Ad
Ad
Ad
Pages: « 1 2 [3] 4 »   Bottom of Page
Print
Author Topic: exposing to the right  (Read 33559 times)
bjanes
Sr. Member
****
Offline Offline

Posts: 2756



« Reply #40 on: April 02, 2008, 07:35:42 PM »
ReplyReply

Quote from: MarkDS,Apr 2 2008, 02:47 PM
I think it would be very helpful to get Jeff's and Andrew's technical input here; and to this objective it could be more useful to encourage them to participate rather than to presume upon their motives vis a vis yourself, the basis of which certainly isn't apparent to me.
[a href=\"index.php?act=findpost&pid=186523\"][{POST_SNAPBACK}][/a]
[/quote

Probably so, but I have had some unpleasant experiences with these characters and have found that they are not open to ideas that differ from their preconceived notions.
 

Quote from: MarkDS,Apr 2 2008, 02:47 PM
So getting back to the substance of the issue, I agree, the Brightness and Contrast settings largely affect the midtones - though both can impact on highlights where there are tones near clipping before these adjustments are made. It is normal to expect that an Exposure change will have a greater impact on highlight clipping. All that said, in my just previous post, based on a contribution from Pano, I suggested an empirical procedure for being able to manage one's exposures between the histogram view, the raw file and ACR. Feedback on these suggestions would be most interesting.
[a href=\"index.php?act=findpost&pid=186523\"][{POST_SNAPBACK}][/a]
[/quote

If you look back to my original post in this thread that is what I did. I found that the D3 luminance histogram is within 0.3 EV of the clipping in the raw channel. That is about where I want it. It leaves just a bit of headroom.

Here is the raw histogram:


and the camera histogram:


and with clipping gone with another 0.3 EV exposure reduction:
« Last Edit: April 03, 2008, 05:40:57 AM by bjanes » Logged
Panopeeper
Sr. Member
****
Offline Offline

Posts: 1805


« Reply #41 on: April 02, 2008, 08:14:42 PM »
ReplyReply

Quote
Therefore the solution to this issue would seem to be to find a JPEG setting for the camera (by making test exposures ostensibly using ETTR) which comes as close as possible to a no-clipping histogram both on the camera LCD and in RawAnalyse. Then open that image in ACR and set ACR's Basic Tab luminosity controls to reference values which come as close as possible to replicating the "correct" histogram in ACR, and use those as one's defaults for that particular camera.
I don't really understand, what problem should be solved this way. The solution for the problem at hand is simple: the auto-adjustment of ACR has to be reserved by the "Exposure" slider, so that one starts at "real zero". From there further adjustments can be done the usual way.

However, this advice does not help those users, who don't know about the auto-adjustment.
Logged

Gabor
Panopeeper
Sr. Member
****
Offline Offline

Posts: 1805


« Reply #42 on: April 02, 2008, 08:17:36 PM »
ReplyReply

Quote
I think it's one of the raw converter's most important features, to compensate for differences among cameras including deficiencies
That's ok. However, we were not talking about a deficiency (except when making perfect shots is a deficiency of the camera).

As to the rest: I don't feel like playing in kindergarten today.
Logged

Gabor
Mark D Segal
Contributor
Sr. Member
*
Offline Offline

Posts: 6819


WWW
« Reply #43 on: April 02, 2008, 09:02:34 PM »
ReplyReply

Quote
I don't really understand, what problem should be solved this way. The solution for the problem at hand is simple: the auto-adjustment of ACR has to be reserved by the "Exposure" slider, so that one starts at "real zero". From there further adjustments can be done the usual way.

However, this advice does not help those users, who don't know about the auto-adjustment.
[a href=\"index.php?act=findpost&pid=186586\"][{POST_SNAPBACK}][/a]

I explained what the problem is about in the previous paragraph and it has been running (or meandering) through the thread, so kindly just remain in kindergarten a bit longer to sharpen your understanding: inconsistency between the histogram reading from the in-camera JPEG and the structure of the raw file, and seeming inconsistency between the latter and the ACR histogram resulting from "zero and linear" default settings. That's how the issue has evolved from the OP. Jonathan and Eric recommended two pieces of software for helping to understand the raw file absent the issues of the JPEG and ACR rendition. Therefore a procedure developed around the use of that software can help one to determine a baseline of exposure practice and default ACR settings that will accomplish ETTR without clipping, as reliably and predictably as possible with present tools and techniques. Clear now?
Logged

