Younger photographer reshoots test shots for IQ180, older photographers give him better advice.

[nikesteam]

Hello,

I think this post will serve as both an introduction and a serious question. I was looking for an IQ1 80 (Contax Mount) for years in the Gear section, finally found one, and now I'm doing test shoots with the family to get an understanding of the best settings.

I'm attaching two pictures to this. They are both shot with the exact same lens, with the exact same settings (A Pentax 90mm @ f2.8 @ ISO50) and were pushed in post about 2 stops. After doing a little bit of searching online, I've heard people say that ISO50 is the sweet spot, so I've been trying to see how far I can push that ISO before I see noise, and for the most part it's been performing pretty well.



This picture came out pretty clean. Again I didn't do anything to it other than push the exposure a couple of stops, and fiddled with the contrast a bit.



This picture, however, didn't do so well. Ignore the warm tones, thats from a filter.

I'm completely lost. I have no idea why the first picture was so much better than the second, when all of the settings were exactly the same. The only thing I can think of is if this back is a little finnicky with temperature like some tend to be. Should I let the back "warm up" before shooting it? I have a few CCD cameras that sometimes show their 'seams' where the sensors were stitched together if I don't let them sit a bit before shooting.

Another thing I thought of was that perhaps the battery power has something to do with it? Maybe my battery was low?

As I'm writing this, I hadn't considered this, but could it be the filter? No way, right?? I'll keep doing tests, but I just wanted to see if you guys had any wisdom to share with me, as to how these two images could be so drastically different under the same exact settings. It's a bit unnerving.

 

[nikesteam]

I just took some more shots just now, and... no noise. Similar performance to that first photo. Maybe it just needed to warm up?

 

[docholliday]

CCD + heat = bad. Cooler is better with much less noise. You'd never want to "warm up" a CCD.

Looks like the second image is way underexposed from the backlighting and your push of shadows is where the noise is. Your settings may have been the "same", but if you had auto metering without spot (center/averaging/matrix), your shot is biased for that window causing massive underexposure. Notice that the highlights have no noise in them. If you'd used spot metering with the spot on the subject, the window would be blown out, but there'd be no noise in the subject since you wouldn't have to push the hell out of it.

 

[dchew]

This picture came out pretty clean. Again I didn't do anything to it other than push the exposure a couple of stops, and fiddled with the contrast a bit.

Pushing the exposure a couple of stops gets you into an area that is pretty marginal for those CCD backs. As doc said, that plus the camera's metering being affected by the window is giving you very little information in those shadows.

Dave

 

[Hasslebad]

I agree with Docholliday. Regarding pushing the image in post, I have a P25 which is an older CCD back by a generation or so from yours. I can shoot at ISO100 with little noticeable noise. Of course when I’m in the studio and have all the light that I want I shoot at ISO50. I feel I can push the image further when I shoot at ISO50 in the studio. Not sure if it’s because of the sheer amount of light from my strobes or my ISO setting.

Regarding batteries, I’ve never seen more noise because of low batteries.

Those are beautiful portraits by the way.

 

[nikesteam]

CCD + heat = bad. Cooler is better with much less noise. You'd never want to "warm up" a CCD.

Looks like the second image is way underexposed from the backlighting and your push of shadows is where the noise is. Your settings may have been the "same", but if you had auto metering without spot (center/averaging/matrix), your shot is biased for that window causing massive underexposure. Notice that the highlights have no noise in them. If you'd used spot metering with the spot on the subject, the window would be blown out, but there'd be no noise in the subject since you wouldn't have to push the hell out of it.

Not always the case. Cold temperatures can definitely make it very obvious where the sensors were stitched together, or cause one section of the sensor to have a slightly different exposure from other sections. I see what you're saying and it is definitely the case 90% of the time, but this comes from owning 7 different CCD cameras / digital backs. I appreciate the contribution, but literally nothing changed outside of adding a filter for these two shots, and shooting on different days (but at the same time of day).

Noise is harder to see in highlights too, btw. It's white. If I pulled this all the way down, the noise pattern would most definitely be in that area as well.

Also, I'm not using the camera's metering system for anything. I shoot fully manually (including focus). Shutter speed and aperture were both done by me... not that it would matter because as I said the ISO was set to 50 in both shots and thus should have the same general level of dynamic range, noise, and texture. Both images were purposefully shot underexposed by around 2 stops for each picture to test the latitude of the sensor in post.

Do you guys own an IQ1 80 or similar back? Hasselblad, I also own the P25, but the two generations are vastly different. Totally different manufacturers even (P25 is Kodak, IQ sereis is Dalsa). Mostly looking for advice from people who actually own this back or a similar one and has experience with it. I think I have the basics of digital photography down, so I'm not looking for the obvious advice on how to expose a shot. Again, trying to figure out why the noise is so much more apparent even though the two shots were taken with identical settings. Looking more to get an understanding of the quirks of this particular model. Thank you though.

