Forgot Password
Pentax Camera Forums Home
 

Reply
Show Printable Version Search this Thread
11-07-2013, 10:40 PM   #61
Pentaxian
audiobomber's Avatar

Join Date: May 2008
Location: Sudbury, Ontario
Photos: Albums
Posts: 6,676
QuoteOriginally posted by ScooterMaxi Jim Quote
Which 6 mp Pentax dSLR was used in the comparison? It would be rather deceptive to make a reference to a comparison that simply does not actually exist in the context of the thread. I'm not aware that anyone here has brought Nikon into the discussion...
I brought Nikon into the discussion. DXOMark measured the Nikon D40 sensor, which is the same sensor used in the Pentax 6mp cameras. Sensor ratings across brands are consistent and immaterial, as shown in many tests. For example check the scores for the D200 vs. K10D, K-x vs. D90, K-5 vs. D7000, etc. All used same generation Sony sensors, and no significant differences between camera brands.

The OP can make up his own mind based on what's been said.

11-08-2013, 10:04 AM   #62
Site Supporter
GeneV's Avatar

Join Date: Oct 2008
Location: Albuquerque NM
Photos: Albums
Posts: 9,777
I don't know that one can assume that a sensor made by the same manufacturer for two different companies is the same or that there aren't other aspects of the construction which change the low light performance. This review even rates the Pentax K100d as having more effective MP than the D40. Nikon D40 and Pentax K100D Compared I've had both Pentax models, and will believe my lying eyes over a test of another camera brand.
11-08-2013, 10:34 AM   #63
Pentaxian
ScooterMaxi Jim's Avatar

Join Date: Jul 2009
Location: Minnesota
Posts: 1,498
QuoteOriginally posted by GeneV Quote
I don't know that one can assume that a sensor made by the same manufacturer for two different companies is the same or that there aren't other aspects of the construction which change the low light performance. This review even rates the Pentax K100d as having more effective MP than the D40. Nikon D40 and Pentax K100D Compared I've had both Pentax models, and will believe my lying eyes over a test of another camera brand.
I completely agree, and will add another wrinkle. That is, the 6 mp sensors were around for quite a long time, and given how quickly developments were taking place 8-10 years ago, it is highly unlikely that the *istD series had the same sensor as the K100 series (or the D70 and the D40 for that matter). I am just as certain that the 10 mp sensors were not a single type from Sony - especially when you examine the Nikon entries. For instance, the f/1.4 chart shows a huge difference between the D60 and the D200 on a very importance characteristic - light acceptance:
F-stop blues - DxOMark
11-08-2013, 11:43 AM   #64
Pentaxian
Na Horuk's Avatar

Join Date: Feb 2012
Location: Slovenia, probably
Photos: Gallery | Albums
Posts: 10,819
So, Im not reading the 5 pages, but if you want great low light performance for cheap, get a K-01. Keep your DSLR, and you've got all your needs covered. Speed and OVF with DSLR, stunning quality and compact build with K-01. And they both accept the same lenses. Bonus points if you get the K-01 with 40mm XS - great lens, super compact, bundled for very cheap

11-08-2013, 11:55 AM   #65
Pentaxian
audiobomber's Avatar

Join Date: May 2008
Location: Sudbury, Ontario
Photos: Albums
Posts: 6,676
Gene, I gave three solid examples that back up my statement of similar performance between brands using the same generation Sony sensors. If there is a difference between manufacturers, where is it shown?

DPR shows 3008x2008 pixels for the Pentax, and 3008x2000 for the Nikon. A small difference in megapixels is irrelevant, it's just how the manufacturer implemented.

QuoteOriginally posted by ScooterMaxi Jim Quote
I am just as certain that the 10 mp sensors were not a single type from Sony - especially when you examine the Nikon entries. For instance, the f/1.4 chart shows a huge difference between the D60 and the D200 on a very importance characteristic - light acceptance:
F-stop blues - DxOMark
What you're alluding to is not a sensor parameter. DXOMark rates the D60, D200 and K10D the same, for DR, SNR and Tone. The graphs for SNR are on top of one another, except where Pentax applied NR to raws.
Compare cameras side by side - DxOMark

Last edited by audiobomber; 11-08-2013 at 12:01 PM.
11-08-2013, 12:09 PM   #66
Veteran Member




Join Date: Jan 2011
Posts: 440
I'll second the K-01 !!
For low light time exposures, there's just no comparison! Get one and you'll be much happier in not having to run images through noise reduction software to eliminate the annoying chroma noise of the old sensors. The LCD finder is a bonus & actually works better imho than optical finders in the dark..
11-08-2013, 02:52 PM   #67
Pentaxian
ScooterMaxi Jim's Avatar

Join Date: Jul 2009
Location: Minnesota
Posts: 1,498
QuoteOriginally posted by audiobomber Quote
Gene, I gave three solid examples that back up my statement of similar performance between brands using the same generation Sony sensors. If there is a difference between manufacturers, where is it shown?

