Forgot Password
Pentax Camera Forums Home
 

Reply
Show Printable Version Search this Thread
12-14-2014, 05:03 PM   #1
Pentaxian




Join Date: Jun 2009
Posts: 11,913
Sigma 8-16 - DxOMark [or Sigma] quality fail?

I was looking at DxOMark.com's reviews of Sigma lenses today, and found some amazing results for their test of the Sigma 8-16 on Pentax

I'm not sure exactly what they indicate - a QC fail on behalf of DxOMark's testing procedure, or a QC failure on behalf of Sigma ...

DxOMark.com test of Sigma 8-16 with K-3

Pic below related, displaying the resolution of the lens when tested on a K-3 at 8mm f4.5 (wide open) and 8mm at f5.6. Something doesn't look right - at f4 the centre of the lens is a total dead zone for sharpness, while all around it everything is nice and sharp. Then at f5.6 the centre is perfectly sharp but now there is a ring of no resolution around the centre



Now the same lens tested on a K-5 at 8mm, f4.5. Looks OK:



and a Nikon D7100 at 8mm, f4.5. Looks pretty good:



I've tried two different browsers and see the same results for the K-3. Could someone pls verify what I am seeing?

12-14-2014, 05:17 PM   #2
Emperor and Senpai
Loyal Site Supporter
VoiceOfReason's Avatar

Join Date: Jun 2010
Location: Mishawaka IN area
Photos: Albums
Posts: 5,975
Mine is pixel peeping sharp in the center at that fl and aperture.
12-14-2014, 05:57 PM   #3
Pentaxian




Join Date: Jun 2009
Posts: 11,913
Original Poster
QuoteOriginally posted by VoiceOfReason Quote
Mine is pixel peeping sharp in the center at that fl and aperture.
According to DxOMark, you should be seeing razor sharp edges all around a very precise central area of total blur wide open at 8mm. Then at f5.6, a nice, big inverse donut.

I think they must have a robot doing their lens tests now. A human would have realised something was amiss with the sharpness test results on K-3.
12-14-2014, 07:30 PM   #4
Emperor and Senpai
Loyal Site Supporter
VoiceOfReason's Avatar

Join Date: Jun 2010
Location: Mishawaka IN area
Photos: Albums
Posts: 5,975
QuoteOriginally posted by rawr Quote
According to DxOMark, you should be seeing razor sharp edges all around a very precise central area of total blur wide open at 8mm. Then at f5.6, a nice, big inverse donut.

I think they must have a robot doing their lens tests now. A human would have realised something was amiss with the sharpness test results on K-3.
I think they may have had too much alcohol and forgot to put their glasses back on or something.

12-14-2014, 11:39 PM   #5
Veteran Member
maxfield_photo's Avatar

Join Date: Mar 2011
Location: Cincinnati, OH
Posts: 1,214
Shows what DxO knows, it's one of the sharpest lenses in my bag. Oh wait, I forgot, the only true measure of a lens is how well it renders two dimensional test charts. [deleted by the author]

By the way, it's probably because they haven't figured out that the K mount doesn't report focal lengths under 10mm.

Last edited by maxfield_photo; 12-16-2014 at 01:38 PM.
12-15-2014, 09:13 AM   #6
Veteran Member
Edgar_in_Indy's Avatar

Join Date: May 2010
Location: Indiana, USA
Photos: Gallery
Posts: 2,679
[Deleted by author]

The 8-16mm test results do make one wonder about the testing methods though. My copy is also very sharp in the entire center region. And when talking about ultra-wides, I would say it's very sharp, period, with no disclaimers.

Last edited by Edgar_in_Indy; 12-16-2014 at 02:23 PM.
12-15-2014, 09:24 AM   #7
Veteran Member
JinDesu's Avatar

Join Date: Jun 2011
Location: New York City
Photos: Gallery
Posts: 5,632
Must have amazing bokeh with those features.
12-15-2014, 11:57 PM   #8
Pentaxian




Join Date: Dec 2009
Location: Dallas / Yucatan
Posts: 1,171
Does anyone care to venture to say why the lens would test so differently on the Nikon than the Pentax?
Is it purely in the fact that Pentax doesn't register under 10mm lenses?

* Yes, it's true, I'm not familiar with DxO tests (obviously). Just going by the last two images above.


Last edited by yucatanPentax; 12-16-2014 at 12:04 AM.
12-16-2014, 01:19 AM   #9
Site Supporter
Site Supporter
Digitalis's Avatar

Join Date: Mar 2009
Location: Adelaide.
Posts: 10,116
Considering that I had to go through several copies of this lens to obtain a single good lens - I say it is sigma that is at fault here.

The current Sigma 8-16mm f/4.5-5.6 that I work with was sent from japan with a statement certifying it was within specifications. If QC was doing their job, that would have been completely unnecessary.

QuoteOriginally posted by rawr Quote
I think they must have a robot doing their lens tests now. A human would have realised something was amiss with the sharpness test results on K-3.
Or this test would harm the reputation of pentax cameras, after all who wants to buy a pentax when third party lens makers can't make a decent lens for it? less experienced people would see these tests, and make the conclusion that the camera was in some way responsible. Therefore publishing tests with such obvious faults increases sales of Nikon and Canon equipment.