 

[nikesteam]

Those are beautiful portraits by the way.

Thank you! I am very happy with how this lens renders on the 180!

 

[docholliday]

Not always the case. Cold temperatures can definitely make it very obvious where the sensors were stitched together, or cause one section of the sensor to have a slightly different exposure from other sections. I see what you're saying and it is definitely the case 90% of the time, but this comes from owning 7 different CCD cameras / digital backs. I appreciate the contribution, but literally nothing changed outside of adding a filter for these two shots, and shooting on different days (but at the same time of day).

Noise is harder to see in highlights too, btw. It's white. If I pulled this all the way down, the noise pattern would most definitely be in that area as well.

Also, I'm not using the camera's metering system for anything. I shoot fully manually (including focus). Shutter speed and aperture were both done by me... not that it would matter because as I said the ISO was set to 50 in both shots and thus should have the same general level of dynamic range, noise, and texture. Both images were purposefully shot underexposed by around 2 stops for each picture to test the latitude of the sensor in post.

Do you guys own an IQ1 80 or similar back? Hasselblad, I also own the P25, but the two generations are vastly different. Totally different manufacturers even (P25 is Kodak, IQ sereis is Dalsa). Mostly looking for advice from people who actually own this back or a similar one and has experience with it. I think I have the basics of digital photography down, so I'm not looking for the obvious advice on how to expose a shot. Again, trying to figure out why the noise is so much more apparent even though the two shots were taken with identical settings. Looking more to get an understanding of the quirks of this particular model. Thank you though.

Well, since you never posted the EXIF of the image nor processing data, we can only guestimate on what we see. "ISO 50 - 2 stops" isn't very functional in determining what has gone wrong with an image.

Yes, I own an IQ180 as well as the IQ250, IQ3100, and older backs as well. I shoot extensively with all the backs in studio and on location. I am a production shooter, so I am well aware of the limitations and specialties of my hardware. I've shot in the field backs that have been cold soaked for hours, both CCD and CMOS, and have never experienced your issue of seeing the sensor wafer seam. But, it seems that you know all there is about the subject owning 7 different cameras / backs so I guess I can't help ya more than you already know...

 

[buildbot]

I have much less experience, but I wonder if the second image is more noisy because the back started saving the image before the shutter was fully closed? It looks almost like it’s readout noise?

 

[nikesteam]

Well, since you never posted the EXIF of the image nor processing data, we can only guestimate on what we see. "ISO 50 - 2 stops" isn't very functional in determining what has gone wrong with an image.

Yes, I own an IQ180 as well as the IQ250, IQ3100, and older backs as well. I shoot extensively with all the backs in studio and on location. I am a production shooter, so I am well aware of the limitations and specialties of my hardware. I've shot in the field backs that have been cold soaked for hours, both CCD and CMOS, and have never experienced your issue of seeing the sensor wafer seam. But, it seems that you know all there is about the subject owning 7 different cameras / backs so I guess I can't help ya more than you already know...

No problem, I didn't realize that the shutter speed and aperture had an effect on the noise levels of digital sensors (/s). Didn't think information like that was needed, since I explained that I was shooting underexposed, to raise the exposure by 2 stops, purely to test the latitude/noise levels in post. Perhaps I didn't explain it well enough the first time though, but I meant no offense, so could you please explain to me how that info would apply to this situation, where the settings are EXACTLY the same, and yet the photo is completely different in terms of noise?

I wasn't trying to boast about having a few CCD cameras. Just trying to learn from people who have had this back longer than myself. Also, I don't know what other CCD cameras you have, but the sensor splitting issue is apparent on some of my stills cameras (verrrrryyy rarely my RD-1), but more so on video cameras (I have an Ikonoskop, and a D16, both share similar S16-sized CCD Kodak chips). When shooting cold it is very clear that some pieces are different colors than the others. Very much a pain in the butt when you need to start shooting immediately in winter.

 

[nikesteam]

I have much less experience, but I wonder if the second image is more noisy because the back started saving the image before the shutter was fully closed? It looks almost like it’s readout noise?

Hmmmm... That is interesting, and it makes sense. Maybe it is random then? Have you had any similar issues?

 

[edmundronald]

A filter will clip a channel. 2 stops under and a filter, we're now 3 under, color noise here we come

 

[drevil]

alan, you mentioned a pentax lens, can you quickly elaborate the full setup?

