Forgot Password
Pentax Camera Forums Home
 

Reply
Show Printable Version Search this Thread
12-13-2009, 04:34 AM   #1
Veteran Member




Join Date: Jul 2009
Location: kobe/japan
Posts: 510
k7 wants double exposure

I mentioned this in another thread, and someone dismissed it as a 'myth'. I remember this was mentioned in dpreview samples too. The issue k7 wants double shutter speeds for same isos. Since dpr might be a mistake i decided to look at imaging resources.

So i opened iso 200 image of still life sections for K7 and KX. Here are exifs of those


K7 file
Sub Information
ExposureTime : 1/15Sec
FNumber : F8.0
ExposureProgram : Manual
ISOSpeedRatings : 200
ExifVersion : 0221
DateTimeOriginal : 2009:07:16 10:39:45
DateTimeDigitized : 2009:07:16 10:39:45
ComponentConfiguration : YCbCr
ExposureBiasValue : EV0.0
MeteringMode : Division

exif from KX:

Sub Information
ExposureTime : 1/30Sec
FNumber : F8.0
ExposureProgram : Manual
ISOSpeedRatings : 200
ExifVersion : 0221
DateTimeOriginal : 2009:11:11 11:06:36
DateTimeDigitized : 2009:11:11 11:06:36
ComponentConfiguration : YCbCr
ExposureBiasValue : EV0.0
MeteringMode : Division

Notce 1/15 seconds for K7 vs 1/30 secs for KX image, everything else is same.

Not so much of myth after all. Troubling part is this has a real world impact on how we photograph. When you are on iso400 on k7 you could do the same with iso 200 on KX. To make matters worse k7 noise-wise is already 1 stop or more worse.

Any body has any sensible explanation of this issue.

Ps: I hope the k7 fanboys does not spoil this thread with protecting k7 at any cost attitude. (it is just a camera).

12-13-2009, 05:24 AM   #2
Senior Member




Join Date: May 2009
Location: Finland
Posts: 191
My guess is.. Different day, different time, maybe different subject and different lightning. This causes the different sutter speed. The same aperture and the same ISO does not lead to the same shutter speed if the lightning is different.
12-13-2009, 05:30 AM   #3
Veteran Member




Join Date: Nov 2006
Location: Norway
Photos: Gallery
Posts: 385
It seems to be dependant on some K-7 special processing which I have not seen on any camera before and which leads to some confusion.

This was discussed in a Norwegian forums when Akam.no published a test where they criticised this. Afterwards some users compared K-7 with other cameras like Canon EOS 5D mkII, EOS 50D and Pentax *istDS and found that K-7 choosed exactly the same ISO, aperture and shutterspeed as those cameras when light level was low. In good daylight, K-7 choosed a slower shutterspeed, but the resulting jpg was still correctly exposed. So it seems that K-7 is exposing more to the right when the light is good enough to minimize noise in dark areas and then darken the jpg result a litle bit. Raw converters like Adobe Camera raw and others are doing the same when processing raw files from K-7, but there are raw converters which do not do this and you have to darken the image by yourself.

In practice the K-7 is selecting just as fast shutterspeed as other cameras when needed in low light, but are processing images a little bit different when shutterspeed isn't that critical low anyway.

Looking at dxomarks result, ISO accuracy is quite good on K-7. It has also been discussed a few times in the dpreview forum.
12-13-2009, 09:05 AM   #4
Dom
Guest




I'll get my K20D and K-7 out and take and do some test shots in the same conditions.

12-13-2009, 10:47 AM   #5
Site Supporter




Join Date: Aug 2007
Location: Dayton, OH
Posts: 365
Were these shots taken with the same lens? I've noticed that different lenses (especially with different maximum apertures) can meter a bit differently. Maybe not a full stop, but there's always some variation.
12-13-2009, 01:50 PM   #6
Pentaxian




Join Date: Sep 2007
Location: Bay Village, Ohio USA
Posts: 1,025
The first thing that jumps out is that these appear to have been taken on different days. In order for the test to be meaningful, it must be done at the same time, under the same lighting conditions, preferably with the same or identical lens.

Metering mode can also affect this, so both cameras should be set to the same mode. For example, if one is set to center point metering, and the other is set to multi-segment, depending on the lighting and dynamic range of the scene, one could easily expect different exposure values. That is the entire point of having different metering modes. If different modes always gave the same exposure, in all light conditions, there would be no point in even having different modes.

