutterSpeed bracketing.

    Im not new to 32bit hdr ACR technique, I learned mine from awesome Gavin Hoey:

    Cameras:E-5; E-3; E-1; E-620; GH-1
    Lens: 12-60mm 50mm f/2; 50-200swd; Ex-25; 70-300; 40-150mkI; 35mm f/3.5; 25mm f/2.8; 14-140 u43
    Lights: Fl36Rx2; Fl50R; FL50; Elinchrom 2x400ws radio controlled.

    Click on the image to see my Facebook page
    And while you are at it, Like it if you like it

  13. #13
    Join Date
    Jan 2007
    Location
    Comox Bristish Columbia
    Posts
    3,222
    Thanks
    0
    Thanked 18 Times in 12 Posts
    Feedback Score
    2 (100%)

    Default Re: ISO Bracketing with E-5: Too good to be true?

    Quote Originally Posted by doofclenas View Post
    Unless it were really windy, then I could see this coming in handy...I wonder how it would have worked for fireworks...
    All three images are captured in the same shutter opening, one shot gives you three bracketed images. This would be ideal for fireworks and very windy days.

    Also, I imagine that it works on a lot of Oly cameras, it certainly works on my e3.
    __________________________________________________ _____________________________________ ​
    ​John Nicklin

    e3, ZD 11-22, ZD 12-60, Sigma 70-200, EC14, FL50R, 5DMkIII, 24-105 L, 70-200 f4 L, RB67, OM1, B&J 8x10, 5DIII, 24-105 f4 L, 70-200 f4 L, 100D.
    www.jnicklin.ca | flicker | SmugMug | Blog

  14. #14
    Join Date
    Sep 2011
    Location
    Naperville, IL 60563
    Posts
    66
    Thanks
    0
    Thanked 0 Times in 0 Posts
    Feedback Score
    0

    Default Re: ISO Bracketing with E-5: Too good to be true?

    Awesome. Still tweaking and testing. For example, did you know that you can actually _stack_ ISO bracketing and AE bracketing??!!! I just tried combining the ISO Bracketing setting with a 3F AE bracket and I got 9 total images, 3 different ISO settings for each exposure. This is nuts. If Oly comes out with the E-5 successor and uses the OMD sensor, this would definitely be a game changer that would make it a one of a kind capability even a $8000 body from Canon can't achieve...

    Again, low key images will definitely have more noise issues than high key (since noise in high key is not as pronounced). Also, because the E-5 (and possibly the OM-D) uses a 12bit DAC vs. the 14bit DAC that modern day DSLRs use, we don't have the Dynamic Range coming out of Raw as everyone else. That said, and by leveraging things like Pre-NR in RAW prior to HDR procesing (still have to see if Adobe LR will pass images to HDR Pro Photomerge if you do RAW treatments first), there could be some really fantastic ways to bring an entirely different class of detail to high resolution images that were otherwise impossible once upon a time.

    Here is the final test composite using a background I shot from a commercial assignment last year in India. I used a few techniques to turn the train lights "on," but the 3-light, edge lighting technique really works nicely for HDR compositing. Make sure to click on the image to see the larger version.


    Click image for larger version. 

Name:	Ajay-Composite-01c.jpg 
Views:	503 
Size:	264.9 KB 
ID:	46367


    Kind Regards,
    ~Ramon

  15. #15
    Join Date
    Jul 2008
    Location
    NH, USA
    Posts
    2,592
    Thanks
    57
    Thanked 30 Times in 26 Posts
    Feedback Score
    13 (100%)

    Default Re: ISO Bracketing with E-5: Too good to be true?

    Quote Originally Posted by findramon View Post
    @RAH- one shutter cycle, 3 images, all varied ISO. It is EXTREMELY useful because you get details across a larger spectrum with no "ghosting" because it was just 1 shutter cycle.
    Now I understand (plus from having read the other replies). I thought you meant you got 9 images. But getting 3 images, each with a different ISO from one shutter activation would indeed be very useful for HDR!

    Can my E-520 do this, or this some new feature they added to the E-5 (and later OM-D and Pens)?
    Rich
    Olympus E-M10; Panasonic GM5
    m4/3 lenses: Oly 75-300; Oly 14-42 f3.5-5.6 II R; Oly 17 f1.8; Oly 40-150 f4.0-5.6 R; Oly WCON-P01 adapter; Rokinon f7.5 fisheye; Sigma 19 f2.8; Pan 20 f1.7; Pan 12-35 f2.8; Pan 12-32

  16. #16
    Join Date
    Mar 2006
    Location
    Hokkaido, Japan
    Posts
    2,989
    Thanks
    0
    Thanked 0 Times in 0 Posts
    Feedback Score
    0

    Default Re: ISO Bracketing with E-5: Too good to be true?

    Quote Originally Posted by findramon View Post
    Hey all!

    Now, the "too good to be true" part. You actually get 3 exposures for every shutter cycle... how is this even possible? Is this considered true bracketing? Why don't you have to go through 3 shutter cycles like you would have to in AE bracketing?

