K10/K100 image quality

[Brad99]

The difference in detail between a K100D with a good lens and any 10MP APS-C camera is absolutely minuscule IMHO, not even worth worrying about. There's only a significant difference once up to 12MP full frame.

 

[neil holmes]

Jpegs from the K10 were dull and lifeless and with an inordinate
amount of fine noise even at ISO400 - I was shooting flowers etc in
slightly overcast and windy conditions, hence the higher ISO
settings. When the sun emerged and the wind died down I tried lower
ISO and adjusting various in-camera settings but still the shots were
mundane at best. So I switched to RAW and P/Pd using both CS2 and RSE
but still there was something fundamentally amiss with most of the
shots. I wondered if it was a duff model as I tried comparison shots
with my DS and the latter's results were much better overall.

Having read so many glowing reports about the K10 and seen so many
examples posted here of really sharp and vibrant images I really
hoped the K10 would be my next purchase and simply couldn't
understand what, if anything, I was doing wrong. I like tro think I;m
reasonably competent but achieving similar quality shots was eluding
my best efforts. I appreciate a day isn't long to get to grips with
any new model but some of the friend's shots, to be honest, looked
equally lifeless to me although he seemed content with the results.

Assuming this could have been a Friday afternoon assembly line job
and most K10s are excellent, should I be seeing a huge difference in
image quality over my DS or am I expecting too much?
I'm really unsure what to do next as I'd like to upgrade from my DS
and the K10 did seem the logical choice until this disappointing
experience. Any guidance or thoughts would be appreciated. Sadly none
of the images survived a recent computer upgrade (lesson learned
there - my other stuff was backed up on CD) or I would post examples.
--

The mind is like a parachute, it works best when open.

Hi

Just curious but was the noise at 400 with the Vivitar? if so could it have been underexposing hence the greater noise?

Most of my photos are 800 iso and up so it was a no brainer for me to get the K100d

I have an ist*D that now gets little use (advantages and all) because I am a jpeg user and love the antishake and larger lcd screen so the K100d is my camera of choice.

If I was shooting at iso 100 or 200 all the time I would have got the K10d.....at those speeds it is simply a far better camera....semi pro versus entry level....800 and up the K100d becomes far more than an entry level camera. (pentax should start selling twin camera kits instead of twin lens kits...would STILL be cheaper than some brands).

I do find that by using +.5 exposure comp (meaning no auto iso for me) iso 3200 gets many more keepers than I otherwise would.

Jetty Road iso 3200 +.5 1/15

neil

link back to flickr
http://www.flickr.com/photos/26884588@N00/

 

[Martyn Welch]

Also, the higher sensor resolution of the K10D makes lens quality
critical. At the 6MP resolution of the K100D lenses need only be
good; to maximize the K10D's IQ they must be excellent.

So buying a K100D (not sure the new K100Super has any advantages for me as the extra features of image doctoring in-camera would never be used) I should see an improvement in jpeg IQ over the DS but comparing this with the K10 using my existing lenses, you believe I'd see little if any improvement?

10 is definitely worse than 6 in terms of noise, but not in terms of
resolving detail.
10 is better at resolving detail on great lenses. Par on average
lenses.

A while back I compared my DS jpeg images (18-55 & 50-200) alongside a borrowed D80's at 200/400 and the D80 seemed to have less noise and a generally smoother appearance, which may well be down to its internal jpeg processing. Its RAW images certainly looked to have more clean and crisp detail, which is why I'd presumed the K10 would perform similarly, but this was proven not to be the case. The D80 required more thought to protect highlights but I was impressed with its qualities using the 18-70 and IS18-200. It would be interesting to compare its results with a prime (expensive) lens and a K10 also so fitted.
--

The mind is like a parachute, it works best when open.

 

[Martyn Welch]

Hi Neil,

Just curious but was the noise at 400 with the Vivitar? if so could
it have been underexposing hence the greater noise?

It was with all lenses. I realise the tendency to underexpose will put shadow detail at risk when brightened and may emphasise noise but I found the borrowed D80 didn't produce these shadow patterns when I deliberately underexposed with that and subsequently brightened the shots.

I believe the Pentax/Nikon 6mp sensors are the same so are their 10mp chips also identical? If so, why does the DS produce these artefacts when I believe the D70 doesn't and likewise the D80/K10? Could this all be down to each company's processing being quite different? I'm purely guessing here as my electronics knowledge is minimal at best.

The mind is like a parachute, it works best when open.

 

[Steve_]

As many on the forums have pointed out, this is simply not true. The
K10D's JPEG engine works wonderfully if you bother to calibrate it
correctly...

Well there you go, somebody on a forum said it. And you said it again (on a forum, no less). Fact established.

Nevermind that I was not comparing your best K10D results with your ACR (or Bibble, for that matter) ones. I was comparing K100D jpeg engine to that of the K10D. If you set both properly (which happens to be bright, -2 saturation for the K100D) and set the K10D as you prescribe, the K100D's results are only slightly worse than is achievable via RAW, where the K10D's are just a bit more off quality RAW results, detail-wise. That is why you are leaning on the sharpening setting so much--the detail just isn't there (in a comparitive sense). And don't even mention VPN...