Mark D Segal (formerly MarkDS)
Author: "Scanning Workflows with SilverFast 8....." http://www.luminous-landscape.com/reviews/film/scanning_workflows_with_silverfast_8.shtml
bjanes
Sr. Member
****
Offline Offline

Posts: 2756



« Reply #44 on: April 03, 2008, 07:06:48 AM »
ReplyReply

Quote
Bill, as we discussed briefly on the Adobe User-to-User forums, the issue here is the baseline exposure differences between various cameras and across vendors. There is a fundamental tradeoff between (a) honoring the original distribution of raw values (between black point and white point) and ( b ) having a given exposure (e.g., f/8, 1/10th sec, ISO 400) be rendered the same way across different cameras. In order to have common controls such as Exposure compensation behave the same way across the different cameras, Camera Raw applies a baseline exposure compensation that varies from model to model in order to get them all to behave similarly when the exposure compensation control is set to its default value of zero.
[{POST_SNAPBACK}][/a]


I think the main problem is that the camera vendors are not using uniform calibration  procedures for the ISO of the sensor and the rest of the camera system and thus the results are not consistent between different cameras of the same or different makes.

The main variables are:

1) Calibration of the light meter itself. According to Phil's tests on DPReview, Nikon has always been spot on for meter calibration and the recent Canon's are also. The calibration is done by comparison with the reading of a calibrated hand held meter. A source of confusion is that the ISO standard for light meters is for the equivalent of around 12% gray, not the commonly assumed value of 18%. See [a href=\"http://www.bythom.com/graycards.htm]Thom Hogan's Post[/url] for details. The upshot is that if you determine exposure by a reading from a Kodak 18% gray card, you have to add 0.5 EV to the exposure if you want the result to come out as middle gray in the photograph.

2) Variations in the shutter speed and aperture control. Electronic shutters in modern digital cameras are quite accurate and repeatable (precise). The aperture control is mechanical and subject to more variability, but the mechanism usually works well unless something is sticking.

3) Absorption of light by multiple lens elements. With complex lens designs, this can be a significant factor. It is taken into account by TTL metering, which will then give more exposure than indicted by the hand held meter.

4) Calibration of the ISO sensitivity of the sensor itself. This is where I think most of the variations arise. Norman Koren explains the ISO 12232:1998 standard for determination of the ISO of digital cameras. The standard defines two methods: saturation based and noise based. the Saturation based standard is usually used. When exposed according to a standard illumination (determined by a meter reading or derived from luminance values directly), an 18% gray card should result in the sensor being 18/106 saturated resulting in a raw pixel value of 18/106 of full scale since the sensor is linear. The denominator of 106 rather than 100 is used to allow some headroom for highlights. If a gamma curve of 2.2 without any other adjustments is applied, this results in a pixel value of 114 in the resulting 8 bit file. Most cameras and raw converters apply additional tone corrections, so it is best to examine the percent saturation in the raw file directly. In ACR you can try a linear setting with all the sliders on the main panes set to zero (exposure, recovery, brightness, contrast, etc) and the point curve set to linear. If you then take card of the baseline exposure factor and take a picture of an 18% card using the exposure indicated the camera meter, you should get a pixel value of about 114 if everything is working correctly and the camera is calibrated to the 1996 standard.

The new ISO standard is 12232:2006 as explained on Wikipedia and it is slightly different. It retains the saturation based method, and amplifies on noise based methods and adds a standard output sensitivity (SOS), which will lead to a pixel value of 118 in a gamma 2.2 space with 8 bit files.

With the saturation method, an exposure of an 18% gray card based on a standard light meter reading will result in an image with a grey level of 18%/√2 = 12.7% of saturation, which corresponds to a pixel value of 100 in a gamma 2.2 space. The square root of 2 is to allow 0.5 EV of headroom for specular highlights. Since the standard light meter is calibrated on the basis of the equivalence of 12% reflectance, you have to add 0.5 EV of exposure if you want the gray card image to have a pixel value of 118 in the gamma 2.2 file.

The Wikipedia article concludes, "Despite these detailed standard definitions, cameras typically do not clearly indicate whether the user "ISO" setting refers to the noise-based speed, saturation-based speed, or the specified output sensitivity, or even some made-up number for marketing purposes."