    I essentially got 3 exposures at ISO 200, 100, and 400. The crazy thing is that because the trio was produced from just one shutter cycle, there is essentially no ghosting or alignment issues. I'm going to email Olympus and see what they have to say. This could be the world's best HDR portrait secret that no other camera system can provide today.
    I think you have answered your own question. If the camera were really doing three different sensor readings in a single exposure, it would be a game changer for many photographers, and Olympus would be shouting it from the rooftops.

    Like Steve said above, this has been discussed a bunch of times on DPR, and two or three years ago someone over there analyzed the files from an ISO bracket sequence out of an earlier camera (E-620?) and found that the camera was only doing one analog reading from a single exposure and then creating three identical ORFs, except the files had different metadata telling the RAW developer and the in-camera JPEG engine to develop each with a different brightness ("ISO"). The data contained in the three "bracketed" ORFs was identical, down to the last byte.

    At the time of the thread, we were discussing the E-620, E-30, and E-3, but unfortunately I don't see why the E-5 would work much differently. Like you, a lot of people would be interested in a true analog ISO bracketing feature. If it existed, most certainly somebody would have discovered it in the three years since the camera came out, and the forum chatter never would have stopped!

    Julie


  17. #17
    Join Date
    Jul 2008
    Location
    NH, USA
    Posts
    2,592
    Thanks
    57
    Thanked 30 Times in 26 Posts
    Feedback Score
    13 (100%)

    Default Re: ISO Bracketing with E-5: Too good to be true?

    Interesting follow-up, Julie. I agree that if this really were happening as we wish, it would be more of a touted feature. If a person can duplicate it in post-processing by just tweaking one raw image, it doesn't seem to have much value.
    Rich
    Olympus E-M10; Panasonic GM5
    m4/3 lenses: Oly 75-300; Oly 14-42 f3.5-5.6 II R; Oly 17 f1.8; Oly 40-150 f4.0-5.6 R; Oly WCON-P01 adapter; Rokinon f7.5 fisheye; Sigma 19 f2.8; Pan 20 f1.7; Pan 12-35 f2.8; Pan 12-32

  18. #18
    Join Date
    Mar 2007
    Posts
    1,652
    Thanks
    0
    Thanked 0 Times in 0 Posts
    Feedback Score
    0

    Default Re: ISO Bracketing with E-5: Too good to be true?

    Quote Originally Posted by windsprite View Post
    I think you have answered your own question. If the camera were really doing three different sensor readings in a single exposure, it would be a game changer for many photographers, and Olympus would be shouting it from the rooftops.

    Like Steve said above, this has been discussed a bunch of times on DPR, and two or three years ago someone over there analyzed the files from an ISO bracket sequence out of an earlier camera (E-620?) and found that the camera was only doing one analog reading from a single exposure and then creating three identical ORFs, except the files had different metadata telling the RAW developer and the in-camera JPEG engine to develop each with a different brightness ("ISO"). The data contained in the three "bracketed" ORFs was identical, down to the last byte.
    although done through the eyeglass of trolling by a known troll that was true,
    there would be no point using ORF RAWs
    but there is still the issue of using jpegs, which are pretty good anyway
    and the results exceed what you can do with a single jpeg, or a single RAW without going to some trouble to repeat what is the same thing

    you have to think about what you would use it for and why

    Quote Originally Posted by windsprite
    At the time of the thread, we were discussing the E-620, E-30, and E-3, but unfortunately I don't see why the E-5 would work much differently. Like you, a lot of people would be interested in a true analog ISO bracketing feature. If it existed, most certainly somebody would have discovered it in the three years since the camera came out, and the forum chatter never would have stopped!

    Julie
    'true analogue ISO bracketing feature' from the people that gave you 'there is no ISO'
    hows that gonna work?
    Riley

    Olympus User, Pro Photographer since 2003

  19. #19
    Join Date
    Mar 2006
    Location
    Hokkaido, Japan
    Posts
    2,989
    Thanks
    0
    Thanked 0 Times in 0 Posts
    Feedback Score
    0

    Default Re: ISO Bracketing with E-5: Too good to be true?

    Quote Originally Posted by RAH View Post
    Interesting follow-up, Julie. I agree that if this really were happening as we wish, it would be more of a touted feature. If a person can duplicate it in post-processing by just tweaking one raw image, it doesn't seem to have much value.
    Hi Rich. I agree, but I can see advantages for some JPEG shooters and for some RAW and HDR workflows.

    Julie


  20. #20
    Join Date
    Jul 2008
    Posts
    839
    Thanks
    0
    Thanked 1 Time in 1 Post
    Feedback Score
    3 (100%)

    Default Re: ISO Bracketing with E-5: Too good to be true?

    I may be wrong but it seems like it should be fairly easy to test to see if the metadata is the only thing being changed. Set up a shot so the base image has clipped highlights or shadows and then see if the +/- frames recover any detail. I would guess that if only the metadata is changed you won't see any more difference than if you take the same frame into an image editor and add or subtract the same amount of exposure. I have to agree that if this was a game changer there would have been a lot made of it long before now since the feature has been around for a few years.