The 18-55 kit lens looks no worse on my K10D than on the DS to me.

Great, glad to hear it. If you read my comments I never addressed this question. What I said is that better lenses are required to really expoit the K10D's detail advantage. Which seems to be exactly what you are stating, as well, below:

The FA43 f1.9 ltd can't fully show it's capabilities on the 6mp, but
I don't think the 10mp harms existing lenses (at least not in a
noticable way.)

Again, I never said that 10mp harms lenses, just that you need really good ones if you hope to exploit what advantage 10MP may bring.

10 is definitely worse than 6 in terms of noise, but not in terms of
resolving detail.

Detail resolution is precisely what noise replaces in a digital image. You can only hope to resolve more detail with a K10D with the aid of both decent light and a really good lens. These are my points. In your haste to refute them, you managed to neatly restate them (see below):

10 is better at resolving detail on great lenses. Par on average
lenses.

Assuming decent light, of course
--
Pentax K100D w/DA18-55 & 50-200, FA50-1.4, and Tamron 70-300Di; Panny LZ3
http://s90223656.onlinehome.us/

 

[Steve_]

So buying a K100D (not sure the new K100Super has any advantages for
me as the extra features of image doctoring in-camera would never be
used) I should see an improvement in jpeg IQ over the DS but
comparing this with the K10 using my existing lenses, you believe I'd
see little if any improvement?

K100D Super brings the flourine anti-dust sensor coating, the sensor-shaker, and compatibility with the new SDM lenses. As far as I know there are no other differences.

I think you will see an improvement in jpeg mode with your existing lenses with either camera as compared to the DS. The DS's jpeg engine is not nearly as good as the K100D's, and although the K10D will be limited by your glass it will be limited at a level of IQ greater than your DS can deliver. It just isn't likely to be a convincing improvement over the K100D where IQ is concerned. And in low-light, it may in fact be worse.

A while back I compared my DS jpeg images (18-55 & 50-200) alongside
a borrowed D80's at 200/400 and the D80 seemed to have less noise and
a generally smoother appearance, which may well be down to its
internal jpeg processing. Its RAW images certainly looked to have
more clean and crisp detail, which is why I'd presumed the K10 would
perform similarly, but this was proven not to be the case.

The D80, as all current Nikons, has a good jpeg machines. That and quality glass, metering and AF makes for great pictures. And it's not that the K10D is an inherently poor performer--quite the opposite, in fact--just that its jpeg engine leaves a bit more on the table (which can be remedied RAW processing).
--
Pentax K100D w/DA18-55 & 50-200, FA50-1.4, and Tamron 70-300Di; Panny LZ3
http://s90223656.onlinehome.us/

 

[Martyn Welch]

K100D Super brings the flourine anti-dust sensor coating, the
sensor-shaker, and compatibility with the new SDM lenses. As far as I
know there are no other differences.
I think you will see an improvement in jpeg mode with your existing
lenses with either camera as compared to the DS. The DS's jpeg engine
is not nearly as good as the K100D's, and although the K10D will be
limited by your glass it will be limited at a level of IQ greater
than your DS can deliver. It just isn't likely to be a convincing
improvement over the K100D where IQ is concerned. And in low-light,
it may in fact be worse.

This would suggest there's little advantage in my going for the K10D at present.

The D80, as all current Nikons, has a good jpeg machines. That and
quality glass, metering and AF makes for great pictures. And it's not
that the K10D is an inherently poor performer--quite the opposite, in
fact--just that its jpeg engine leaves a bit more on the table (which
can be remedied RAW processing).
--
Pentax K100D w/DA18-55 & 50-200, FA50-1.4, and Tamron 70-300Di; Panny
LZ3

Interested to see your lens line-up. As mentioned, I have the first two on your list and although I've been seriously looking at my Pentax body options, my instincts are starting to direct me toward that D80 with suitable glass and a decent Nikon flashgun as my Vivitar flash is a very hit and miss affair. Even if I went for the genuine Pentax guns, threads here would suggest these leave a lot to be desired whereas the Nikon system seems highly rated by all and sundry.

If money were no object, I'd simply utilise both systems but I don't have the funds to splash out on a change of system and keep the old, hence my dilemma:-

1. Buy a new Pentax body and stay with what I have and maybe buy better glass as cash becomes available.

2. Sell the lot (not likely to retrieve much in p/exchange or even ebay sales) and jump ship, which would be more costly.

My photos are occasionally used in a national UK modelmaking magazine and one could argue that since they're seldom reproduced larger than 10x8, nobody would probably be any the wiser whichever system or body I used, but photography's my main hobby as well and I just want to produce the best pictures I can as a matter of personal pride. It would be great to have total confidence in the equipment I use and at the moment I'm struggling a bit.

The mind is like a parachute, it works best when open.

