Web Analytics RiceHigh's Pentax Blog: 12(.3)M Vs 12(.8)M: APS-C K-x Vs FF 5D ISO Measurbation

Friday, January 01, 2010

12(.3)M Vs 12(.8)M: APS-C K-x Vs FF 5D ISO Measurbation

First of all, I would like to say HAPPY NEW YEAR to all of YOU who read here!

As a New Year Gift, I have this measurbation shootout, with all full sized original samples directly out of the cameras posted, as below!

The shootout was carried out at almost the same time, for the same target at the same distance, with (almost) the same lighting condition (diffused daylight indoor). The cameras were mounted on a sturdy tripod, shutter fired with IR remote controller (for K-x) or cable switch (for 5D). Other key essential shooting information as follows. And of course, full EXIF data preserved:

- Pentax K-x with FA 31/1.8 Limited (Effective Film Focal/AOV: 47.6mm), Av Exposure Mode at f/8, Single AF Mode Focusing on the Centre, Finest JPEG, "Bright" Custom Image Mode, Default Settings 0, 0, 0, 1, 1 for Saturation, Hue, High/Low Key Adj, Contrast and *Fine* Sharpness, Auto White Balance, Evaluative Metering, Exposure Compensation +0.5EV, High ISO NR Starting from ISO 800 and Up at Medium (Factory Default), SR Off;

- Canon 5D with EF 50/1.8II, Av Exposure Mode at f/8, Single AF Mode Focusing on the Centre, Finest JPEG, "Portrait Snapshot" Picture Style, Default Settings 3, 0, 0, 0 for Sharpness, Contrast, Saturation and Hue, Auto White Balance, Evaluative Metering, No Exposure Compensation;

Do note that it's my sole intention to keep the colour response, exposure and sharpening etc. as close as possible for the two cameras, so that a more useful and meaningful shootout and comparisons can be achieved.

ISO
K-x
5D
50
Not Available
100
200
400
800
1600
3200
6400
Not Available
12800
Not Available

Remarks:-

1. Click on any of the Thumbnails will download
Large Files in size near 12MPs each;
2. If you still want to view original un-retouched full sized sample, see them here (under the "download" option, select "high res image").

Well, I don't comment anything this time on all the IQ performance, exposure accuarcy and focusing accuracy etc. YOU are the one who measurbate, observe, compare and re-measurbate again, inspect and make up your own conclusions this time! (could be endlessly until the end of the days, forever! ;-D)

P.S. and Acknowledgement: My Special Thanks to Fred, who is a regular reader here, for kindly lending me his K-x with his FA 31 Limited (Vietnam made version), for doing this test. And, a Full Review with Tips and Tricks (and my Recommended Best Settings) for the K-x will follow, at a later stage.

Read Also:-

K-m Olive Vs 5D Pixel Peeping

K-7 Vs 5D Pixel Peeping

K-x Vs K-m ISO Measurbation

My K-m Review, Tips and Tricks

Comments (17)

Loading... Logging you in...
  • Logged in as
Measurbation? Sounds like you're trying to mimic 1001 Noisy Cameras' wit. But, you don't.
3 replies · active 795 weeks ago
Are you kidding? Do you know who invented the term? (And how long that I have been using it?)
I am not sure where Bob got that impression.I don't use the word measurbation or its derivatives and I certainly did not create it. Bob may be confused or perhaps he is trying to create trouble/drama with a hit-and-run provocative and trollish comment. Perhaps Bob had too much wine during the new year celebrations.
Possibly! :-)

As for the term "measurbation", it is derived from the word "measurbator", which was created by our Master Ken Rockwell !
Michael A.'s avatar

Michael A. · 796 weeks ago

Thanks much for this posting and your effort/time.

The FA31, 43 and 77mm FF lenses are really awesome. I'm lucky to have all 3 (& K-x) but still Made in Japan... :o) Have a great 2010 as well!!!