  21. #21
    Join Date
    Jan 2007
    Location
    Comox Bristish Columbia
    Posts
    3,222
    Thanks
    0
    Thanked 18 Times in 12 Posts
    Feedback Score
    2 (100%)

    Default Re: ISO Bracketing with E-5: Too good to be true?

    I shot a few frames to see what happens. The thumbnails for each triplet look different, one dark, one light, one just right. When I take the orfs into LR4, the resulting images are identical, no difference from one to the next. The metadata says that they were shot at three different ISO values with the same shutter and f-stop as would be expected. Curious.
    __________________________________________________ _____________________________________ ​
    ​John Nicklin

    e3, ZD 11-22, ZD 12-60, Sigma 70-200, EC14, FL50R, 5DMkIII, 24-105 L, 70-200 f4 L, RB67, OM1, B&J 8x10, 5DIII, 24-105 f4 L, 70-200 f4 L, 100D.
    www.jnicklin.ca | flicker | SmugMug | Blog

  22. #22
    Join Date
    Mar 2007
    Posts
    1,652
    Thanks
    0
    Thanked 0 Times in 0 Posts
    Feedback Score
    0

    Default Re: ISO Bracketing with E-5: Too good to be true?

    thats the case
    thumbnails are derived from the jpegs not the ORFs
    Riley

    Olympus User, Pro Photographer since 2003

  23. #23
    Join Date
    Jan 2007
    Location
    Comox Bristish Columbia
    Posts
    3,222
    Thanks
    0
    Thanked 18 Times in 12 Posts
    Feedback Score
    2 (100%)

    Default Re: ISO Bracketing with E-5: Too good to be true?

    Quote Originally Posted by Riley View Post
    thats the case
    thumbnails are derived from the jpegs not the ORFs
    So the feature is only usable if one shoots jpegs or RAW + jpeg. That's something to keep in mind at least.
    __________________________________________________ _____________________________________ ​
    ​John Nicklin

    e3, ZD 11-22, ZD 12-60, Sigma 70-200, EC14, FL50R, 5DMkIII, 24-105 L, 70-200 f4 L, RB67, OM1, B&J 8x10, 5DIII, 24-105 f4 L, 70-200 f4 L, 100D.
    www.jnicklin.ca | flicker | SmugMug | Blog

  24. #24
    Join Date
    Sep 2011
    Location
    Naperville, IL 60563
    Posts
    66
    Thanks
    0
    Thanked 0 Times in 0 Posts
    Feedback Score
    0

    Default Re: ISO Bracketing with E-5: Too good to be true?

    Quote Originally Posted by windsprite View Post
    Hi Rich. I agree, but I can see advantages for some JPEG shooters and for some RAW and HDR workflows.

    Julie
    @ Julie, I think your contribution comes across as downplaying the significance of not only ISO Bracketing (which any modern Pro Canikon can do by the way), but ISO bracketing for HDR. HDR is not limited to landscapes and interiors, it is an innovative way for portraits. AE Bracketing by varying shutter speed is simply not possible because of flash sync speed limitations topping off at 1/250 second (maybe 1/320 if you are using a leaf shutter). Even with ghosting reduction, there will always be a slight amount of blur when you are interpolating and blending all the frames into a subject (i.e. nose and ears). ISO bracketing alleviates the constraint of needing to vary aperture or shutter speed, but it will be at the cost of noise. And, even with noise, camera raw processing has evolved to really accommodate higher ISO level ranges and their corresponding noise- in a non-destructive manner since it is in camera raw.

    If you looked at the example images that I uploaded, you would see that the exposures coming out of the raw files are inherently different with respect to detail. Furthermore, 32 bit processing didn't exist 3 years ago, much less 2 years ago. So, the 20 stops of exposure latitude you can work with using Photoshop CS6 or CC is very significant.

    Answering my own question rarely comes from posing on a thread such as this. Only a technical rep from Olympus could truly answer what is going on. Nevertheless, I'll definitely start using this feature moving forward, since even at ISO 100 & 200 with AE Bracketing, I'll be able to get a 2nd dimension of detail from an exposure string.

  25. #25
    Join Date
    Sep 2011
    Location
    Naperville, IL 60563
    Posts
    66
    Thanks
    0
    Thanked 0 Times in 0 Posts
    Feedback Score
    0

    Default Re: ISO Bracketing with E-5: Too good to be true?

    Quote Originally Posted by RAH View Post
    Interesting follow-up, Julie. I agree that if this really were happening as we wish, it would be more of a touted feature. If a person can duplicate it in post-processing by just tweaking one raw image, it doesn't seem to have much value.
    @RAH. Since when can you duplicate a RAW image file and change a value on top of that? I don't understand how that is possible. The goal for 32bit HDR processing requires only RAW files to be used to get the best details from each exposure.

Posting Permissions

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