Approve the Cookies
This website uses cookies to improve your user experience. By using this site, you agree to our use of cookies and our Privacy Policy.
OK
Index  •   • New posts  •   • RTAT  •   • 'Best of'  •   • Gallery  •   • Gear  •   • Reviews
Guest
New posts  •   • RTAT  •   • 'Best of'  •   • Gallery  •   • Gear  •   • Reviews
Register to forums    Log in

 
FORUMS Post Processing, Marketing & Presenting Photos RAW, Post Processing & Printing 
Thread started 17 Sep 2014 (Wednesday) 16:43
Search threadPrev/next
sponsored links
(this ad will go away when you log in as a registered member)

7Dii "Noise" - Imaging Resource raw files

 
TeamSpeed
01010100 01010011
Avatar
37,774 posts
Gallery: 111 photos
Best ofs: 2
Likes: 6410
Joined May 2002
Location: Midwest
     
Sep 23, 2014 11:51 |  #151

I agree, it is easier to convince people they are lacking in something, than to convince them they have enough to get the job done. Most tech companies utilize that psychology tendency very well, like Apple, etc. Just look at all the folks that suddenly are aware their images are not sharp, after reading some threads with complaints thrown all over, where beforehand, they thought their images were very good. ;)


Past Equipment | My Personal Gallery (external link) My Business Gallery (external link)
For Sale: Sigma USB Dock

  
  LOG IN TO REPLY
sponsored links
(this ad will go away when you log in as a registered member)
davesrose
Title Fairy still hasn't visited me!
4,240 posts
Likes: 727
Joined Apr 2007
Location: Atlanta, GA
     
Sep 23, 2014 12:03 |  #152

TeamSpeed wrote in post #17173072 (external link)
If we were to look at what DR would be "good enough" for nearly 90% of shots, what would that DR range have to be? Would 16-17 stops be that goal, or is that too low?

I shoot alot of low light high ISO shots, and clean 12800 is my personal goal of what I need. If I need 25600, as long as I could post process and end up with something that still has detail and looks decent, I won't need more. It would be interesting to see practical input on what the ultimate camera would be capable of, and why.

I would think that's like asking is there ever a limit on resolution. When blu-ray came out, many people said they didn't need it: that upscaled DVD was just fine. Then with that new 65" HD TV, well better HD sources just took off. Now we're seeing 4K technologies....it's never ending.

For taking photos of lightmaps, I wouldn't mind if cameras could get all the way to 32bit tonal DR (that's the top tonal intensity software uses for simulating light). An ideal recorded DR is one that can record the whole luminosity of the scene. Some photographers may do fine with 10 stop DR if they're doing strict indoor photography. Some may need over 20 if they want to photograph extreme outdoor window/ indoor type photos (without the use of bracketing). And those involved with 3D graphics wouldn't mind one that could go all the way to 32:)

I think you also raise another subject of high ISO. Do you have many indoor sports shots where you're having clipping? DR gets diminished the higher in ISO you go...even though you might not need 16 or 18 stops of light, it may also help with the lower DR of high ISOs.

I appologize to those who I might of offended with our arguments of DR. I have learned now how different DR of sensor is compared to tonal DR of a file (and thank you for bearing with me). It seems like there was a different POV, and I have been under the assumption that the recording DR worked the same way as the recorded DR. Being involved with graphics, when I use a 16bit file, I'm expecting 65536 levels of tone (simulated light). For simulating daylight, a 32bit file gives me 4.2 million (so that's the limit computer graphics folks have set). When I saw sensor DR as 12stops vs 14stops, I was equating the 12 with recording up to 4096 tonal DR vs 16384 (being involved with graphics and photography, I've been thinking only in terms of light;) ). I can see why it takes a lot to develop a sensor that can bump up saturation a full stop (because each time it's exponential).

That the Sony sensors have low noise floors, means they may be able to get closer to filling a full 14bit space. Wouldn't a brand start considering going to a higher bit processor when they start reaching a saturation limit that can fill well into the next stop of light? If it's that much harder to develop sensors that can have higher saturation limits, then it might take quite a bit of time for one to fully utilize 16bits. I suppose the extra bits could help with precision, or is there such a thing as a 15bit ADC? There doesn't seem to be one consistant measure of sensor DR, but I would assume camera manufacturers know their sensors the best and would consider going higher when the sensor demands it.


