In praise of the K100D's low-light performance

Alfisti · 09-04-2009, 10:19 AM

Originally posted by aerodave

Well you're no fun.

Again, let me say that I completely understand that it's an application-dependent thing. Sometimes having the same display size is the fairest comparison. Sometimes it's viewing 100%. It completely depends on what your objectives are and what you want to do with the images.

It's like the DOF/cropped sensor debate raging in another thread here. There's no way to come up with a one-size-fits-all measure of merit for anything related to a tool's performance, because the things that matter change with how that tool is being used.

I'm really not trying to be argumentative. My original point was that, for the way I use my cameras, I find the K100D to have much more satisfying low-light performance. How I am using my equipment is leading me to that conclusion, so it's a valid one...for me at least.

Gotta be honest, there is really no "how i use my images" or 'the way i view them' discussion, you're either printing to a certain size or viewing ont he web to a certain size and either one you're going to have to re-sample the 100D up or the 20D and i'm absolutely certain the 20D is very similar to the 100D.

Don't get me wrong, at 100% I nearly had a freakin heart attack but at 6mp each or at 11*14 ...... they look the same.

NewportPreacher · 09-04-2009, 03:03 PM

Very interesting discussion and timely, since I've had my K20D for a week now & moved up from the K100D. I noticed that, when viewed at 100%, noise at ISO 400 on the K20D looked a little worse than a comparable photo from the K100D, shot at ISO 800 & viewed at 100%. But, I rationalized that nobody's going to be viewing it at 100% "in the real world" & even larger prints would give me more than acceptable results.

Unless my quick reading of this thread was way off, that seems to fall in line with what most all of you are saying.

SteveB · 09-05-2009, 02:48 AM

Low light AF performance on the K100D is its only problem really for a 3yr old camera. Noise levels aside, one very good low light characteristic of the K100D is its ability to maintain good colours (assume RAW as AWB isn't great in low light). A lot of cameras really fall down here. Is the K20D (or K7) as good as the K100D in this respect?

kittykat46 · 09-05-2009, 07:36 AM

Originally posted by SteveB

Low light AF performance on the K100D is its only problem really for a 3yr old camera. Noise levels aside, one very good low light characteristic of the K100D is its ability to maintain good colours (assume RAW as AWB isn't great in low light). A lot of cameras really fall down here. Is the K20D (or K7) as good as the K100D in this respect?

The K100D at ISO1600 is good in RAW. Work some Noise Reduction on the RAW file, and you have a very usable photo.
The K100D's shortcoming in low-light is its AF. Many times I failed to get an AF lock even in decent indoor lighting....OK for static objects not for anything which may move away.

The K-7 looks noisier if you compare 100% crops for ISO1600. But a fairer real-world comparison is to compare print photo to print, or full screen to full screen...No, from that point of view, the K-7 is definitely better with more detail and less visible noise at ISO1600 and ISO3200.

**Rondec** · 09-05-2009, 07:50 AM

I have owned the K100 and currently own the K10 and K20. Looking back, I actually shot to iso 3200 with K100 -- a little noisy, but certainly acceptable photos. The K10 can be shot to iso 800, but past that is definitely rough. A lot of noise and sometimes banding crop up. The K20 is really strong up to iso 2000.

It is a little difficult comparing the photos as mentioned above. 6 megapixel files are a lot smaller than 14 megapixel files. The question is really one of printing size -- most photos look decent if printed at 4 by 6 size, while if you push the size up -- noise crops up too.

ricardobeat · 09-05-2009, 11:35 PM

High ISO on the K100D really looks much better than on the K10D. I have no fear of using 1600, even going up to 3200 when needed. The grain looks very nice, not like digital blotches, and color is good, you just lose some dynamic range.

The K10D (GX-10 in my case) is a whole other story. Shooting at ISO 800 could ruin colors and make for horrible images. That's one of the main reasons I never went past 6000 shots on it in 2 years.

aerodave · 09-06-2009, 10:06 AM

Inspired by the noise/zoom discussion in the first part of this thread, I decided to conduct a an objective trial. Seems that people on both sides of the discussion were talking in hypotheticals without any pictures to back up their points. And I admit I was equally guilty in that regard.