Paul Noble
12-13-2009, 03:31 PM   #7
Veteran Member




Join Date: Jul 2009
Location: kobe/japan
Posts: 510
Original Poster
QuoteOriginally posted by StigVidar Quote
It seems to be dependant on some K-7 special processing which I have not seen on any camera before and which leads to some confusion.

This was discussed in a Norwegian forums when Akam.no published a test where they criticised this. Afterwards some users compared K-7 with other cameras like Canon EOS 5D mkII, EOS 50D and Pentax *istDS and found that K-7 choosed exactly the same ISO, aperture and shutterspeed as those cameras when light level was low. In good daylight, K-7 choosed a slower shutterspeed, but the resulting jpg was still correctly exposed. So it seems that K-7 is exposing more to the right when the light is good enough to minimize noise in dark areas and then darken the jpg result a litle bit. Raw converters like Adobe Camera raw and others are doing the same when processing raw files from K-7, but there are raw converters which do not do this and you have to darken the image by yourself.

In practice the K-7 is selecting just as fast shutterspeed as other cameras when needed in low light, but are processing images a little bit different when shutterspeed isn't that critical low anyway.

Looking at dxomarks result, ISO accuracy is quite good on K-7. It has also been discussed a few times in the dpreview forum.
thank you, i will try to understand what you said.
12-13-2009, 03:35 PM   #8
Veteran Member




Join Date: Jul 2009
Location: kobe/japan
Posts: 510
Original Poster
About lighting.

Imaging resources make sure that the lighting stays as constant as possible. Since it is inside room and done with same set up, i think lighting is not the issue. Because if they can not control it then all their samples are simply useless for comparison.

To check this theory i picked another camera, nikon d700 and opened the still life iso200 image. Again shutter is 1/30 sec same as k-x

Here is exif

Filename : D700hSLI00200.JPG
JFIF_APP1 : Exif
Main Information
Make : NIKON CORPORATION
Model : NIKON D700
Orientation : left-hand side
XResolution : 300/1
YResolution : 300/1
ResolutionUnit : Inch
Software : Ver.1.00
DateTime : 2008:07:15 13:17:20
Artist :
YCbCrPositioning : co-sited
Copyright :
ExifInfoOffset : 348
GPSInfoOffset : 35264
Sub Information
ExposureTime : 1/30Sec
FNumber : F8.0
ExposureProgram : Manual
ISOSpeedRatings : 200
ExifVersion : 0221
DateTimeOriginal : 2008:07:15 13:17:20
DateTimeDigitized : 2008:07:15 13:17:20
ComponentConfiguration : YCbCr
CompressedBitsPerPixel : 4/1 (bit/pixel)
ExposureBiasValue : EV0.0
MaxApertureValue : F2.8
MeteringMode : Division
LightSource : Unidentified
Flash : Not fired
FocalLength : 70.00(mm)
MakerNote : Nikon COOLPIX Format : 34248Bytes (Offset:992)
UserComment :

12-13-2009, 11:16 PM   #9
Site Supporter




Join Date: Aug 2007
Location: Dayton, OH
Posts: 365
I agree that the vastly different dates in these comparisons makes them essentially meaningless. And we still don't have an answer to whether the same lens was used on the K-7 and K-x.

One reason that a few months' time might make a difference is in how much control you have over the light source. A light bulb can decline in brightness pretty significantly over its life. That's why replacing an old bulb with a fresh one can really seem to brighten up a room. I have no doubt that time alone could account for the different shutter speeds.

You may be thinking "I could tell if a light was 50% dimmer than it used to be...that's a lot!" But you really can't. A whole stop of light may seem like a lot, but it's not something easily noticed by the eye without direct and immediate comparison. The combination of the eye's non-linear sensitivity and the brain's highly subjective memory means that it's impossible to compare two similar lights (or even the same light) at different times. And I'm talking minutes, never mind months.

I'm not saying that the K-7 isn't doing some odd things with metering. But to convince me of that, you're going to have to eliminate more variables in the setup. Run a more controlled, repeatable experiment that doesn't involve months of elapsed time between trials. Use the same light, the same lens, and all the same camera settings. Perhaps use a separate light meter concurrent with each shot to show that the incident light on the scene is the same. If you can eliminate any alternative explanation for different metering, then you can safely say that it's due to the camera itself.
12-13-2009, 11:26 PM   #10
Veteran Member




Join Date: Jul 2009
Location: kobe/japan
Posts: 510
Original Poster
QuoteOriginally posted by aerodave Quote
I agree that the vastly different dates in these comparisons makes them essentially meaningless. And we still don't have an answer to whether the same lens was used on the K-7 and K-x.