Bill Claff, a respected Nikon guru, has published ISO values necessary to attain 18% saturation of the sensor (essentially the SOS standard) on his Web Site. For example the D3 is rated at ISO 200 but an ISO setting of 138 is needed to attain 18% saturation. By contrast, the D80 is rated at ISO 200 and Bill has rated the native ISO at 93. The high ISO performance of the D3 may not be quite so good as it is cracked up to be.

Bill
Logged
Mark D Segal
Contributor
Sr. Member
*
Offline Offline

Posts: 6819


WWW
« Reply #45 on: April 03, 2008, 08:38:02 AM »
ReplyReply

Bill,

Thanks very much for posting this clear and detailed explanation. I shall retain it for future reference re some tests on the 1DsMk3 I intend to undertake later this month in connection with another exercise. Meanwhile, I notice the standards and the math are based on 8-bit depth. Our current crop of high-end DSLRs are working in 14 bit depth. Does this affect the numbers much?

Mark
Logged

Mark D Segal (formerly MarkDS)
Author: "Scanning Workflows with SilverFast 8....." http://www.luminous-landscape.com/reviews/film/scanning_workflows_with_silverfast_8.shtml
bjanes
Sr. Member
****
Offline Offline

Posts: 2756



« Reply #46 on: April 03, 2008, 09:54:17 AM »
ReplyReply

Quote
Bill,

Thanks very much for posting this clear and detailed explanation. I shall retain it for future reference re some tests on the 1DsMk3 I intend to undertake later this month in connection with another exercise. Meanwhile, I notice the standards and the math are based on 8-bit depth. Our current crop of high-end DSLRs are working in 14 bit depth. Does this affect the numbers much?

Mark
[{POST_SNAPBACK}][/a]

Mark,

You are welcome and I look forward to seeing the results of your testing. I would also suggest that you participate in some of Bill Claff's collaborations as he is now extending his methods to Canon cameras. He does a very sophisticated analysis at multiple ISOs with automated programs that he has devised. Also it serves as a good check on your own results.

[a href=\"http://home.comcast.net/~nikond70/Collaborations/ISO_Collaboration.htm]Claff ISO Collabroations[/url]

Here are some results including the Canon 1Ds Mark II, which does quite well considering its pixel count: Results


If you go to the trouble of testing, I think it makes sense to use the highest bit depth possible so as to get the best results. Thus far the extra two bits do not seem to make much difference. If you normalize the results to 1.0 (e.g. for 14 bit normalized pixel value = observed pixel value/16383) the results are the same for different bit depths.

Bill
Logged
Mark D Segal
Contributor
Sr. Member
*
Offline Offline

Posts: 6819


WWW
« Reply #47 on: April 03, 2008, 10:21:53 AM »
ReplyReply

Thanks Bill, I'll keep this on referral as it will be a bit of time before I can get down to this work. I'll report progress as and when........

Mark
Logged

Mark D Segal (formerly MarkDS)
Author: "Scanning Workflows with SilverFast 8....." http://www.luminous-landscape.com/reviews/film/scanning_workflows_with_silverfast_8.shtml
Panopeeper
Sr. Member
****
Offline Offline

Posts: 1805


« Reply #48 on: April 03, 2008, 06:41:26 PM »
ReplyReply

Quote
inconsistency between the histogram reading from the in-camera JPEG and the structure of the raw file, and seeming inconsistency between the latter and the ACR histogram resulting from "zero and linear" default settings
These two issues are totally independent.

The former is an old hat, the solution has been discussed in a thread started by Guillermo for many weeks ago. Since then I have been using a special WB preset "template", which makes the previews ridiculously off-color, but I achieve ETTR always exactly (i.e. within 1/3 EV).

The latter issue is an ACR-specific problem, the solution of which lies with Adobe.

Your post was in response to the description of ACR's unwanted adjustment, thus I did not understand, what you want to achieve.
Logged

Gabor
Mark D Segal
Contributor
Sr. Member
*
Offline Offline

Posts: 6819


WWW
« Reply #49 on: April 03, 2008, 11:54:50 PM »
ReplyReply

Quote
These two issues are totally independent.

The former is an old hat, the solution has been discussed in a thread started by Guillermo for many weeks ago. Since then I have been using a special WB preset "template", which makes the previews ridiculously off-color, but I achieve ETTR always exactly (i.e. within 1/3 EV).