So the attached pics are as close to an apples-to-apples comparison as I could do. You're looking at crops of an identically-framed shot with identical exposure, taken with both the K10D and K100D. What you are seeing is a crop, but not a 100% crop from either camera. It's the same angle of view taken from each identical original picture. The reason I didn't use the whole image is because resizing the whole image to fit as a forum attachment would have resulted in far more downsampling for both cameras than I would normally do, and so would be a pointelss comparison. This section exported to about 800 by 560 pixels represents a comparable amount of resizing compared to what I might publish as a large web image (say, 2000 pixels on the long side for the whole image).

Also attached is the original shot as it came out of the K10D, just so you can see how much of a crop we're talking. The original from the K100D looks so close to the same it's not worth taking up space with.

Both shots were RAW captures at ISO 1600, 1/15 sec, f/2, using an FA 35 f/2 AL. Lightroom settings were identical between them, and no noise reduction was performed.

As you can see, the K10D has a lot more chroma noise, especially visible in the shadow on the side of the amp. Subjectively, I'd say the K100D is less noisy overall, and what noise does exist is a little more appealing.

I don't have any noise reduction tools outside what Lightroom and Photoshop offer. But attempts to apply noise reduction to both images never made the K10's picture look as good as the K100's. The extra resolution doesn't seem to help the Adobe programs' efforts. Maybe a third-party application would work more magic...I'm open to that possibility.

Sure this is one shot, and doesn't tell the whole story. But I think it somewhat refutes the idea that the noise differences between the two cameras even out when you output to the same size. That's exactly what I've done here, and the K10D is still worse.

Attached Images

View Picture EXIF

View Picture EXIF

Marc Sabatella · 09-06-2009, 09:10 PM

Originally posted by aerodave

So the attached pics are as close to an apples-to-apples comparison as I could do.

I agree; what you did here is the best way to do these sorts of tests.

Quote:

Sure this is one shot, and doesn't tell the whole story. But I think it somewhat refutes the idea that the noise differences between the two cameras even out when you output to the same size. That's exactly what I've done here, and the K10D is still worse.

Agreed, but you'll also note that I at least have tried to be careful and always mention in my remarks that, due to an unfortunate quirk in the implementation of the K10D (omitting the analog signal amplification and relying entirely on an apparently questionable analog-digital-converter), the K10D really is worse than the K100D, as well as being worse than other Pentax cameras with the same sensor (K200D and K-m). That's what I've seen in m own looking at well-done tests like this, and I believe that's what the folks who've actually measured the noise statistically have found as well.

Still, I also think this comparison does illustrate my point pretty well - while the K10D *is* indeed worse, the difference is rather less when comparing apples to apples as you are here than when comparing at 100%. And showing the full images too really helps put into context just how small those differences actually will turn out to be in practice. And yes, I do think programs with better NR algorithms than LR could probably bring the K10D image more in line with the K100D, but not to the same extent that would be possible for the K200D or K-m.

gfmucci · 09-06-2009, 09:29 PM

Aerodave:

Thanks for taking the time to give us that apples to apples comparison. It shows how much better "doing it" is than just "talking about it." I appreciate that.

shaolin95 · 09-07-2009, 06:22 AM

Very nice.
I need to do the same but I had a rough weekend since I just moved.
I will test the K20D and the K100D both with the K20D kit lens.

aerodave · 09-07-2009, 06:26 AM

Marc,

I think you're right about the K10D being a special case and unnecessarily worse than pretty much any other Pentax with the same sensor. And I also believe that I'd get better performance out of the K20D or K-7. Can't wait to see first-hand how the K-7 does. =)

**interested_observer** · 09-07-2009, 07:19 AM

Several months ago I upgraded from the K100 to the K20 for a number of reasons, including the ISO100, additional control (the 2 wheels), more room to crop, along with better menu controls - plus just the price of the K20. I did keep the K100 and the more I use the K20, the more appreciation I have for the K100. Its a great little camera, I really still like the size and it still takes wonderful images. I have also shied away from SDM lenses, so that they are usable across both camera bodies (maybe I should go find a super body - but what I have is more than sufficient, and my lens acquisition is essentially over, other than a DA 21 limited - maybe).

On using ambient light, I have gone the other way, rather than going up in ISO speed, I am using slower ISO speeds and taking the hit in shutter speed. This has caused me to significantly upgrade my tripod and head, but I was going to have to do that anyway.