 

[docholliday]

No problem, I didn't realize that the shutter speed and aperture had an effect on the noise levels of digital sensors (/s). Didn't think information like that was needed, since I explained that I was shooting underexposed, to raise the exposure by 2 stops, purely to test the latitude/noise levels in post. Perhaps I didn't explain it well enough the first time though, but I meant no offense, so could you please explain to me how that info would apply to this situation, where the settings are EXACTLY the same, and yet the photo is completely different in terms of noise?

I wasn't trying to boast about having a few CCD cameras. Just trying to learn from people who have had this back longer than myself. Also, I don't know what other CCD cameras you have, but the sensor splitting issue is apparent on some of my stills cameras (verrrrryyy rarely my RD-1), but more so on video cameras (I have an Ikonoskop, and a D16, both share similar S16-sized CCD Kodak chips). When shooting cold it is very clear that some pieces are different colors than the others. Very much a pain in the butt when you need to start shooting immediately in winter.

The settings *to you* may be the exact same. Obviously, something wasn't. Also, since the images aren't the same, post processing can also affect the data. So, to eliminate any other things that could've happened, uploading the raw files to somewhere like a dropbox or onedrive would be useful so that the capture data could be evaluated in detail. It may be something as simple as a power dip causing the capture to cutoff early. The only way to be able to truly establish quantifiable data is to have the raw materials so that testing and evaluation can be done on them.

You still haven't said what your actual exposure settings, hardware configuration, and filtration pack was.

You mention Contax mount, so I'm assuming you're on a C645? I had 3 of them and they all pissed me off badly. Prism meters failing, lens apertures not stopping down completely, and one that suddenly died right before a production shoot due to a power regulator. Easy fix in my lab, but not when in the field. Nice when it worked, produced great Acros files. The only camera I hated more was the Rollei 6008 with it's battery-draining, sliding-darkslide-jamming crap. May be fine for hobbyists but that 6008 was not a production grade tool unless you needed a sandbag/weight for your tripod or lightstand.

 

[nikesteam]

You guys have all got me thinking a lot more now. I will definitely run some more tests tomorrow to see what could be causing it. I will try out a bunch of different things and let you know if I can replicate the issue.

@drevil Good to see you on here! Here was the setup:

Body: Contax 645
DB: IQ1 80
ISO: 50
Lens: Pentax 6x7 90mm F2.8
Shutter speed: 1/60th
Aperture: F2.8
Filter (for 2nd photo): Tiffen Warm Black Pro Mist 1/8th

Didn't use a light meter, but if I had to guess, I'd say that it would've given me an F1.4 reading with the 50 ISO at shutter priority for both days.

@docholliday

When I run this again tomorrow, I will add a link to the RAW files (assuming I can replicate the issue again). I see what you're saying about the metadata. Sorry for the confusion. Haha @ the sandbag comment. Funny, I was tempted to buy a 3008 that was being sold at a local shop the other day, but the shop owner even told me not to buy it! I'm terrified of having an issue with the C645. It's my first one, and I know parts are dwindling. I'm in Japan though, and there are a LOT of Contax lovers, and still plenty of repair shops.

Quick follow-up question:

Let's say that this is something that the filter caused as @edmundronald has suggested. What noise reduction software do you suggest. I use C1, because I really like how it interprets RAW files, but I haven't had any luck with the noise reduction. LR is better, but kind of kills the crispness of the photo. If this is something I will just have to live with, what's the best way to remedy it in post?

 

[chrismuc]

I also have a Contax 645 with IQ180. From my experience I can say: The dynamic range of the CCD sensor is not on latest Sony level but it is good, if ISO is at 35 or maximum 50, not above, exposure time absolute limit is 4-8 seconds, and exposure ETTR, avoid under exposure.

 

[chrismuc]

The sweet spot is certainly ISO 35 and the back should not be warm.

 

[docholliday]

You guys have all got me thinking a lot more now. I will definitely run some more tests tomorrow to see what could be causing it. I will try out a bunch of different things and let you know if I can replicate the issue.

@drevil Good to see you on here! Here was the setup:

Body: Contax 645
DB: IQ1 80
ISO: 50
Lens: Pentax 6x7 90mm F2.8
Shutter speed: 1/60th
Aperture: F2.8
Filter (for 2nd photo): Tiffen Warm Black Pro Mist 1/8th

Didn't use a light meter, but if I had to guess, I'd say that it would've given me an F1.4 reading with the 50 ISO at shutter priority for both days.

@docholliday

When I run this again tomorrow, I will add a link to the RAW files (assuming I can replicate the issue again). I see what you're saying about the metadata. Sorry for the confusion. Haha @ the sandbag comment. Funny, I was tempted to buy a 3008 that was being sold at a local shop the other day, but the shop owner even told me not to buy it! I'm terrified of having an issue with the C645. It's my first one, and I know parts are dwindling. I'm in Japan though, and there are a LOT of Contax lovers, and still plenty of repair shops.