The latter issue is an ACR-specific problem, the solution of which lies with Adobe.

Your post was in response to the description of ACR's unwanted adjustment, thus I did not understand, what you want to achieve.
[a href=\"index.php?act=findpost&pid=186892\"][{POST_SNAPBACK}][/a]

The two issues are separable but related in terms of the end objective. Thanks for your guidance about what is "old hat". Whatever Adobe does or doesn't do, to the extent any workaround is needed, I believe that is wha some of us have been discussing.
Logged

Mark D Segal (formerly MarkDS)
Author: "Scanning Workflows with SilverFast 8....." http://www.luminous-landscape.com/reviews/film/scanning_workflows_with_silverfast_8.shtml
Jonathan Wienke
Sr. Member
****
Offline Offline

Posts: 5759



WWW
« Reply #50 on: April 04, 2008, 06:59:27 AM »
ReplyReply

I really don't see a problem with ACR's exposure bias adjustment. It is nothing more than an offset between the actual exposure value and the displayed exposure value. Under no circumstances does it cause unrecoverable clipping. If the exposure bias is +.5 stops, then the actual exposure adjustment is +.5 stop when the displayed exposure adjustment is 0. If you have a perfect ETTR shot, sometimes you'll see clipping when the exposure slider shows and adjustment of zero, but in all cases, moving the slider to -.5 will eliminate the clipping. There is no case where the exposure bias adjustment will unrecoverably clip channels in a RAW image. All you need to do is move the exposure slider to the offsetting value, and the "issue" is completely eliminated.
Logged

Mark D Segal
Contributor
Sr. Member
*
Offline Offline

Posts: 6819


WWW
« Reply #51 on: April 04, 2008, 10:14:20 AM »
ReplyReply

Jonathan, yes, that's fine. But to remind about the main point of this discussion, starting from the OP, it's about the predictability of getting a good ETTR: <<do you develop a feel for how much to expose to the right, getting non-clipped shots most of the time?>> In this context, ideally one would want a set of defaults for both the in-camera JPEG and ACR which start by giving a reasonably reliable read of how the camera histogram portrays the raw data, and then in ACR, hoiw its histogram portrays the same data. The idea - and ideal - is to operate in both stages with as little custom adjustment as possible after say the second capture, once the first histogram is available - or even the first capture if we're talking live-view with live histogram. In this regard, the comments by Andrew on the camera's JPEG settings, and by Eric, Bill and yourself on raw file programs for evaluating the raw data apart from Camera Raw have been very helpful. I think the answer to the OP is that one needs to work with these tools and experiment till one finds the settings that work best for various kinds of imaging conditions.
Logged

Mark D Segal (formerly MarkDS)
Author: "Scanning Workflows with SilverFast 8....." http://www.luminous-landscape.com/reviews/film/scanning_workflows_with_silverfast_8.shtml
NikosR
Sr. Member
****
Offline Offline

Posts: 622


WWW
« Reply #52 on: December 31, 2008, 01:48:47 AM »
ReplyReply

I've just noticed this interesting old thread.

Let me add my opinion about Adobe's implementation of the baseline exposure adjustment. I think it s flawed because, as a matter of principle, these things should be DOCUMENTED and USER SELECTABLE. There are good arguments for not applying baseline exposure compensation and, IMO less good, arguments, for doing so. So, just document the damn thing and let the user decide whether he needs this function or not. Just my 2 cents worth.
« Last Edit: December 31, 2008, 10:07:42 AM by NikosR » Logged

Nikos
bjanes
Sr. Member
****
Offline Offline

Posts: 2756



« Reply #53 on: December 31, 2008, 07:39:52 AM »
ReplyReply

Quote from: NikosR
I've just noticed this interesting old thread.

Let me add my opinion about Adobe's implementation of the baseline exposure adjustment. I think it s flawed because, as a mater of principle, these things should be DOCUMENTED and USER SELECTABLE. There are good arguments for not applying baseline exposure compensation and, IMO less good, arguments, for doing so. So, just document the damn thing and let the user decide whether he needs this function or not. Just my 2 cents worth.


