No. The base ISO is 200 because each pixel on the sensor is relatively large (it is 6MP sensor, as opposed to 10MP), so catches a lot of light. If the base ISO were 100 the sensor would saturate. The Nikon D40 and D40x are, respectively, 6MP and 10Mp (for the same size sensor) and have base ISOs of 200 and 100 for this reason..

I don't find it a significant issue for normal use... if you find yourself wanting to use a wide aperture for narrow depth of field, in bright sunlight, you have a shutter speed of 1/4000 sec available. Using the 'sunny 16' rule, in bright sunlight at ISO100 you would get a correct exposure with about 1/125 at f/16. At ISO200 that translates to 1/4000 sec and f/4..

The advantage at the other end of the scale is that the K100D has excellent high-ISO performance, giving excellent images at ISO800 and perfectly usable ones at ISO-1600, so that needs to be balanced against the above problem..

If you want to use really long shutter speeds (e.g. to get blurred water motion) you'd need a neutral density filter anyway..

Best wishesMike..

John Schranz wrote:.

I'm thinking of buying my forst DSLR, the Pentax K100D but am a bitworried that it's lowest ISO is 200. I primarily take outdoor photoswith often bright natural light conditions. Should I be worried thatthere isnt ISO 100?.

No, no reasons to worry..

Edvinas..

Mike703 wrote:.

No. The base ISO is 200 because each pixel on the sensor isrelatively large (it is 6MP sensor, as opposed to 10MP), so catches alot of light. If the base ISO were 100 the sensor would saturate.The Nikon D40 and D40x are, respectively, 6MP and 10Mp (for the samesize sensor) and have base ISOs of 200 and 100 for this reason.Mike.

This makes no sense. You are stating that at ISO 100 the sensor would "saturate" whereas at ISO 200 it would not..

First off, the metering system is designed to (unless you manually choose otherwise) make sure the sensor does not "saturate", which I take it that you mean that it would collect too much light and overexpose the image. The metering system would increase or decrease the aperture and/or shutter speed to keep the exposure accurate..

Second, ISO 100 is less sensitive than ISO 200, by a full stop (half as sensitive). It would take twice the amount of light captured by the sensor at ISO 100 to match the exposure of the same sensor at ISO 200 (all else being equal, such as shutter speed, aperture, and ambient light). Therefore, ISO 100 has less of a change of overexposure and blown highlights, given any improper settings for the other variables to exposure..

BTW, the Canon 300d, the Konica Minolta Maxxun 7D and 5D all are 6MP SLRs with ISO100 settings, so that shows that there are no overall technical reasons to exclude ISO100..

All in all, I believe it was just a decision made by Pentax and Nikon because they may have believed that their ISO 200 was "clean" enough, no need for ISO 100, when they had 6MP sensors. Another possibility is that their hardware at the time could not reduce the current (or voltage, not sure shich) accurately enough for an ISO 100 capture..

Now, as their new sensors have 10MP and higher resolution, the noise might be more pronounced and noticeable, therefore they had to include ISO 100 in order to keep the base ISO noise to a minimum and compete with others who do include ISO 100. BTW, the Pentax 6MP K100 and K110 are relying on older tech for those sensors. Take a look at the noise they create in Phil's long exposure tests)...

The base sensitivity of the sensor is ISO 200. That means, it requires no gain in order to get that ISO setting. If you want ISO 100, then you need to do some unnecessary fanagling to do it. IMHO, why bother? It really is clean at ISO 200 with excellent rendering. The sensor is also very low in noise. I do not know why Nikon needs to use NR but Pentax uses none.



ISO 3200 IstDS.

Image control:Zoom outZoom 100%Zoom inExpand AllOpen in new window..

... then why are there no settings such as ISO 12 / 25 / 50 on a DSLR? They could be useful in bright light, but the resulting long exposure times (large amounts of light hitting the sensor) would saturate the sensor elements..

Mike..

There is this impression that a DSLR will be like a compact. It will have a low starting ISO and be noisy when increased. Compacts need noise reduction at ISO 100. The Sony 6MP senso in the K100Dr needs none at ISO 1600. ISO 3200 is better than many compacts at ISO 200...

Mike703 wrote:.

... then why are there no settings such as ISO 12 / 25 / 50 on aDSLR? They could be useful in bright light, but the resulting longexposure times (large amounts of light hitting the sensor) wouldsaturate the sensor elements..

Mike.

Notice that even the new Canon 1Ds MarkIII has an ISO 50 setting. They needed to add it because it has freakishly high resolution, thereby needing a lower ISO baseline to give it comparable noise to what they had delivered with lower resolution SLRs..