Quick follow-up question:

Let's say that this is something that the filter caused as @edmundronald has suggested. What noise reduction software do you suggest. I use C1, because I really like how it interprets RAW files, but I haven't had any luck with the noise reduction. LR is better, but kind of kills the crispness of the photo. If this is something I will just have to live with, what's the best way to remedy it in post?

Oh, boy...where to start. Since you didn't take a meter reading, it could be that the second image was massively underexposed (beyond your settings). It could also be that the filter is causing a form of flare, reducing contrast in the shadows, and resulting in noise when your pushed them in post. I've also seen faint streaking on a sensor that was cleaned (not visible to the naked eye, but under a microscope at an oblique angle). The streaks impressed a "pattern" on the image that were flaring when light hit them and they weren't noticable on most images, until pushed and then it showed in the shadows like an overlay of noise (think noisy, random screen door pattern).

You also mention shutter priority? So were you manually setting the camera settings, or were you using shutter priority? If you were not in full manual, the camera would've compensated for the brightness from the window and underexposed more than you expected. If you were in full manual, your exposure should've been set much different than the first shot with the extra light in the room or your underexposure may actually be -3 or -4.

Tiffen filters are horrid. They flare like there is no other and their "mist" filters tend to reduce contrast and induce a touch of flare (it's part of the effect). That flare could be exasperated by the depth (or lack) of hood, angle to specular sources, or in this case, bright white blob in frame. Great for video and when exposure is compensated for (which the camera usually would in aperture or shutter priority). Depending on the filter strength and flaring, you may be causing up to a 50% contrast loss and/or further 1 stop underexposure. Flare of any kind can cause loss of contrast that makes pulling shadows much noiser than normal.

When I shoot any of my CCD backs, I tend to bias the exposure for +1/3 or even +2/3 to be sure that there is enough data for the shadows. I can always pull the highlights back in post. When there is no data, the processor can only compensate by extrapolating mathematically from surrounding pixels and that usually results in noise. I also spot meter, either by hand or using the in-camera meter, for the shadows that require clean detail - just like a Zone System user would. Under studio light, it's not a problem, but in the field, CCD needs much more careful metering like one was exposing transparency film.

When you run the test again, be sure to meter properly (before you determine your 2 stop underexposure). Record the actual meter reading (in the shadow area where noise would form), then the exposure settings. i.e. meter = ISO50, 1/15s @ f/2.8, camera = ISO50, 1/60s @ f/2.8, metered @ below subject left eye shadow area. To truly determine the cause (if it occurs again), you'll need to eliminate as many variables as possible and record the conditions so that the data can be analyzed with knowns.

If you did actually underexpose more than expected, there's really no way to "clean up" what's not there. Noise reduction is typically smearing the noise pixel to the surrounding, using the surrounding to "create" what the algorithm thinks is there. A large patch of bad/missing data can't just be created out of thin air properly.

 

[darr]

A dark diffusion filter on top of an exposure made with an incorrect meter reading, or deliberately under exposing will not produce good results. Then pushing the same exposure in post will show its deficiencies (grainy results). That exposure equation is marked for dissatisfaction. I have a CCD back for backup and would never do any of this. Getting a proper exposure while being creative with filters is essential. Overexposure is better to recover from if lazy with exposure readings. Yeah, I read you do not need to hear the basics, but sometimes we need to remember why they are basics.

 

[nikesteam]

@darr

A dark diffusion filter on top of an exposure made with an incorrect meter reading, or deliberately under exposing will not produce good results.

What am I not saying correctly to have everyone misinterpret my post? Incorrect meter reading? There was no meter reading at all. Did you read anything before commenting? Did you see the pictures? Do you understand that BOTH pictures were taken with the exact same settings (other than the filter)?

I purposefully underexposed the photo to test the sensor's latitude. How do you test noise floors with your cameras? Do you bother? When I get new equipment, my first instinct is to test its limits so that I know what it can and cannot handle.

Let me put this in caps so that no one else can misunderstand what is happening here, because the (willful?) misinterpretation of this post has me questioning my sanity.

THIS IS A TEST/EXPERIMENT TO DETERMINE THE BEST SETTINGS FOR THE LOWEST NOISE POSSIBLE ON AN IQ1 80, and get properly acquainted with how much the files can handle in post, and determine why the second photo's noise level was so much higher than the the first. If you don't have experience with this back, or similar ones. I have a good understanding of photography basics, please don't treat me like I'm stupid. Do you personally own this digital back? Have you ever used one?

I appreciate the response, but I don't think you read the previous comments. This isn't a professional shoot, or anything remotely close to that. I'm trying to see how far I can push the files, and also trying to troubleshoot a strange noise issue. BTW, are you still trying to sell all three Merrills at once? I still want to buy the DP3.