One reason that a few months' time might make a difference is in how much control you have over the light source. A light bulb can decline in brightness pretty significantly over its life. That's why replacing an old bulb with a fresh one can really seem to brighten up a room. I have no doubt that time alone could account for the different shutter speeds.

You may be thinking "I could tell if a light was 50% dimmer than it used to be...that's a lot!" But you really can't. A whole stop of light may seem like a lot, but it's not something easily noticed by the eye without direct and immediate comparison. The combination of the eye's non-linear sensitivity and the brain's highly subjective memory means that it's impossible to compare two similar lights (or even the same light) at different times. And I'm talking minutes, never mind months.

I'm not saying that the K-7 isn't doing some odd things with metering. But to convince me of that, you're going to have to eliminate more variables in the setup. Run a more controlled, repeatable experiment that doesn't involve months of elapsed time between trials. Use the same light, the same lens, and all the same camera settings. Perhaps use a separate light meter concurrent with each shot to show that the incident light on the scene is the same. If you can eliminate any alternative explanation for different metering, then you can safely say that it's due to the camera itself.
1. Not all the cameras are released at same date and not all the samples could be shot at same time. This means, by the logic you have presented all the comparisons of images of sample done by all the review sites are just useless.
In other words all these review sites is run by stupids who could not even get basic stuff right. (yes there are some times mistakes but not blunders like you are suggesting).

2. Logic: Light intensity could fall with time of light source.

Now lets look at time line of cameras D700 -> K7 -> Kx

d700 : exposure : 1/30 sec
k7 : exposure : 1/15 (intensity fell).
kx : exposure : 1/30 (intensity again back to d700' time level).

So when it is shot with k7, all the factors other than camera come into play, but when samples are taken with d700, kx and other cameras things become alright and equal.

What you wrote is not logical. I do not buy it.
12-13-2009, 11:29 PM   #11
Veteran Member
KungPOW's Avatar

Join Date: Nov 2007
Location: Vancouver, Canada
Photos: Gallery
Posts: 1,702
Without the internet, there would be nothing to worry about.
12-14-2009, 12:54 AM   #12
Site Supporter




Join Date: Aug 2007
Location: Dayton, OH
Posts: 365
QuoteOriginally posted by zxaar Quote
1. Not all the cameras are released at same date and not all the samples could be shot at same time. This means, by the logic you have presented all the comparisons of images of sample done by all the review sites are just useless.
No, it just means that the test setup they used is not a carefully-calibrated test of exposure. The purpose of the shots is to assess image quality, not metering. And for the things that usually matter in an image test, it's not a big deal if their light source varies by an EV or so. As I said before, that's not as big a difference as it seems.

You're trying to use their results to draw conclusions on your own, but you don't know what steps may have been taken to ensure matching incident light levels. There likely was no special effort aside from using the same lamp, which is insignificant from a photometric standpoint. Their image results are surely not intended to be used as specific examples of metering accuracy, because that would require very specific measurements to be taken, independent of the camera. We're talking about possible differences in light levels that amount to a factor of 2 on sensors that measure over a factor of 400. To resolve that small deviation with confidence requires careful control.

But in this case, you don't know the specifics of how things were set up, and that's why you can't draw strict conclusions.

And, even if the metering results are totally accurate, why might a measured EV go down and then back up over time? Ever change a light bulb?
12-14-2009, 02:15 AM   #13
Veteran Member




Join Date: Jul 2009
Location: kobe/japan
Posts: 510
Original Poster
QuoteOriginally posted by aerodave Quote
No, it just means that the test setup they used is not a carefully-calibrated test of exposure. The purpose of the shots is to assess image quality, not metering. And for the things that usually matter in an image test, it's not a big deal if their light source varies by an EV or so. As I said before, that's not as big a difference as it seems.

You're trying to use their results to draw conclusions on your own, but you don't know what steps may have been taken to ensure matching incident light levels. There likely was no special effort aside from using the same lamp, which is insignificant from a photometric standpoint. Their image results are surely not intended to be used as specific examples of metering accuracy, because that would require very specific measurements to be taken, independent of the camera. We're talking about possible differences in light levels that amount to a factor of 2 on sensors that measure over a factor of 400. To resolve that small deviation with confidence requires careful control.

But in this case, you don't know the specifics of how things were set up, and that's why you can't draw strict conclusions.

And, even if the metering results are totally accurate, why might a measured EV go down and then back up over time? Ever change a light bulb?
Here is the set up:

Digital Camera Test Lighting - HMIs!

read through it and let me know of the objections to it, and please let us also know that what they should do so that k7 also exposes same as other cameras.