I agree that better documentation, not only from Adobe but also the camera makers, would be helpful. After all, it is the camera makers decision to allow for varying amounts of headroom for highlights that has prompted Adobe to use the baseline exposure adjustment. If a maker allows extra headroom for the highlights, it will likely use a baseline exposure adjustment so the previews do not appear dark. Since a common complaint about ACR is the images do not match the camera JPEG previews, Adobe is more or less forced to use a similar offset. As Eric Chan has explained, the baseline adjustment is user adjustable, since the user can make an exposure adjustment a part of his/her ACR defaults.

Bill
Logged
NikosR
Sr. Member
****
Offline Offline

Posts: 622


WWW
« Reply #54 on: December 31, 2008, 09:14:46 AM »
ReplyReply

Quote from: bjanes
As Eric Chan has explained, the baseline adjustment is user adjustable, since the user can make an exposure adjustment a part of his/her ACR defaults.

It is no good being user adjustable if one doesn't know it's there in the first place, is it?  How many LR/ACR users have read this thread (or any other thread / forum where somebody might have mentioned it) to know about it? Suppose they've heard about it, how are they to know by how much Mr. Knoll has decided to adjust each camera's output?

I've had it with all sorts of undocumented processing and features by Adobe. Yes, other vendors might do similar things but it's the Adobe tools I use. Camera vendors not being documented enough is no excuse for third party sw vendors to be undocumented also. And no, based on the explanations given above, this is not done to match the out of camera jpg, it is done to match different cameras' output. I bet the users who have a need for that feature are in a minority and since, most probably, they are advanced pro users they will run their tests and know how to adjust the ACR defaults to provide them with a uniform exposure across cameras. Nobody needs Mr. Knoll deciding what's best for them without even letting them know about it.
« Last Edit: December 31, 2008, 10:17:38 AM by NikosR » Logged

Nikos
Panopeeper
Sr. Member
****
Offline Offline

Posts: 1805


« Reply #55 on: December 31, 2008, 12:26:54 PM »
ReplyReply

Quote from: bjanes
I agree that better documentation, not only from Adobe but also the camera makers, would be helpful. After all, it is the camera makers decision to allow for varying amounts of headroom for highlights that has prompted Adobe to use the baseline exposure adjustment
The so-called "headroom" is a myth. It relates to a certain way of metering in a certain setting; as such it is a concept of very limited use. Tell about the headroom to all those, who sometimes overexpose when metering with the camera despite the headroom.

Quote
If a maker allows extra headroom for the highlights, it will likely use a baseline exposure adjustment so the previews do not appear dark. Since a common complaint about ACR is the images do not match the camera JPEG previews, Adobe is more or less forced to use a similar offset
I don't remember to have read anywhere, that Adobe had been asked to "correct" camera makers' mistakes.

Quote
As Eric Chan has explained, the baseline adjustment is user adjustable, since the user can make an exposure adjustment a part of his/her ACR defaults
1. The baseline exposure adjustment is paternalistic. Who the hell asked Adobe to make such decisions on part of the photographer? What is Adobe's aim? To shild the camera makers from the justified criticism?

2. The baseline exposure adjustment is clandestine. The users have no idea that this happens, thus they can not react by adjusting the ACR defaults. They are plainly misled by ACR.

3. The baseline exposure adjustment is counterproductive, for it hides the real exposure issues; in some cases it simply defeats the purpose. Example: Highlight Tone Protection with Canon cameras. Such a shot is underexposed by 1 EV (relative to the user's intention/camera metering). The camera's JPEG processing and Canon's raw processing software corrects this underexposure, in a non-linear fashion; this is the essence of HTP. What is ACR doing? It increases the intensity in a linear fashion without telling about it. The result is the worse of both worlds.

4. The baseline exposure adjustment is incorrect in several cases.

All in all: Adobe should get out of the business of fooling the photographers.
Logged

Gabor
Tim Lookingbill
Sr. Member
****
Offline Offline

Posts: 1147



WWW
« Reply #56 on: December 31, 2008, 03:19:01 PM »
ReplyReply

When talking about baseline exposure does this include the base contrast curve applied to the demosaiced sensor data that turns the dark linear preview AFTER demosaicing into a rendered image that somewhat resembles what the eye sees give or take varying contrast levels?