DPR shows 3008x2008 pixels for the Pentax, and 3008x2000 for the Nikon. A small difference in megapixels is irrelevant, it's just how the manufacturer implemented.



What you're alluding to is not a sensor parameter. DXOMark rates the D60, D200 and K10D the same, for DR, SNR and Tone. The graphs for SNR are on top of one another, except where Pentax applied NR to raws.
Compare cameras side by side - DxOMark
Absolutely incorrect. Light acceptance is entirely related to how the sensor responds to oblique lighting angles from wide apertures. You are confusing the lack of sensor testing thoroughness by disregarding the more esoteric aspects of sensor design. The sensor does not exist in isolation, but responds to various inputs unique to what each lens presents. I am using your beloved and slavish devotion to DXOmark to expose the folly of the argument. Please read (and comprehend) what DXOmark has stated regarding "F-stop Blues" and its appropriate analysis regarding sensor response. Take special note of what the chief scientist stated as "some of the gain from wider lens openings seems to be offset by the present state of sensor technology." The sensor technology difference (related to microlens design and light well capacity) between the D200 and D60 is marked - 45% of a stop loss versus 75% of a stop loss at f/1.4. This is outside the realm of the standard DXOmark sensor testing procedure - as implied in the article.
11-08-2013, 07:35 PM   #68
Pentaxian
audiobomber's Avatar

Join Date: May 2008
Location: Sudbury, Ontario
Photos: Albums
Posts: 6,676
QuoteOriginally posted by ScooterMaxi Jim Quote
This is outside the realm of the standard DXOmark sensor testing procedure - as implied in the article.
Yes, and only relevant at f1/4 and wider, not to do with this discussion.

11-08-2013, 10:20 PM   #69
Pentaxian
ScooterMaxi Jim's Avatar

Join Date: Jul 2009
Location: Minnesota
Posts: 1,498
QuoteOriginally posted by audiobomber Quote
Yes, and only relevant at f1/4 and wider, not to do with this discussion.
Well, the discussion was regarding the 6 and 10 mp Pentax sensors, and you quietly brought Nikon into the discussion, which might not match the spec of the Pentax sensors. Realizing that a few of us didn't think you should assume that the sensors are the same you then indicated that you were sure they are. At that point, I gave you the full explanation based entirely on DXOmark sensor analysis data that indeed not even all Nikon 10 mp are the same. Apparently you now understand they are indeed different.

But now a new misstatement that this is relevant only at f/1.4 and wider. No, the article clearly shows the trend line of differing light acceptance among sensors starts before f/2.8 - and the problem is an extreme 66% worse when comparing those two 10 mp sensors - and that is a huge comparative deficit when shooting wide open (low light capability being the reason for the OP). You are right about one thing - and that is the standard DXOmark rating shows no apparent difference between these two sensors with its own standard measurement system.

Dan, you have been very helpful as foil in proving my point that the DXOmark standard of measuring sensors simply isn't adequate at covering all the angles. At this point, I am pretty sure everyone else reading this thread gets it.
11-09-2013, 09:21 AM   #70
Veteran Member




Join Date: Mar 2007
Location: U.K.
Posts: 685
QuoteOriginally posted by Darley Quote
Another example - compare low light performance of a compact with a DSLR. No comparison. My brand new Canon SX160IS 16MP takes great pics in good light (brilliant for macro) but is a total waste of time when the light fades. Look also at Olympus 4/3 sensors which were too small and led to poor dynamic range when they tried upping the number of pixels. Where are Olympus DSLRs now?

Old full-frame cameras like the Canon 5D with only 12MP easily out-perform modern 20MP APS cameras because they gather more light. No matter how good the image processing or ISO performance, surely size matters?