For me their set up is good enough, if the exposure values from different cameras were very much random, i would doubt the set up. So far lighting set up is only acting up when you shoot with k7 and not with others.

Plus why dpreview's sample also show same behaviour, i do not think they are using imaging-resource's light set up. Or that is screwed up too.

All the things indicate that it is something to do with sensor. K7's base iso is 100, it might have something to do with it.

PS: not saying that base iso 100 is bad, there are people who would like base iso to be low as 50 or 25. They value this very much.

PS: Here is quote from that page:

"the downsides of our original light setup were that changing bulbs and re-balancing the illumination was a tedious, painful process, and the diffusion gels we had to use to achieve even illumination with them introduced a slight color shift toward warmer tones."

What it means they go great lengths to ensure that lighting remains uniform. Meaning they calibrate it. Not so stupid guys it seems they are.
12-14-2009, 06:09 AM   #14
Site Supporter




Join Date: Aug 2007
Location: Dayton, OH
Posts: 365
I read that, and it sounds to me like they care about color temperature far more than the irradiance of their subject. At no point to they talk about using a light meter for calibrating intensity. And that's perfectly fine for what they're doing, because they're not making objective comparisons of camera metering. You are.

Look, I'm perfectly willing to accept the the K-7 has funny metering under some conditions. But your meta-analysis isn't cutting it from a "burden of proof" standpoint. If you tried to publish your conclusions in a scientific paper, they'd be completely ignored--if not laughed at--because of the inability to ensure experimental control. There's too much uncertainty. And being anti-intellectual about it, thumbing your nose at the scientific method, isn't helping.
12-14-2009, 06:45 AM   #15
Pentaxian
Wheatfield's Avatar

Join Date: Apr 2008
Location: The wheatfields of Canada
Posts: 10,042
QuoteOriginally posted by zxaar Quote
I mentioned this in another thread, and someone dismissed it as a 'myth'. I remember this was mentioned in dpreview samples too. The issue k7 wants double shutter speeds for same isos. Since dpr might be a mistake i decided to look at imaging resources.

So i opened iso 200 image of still life sections for K7 and KX. Here are exifs of those




Notce 1/15 seconds for K7 vs 1/30 secs for KX image, everything else is same.

Not so much of myth after all. Troubling part is this has a real world impact on how we photograph. When you are on iso400 on k7 you could do the same with iso 200 on KX. To make matters worse k7 noise-wise is already 1 stop or more worse.

Any body has any sensible explanation of this issue.

Ps: I hope the k7 fanboys does not spoil this thread with protecting k7 at any cost attitude. (it is just a camera).
I'm thinking that Pentax has recognized that the Samsung sensor isn't working as advertised, and changed sensor manufacturers when they went to the K-x.
The K-20 had some really wonky sensor issues when first released (random moving hot pixels) that they ended up masking with software in their first firmware revision. I suspect that Samsung has fixed one problem at the expense of something else.
BTW, the dpreview discussion may just be a regurgitaion of the image resource test, so you may not really have multiple sources of information.
Anyway, if it's a problem for you, then buy a K-x. If the K-7 meets more of your needs by a K-7.
I'm really not seeing what the deal is here.
Reply

Bookmarks
  • Submit Thread to Facebook Facebook
  • Submit Thread to Twitter Twitter
  • Submit Thread to Digg Digg
Tags - Make this thread easier to find by adding keywords to it!
camera, division, dslr, f8.0, k7, kx, photography
Thread Tools Search this Thread
Search this Thread:

Advanced Search


Similar Threads
Thread Thread Starter Forum Replies Last Post
Double Exposure dbay150 Digital Processing, Software, and Printing 1 05-30-2010 09:49 PM
K20D Double-Exposure Adam_T4 Post Your Photos! 0 09-02-2009 04:13 PM
Double Exposure With a Spotmatic2 (help) Miserysmalice Pentax Film SLR Discussion 7 01-28-2009 10:25 AM
double (or more) exposure on the same frame? Gooshin Pentax DSLR Discussion 25 12-23-2008 01:35 PM
In-Camera Double Exposure awjweb Post Your Photos! 7 02-28-2007 10:05 PM



All times are GMT -7. The time now is 03:53 AM. | See also: NikonForums.com, part of our network of photo forums!
  • Red (Default)
  • Green
  • Gray
  • Dark
  • Dark Yellow
  • Dark Blue
  • Old Red
  • Old Green
  • Old Gray
  • Dial-Up Style
Hello! It's great to see you back on the forum! Have you considered joining the community?
register
Creating a FREE ACCOUNT takes under a minute, removes ads, and lets you post! [Dismiss]
Top