Forgot Password
Pentax Camera Forums Home
 

Reply
Show Printable Version Search this Thread
08-09-2011, 01:09 PM   #16
Pentaxian
audiobomber's Avatar

Join Date: May 2008
Location: Sudbury, Ontario
Photos: Albums
Posts: 6,631
"Such lenses as the Sigma 50 mm f/1.4 EX DG HSM or the Sigma 85 mm f/1.4 EX DG HSM, launched not so long ago, turn out to be optically better than their equivalents produced by Canon, Nikon, Pentax, or even Zeiss."

This seems a provocative statement. I didn't see the proof in the article. It looks to me like the DA 40mm graph was superior to the other lenses on the page. If the Sigma is so much better it would have been interesting to see the proof.

08-09-2011, 03:19 PM   #17
Pentaxian




Join Date: Jun 2009
Location: GMT +10
Photos: Albums
Posts: 10,576
Original Poster
QuoteOriginally posted by audiobomber Quote
I didn't see the proof in the article.
I think they were probably referencing more than just the colour responsiveness of those lenses. They have reviewed those Sigmas and similar Nikons etc in more detail elsewhere, so maybe that previous work was in the back of their mind. Ditto for their SamYang comments:
Lenses reviews - Lenstip.com
08-09-2011, 03:28 PM   #18
Pentaxian
jsherman999's Avatar

Join Date: Dec 2007
Photos: Gallery
Posts: 8,228
QuoteOriginally posted by audiobomber Quote
"Such lenses as the Sigma 50 mm f/1.4 EX DG HSM or the Sigma 85 mm f/1.4 EX DG HSM, launched not so long ago, turn out to be optically better than their equivalents produced by Canon, Nikon, Pentax, or even Zeiss."

This seems a provocative statement. I didn't see the proof in the article. It looks to me like the DA 40mm graph was superior to the other lenses on the page. If the Sigma is so much better it would have been interesting to see the proof.
.


With regards to color transmition - here's one comparison they gave:





Keep in mind that the visible spectrum ends at 700nm, so the Sigma beats or matches the Nikon across the entire visible range.





Of course, right off the top they showed what's possibly the best color transmitter they tested - the DA 40ltd.



Last edited by jsherman999; 08-09-2011 at 09:47 PM.
08-09-2011, 04:42 PM   #19
Veteran Member
Laurentiu Cristofor's Avatar

Join Date: Nov 2009
Location: WA
Photos: Albums
Posts: 3,044
QuoteOriginally posted by northcoastgreg Quote
Using AWB with old lenses sometimes leads to curious results. AWB with the Pentax K 28/3.5 tends to over-compensate with magenta tint, leading to rich, purplish blues. Of course, these can be corrected in post, where one can achieve distinctive and stunning, if not always fully accurate, results. Another two lenses that affect AWB in non-intuitive ways are the M 50/1.7 and, to a lesser extent, the K 35/3.5. In my experience, both of these lenses cause the AWB to over-compensate with a cooler temperature, giving a slighty cold, bluish cast to the image.
Yes, I noticed similar results with adaptall lenses. Now I have an explanation of why AWB can get confused when using certain lenses.

QuoteOriginally posted by bdery Quote
I thought the Samyag sucked based on those graphs. sure it has a great peak value, but the curve is far from being flat, so colours will be uneven.
You just have to check the graph between 400 and 700nm - the rest of the curve would not matter for regular photography (the curve over 700nm will matter for IR photography only). The Samyang has excellent transmission from 420 to 650nm and it's not bad in the rest of the range either. The DA40 curve looks better, but note that at 500nm, the Samyang has better transmission even than the DA. Also, the DA is a 5 element optical formula, while the Samyang has 8 elements, so it's not unexpected for the DA to have an overall higher transmission curve.

08-09-2011, 06:41 PM   #20
Pentaxian
bdery's Avatar

Join Date: Mar 2008
Location: Quebec city, Canada
Photos: Gallery
Posts: 5,654
QuoteOriginally posted by Laurentiu Cristofor Quote
The Samyang has excellent transmission from 420 to 650nm and it's not bad in the rest of the range either. The DA40 curve looks better, but note that at 500nm, the Samyang has better transmission even than the DA. Also, the DA is a 5 element optical formula, while the Samyang has 8 elements, so it's not unexpected for the DA to have an overall higher transmission curve.
Trust me, I know how to read a curve

Having a better peak peformance is less important than having an even performance.