Canon 5D mk IV
EF 135mm 2.0L, EF 70-200mm 2.8L IS II, EF 24-70 2.8L II, EF 50mm 1.4, EF 100mm 2.8L Macro, EF 16-35mm 4L IS, Sigma 150-600mm C, 580EX, 600EX-RT, MeFoto Globetrotter tripod, grips, Black Rapid RS-7, CAMS plate and strap system, Lowepro Flipside 500 AW, and a few other things...
smugmug (external link)

  
  LOG IN TO REPLY
BigAl007
Cream of the Crop
Avatar
8,026 posts
Gallery: 543 photos
Best ofs: 1
Likes: 1633
Joined Dec 2010
Location: Repps cum Bastwick, Gt Yarmouth, Norfolk, UK.
     
Sep 23, 2014 12:09 |  #153

pwm2 wrote in post #17173216 (external link)
The debate about more DR was almost zero when the 5D2 was new. Anyone pointing at the performance got a "are you stupid" message from just about everyone. Now, it seems to be the reverse. A huge amount of people demands more. But I don't think that a large percent of POTN members have suddenly changed their photography style in a way that actuall makes 14 stops of DR as important as the debates seems to indicate.

Well if Sony/Nikon didn't have a sensor that could get a bit more out of the bottom end, coupled with a big move in the megapixel war, remember that? We would not be having this argument, at least not on a Canon centric bulletin board. The odd thing is that because of my personal situation I have only very recently moved from a 12 bit sensor to a 14 bit one. I really did not think that it would make a huge amount of difference.

Actually it is taking me a lot longer to get ontop of how to work with my new 50D than it did with going from my 300D to the 20D, that change was pretty uneventful, sensor performance was actually very close, at least as far as working out the optimum exposure levels, mated with my RAW processing. I'm still not really confident with just where to put the exposure in my attempts to ETTR as far to the right as possible. I think this could also be in part thanks to using a cheap lens a lot of the time. Most of the time 12 stops of DR is plenty, actually 10 or 11 would probably be OK much of the time.

It has only been a couple of times where conditions have been particularly difficult, the scene has contained about 16 stops of DR, possibly more. If you take a very thin layer of cloud, back light it with the sun, so that it is still bright enough that shadows are cast on the ground, and try to shoot an aircraft against that. You are not getting good detail in the underside of the aircraft. this is especially true if you expose for the shadows. I am using a cheap lens (Sigma 28-300mm) and the flare and blooming effects that I see on the sensor make it completely unusable.

This is a subject with the same level of illumination falling on it, and where I do not mind too much that I clip the sky detail. Changing the camera orientation by 90 degrees results in a three stop change in the background brightness. That moves the shot from being just about perfectly ETTR to one where the sensor seems overwhelmed with light.

It is only when in this situation that I really want all of the extra DR on the sensor that I can get. When you can no longer slide the "exposure window" to where you really want it, because the rest of the sensor can't handle the light levels.

Alan


My Flickr (external link)
My new Aviation images blog site (external link)

  
  LOG IN TO REPLY
pwm2
"Sorry for being a noob"
Avatar
8,626 posts
Likes: 3
Joined May 2007
Location: Sweden
     
Sep 23, 2014 12:28 |  #154

davesrose wrote in post #17173276 (external link)
I suppose the extra bits could help with precision, or is there such a thing as a 15bit ADC? There doesn't seem to be one consistant measure of sensor DR, but I would assume camera manufacturers know their sensors the best and would consider going higher when the sensor demands it.

It's fully possible to make 15-bit ADC. For most situations, there are no technical reasons to only make 8-bit, 10-bit, 12-bit, 14-bit, 16-bit, ... ADC. In some situations, the use of an even number of bits gives the advantage that the ADC would consume an even number of clock cycles for doing the conversion. But that isn't applicable to all types of ADC, and not all systems has an advantage of the conversion taking an even number of clock cycles. So it's mostly tradition that makes them have an even number of bits.

For 3D graphics, the logical step when finding that 16 bits was too little is to select 32 bits. Few processors can handle 20 bits or 24 bits faster than they handle 32-bit. So selecting 24-bit color channel resolution would just mean that 8 bits would have to be thrown away.