Let us assume I'm right for now (!) and get back to the question of which is the better of the 6MP cameras. I've seen a vote for the 100D. Any more? I'll get one and if I'm wrong I'll come back and say so
A friend until recently used a 5D for wedding photography. Nice images, he gave me some RAW files, but you couldn't hardly push the exposure in PP because the noise in dark areas was quite bad, and I'm talking about dark suits in broad daylight outdoors. My K100D wasn't any worse shall we say.
11-09-2013, 10:09 AM   #71
Pentaxian
ScooterMaxi Jim's Avatar

Join Date: Jul 2009
Location: Minnesota
Posts: 1,498
QuoteOriginally posted by SteveB Quote
A friend until recently used a 5D for wedding photography. Nice images, he gave me some RAW files, but you couldn't hardly push the exposure in PP because the noise in dark areas was quite bad, and I'm talking about dark suits in broad daylight outdoors. My K100D wasn't any worse shall we say.
As a former Canon 5D shooter, I recall that underexposure caused big problems. My biggest problem was the softness at high ISOs - anything higher than about 800. Properly exposed the camera was great at 100-200 ISO as you would expect from full frame. However, my only Canon body now is the 20D which really wasn't far off in image quality compared to the 5D on a per-pixel basis (but on a small sensor camera). It still would be my preference for some action shooting and more-difficult TTL flash shooting (especially macro). Those situations are pretty few and far between these days. More than anything, I keep it because the 200 f/2.8 is an absolutely great lens - even with a TC, and on occasion I'll have a potential customer asking what I shoot, and its best to say Pentax and Canon.
11-09-2013, 06:01 PM   #72
Site Supporter
GeneV's Avatar

Join Date: Oct 2008
Location: Albuquerque NM
Photos: Albums
Posts: 9,777
QuoteOriginally posted by audiobomber Quote
Yes, and only relevant at f1/4 and wider, not to do with this discussion.
Dan, other than that it puts a big dent in your assumption that all contemporaneous sensors are the same, no. You can come up with examples where sensors made for different manufacturers behave similarly and examples where they don't. There is no hard evidence or rigorous logic behind that assumption, yet you tell us we should not believe what we see.
11-09-2013, 11:30 PM   #73
Veteran Member




Join Date: Jan 2010
Location: Southern England
Posts: 495
As others have mentioned, larger pixels will collect and convert more photons, so there will be lower "photon noise" associated with each pixel (this assumes identical technology). However, if you combine (or share) the output from several small pixels, you end up with the same noise characteristics as a single large pixel. This is known as "binning". And this is effectively what happens when you downsize an image in order to display it on a computer screen or when it gets printed.

So, for any given displayed image size, there will be near-identical levels of noise. Subjectively, though, the higher pixel count sensor will yield a more acceptable appearance to the noise. There is everything in favour of high pixel counts - there are no drawbacks, only gains (better resolution, less susceptibility to moire). That is, apart from the problems associated with handling large amounts of data - and this is probably the main reason why we don't have higher pixel counts than we already do.
11-11-2013, 04:30 PM   #74
Senior Member




Join Date: Feb 2009
Location: Lévis, Canada (Québec)
Posts: 143
QuoteOriginally posted by Darley Quote
My question is therefore two-fold: do people agree that fewer pixels gives better low-light performance, and if so which is the best of the 6MP bodies to go for? I'm also looking at Samsung (dunno if I'm allowed to use that word here!) seeing as they were basically the same.

So far I'm looking at cheapish bodies on feebay and have spotted a K110D, and *ist DS and a SS GX 1L. Any one of these better than another or is there something else I should be looking to get? Or am I deluded about the low-light performance of older cameras?

Thanks in anticipation
Well, lower pixel count means larger pixels, which usually leads to a better signal/noise ratio, which is good for low noise.

However, technology improves fast, so fast that today's sensors with larger pixel count actually produce better images than sensors with bigger pixels, unless these sensors with bigger pixels are those of recent cameras (like the sensors found in the D4 or 1DX, which have big pixels compared to the APS-C D7100 or 70D).

I have the K10D, K-7 and K-5, and I can say the noise per pixel (not the overall noise, but the noise when looked at 100% magnification) of the K-5 beats hands down the noise per pixel of the K-7, which in turns beats the K10D hands down too, LoL.