I am still learning and the K20 is a wonderful tool to extend my skills with. It is enough camera to satisfy me past the K7 and on to its replacement.

Similar Threads
Thread	Thread Starter	Forum	Replies	Last Post
Has any preview mentioned low light performance on the K-7?	NothingInCommon	Pentax News and Rumors	11	06-23-2009 01:09 PM
K200D low light/high iso performance	indytax	Pentax DSLR Discussion	22	02-07-2009 07:49 PM
K20D low-light performance	Stoosh	Pentax DSLR Discussion	10	03-03-2008 03:44 AM
A Low-Light Performance	paden501	Post Your Photos!	3	01-29-2008 12:18 PM
Low light performance of 2 kit lenses	fevbusch	Pentax SLR Lens Discussion	12	05-23-2007 07:30 PM

09-04-2009, 10:19 AM	#16
Alfisti Veteran Member Join Date: Mar 2007 Location: Toronto (for now) Posts: 1,748	Originally posted by aerodave Well you're no fun. Again, let me say that I completely understand that it's an application-dependent thing. Sometimes having the same display size is the fairest comparison. Sometimes it's viewing 100%. It completely depends on what your objectives are and what you want to do with the images. It's like the DOF/cropped sensor debate raging in another thread here. There's no way to come up with a one-size-fits-all measure of merit for anything related to a tool's performance, because the things that matter change with how that tool is being used. I'm really not trying to be argumentative. My original point was that, for the way I use my cameras, I find the K100D to have much more satisfying low-light performance. How I am using my equipment is leading me to that conclusion, so it's a valid one...for me at least. Gotta be honest, there is really no "how i use my images" or 'the way i view them' discussion, you're either printing to a certain size or viewing ont he web to a certain size and either one you're going to have to re-sample the 100D up or the 20D and i'm absolutely certain the 20D is very similar to the 100D. Don't get me wrong, at 100% I nearly had a freakin heart attack but at 6mp each or at 11*14 ...... they look the same.

09-04-2009, 03:03 PM	#17
NewportPreacher Senior Member Join Date: Jul 2007 Location: Phnom Penh, Cambodia Photos: Gallery Posts: 150	Very interesting discussion and timely, since I've had my K20D for a week now & moved up from the K100D. I noticed that, when viewed at 100%, noise at ISO 400 on the K20D looked a little worse than a comparable photo from the K100D, shot at ISO 800 & viewed at 100%. But, I rationalized that nobody's going to be viewing it at 100% "in the real world" & even larger prints would give me more than acceptable results. Unless my quick reading of this thread was way off, that seems to fall in line with what most all of you are saying.

09-05-2009, 02:48 AM	#18
SteveB Veteran Member Join Date: Mar 2007 Location: U.K. Posts: 685	Low light AF performance on the K100D is its only problem really for a 3yr old camera. Noise levels aside, one very good low light characteristic of the K100D is its ability to maintain good colours (assume RAW as AWB isn't great in low light). A lot of cameras really fall down here. Is the K20D (or K7) as good as the K100D in this respect?

09-05-2009, 07:36 AM	#19
kittykat46 Veteran Member Join Date: Dec 2007 Posts: 923	Originally posted by SteveB Low light AF performance on the K100D is its only problem really for a 3yr old camera. Noise levels aside, one very good low light characteristic of the K100D is its ability to maintain good colours (assume RAW as AWB isn't great in low light). A lot of cameras really fall down here. Is the K20D (or K7) as good as the K100D in this respect? The K100D at ISO1600 is good in RAW. Work some Noise Reduction on the RAW file, and you have a very usable photo. The K100D's shortcoming in low-light is its AF. Many times I failed to get an AF lock even in decent indoor lighting....OK for static objects not for anything which may move away. The K-7 looks noisier if you compare 100% crops for ISO1600. But a fairer real-world comparison is to compare print photo to print, or full screen to full screen...No, from that point of view, the K-7 is definitely better with more detail and less visible noise at ISO1600 and ISO3200.

09-05-2009, 07:50 AM	#20
Rondec Loyal Site Supporter Join Date: Mar 2009 Location: Gladys, Virginia Photos: Gallery Posts: 27,652	I have owned the K100 and currently own the K10 and K20. Looking back, I actually shot to iso 3200 with K100 -- a little noisy, but certainly acceptable photos. The K10 can be shot to iso 800, but past that is definitely rough. A lot of noise and sometimes banding crop up. The K20 is really strong up to iso 2000. It is a little difficult comparing the photos as mentioned above. 6 megapixel files are a lot smaller than 14 megapixel files. The question is really one of printing size -- most photos look decent if printed at 4 by 6 size, while if you push the size up -- noise crops up too.