5DMk2 + BG-E6 | 40D + BG-E2N | 350D + BG-E3 + RC-1 | Elan 7E | Minolta Dimage 7U | (Gear thread)
10-22 | 16-35/2.8 L II | 20-35 | 24-105 L IS | 28-135 IS | 40/2.8 | 50/1.8 II | 70-200/2.8 L IS | 100/2.8 L IS | 100-400 L IS | Sigma 18-200DC
Speedlite 420EZ | Speedlite 580EX | EF 1.4x II | EF 2x II

  
  LOG IN TO REPLY
davesrose
Title Fairy still hasn't visited me!
4,240 posts
Likes: 727
Joined Apr 2007
Location: Atlanta, GA
     
Sep 23, 2014 12:32 |  #155

pwm2 wrote in post #17173317 (external link)
For 3D graphics, the logical step when finding that 16 bits was too little is to select 32 bits. Few processors can handle 20 bits or 24 bits faster than they handle 32-bit. So selecting 24-bit color channel resolution would just mean that 8 bits would have to be thrown away.

Yes, for rendering, I've found it true that 32bit can be quite a bit faster then 16bit. It just can take up a lot of hard drive space if you're trying to use an image sequence of 32bit TIFs (let alone EXR files that can have any number of channels) :)


Canon 5D mk IV
EF 135mm 2.0L, EF 70-200mm 2.8L IS II, EF 24-70 2.8L II, EF 50mm 1.4, EF 100mm 2.8L Macro, EF 16-35mm 4L IS, Sigma 150-600mm C, 580EX, 600EX-RT, MeFoto Globetrotter tripod, grips, Black Rapid RS-7, CAMS plate and strap system, Lowepro Flipside 500 AW, and a few other things...
smugmug (external link)

  
  LOG IN TO REPLY
davesrose
Title Fairy still hasn't visited me!
4,240 posts
Likes: 727
Joined Apr 2007
Location: Atlanta, GA
     
Sep 23, 2014 12:34 |  #156

BigAl007 wrote in post #17173289 (external link)
Actually it is taking me a lot longer to get ontop of how to work with my new 50D than it did with going from my 300D to the 20D, that change was pretty uneventful, sensor performance was actually very close, at least as far as working out the optimum exposure levels, mated with my RAW processing. I'm still not really confident with just where to put the exposure in my attempts to ETTR as far to the right as possible. I think this could also be in part thanks to using a cheap lens a lot of the time. Most of the time 12 stops of DR is plenty, actually 10 or 11 would probably be OK much of the time.

I had some P&S digital cameras before getting my first DSLR. It took me awhile to get used to the noise and clipping on my 5Dc when I got it new. Any landscape scenes, I would try out ETTR as much as I could tell from the histogram. After pulling levels from the RAW, I always found I had more leeway with the highlights and not any with the shadows. Maybe it was good to first learn on an older 12bit system, as when I got the 5DmkIII, I immediately noticed the change in DR. I have a lot more leeway in the highlights (and the shadows may be a bit cleaner due to slightly better noise handling/extra resolution as compared to the 5D). It was very easy for me to get used to the better DR in the 5DmkIII.

To try to determine where my clipping point is, I tend to try to see if I'm clipping into black areas and white areas. Then I know I'm reaching or past the recorded DR. I'll gauge if there's large areas with "blinkies" and if I need more contrast in there (vs blocked up shadows). Or if it's a scene without a person, I'll take out the tripod and bracket. You can get a histogram showing more of the right end of clipping by making sure you have a "Natural" profile with contrast set all the way down to -4. My own settings are different, but that one shows you the most levels on the right side.


Canon 5D mk IV
EF 135mm 2.0L, EF 70-200mm 2.8L IS II, EF 24-70 2.8L II, EF 50mm 1.4, EF 100mm 2.8L Macro, EF 16-35mm 4L IS, Sigma 150-600mm C, 580EX, 600EX-RT, MeFoto Globetrotter tripod, grips, Black Rapid RS-7, CAMS plate and strap system, Lowepro Flipside 500 AW, and a few other things...
smugmug (external link)

  
  LOG IN TO REPLY
pwm2
"Sorry for being a noob"
Avatar
8,626 posts
Likes: 3
Joined May 2007
Location: Sweden
     
Sep 23, 2014 12:38 |  #157

davesrose wrote in post #17173328 (external link)
Yes, for rendering, I've found it true that 32bit can be quite a bit faster then 16bit. It just can take up a lot of hard drive space if you're trying to use an image sequence of 32bit TIFs (let alone EXR files that can have any number of channels) :)

I have a home-written ray tracer that likes to consume quite a lot of disk. I use a variant of PNG files, where I store a large number of color channels - so I can post-process the RGB contribution from the different light sources. Way faster than tweaking the intensity or color of one light source and re-render again and again. But quite big files if there are lots of global light sources.