You logic is faulty. None of us really knows why Pentax and Nikon chose ISO 200 as the baseline, but is surely is not because of lower ISO value "saturating" the sensor due to longer exposures. If that were the case, why do they have 30 second or longer long exposure times? You know that those settings are available at all times, not just when it's dark outside, don't you?.

Tell me, how does a lower ISO (which by definition means lower sensitivity) get "saturated" with a properly exposed image, especially when you are saying that a higher ISO setting would not "saturate" the sensor. Please explain your logic..

I have cited 3 6MP SLRs and 1 21MP SLR that have ISO 100 or lower settings. Also, ALL 10MP SLRs have ISO 100 settings. It is therefore not an issue of "saturation" as you say, because they obviously are doing it..

Further, look at these dynamic range tests, notice how in almost every instance had the lower ISO setting giving more dynamic range than the higher ones, which is the opposite of this "saturation" you are stating..

Http://www.dpreview.com/reviews/canoneos30d/page21.asphttp://www.dpreview.com/reviews/CanonEOS400D/page19.asphttp://www.dpreview.com/reviews/fujifilms5pro/page18.asphttp://www.dpreview.com/reviews/nikond40/page18.asphttp://www.dpreview.com/reviews/nikond40x/page18.asphttp://www.dpreview.com/reviews/nikond80/page19.asphttp://www.dpreview.com/reviews/nikond200/page22.asphttp://www.dpreview.com/reviews/OlympusE330/page19.asphttp://www.dpreview.com/reviews/OlympusE410/page17.asphttp://www.dpreview.com/reviews/panasonicdmcl1/page17.asphttp://www.dpreview.com/reviews/pentaxk100d/page15.asphttp://www.dpreview.com/reviews/pentaxk10d/page15.asphttp://www.dpreview.com/reviews/sonydslra100/page21.asp.

The ONLY exception was the Pentax K100D. Its ISO 400 had better DR than it's ISO 200. Again, as my first post indicated, Phil commented on the aging sensor technology in his long exposure test:.

"Maybe it's a sign of the sensor's age (in development terms) but using the K100D without noise reduction for a typical thirty second exposure will lead to a liberal sprinkling of 'hot pixels' all over the image, some more pronounced than others. With noise reduction enabled however these are virtually all removed, with only one or two leading to black 'pits'.".

That Pentax sensors, which have some of the lowest dynamic range values on the entire list, obviously need more work..

So, tell us why you think that LOWER ISO values will result in "saturation" of the sensor whereas HIGHER ISO values would not...

I do take reviews and tests with a grain of salt. My experience with that lowly 6MP sensor is that the DR is pretty dog gone good. Noise is as low as it gets and it has excellent tonality. Yes, dark frame subtraction is a nice thing to use on very long exposures. You have to realize that is the only NR available on the 6MP Pentax DSLR's, yet it is quite clean at ISO 1600. This does not seem to be sinking in. If you want better high ISO abilities, you need to get a Canon 5D which is several times the price...

Jd73 wrote:.

You logic is faulty. None of us really knows why Pentax and Nikonchose ISO 200 as the baseline, but is surely is not because of lowerISO value "saturating" the sensor due to longer exposures..

How would you do it? ISO 200 is that sensor's base. It requires no amplification at that point. You mention other sensors but they are less sensitive and have a lower base ISO. Unless you are willing to do something funky, you will start at ISO 200 with that sensor...

I've got a Nikon D50 that also has it's lowest ISO setting at 200. There's no need to worry, it produces a nice clean shot with good detail. A co-worker of mine has the K100D and I've seen nothing from his shots that would suggest that there is anything for you to worry about. It's a nice camera with great features for the price. Go for it.My humble photo gallery: http://ntotrr.smugmug.com.

Image control:Zoom outZoom 100%Zoom inExpand AllOpen in new window..

Hey there K1000Photographer,.

My comments are not directed toward you, but to Mike. I am not commenting on the relative noise of each sensor (except as example), but on his assertion that the lack of lower ISO settings is because it would cause the exposures to "saturate" the sensor. That is a load of bunk. It would just be wrong to not correct such an obviously wrong assertion when it is presented as fact in response to the OP's question..

I agree with you that the Pentax sensors are great when compared to their contemporaries. I used Pentax as an example because the OP questioned that particular camera, and because it was one of the few 6MP SLRs being sold now. As a matter of fact, didn't Phil think that the K100D was the best 6MP being sold, until the Nikon D40 that is (even that can be argued)...