09-05-2009, 11:35 PM	#21
ricardobeat Veteran Member Join Date: Aug 2007 Location: Brazil Posts: 376	High ISO on the K100D really looks much better than on the K10D. I have no fear of using 1600, even going up to 3200 when needed. The grain looks very nice, not like digital blotches, and color is good, you just lose some dynamic range. The K10D (GX-10 in my case) is a whole other story. Shooting at ISO 800 could ruin colors and make for horrible images. That's one of the main reasons I never went past 6000 shots on it in 2 years.

09-06-2009, 09:10 PM	#23
Marc Sabatella Veteran Member Join Date: Sep 2006 Location: Denver, CO Photos: Gallery Posts: 10,685	Originally posted by aerodave So the attached pics are as close to an apples-to-apples comparison as I could do. I agree; what you did here is the best way to do these sorts of tests. Quote: Sure this is one shot, and doesn't tell the whole story. But I think it somewhat refutes the idea that the noise differences between the two cameras even out when you output to the same size. That's exactly what I've done here, and the K10D is still worse. Agreed, but you'll also note that I at least have tried to be careful and always mention in my remarks that, due to an unfortunate quirk in the implementation of the K10D (omitting the analog signal amplification and relying entirely on an apparently questionable analog-digital-converter), the K10D really is worse than the K100D, as well as being worse than other Pentax cameras with the same sensor (K200D and K-m). That's what I've seen in m own looking at well-done tests like this, and I believe that's what the folks who've actually measured the noise statistically have found as well. Still, I also think this comparison does illustrate my point pretty well - while the K10D is indeed worse, the difference is rather less when comparing apples to apples as you are here than when comparing at 100%. And showing the full images too really helps put into context just how small those differences actually will turn out to be in practice. And yes, I do think programs with better NR algorithms than LR could probably bring the K10D image more in line with the K100D, but not to the same extent that would be possible for the K200D or K-m.

09-06-2009, 09:29 PM	#24
gfmucci Forum Member Join Date: Jun 2009 Location: The Villages, FL Posts: 70	Aerodave: Thanks for taking the time to give us that apples to apples comparison. It shows how much better "doing it" is than just "talking about it." I appreciate that.

09-07-2009, 06:22 AM	#25
shaolin95 Senior Member Join Date: Jan 2008 Posts: 201	Very nice. I need to do the same but I had a rough weekend since I just moved. I will test the K20D and the K100D both with the K20D kit lens.

09-07-2009, 06:26 AM	#26
aerodave Veteran Member Join Date: Aug 2007 Location: Dayton, OH Posts: 365 Original Poster	Marc, I think you're right about the K10D being a special case and unnecessarily worse than pretty much any other Pentax with the same sensor. And I also believe that I'd get better performance out of the K20D or K-7. Can't wait to see first-hand how the K-7 does. =)

Thread Tools	Search this Thread
	Search this Thread: Advanced Search

09-07-2009, 07:19 AM	#27
interested_observer Loyal Site Supporter Join Date: Jun 2009 Location: Tumbleweed, Arizona Photos: Gallery \| Albums Posts: 5,707	Several months ago I upgraded from the K100 to the K20 for a number of reasons, including the ISO100, additional control (the 2 wheels), more room to crop, along with better menu controls - plus just the price of the K20. I did keep the K100 and the more I use the K20, the more appreciation I have for the K100. Its a great little camera, I really still like the size and it still takes wonderful images. I have also shied away from SDM lenses, so that they are usable across both camera bodies (maybe I should go find a super body - but what I have is more than sufficient, and my lens acquisition is essentially over, other than a DA 21 limited - maybe). On using ambient light, I have gone the other way, rather than going up in ISO speed, I am using slower ISO speeds and taking the hit in shutter speed. This has caused me to significantly upgrade my tripod and head, but I was going to have to do that anyway. I am still learning and the K20 is a wonderful tool to extend my skills with. It is enough camera to satisfy me past the K7 and on to its replacement.