5DMk2 + BG-E6 | 40D + BG-E2N | 350D + BG-E3 + RC-1 | Elan 7E | Minolta Dimage 7U | (Gear thread)
10-22 | 16-35/2.8 L II | 20-35 | 24-105 L IS | 28-135 IS | 40/2.8 | 50/1.8 II | 70-200/2.8 L IS | 100/2.8 L IS | 100-400 L IS | Sigma 18-200DC
Speedlite 420EZ | Speedlite 580EX | EF 1.4x II | EF 2x II

  
  LOG IN TO REPLY
AJSJones
Goldmember
Avatar
2,647 posts
Gallery: 6 photos
Likes: 92
Joined Dec 2001
Location: California
     
Sep 23, 2014 13:34 |  #158

davesrose wrote in post #17173276 (external link)
I appologize to those who I might of offended with our arguments of DR. I have learned now how different DR of sensor is compared to tonal DR of a file (and thank you for bearing with me). It seems like there was a different POV, and I have been under the assumption that the recording DR worked the same way as the recorded DR. Being involved with graphics, when I use a 16bit file, I'm expecting 65536 levels of tone (simulated light). For simulating daylight, a 32bit file gives me 4.2 million (so that's the limit computer graphics folks have set). When I saw sensor DR as 12stops vs 14stops, I was equating the 12 with recording up to 4096 tonal DR vs 16384 (being involved with graphics and photography, I've been thinking only in terms of light;) ). I can see why it takes a lot to develop a sensor that can bump up saturation a full stop (because each time it's exponential).

Thanks for posting that - I think it also helps me understand more clearly what you meant by tonal range in your (noise-free unless you add it) computed world:D


My picture galleries (external link)

  
  LOG IN TO REPLY
AJSJones
Goldmember
Avatar
2,647 posts
Gallery: 6 photos
Likes: 92
Joined Dec 2001
Location: California
     
Sep 23, 2014 13:44 |  #159

BigAl007 wrote in post #17173289 (external link)
...
I'm still not really confident with just where to put the exposure in my attempts to ETTR as far to the right as possible. I think this could also be in part thanks to using a cheap lens a lot of the time. Most of the time 12 stops of DR is plenty, actually 10 or 11 would probably be OK much of the time.

...
It has only been a couple of times where conditions have been particularly difficult, the scene has contained about 16 stops of DR, possibly more.
This is a subject with the same level of illumination falling on it, and where I do not mind too much that I clip the sky detail. Changing the camera orientation by 90 degrees results in a three stop change in the background brightness. That moves the shot from being just about perfectly ETTR to one where the sensor seems overwhelmed with light.

...
It is only when in this situation that I really want all of the extra DR on the sensor that I can get. When you can no longer slide the "exposure window" to where you really want it, because the rest of the sensor can't handle the light levels.

Alan

In those situations where the scene DR exceeds your sensor DR and you want to capture as much as possible, it's important to ETTR as much as possible - any unused levels on the right of the histogram translate directly into lost DR, no matter where the meter might put 18% gray. Two things are critical for this: 1) deciding on which high luminance sensels are unnecessary (specular highlights etc) and 2) the ability to see when a sensel reaches saturation - that means before any processing like WB or gamma curve is applied. I am hoping that as DR increases (or even if it hadn't) that manufacturers would include the ability to assess raw luminance values to allow us to make sure they don't saturate. Right now it's a workaround (UniWB Magic Lantern etc), because the histograms are generated from the in-camera jpeg used for the display. It would be a shame if the manufacturers achieved greater DR but didn't make it easier to "use" it all. Something as simple as a CFn toggle between jpeg and raw for displaying blinkies would do it for me:D


My picture galleries (external link)