I even used to have a K100D, but I sold it shortly after I had bought it used, because the shutter was much louder than that of the K10D (thanks to the K10D's weather seals, most probably). Was the K100D better than the K10D? Yes, the K100D was as good at ISO 3200 as the K10D was at ISO 1600, which proves the lower pixel count does matter to some point. But the lower pixel count of the K100D meant the noise became visible much sooner as I enlarged prints, and the K10D would beat the K100D as soon as I produced to anything equal or above the 8x12 inches size print. And when I print, I print at least 8x12 inches, so... I sold the louder K100D.

Most important, keep in mind that both cameras were released pretty much at the same time, so they had similar technology in them. But newer technology means better pictures (noise levels-wise), and pics of my K-7 at ISO 1600 still look better than those of the K100D at the same ISO.

In fact, while I used my K10D at ISO 1600 with mixed feelings (I often had no choice, being a stage performance photographer), I did use the K-7 at ISO 1600 without too much worry and I can now use my K-5 at ISO 6400 with pleasure and make poster size prints from these files (but then, I shoot RAW and process my images to remove the unwanted noise, and with the best software for that IMHO).

Even better, more pixels means the noise, if any, will look finer on prints (and on almost all LCD monitors), except when reaching the limits of the sensor's resolution potential.

If you want a camera that has a very good noise control and not too many pixels, look for some used D700s. They're much cheaper now (for FF DSLR cameras), and they have "only" 12 Mpix. But they're Nikons, so you'll need new lenses...

Bottom line, used K100Ds are so cheap, I'd recommend you go for it if you want less pixels. Worse case scenario, you'll be left with a backup camera that doesn't perform as well as you thought. Best case scenario, you'll have the resolution / file size compromise you're looking for.

Last edited by tigrebleu; 11-12-2013 at 11:10 PM.
11-14-2013, 01:01 PM - 1 Like   #75
Forum Member




Join Date: Oct 2013
Posts: 93
Original Poster
Follow-Up (Unscientific Experiment!)

I thank you all for some interesting and informative posts. I started this thread believing that CCD sensors are better than CMOS, fewer pixels are better than more (to a point) and learned that image processing is perhaps more important than both.

As a result I decided to do a direct comparison between a K200D and a K-x. The K-x is newer, has more pixels, and of course a CMOS sensor. I set up the two cameras and used one lens (18-55 DA II) and took about 50 shots at various settings, using identical settings where possible (the two cameras have different pre-set shutter speeds so this complicated things). I took the RAW images and converted them to JPEG in Photoshop, then resized them. There is no cropping or editing of any sort - basically SOOC images. I looked at all the shots over and over and came to a conclusion, but I won't say what yet! I have uploaded 8 sample images taken at f/11 and ISO400 and ISO1600 so you can draw your own conclusions (if you're interested!) as to whether the smaller, older CCD sensor is any better or worse in low light than the newer, larger CMOS sensor. There will be slight differences in focus as it's hard to focus both cameras the same way, but I am more interested in the way light and shadow are captured - dynamic range? The larger images are on flickr K-x / K200D - a set on Flickr The images are at 18mm and 55mm.







The K-x definitely gives more accurate colours/colors (both cameras were set to AWB). However in all other respects I think the images are inferior to the K200D's. I think the older camera shows more detail especially in darker areas. The range of light levels captured is greater, which I assume is what dynamic range refers to. However the K200D stops at ISO1600 and in very low light the extra sensitivity of the K-x is not just better, it blows the older camera away - there is no point showing the images as it would be like comparing a toy camera to the K-x. The newer camera captures details in situations where even human eyes struggle, where the K200D just captures fuzzy blobs of weird colour. So I still don't really know which I would find more useful!

Last edited by Darley; 11-14-2013 at 01:28 PM.
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!
bodies, camera, dslr, light, low-light, performance, photography, pixels, question
Thread Tools Search this Thread
Search this Thread:

Advanced Search


Similar Threads
Thread Thread Starter Forum Replies Last Post
Q's low light performance compared to Olympus XZ-1 (or other cams) ChopperCharles Pentax Q 8 12-21-2012 08:41 AM
Question about the DA 14mm for very low light jeepson33s Pentax SLR Lens Discussion 3 06-30-2009 06:01 AM
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



All times are GMT -7. The time now is 01:09 PM. | See also: NikonForums.com, CanonForums.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