The D70, D50, D40, *istD, *istDL, *istDS, GX1-s, GX1-L and others use the same 6MP Sony sensor. The base ISO is 200. How would you force it to do ISO 100?.

IMHO, Nikon does OK with it but it bothers me that they need ot use NR and Pentax does not...

BTW, in case it gets lost in the other "sub-thread" discussion; the K100D is a great camera. I would consider it over the Nikon D40 for 3 reasons..

1 - CCD shift IS, which is great when coupled to ANY of the lenses..

2 - Can use any lens, to their fullest extent, that uses the same mount. Some of those older lenses are cheaper too.3 - The price makes it an incredible value..

If you think you will stay with entry level SLRs (ie, below $800) then this might be the most compelling choice...

There is no reason to force an ISO 100 on it. My point against Mike is his reasoning behind why there is no ISO 100. He says it's because of "saturating" the sensor. It's not that because that explanation makes no sense (technically wrong logic)..

Perhaps the ISO 200 baseline was Sony's choice, which the others had no power to change. Perhaps the engineers just said "ISO 200 is darn clean, 100 is not better, let's not put it in." That too is a plausible and satisfactory answer. Either is better than saying "because it would saturate the sensor.".

Who knows. Would ISO 100 be nice? Sure, why not. Is it needed? Not for 99.9% of the shots most people would take...

This makes no sense. You are stating that at ISO 100 the sensorwould "saturate" whereas at ISO 200 it would not..

First off, the metering system is designed to (unless you manuallychoose otherwise) make sure the sensor does not "saturate", which Itake it that you mean that it would collect too much light andoverexpose the image. The metering system would increase or decreasethe aperture and/or shutter speed to keep the exposure accurate..

No - you've got the wrong end of the stick. The relationshop between how much light hits the sensor, and what is a correct exposure, is not fixed but is variable according to the ISO setting. Saturation is nothing to do with overexposing the image..

Look at it like this..

Suppose, to get a correct exposure at ISO1600, a sensor requires on average 10 photons per pixel (just a for-instance value but probably not massively out)..

If the sensor gets a longer exposure (20 photons per pixel) the picture will be overexposed by one stop - unless the user changes the ISO setting to 800. So whether or not the picture is correctly exposed depends on the ISO setting..

Similarly, a correct exposure at ISO 400 requires 40 photons per pixel; ISO 200 requires 80 photons per pixel; and ISO100 requires 160 photons per pixel for a correct exposure. You can keep going like this as far as you like up to 1280 photons per pixel for ISO12....

There comes a point when a pixel on a sensor cannot detect any more photons. The light flux causes an electrical charge to build up - to a certain physical limit which depends on the material used, dimensions, and all sorts of other things. Beyond that limit, more photons will not cause a corresponding increase in charge - and the sensor is saturated and useless from the point of view of recording an accurate image..

The base ISO of a sensor will be dependent on this. In the above example, if each pixel becomes 'full' of charge after 100 photons, it will have a base ISO of about 200. If it becomes full of charge after 1000 photons, it will have a base ISO of about 25. Different sensors use different materials and have different dimensions so will vary, but the principle remains. The sensor used in the K100D has been used in several cameras - always, as far as I know, with a base ISO of 200..

Saturation of CCD sensors is well known and understood, and is what limits their dynamic range in any application, be it a camera, mnicroscope, or other imaging device. For a more detailed explanation, have a look at.

Http://www.microscopy.fsu.edu/...gitalimaging/concepts/ccdsatandblooming.html.

If there is anything in the above that is factually incorrect I would be very pleased to know what it is. There is no need (having read some of your earlier posts) to be offensive.Mike..

Further, look at these dynamic range tests, notice how in almostevery instance had the lower ISO setting giving more dynamic rangethan the higher ones, which is the opposite of this "saturation" youare stating..

Http://www.dpreview.com/reviews/canoneos30d/page21.asphttp://www.dpreview.com/reviews/CanonEOS400D/page19.asphttp://www.dpreview.com/reviews/fujifilms5pro/page18.asphttp://www.dpreview.com/reviews/nikond40/page18.asphttp://www.dpreview.com/reviews/nikond40x/page18.asphttp://www.dpreview.com/reviews/nikond80/page19.asphttp://www.dpreview.com/reviews/nikond200/page22.asphttp://www.dpreview.com/reviews/OlympusE330/page19.asphttp://www.dpreview.com/reviews/OlympusE410/page17.asphttp://www.dpreview.com/reviews/panasonicdmcl1/page17.asphttp://www.dpreview.com/reviews/pentaxk100d/page15.asphttp://www.dpreview.com/reviews/pentaxk10d/page15.asphttp://www.dpreview.com/reviews/sonydslra100/page21.asp.