P.S. Landscape color setting on K-x looks best to me on most occasions...
I prefer the edge clarity of the 5D, also the brightness (my eldest son preferred "the brighter one").

However, I see more texture with the K-x and the chroma noise is better.

I look forward to your further tests!
With these particular settings it is noticeable that both use lots of chroma noise reduction (way too much in some places) - 5D has a lousy algorithm. Also the K-x noise reduction pays special attention to edges - makes small detail look sharper, but at the cost of digital artifacts.

I don't like either images.
It would be better if you have shot the test in RAW files and then export to JPG without post processing.
TheGuest(TM)'s avatar

TheGuest(TM) · 795 weeks ago

Well, apparently the 5D has lower Chroma noise reduction than Kx, boiling down to 1 fstop advatage of the 5D. As it should be... Nothing special at all.
Hello Rice, thank you for detailed comparison. What is your opinion on K-x exposure consistency, is it better than K20? (as you increased the Exposure compensation by half a stop, is it at least consistent? with such a compensation, the images taken with K-x still looks darker than that from 5D). Thanks.
3 replies · active 795 weeks ago
The +0.5EV was required as the K-x was not intelligent enough to compensate for the light background. But in general for the metering accuracy and consistency, I am quite sure that it is undoubtedly better than the K20D and actually even a bit better than my K-m. Waiting for my full K-x review and I shall discuss more about its performance and accuracy aspects and overall reliability. Stay tuned. :-)
Rice, when you do a full review of K-x, please check the reliability of PTTL also. I really wonder if PTTL is improved this time.
Checked already in the field ! Not so favourable so far, overall usually too bright and overexposed, too heavily centre-weighted ( but unable to detect dark centred positioned objects - the overexposures are just too terrible sometimes :-( ).
de lageneste's avatar

de lageneste · 795 weeks ago

What I see:

Slightly sharper K-x images (higher shapening?).
Less chroma noise in K-x images (better NR algorithms, the 5D jpeg engine is showing its age).
Complete detail obliteration in the red tones on K-x images (better NR yes, but completely messes up anything red...?)
Presence of heavy Moiré patterns on the 5D images (very light LP filter).

My suggestion:
Just redo the same comparison test in raw with Rawtherapee and everything set to default if you want to do any real IQ comparison.

My conclusion:
The K-x is probably very close to the 5D if not better above ISO 3200 (absence of banding that limits the 5D to ISO 3200 for all practical purposes).
When shot in raw, the 5D still has about 1 stop better high ISO noise and will show marginally better details at the expense of frequent Moiré patterns on fine meshed fabrics.
1 reply · active 782 weeks ago
Reds are oversaturated or there is too much contrast added in PP on K-x.
Bill Merritt's avatar

Bill Merritt · 769 weeks ago

Two month old K-x is giving me fits with blur issues. Seems to be at worst at shutter speeds between 1/60 - 1/125. Above 125 makes sense for obvious reason, but below 1/60? I have tried everything I know about in attempts to correct without success. Two days ago, ran across a site that said Pentax did have a batch of K-x's go out with AF issues. I contacted Pentax who did not admit to that, but did acknowledge they felt I did all I could with it and suggested shipping to their repair facility. We happen to have one about 40 minutes away from our home, so I will be taking it back next week for repair or adjustment. I really like most of the other features on this camera and if not for the current issue with focusing, would agree this is the perfect camera for me. For what it is worth, I have tried using all modes, including M, used SR in ON and OFF mode ( I agree with Ricehigh, this SR is not worthy of a Pentax product, but slightly better than nothing at all), used some of my old M, and A lenses from an old K1000 as comparisons and have concluded it is a camera issue, but no technical testing was used to support my claim. On the other hand, Using this camera with 18-55 kit lens outside in bright sunlight has on occasion produced very nice results. Fingers crossed they can fix it.
2005 vs 2009 years? хД

Post a new comment

Comments by