  
  LOG IN TO REPLY
davesrose
Title Fairy still hasn't visited me!
4,240 posts
Likes: 727
Joined Apr 2007
Location: Atlanta, GA
     
Sep 23, 2014 13:47 |  #160

AJSJones wrote in post #17173425 (external link)
Thanks for posting that - I think it also helps me understand more clearly what you meant by tonal range in your (noise-free unless you add it) computed world:D

Yes, I think we'll have some better interactions from here on out: we were seeing things from the opposite ends (quite literally in terms of DR) :D


Canon 5D mk IV
EF 135mm 2.0L, EF 70-200mm 2.8L IS II, EF 24-70 2.8L II, EF 50mm 1.4, EF 100mm 2.8L Macro, EF 16-35mm 4L IS, Sigma 150-600mm C, 580EX, 600EX-RT, MeFoto Globetrotter tripod, grips, Black Rapid RS-7, CAMS plate and strap system, Lowepro Flipside 500 AW, and a few other things...
smugmug (external link)

  
  LOG IN TO REPLY
davesrose
Title Fairy still hasn't visited me!
4,240 posts
Likes: 727
Joined Apr 2007
Location: Atlanta, GA
     
Sep 23, 2014 14:01 |  #161

AJSJones wrote in post #17173444 (external link)
Something as simple as a CFn toggle between jpeg and raw for displaying blinkies would do it for me:D

Or if instead of a jpeg profile that used any gamma curves, linearly convert to 8bpc. The Neutral -4 settings push down highlight info the most, so you're getting more spread in your topmost spots.

With such a small LCD screen it might be hard to see full stops of captured light (if Sony were to come out with a 16bit ACD, that's 32768 levels right at the extreme right of the histogram), I think it's pretty difficult to fully preview that clipping point. Not unless they have in internal RAW processor that can let you have level adjustment right in camera. Actually, just thought about it, the 5DmkIII does let you adjust the exposure settings through the RAW and it resaves the jpeg....seems like maybe one of the best tools for really judging your clipping point.


Canon 5D mk IV
EF 135mm 2.0L, EF 70-200mm 2.8L IS II, EF 24-70 2.8L II, EF 50mm 1.4, EF 100mm 2.8L Macro, EF 16-35mm 4L IS, Sigma 150-600mm C, 580EX, 600EX-RT, MeFoto Globetrotter tripod, grips, Black Rapid RS-7, CAMS plate and strap system, Lowepro Flipside 500 AW, and a few other things...
smugmug (external link)

  
  LOG IN TO REPLY
TeamSpeed
01010100 01010011
Avatar
37,774 posts
Gallery: 111 photos
Best ofs: 2
Likes: 6410
Joined May 2002
Location: Midwest
     
Sep 23, 2014 14:17 |  #162

davesrose wrote in post #17173478 (external link)
Or if instead of a jpeg profile that used any gamma curves, linearly convert to 8bpc. The Neutral -4 settings push down highlight info the most, so you're getting more spread in your topmost spots.

With such a small LCD screen it might be hard to see full stops of captured light (if Sony were to come out with a 16bit ACD, that's 32768 levels right at the extreme right of the histogram), I think it's pretty difficult to fully preview that clipping point. Not unless they have in internal RAW processor that can let you have level adjustment right in camera. Actually, just thought about it, the 5DmkIII does let you adjust the exposure settings through the RAW and it resaves the jpeg....seems like maybe one of the best tools for really judging your clipping point.

maybe its time for a better histogram option in camera? I know I can get both histograms up on the screen, but you cannot tell much. If we are able to edit raws in camera, I would like a few extra tools at my disposal. There are times I am sitting idly by and it would be nice to do some raw editing while sitting there. Obviously this would be a very niche request though, and won't see the light of day. :)


Past Equipment | My Personal Gallery (external link) My Business Gallery (external link)
For Sale: Sigma USB Dock

  
  LOG IN TO REPLY
davesrose
Title Fairy still hasn't visited me!
4,240 posts
Likes: 727
Joined Apr 2007
Location: Atlanta, GA
     
Sep 23, 2014 14:24 |  #163

TeamSpeed wrote in post #17173499 (external link)
There are times I am sitting idly by and it would be nice to do some raw editing while sitting there. Obviously this would be a very niche request though, and won't see the light of day. :)