Oh, dear. This issue (reduced dynamic range at high ISO) is nothing whatsoever to do with CCD saturation at low ISO..

Since you have brought it up, it works like this..

Imagine the sensor described in my previous post. Let's say that at ISO 100, for a correctly exposed picture, the brightest highlight (pure white) has 160 photons hitting a pixel; zero photons corresponds, obviously, to complete black. In between the sensor can record 158 different grey levels, from 1, 2, 3... up to 159. Non-integral levels (say, 100.5) can't be measured because photons come in complete units; you can't have half of one..

The noise is, let's say, 1 unit, arising from thermal motion of electrons in the camera's circuits. So an intensity reading that should be 100 may come out as 100, or 99, or 101. Not much variation and the result will be a clean, low-noise image with lots of different grays accurately rendered (good tonal range)..

Now imagine setting the ISO on the camera to 1600, and recording the same scene. We have effectively reduced the exposure by 4 stops, from (say) 1/60 at f/8 to 1/1000 at f/8 - only one-sixteenth as much light is hitting the sensor. So the brightest part of the scene will now only result in 10 photons hitting a pixel. To compensate for this the weak signal is multipled by 16 (a gain is applied) to bring the maximum signal level up to 160, as before. Black is still zero of course. Note that slight overexposure (a pixel collecting 11 photons) will result in that part of the picture being burnt out, because 11 scales up to 176 and the maximum value for white is set to 160.

This has nothing whatever to do with saturation; the light flux on the sensor is still very low and it is operating well within it's range..

This reduction in signal intensity, and compensatory gain to boost the signal, Have two undesirable results..

Firstly, we now get a much reduced tonal gray scale. Since photons can only be measured in ones, and a reading of 10 corresponds to full brightness after the gain has been applied, we can only get readings of 0, 1, 2, 3... up to 10 fr the brightest part of the image; these are scaled up to 0, 16, 32, 48... 160. So the number of mid-range gray tones that can be accurately rendered is reduced..

Secondly, the noise was still 1, but this is now multiplied by a factor of 16 as the weak signal is scaled up. A sensor reading for mid-gray of 51 is scaled up to 8016. There will therefore be a lot of noise in the mid-range grays, as a reading that is supposed to be 80 could equally well be 64 or 96 - a large random varation..

HOWEVER - and this is the important bit - the noise is not applied equally in either direction at the extremes, but only works in one direction. A white pixel (reading of 160 after the gain is applied) has an error of 16. But an error of +16 is meaningless: if 160 is bright white, so is 161 ,162 and every value up to 176 (and beyond). So the white pixels cannot increase in brightness. But if the error is negative, they could be darkened down to 144. Similarly at the other end: a part of the picture that is supposed to be black wil have an actual reading of 0 16.



Overall, noise in the mid-range applies equally in either direction. But white areas will be slightly darkened and black areas will be slightly lightened. Result: reduced dynamic range at high ISO, a simple and inevitable consequence of the statistical effects of adding noise to a weak signal..

The fact that you think that this has anything to do with the phenomenon of sensor saturation under high light flux that you were disagreeing with earlier illustrates clearly your confusion on the subject..

Well, in this post (and the previous one) I have explained the issues as simply as I can, which is far more than you deserve given the tone of your comments. If after reading them - and the article on sensor saturation that I gave the link to in the previous post - you still don't get it, there is nothing more I can say. Except to suggest that you go back to school and take remedial classes in maths and physics, and try to stay awake this time..

I'm done with this thread. Have a nice day.Mike..

Mike,.

Your explanations couldn't be clearer. Thanks for the lesson..

- Seb..

Thanks everyone! I just got the K100D Super (it was cheaper than the K100D being sold in the shop right next door believe it or not). Havent yet had much chance to try it out but I'm sure I wont be let down..

John..

How much was it, if you don't mind to tell? I have an old DS and love the handling and IQ. The AS, faster AF and ability to use the SDM lenses appeals to me. Thanks..

Well I live in Brussels which maybe, in a way is not the best place for deals since many online shops are not available since they do not operate in Belgium. Still I managed to find it for 550 EUR (camera plus 18-55 lens) which I think was a very respectable deal..

John..

Congratulations and good luck with it. Post some pictures...Best wishesMike..