wetransfer.com offers free 2GB file sharing - useful for files larger than 100MB which many other services don't support
http://www.graham-mitchell.com Graham Mitchell
Yousendit works fine
Try something for me on your daylight shots drop down to 4900 Kelvin . What temp is it coming in
Not received any raw yet, but thought of one other thing to try --- go into prefs and turn off open cl. I would begin a new session and re-import as well to clear all previous cache and settings for each image.
As a newbie to Capture One and the Phase One IQ180, I've been following this thread with great interest. I'm curious why you suggest turning off OpenCL. Is that something you suggest in general, or is it a suggestion in this case to remove one of the variables in the color management chain? Inquiring minds want to know.
Joe Colson Photography
It is a system -- actually video card interacting with the system -- dependent option, and on some machines it improves performance measurably with no ill effects; on others it can create all sorts of weird anomalies and/or actually slow performance. So that's why I recommend turning it off here and trying it.
I am convinced that Geoff's images are not getting properly tagged with the profile.
Sorry about this, but I hadn't realized this had all gone to a new page (2), and it looked like nobody had responded to my last posting. I did warn you that this was my first experience on this website!
Anyway, back the topic in hand, Jack & Guy will you please forward your emails addresses to [email protected], and I will send over one of the RAW files to you.
I got your link and quickly processed the image on my laptop. Bottom line is I think your original is about 1/2 to 2/3rds stop under-exposed, and that's all that's causing the red over-saturation you are seeing. Here is your file processed with outdoor DL profile and as a 16-bit tiff, WB'd off the gray pavement under the bricks which was very close to my normal WB, and then I saturation left at zero which I would normally bump about 3 points. Even with a 1/2 stop bump in exposure, I needed to add about 15 points to the shadow slider, so it could be as much as 2/3rds under -- histo remained good at +2/3rds, FWIW. I then downsized and converted to a jpeg in CS5 for this result:
Here it is with a +1 stop exposure bump, +4 sat, and shadow slider back to 0, all else the same. Histo looked fine even at +1 -- image still looks okay to me though a bit hot, but again I am on my laptop:
I just thought I would enlighten you about the final results of the saturation issue.
Jack Flesher has been very helpful indeed. I sent a few RAW files directly to him to examine on his computer, and the result is that there is two sides to the problem. One is that my camera was underexposing slightly, which of course increases colour depth, but there also is a saturation issue, and more so with the orange-red tones. Both these things we have agreed upon, but what doesn't make sense is why we have had 3 different backs have the same issue, and Jack has not reported any such issue with the backs he has used.
My dealer has sent some sample images to Phase One for assessment, but I haven't heard any results yet. My solution (thanks to Jack) has been to modify in Capture One's colour editor a camera profile to fix the problem, which I now use. So yes the issue has been got around, but it would be nice to know the real cause.
Again, thanks Jack for your assistance!
We should add that the one constant between the three backs you tried were you used the same body and lens. My initial suspicion is that there are three separate issues possibly contributing:
1) Slight body under-exposure, on average about 1/2 stop;
2) Possible lens-cast and/or lens induced contrast issue;
3) It is winter and overcast/raining in Geoff's half of the world right now, and wet and overcast both can increase relative saturation.
Jack, I still think there is something for Phase to investigate in the proportional red rise with underexposure. Certainly, all sats rise under those conditions but red rises more than green or blue. I see similar effects with low sun angles of course (different in Los Altos and Melbourne?). Not much to be done as flux dependent balance of the primaries is a fundamental sensor property. However, a red-reduce C1 style might be handy.
I don't have a quantum efficiency chart of the Dalsa chip but I would suspect its red response is extended relative to green/blue and to our eyes. So when all primary fluxes are low or when red light is prevalent (low sun eg), the camera sees more of it than we do visually. Wonder who has the chip response data?
Forgive me for off topic but
does your IQ comes with the plastic thingie that cover the cable release etc (below IQ180 sign). Mine does not and my dealer told me that Phase One sent the back without it.
For some reason all IQ sent to my dealer does not have the rubber gasket.
what in the world!!!!
Like Terry says, you'll lose it in moments. Whenever I peel the top half off to stick the tech cam shutter sync thingie in, it is a bugger to get back in afterwards. I anticipate getting through a few of these.