 

[JerryFn]

Almost there and about to jump for the K10. It's the robust build and the seals that I want, so I can use it in this English weather. Prices look good now at £440

However.. the comparison images here on DP review don't look good, even when compared to the 400xti. Yes they look soft. I also reviewed and downloaded images on Photographyblog and there was a nasty and very noticable purple fringe at high contrast on the top left.. not there on the xti.....

How do folk find the kit lens, I really like to fact that I can use all my old K lenses, but maybe this is a bit academic or stoogy and clearly new lenses will have moved on.

I am not into status kit display or collecting kit, what I want is a slick usable, durable camera that will not let me down and feels transparent in use.

In the end it is the pictures that count.

 

[RickS]

Jerry,

Martyn's findings and experiences are his own and certainly don't reflect my results.

I have an istDS and a K10D, use the same set of about a dozen lenses on both, and have formed a completely different opinion from Martyn in his October 2007 post.

The results from my K10d, both raw and jpeg, are not excessively noisy, are not flat and are definitely sharper than my DS, which has performed well for me over the last three and a half years.

I still use both cameras but the K10D is way out of the DS's league in most areas.

As for the 18-55mm kit lens, mine is extremely good, as are most, judging by the historical comments on this forum. (Don't know about the new 18-55 II, though.)

Rick

 

[JerryFn]

Ok thanks for your assurances, but do I get the feeing that there is a variation in camera quality or is it folks ability to use it.

This image from the samples on Photogra0phy blog does give me some misgivings. It is a full size image but you only need to download the top portion which has considerable purple fringing on the left corner.



Is this the lens or the sensor?

Also with patience I downloaded sample images from the K10 review page on DP review.

Am I being too critical, or does the £350 Canon 400 xti seems much sharper?

Thanks Jerry

 

[Paul Hunt]

I'll be the first to say that the shot demonstrated is not a good example of K10D image quality.

For the moment, we'll leave aside the purple fringe issue.

The thing to realize is that the K10D jpegs, by default and by design, are softer (ie lower degree of edge acutance) as they are intended for further processing. This, for right or wrong, was a Pentax design decision. You can easily get sharper pics out of the camera by changing the image tone from Natural (default) to Bright. Fiddle with Sharpness, Saturation and Contrast slides to suit.

OR shoot raw and have full control over the final appearance.

Don't mistake the softer edge acutance for lack of detail.

Purple fringe is, I think, more a function of the lens - some folk claim that Pentax lenses may be more susceptible to it than some other brands. It will usually occur in high contrast scenes with abrupt changes in light/dark - branches against a bright sky is a typical situation. Purple fringe can be ameliorated, if not removed, by software plugins.

You could look at the K10D with the DA16-45.

Anyway, there are lots of posts on this board with samples of razor-sharp images.

But maybe you prefer the Canon image look? Your choice!

Paul

 

[JerryFn]

Yes well as the fringe is on the top left, it would indicate that it is a problem with the lens. I also did a 'purple fringe' search and that was informative. It seems it is everywhere!

Not a particular fan of EOS 400d but I don't want to spend an unneccesary £100.

I guess it's time for a touchy feely session and that will clinch it!

Thanks Jerry

 

[topgun2007]

I have seen great pictures posted from users of the DS, K100D and K10D. Don't know if you were using a dud K10D or not. If you are only shooting JPEG's and not cropping a lot or producing large pictures, you may not notice an improvement, particularly if the owner of the K10D you used had not adjusted the settings for his JPEG's as suggest by another poster. I had the K100D and actually liked the JPEG's from it, but the difference isn't as large after I made adjustments on my K10D. I have also started shooting more RAW so that isn't as large an issue anyway. I have not noticed an excessive amount of noise at ISO 400 as you mentioned and most of the time I try to shoot between 100 to 400. I do love how easy the K10D is to use over the K100D and having made the switch, don't miss the K100D at all. Having said this, if you are upgrading solely for image quality, but aren't cropping or blowing up your photos or printing larger sizes, you may want to wait until the K20D's price decreases. Or you may just want to try a different K10D, adjust the setting, and see if the results are different. Best of luck.

Regards, Jim

 

[Rupert60]

I have a DS that I love and think it takes every bit as fine a shot as my K10D or even the new K20D, from what I've seen. My DS is in like-new condition yet I almost never use it and instead use the K10D. So why?
I shoot RAW +Jpeg and if set correctly the K10D Jpegs are equal to the DS.

The settings on the K10D are far superior to the DS in many, many ways, giving you better control and more choices. The sealed body gives a little comfort in bad weather.
The extra resolution is great for cropping, and I crop a lot. IMO the colors are better than even the DS, if processed with Silkypix.

It is a little bigger than the DS and easier to hold and work the dials/buttons.

If you get a K10D or, from what I see so far a K20D, I doubt that you will use your DS much afterwards. It will take you a little time to get used to it, as you noted your first results were not satisfactory, but when you "get it" you will be well pleased, I guarantee.
--
'This is more serious than I thought.....but it is still fun!
http://www.pbase.com/rupertdog Take a look- It's Free!