Have you ever examined this curve? It's not an easy curve to tweak when compensating for highlights or contrast because of its huge arc. One slight nudge or tweak to the curve and the highlights either go too dark or blow completely out. Same happens adjusting contrast. You can even see this artificial software driven simulation of light increase within the in-camera processor as it tries to deal with this crappy tone curve with each 1/3 to 1/2 step increase in exposure. One exposure shows the highlights located in the 1/4 tone section of the in-camera histogram and then the next step bunches it up on the right wall. WHAT? NO IN BETWEEN! LIGHT DOESN'T BEHAVE THAT WAY AND I DON'T THINK SENSORS DO EITHER!

So on top of this major base tone curve are the sliders and curve tools to apply finer tweaks. I think ACR does a remarkable job of allowing the user to smoothly and finely remap all tones without having to deal with this unruly base curve.

You could work directly off this dark linear data from scratch by editing this big arcing base curve or creating and applying a 1.0 gamma profile after the raw file has been rendered to 16 bit tiff and edit from there. Some raw converters provide this option. But from my experience fiddling with two raw converters that claim to deliver TRUE LINEAR data, I've gotten two different previews of the same raw file between them. So even at this level it's still interpretive and so it now becomes a question of not of who's right but what works.

No software be it the camera vendor's or a third party raw converter is going to get it right all the time because the data is interpretive right from the moment light hits the sensor.
Logged
JeffKohn
Sr. Member
****
Offline Offline

Posts: 1671



WWW
« Reply #57 on: December 31, 2008, 05:42:26 PM »
ReplyReply

I agree with Panopeeper, Adobe's implementation of baseline exposure just flat-out sucks. If their supposed goal is to match the tonality of in-camera JPEG then they have failed, at least in the case of my D300.

My approach to exposure for landscape work is to use manual exposure mode and UniWB custom white balance, exposing as far to the right as possible without blowing important highlights. When I first got the D300 I could not figure out why all of my NEF's were opening up in ACR signifcantly over-exposed, with blown highlights that weren't there in-camera. I'd still be in the dark if not for finding out about this baseline exposure adjustment. Not only do D300 NEF's look overexposed by default in ACR, but I think this behavior probably also accounted for some of the early criticisms about shadow noise at ISO 200 on the D300.

Now that I know about this baseline adjustment crap I've set my camera defaults to automatically apply -0.5 EV adjustment to offset it, and whenever I shoot with a new camera I'll be sure to use Rawnalyze to figure out what the baseline adjustment is so that I can 'undo' it. But I shouldn't have to do that, and the fact that Adobe was doing this behind my back because they think they know better than me what my exposures should look like is aggravating and insulting.
« Last Edit: December 31, 2008, 05:48:30 PM by JeffKohn » Logged

JeffKohn
Sr. Member
****
Offline Offline

Posts: 1671



WWW
« Reply #58 on: December 31, 2008, 05:45:42 PM »
ReplyReply

Quote
As Eric Chan has explained, the baseline adjustment is user adjustable, since the user can make an exposure adjustment a part of his/her ACR defaults.
How many people know how to determine the correct offset? You have to convert your file to DNG and open it in Rawnalyze to see what the baseline adjustment is, Adobe completely hides this information in their own UI and metadata displays. And to be thorough, you have to do this at every ISO you use, since the baseline adjustment can vary.

For that matter, how many people who don't hang out in technical photography forums even know this baseline adjustment even exists?
Logged

Panopeeper
Sr. Member
****
Offline Offline

Posts: 1805


« Reply #59 on: December 31, 2008, 06:25:18 PM »
ReplyReply

Quote from: JeffKohn
My approach to exposure for landscape work is to use manual exposure mode and UniWB custom white balance
Have you verified your Uni-WB? Rawnalyze from version 2.9.9.3 displays the WB coefficients on the histogram panel. They should not be farther apart than 5%.

Quote
Now that I know about this baseline adjustment crap I've set my camera defaults to automatically apply -0.5 EV adjustment to offset it
Are you aware of the fact, that the baseline exposure for the D300 is MINUS 0.5 EV @ ISO 100? Accordingly, your adjustment should be +0.5 EV.

This is one of the cases of incorrect adjustments I mentioned above; it should be about -0.15 (it should be zero, but compared to  +0.5 EV @ ISO 200, the -0.5 EV @ ISO 100 is incorrect).
Logged

Gabor
Pages: « 1 2 [3] 4 »   Top of Page
Print
Jump to:  

Ad
Ad
Ad