Last edited by Digitalis; 12-16-2014 at 01:27 AM.
12-16-2014, 02:40 AM   #10
Pentaxian




Join Date: Jun 2009
Posts: 11,913
Original Poster
QuoteOriginally posted by Digitalis Quote
Considering that I had to go through several copies of this lens to obtain a single good lens - I say it is sigma that is at fault here.
Hmmm. I expect that DxO have a 'library' of lenses that they accumulate over time and test against different bodies as they come to market. I doubt they buy or borrow a new lens for each test set (which gets harder of course once lenses get discontinued).

On that assumption, the same lens should have been tested on both the K-5 and K-3. Which makes the massive (and simply weird) difference in the test results between the two bodies hard to account for. I say DxO just bungled the test. Someone in the lab bumped the test bench, or the spreadsheet data that feeds their charting software got mangled somehow.
12-16-2014, 03:22 AM   #11
Site Supporter
Site Supporter
Digitalis's Avatar

Join Date: Mar 2009
Location: Adelaide.
Posts: 10,116
I also think we should consider the difference between the K5 and the D7100 - the Nikon D7100 DSLR doesn't have an AA filter, the K5 does. The K3 doesn't have a built in AA filter but that still doesn't explain the bizarre test result.

QuoteOriginally posted by rawr Quote
. I expect that DxO have a 'library' of lenses that they accumulate over time
I have a library of 50mm lenses - I don't have all of them I started out with, I eliminated sub-standard copies and only kept the good copies. The Copy of the Sigma 8-16mm DxO used on the K3 is...well i'm not sure what to make of it... for normal photographic purposes it is nothing short of a complete disaster*.

QuoteOriginally posted by rawr Quote
I say DxO just bungled the test. Someone in the lab bumped the test bench, or the spreadsheet data that feeds their charting software got mangled somehow.
or the software freaked out because of the barrel distortion at 8mm**...it wouldn't be the first time something like that has happened.

* i'm sure someone could make an interesting image or two out of it.
** Let me put it this delicately: at the 8mm setting it is noticeable even in thumbnails.

Last edited by Digitalis; 12-16-2014 at 03:41 AM.
12-16-2014, 03:37 AM   #12
Site Supporter
Site Supporter




Join Date: Mar 2009
Location: Gladys, Virginia
Photos: Gallery
Posts: 22,362
Clearly they have a better copy for Nikon than for Pentax. I don't know that you can say much more than that. They might make better effort to get good copies for Nikon and Canon -- that would make sense, since more of those who purchase their software use Canon and Nikon than Pentax. With the Sigma 35mm f1.4, it tested quite a bit better on the D7100 than the K3. I had always chalked that up to copy variation.

It is a really weird test result. The concept that the border is sharper by a considerable margin than the center is unheard of in lens design. Even things like decentering shouldn't give you that kind of a result. Maybe they drew in magic marker on the center of the front element and then tested it, just to see what would happen.
12-16-2014, 03:47 AM   #13
Site Supporter
Site Supporter
Digitalis's Avatar

Join Date: Mar 2009
Location: Adelaide.
Posts: 10,116
QuoteOriginally posted by Rondec Quote
Maybe they drew in magic marker on the center of the front element and then tested it, just to see what would happen.
If they did that, it would throw the accuracy of every test they have ever done into question. If they posted results like this on a current Canon, Nikon, Zeiss or Leica lens their reputation would take a severe blow.
12-16-2014, 03:50 AM   #14
Site Supporter
Site Supporter




Join Date: Mar 2009
Location: Gladys, Virginia
Photos: Gallery
Posts: 22,362
QuoteOriginally posted by Digitalis Quote
If they did that, it would throw the accuracy of every test they have ever done into question. If they posted results like this on a current Canon, Nikon, Zeiss or Leica lens their reputation would take a severe blow.
Sure. I'm joking. I just can't imagine a situation in which a lens tests with a big soft spot right in the middle of the image circle, but with relatively intact borders. Anything I can think of would tend to effect the image borders more than the center.
12-16-2014, 04:22 AM   #15
Pentaxian




Join Date: Jun 2009
Posts: 11,913
Original Poster
I just noticed that the K-3 test result at 8mm and f4.5 is identical to the flag of Bangladesh:



Perhaps it's a not-so-subtle hint that DxO have outsourced their lens testing to a company in Dhaka.
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!
8mm, centre, f4.5, f5.6, k-3, k-mount, lens, pentax lens, resolution, results, sigma, sigma 8-16 dxomark, slr lens
Thread Tools Search this Thread
Search this Thread:

Advanced Search


Similar Threads
Thread Thread Starter Forum Replies Last Post
Sigma 85 f1,4 or Sigma 70-200 f2,8? Cavalier Pentax SLR Lens Discussion 11 03-22-2014 08:57 PM
Sigma or Tamron all around lens (picture quality) jrcastillo Pentax SLR Lens Discussion 18 10-02-2013 08:09 PM
Sigma 8-16 f4.5-5.6 or Sigma 10-20 f3.5 Cucho Pentax SLR Lens Discussion 9 05-14-2013 12:31 PM
For Sale - Sold: Sigma 50-150mm f/2.8 - $750 or trade for Sigma 85mm 1.4 or Sigma 17-50mm f/2.8 Edgar_in_Indy Sold Items 4 03-29-2013 09:10 PM
Sigma 8-16 or 10-24?? Unruely Pentax SLR Lens Discussion 8 11-11-2010 01:01 PM



All times are GMT -7. The time now is 08:20 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