Geoff i got your file but what lens are you using here . Its marked 75mm. Phase does not have a 75mm lens unless it is the zoom
Okay Geoff sent me a file but first you need to understand where the light is coming from which is directly overhead. Here is the problem the brick in the front is really in the shadows and you really can only go by the top of the bricks. Plus the color chart is on the shadow side. Plus you are .5 underexposed right out of the gate. So I boosted the exposure and only go by the top bricks as far as color because the bottom bricks are underexposed and in the next frame I boosted the shadow side.
Now going in and boosting the shadow side the bottom bricks are still under but not as bad. Honestly this is not a great file . Really the sun should be hitting the whole scene directly not from the top plus it is overcast which causes more saturation. I'm just not convinced given this file.
I see where you are coming from, but after shooting in red rock country all weekend, I'm seeing some red saturation issues, and I know it's not underexposure or shadows. Still seeing if I've got something in my workflow that might explain it. but if I want saturation to all colors, I'm having to pull the reds back down sometimes.
I'll work on in more and get back with you.
BTW, I understand about it being underexposed maybe contributing to this, but isn't this "new"? I don't recall issues like this with the p65+, and I know reichmann and dubovoy both mentioned some red issues.
Sorry about the delay in responding, but I haven't had much time lately.
I can send to you an identically positioned Canon image of the same subject matter, and I can assure you the Canon doesn't have the same reds saturation problem, also being opened in Capture One using a generic Canon profile.
Anyway, I know there is a problem, and I have taken steps to solve it, by utilizing a modified camera profile. So as far as I am concerned I have done my best to alert others to it, and each person can make their own judgement. If you are happy with what you are seeing for your own images, then that's great, but I wasn't.
It is interesting that it appears that Wayne has also identified the same thing.
One other thing that occurred to me was that maybe others aren't viewing on a full gamut monitor. I use a top of the range Eizo, which pretty well covers the full Adobe RGB gamut. So it is possible that the issue isn't so obvious for others who have monitors with a more restricted colour gamut. It would explain it.
I did a quick test via IQ LCD, did some bracket with red clothes, and I found the color in the LCD is darker/more saturated than the actual clothes even when I overexposed it.
I will check more in my PC later. This may be because of inaccurate LCD coloring.
During my last studio shoot, I notice that overexpose image gives washout color in the skin (LCD), while tethered the color looks nice in my monitor. I suspect the LCD's color is inaccurate.
I'm on a Nec wide gamut monitor as well as Jack is
Does anyone with this back have the facilities to do a proper color check? The testing here is too subjective to pin point the problem. At least a controlled environment would help.
Just a thought.....
You guys who are having this red problem, what firmware version are you running?
The reason I ask is that the first time I shot with an IQ180 was when I did the demo for the global dealer launch here in Dubai. Those backs we used were pre-prod prototypes and weren't fully profiled. They had a terrible over saturation problem in the mid to three-quarter tone part of the red channel. The shadows were also slightly green.
My production back does not have this problem at all. We have had a firmware update since we received our backs, so I am curious to know if you are running older firmware. At the very least it would be good to remove that as one of the variables in this discussion. I am running version 1.19 from May26 2011.
For the record, my IQ180 is not showing any sign of this red over-saturation. Nor is my Aptus 12.
The accuracy in the Khaki/green area is also better than my P65+.
I am attaching two images taken a short time ago, and both within a couple of minutes of each other. Unfortunately it is still raining here, which affects the real colour judgement of things, but in this instance it is more a comparison of colourv rendering between a Canon image, and the Phase One image. Unless people have difficulty distinguishing colours, then I think this example will clearly show what I have been referring to. The images were both processed in Capture One and the only change has been the standard sharpening. The Phase One rendering of the bricks is certainly over-saturated. Even the Canon is more saturated than the actual bricks are.
I will leave it for you to make your own judgement.
Try again! I Forgot to covert to SRGB, sorry.
and the Canon image
One more time!
I don't know what is happening here, I can't get both images to transfer. It all looks correct with both images prior to posting.
Geoff, the P1 file you are trying to attach is a tiff, and probably too large.
Simple exposure correction algorithms do not model this effect so the corrected image keeps the inaccurate red cast. Perhaps this is normal and we are expected to establish our own camera profiles for different lighting conditions? That's fine if the effect is subtle. A few profiles will cover most conditions. If the effect is substantial, however, modeling becomes a more effective way of handling it and that is up to Phase.