Yeah, and then we have so many other options for really editing. If the 5D or 7D series had a way to wirelessly transfer RAWs to an ipad, that would be a very nice feature. Would save time if I was out and about to be able to process in Lightroom on the tablet. One can always dream at least;)


Canon 5D mk IV
EF 135mm 2.0L, EF 70-200mm 2.8L IS II, EF 24-70 2.8L II, EF 50mm 1.4, EF 100mm 2.8L Macro, EF 16-35mm 4L IS, Sigma 150-600mm C, 580EX, 600EX-RT, MeFoto Globetrotter tripod, grips, Black Rapid RS-7, CAMS plate and strap system, Lowepro Flipside 500 AW, and a few other things...
smugmug (external link)

  
  LOG IN TO REPLY
pwm2
"Sorry for being a noob"
Avatar
8,626 posts
Likes: 3
Joined May 2007
Location: Sweden
     
Sep 23, 2014 14:25 |  #164

AJSJones wrote in post #17173444 (external link)
Something as simple as a CFn toggle between jpeg and raw for displaying blinkies would do it for me:D

There shouldn't be a need for a toggle. The camera should be able to figure out that whatever white balance you might want to use, you still want to see zebra or similar based on actual sensor/raw clipping. And there is nothing that stops the manufacturers from storing two separate histograms in the Exif data - one based on the JPEG data, and one based on the unprocessed raw data.


5DMk2 + BG-E6 | 40D + BG-E2N | 350D + BG-E3 + RC-1 | Elan 7E | Minolta Dimage 7U | (Gear thread)
10-22 | 16-35/2.8 L II | 20-35 | 24-105 L IS | 28-135 IS | 40/2.8 | 50/1.8 II | 70-200/2.8 L IS | 100/2.8 L IS | 100-400 L IS | Sigma 18-200DC
Speedlite 420EZ | Speedlite 580EX | EF 1.4x II | EF 2x II

  
  LOG IN TO REPLY
TeamSpeed
01010100 01010011
Avatar
37,774 posts
Gallery: 111 photos
Best ofs: 2
Likes: 6410
Joined May 2002
Location: Midwest
     
Sep 23, 2014 14:31 |  #165

davesrose wrote in post #17173518 (external link)
Yeah, and then we have so many other options for really editing. If the 5D or 7D series had a way to wirelessly transfer RAWs to an ipad, that would be a very nice feature. Would save time if I was out and about to be able to process in Lightroom on the tablet. One can always dream at least;)

I can transfer the raws (I think, I will have to check) to my Android tablet from the 7D, but I don't have any raw editors installed.


Past Equipment | My Personal Gallery (external link) My Business Gallery (external link)
For Sale: Sigma USB Dock

  
  LOG IN TO REPLY
sponsored links
(this ad will go away when you log in as a registered member)

25,530 views & 0 likes for this thread
7Dii "Noise" - Imaging Resource raw files
FORUMS Post Processing, Marketing & Presenting Photos RAW, Post Processing & Printing 
AAA
x 1600
y 1600

Jump to forum...   •  Rules   •  Index   •  New posts   •  RTAT   •  'Best of'   •  Gallery   •  Gear   •  Reviews   •  Member list   •  Polls   •  Image rules   •  Search   •  Password reset

Not a member yet?
Register to forums
Registered members may log in to forums and access all the features: full search, image upload, follow forums, own gear list and ratings, likes, more forums, private messaging, thread follow, notifications, own gallery, all settings, view hosted photos, own reviews, see more and do more... and all is free. Don't be a stranger - register now and start posting!


COOKIES DISCLAIMER: This website uses cookies to improve your user experience. By using this site, you agree to our use of cookies and to our privacy policy.
Privacy policy and cookie usage info.


POWERED BY AMASS forum software 2.1forum software
version 2.1 /
code and design
by Pekka Saarinen ©
for photography-on-the.net

Latest registered member is RonWatson
922 guests, 271 members online
Simultaneous users record so far is 15144, that happened on Nov 22, 2018

Photography-on-the.net Digital Photography Forums is the website for photographers and all who love great photos, camera and post processing techniques, gear talk, discussion and sharing. Professionals, hobbyists, newbies and those who don't even own a camera -- all are welcome regardless of skill, favourite brand, gear, gender or age. Registering and usage is free.