The number of elements, if well coated, will have little impact on the transmission curve. And if the right types of glass are used, it can improve the other optical aspects that make a lens good or bad.
08-09-2011, 07:56 PM   #21
Veteran Member
Laurentiu Cristofor's Avatar

Join Date: Nov 2009
Location: WA
Photos: Albums
Posts: 3,044
QuoteOriginally posted by bdery Quote
Having a better peak peformance is less important than having an even performance.
The Samyang graph is not uneven in any significant way. It pretty much spreads over the same 10% MAX-MIN difference as other modern lenses if you look between 450 and 650nm, and it's pretty much always over 80% - that is pretty even. Just the fact that the DA40 has a smoother looking graph doesn't mean the Samyang is poor.
08-09-2011, 09:51 PM   #22
Pentaxian
jsherman999's Avatar

Join Date: Dec 2007
Photos: Gallery
Posts: 8,228
How difficult is it to come up with these?

You know... do these curves appear on any other lens review site? It seems they'd be a very popular data item on something like photozone, topics for endless mine-is-better-see? discussions, linking, etc.

I'd love to see the curve for each Pentax lens, for example, and it would be very interesting to compare equiv focal lengths from different eras directly to see improvements coatings brought...
08-10-2011, 04:51 AM   #23
Pentaxian
bdery's Avatar

Join Date: Mar 2008
Location: Quebec city, Canada
Photos: Gallery
Posts: 5,654
QuoteOriginally posted by jsherman999 Quote
How difficult is it to come up with these?
Not very difficult if you have the correct rig : monochromator, reliable "white" light source. Actually the manufacturers have the data for all the lenses they design, it's a by-product (or a tool) of numerical lens design.

08-10-2011, 05:08 AM   #24
Pentaxian




Join Date: Jun 2009
Location: GMT +10
Photos: Albums
Posts: 10,576
Original Poster
QuoteOriginally posted by jsherman999 Quote
I'd love to see the curve for each Pentax lens, for example,
Well, Lenstip has reviewed about 10 Pentax lenses, and they have transmission graphs for most of them it seems - eg here's a link to graph for the the DA35 2.4 AL:
Pentax smc DA 35 mm f/2.4 AL review - Ghosting, flares and transmission - Lenstip.com

the FA43:
Pentax smc FA 43 mm f/1.9 Limited review - Ghosting, flares and transmission - Lenstip.com

the FA 50 1.4:
Pentax smc FA 50 mm f/1.4 review - Ghosting, flares and transmission - Lenstip.com

etc
08-10-2011, 10:50 AM   #25
Veteran Member
Laurentiu Cristofor's Avatar

Join Date: Nov 2009
Location: WA
Photos: Albums
Posts: 3,044
QuoteOriginally posted by jsherman999 Quote
You know... do these curves appear on any other lens review site? It seems they'd be a very popular data item on something like photozone, topics for endless mine-is-better-see? discussions, linking, etc.

I'd love to see the curve for each Pentax lens, for example, and it would be very interesting to compare equiv focal lengths from different eras directly to see improvements coatings brought...
lenstip has a spectroscope to do this analysis. Regular reviewing is done just by shooting charts. So I don't expect anyone else to start doing this, though I share your interest in seeing transmission charts for more lenses. On the other hand, not having this information means we'll not have nitpicking discussions around lens coatings and such, so it may all be for the best.
08-10-2011, 07:08 PM   #26
Site Supporter




Join Date: Jul 2008
Location: Detroit MI, USA
Posts: 502
QuoteOriginally posted by bdery Quote
Not very difficult if you have the correct rig : monochromator, reliable "white" light source. Actually the manufacturers have the data for all the lenses they design, it's a by-product (or a tool) of numerical lens design.
Your job sounds fun now that I am into lens . However it seems good optical lens design as far as light transmission in the visible spectrum is still somewhat of an art. Or in other words you have a lot to discover

I know I see better color from my simple Sears made by Ricoh MC 50mm f/1.7 lens versus my Sigma 17-70mm f/2.8-f4.5. I can't put my finger on it but its there.

I don't agree with DPR or any other entity saying WB can be made to match between two different lens, say the above lens. If you made the white balance the same for both the lens every single color/pixel will be effected and that's not a good thing. In fact that may be the very problem. Some P/S cameras seem to have a magic with their color response (look at the Kodak P880; has a cult following). Kodak could tune the colors/hues in PP for that lens. It does not need to worry about any other lens. a dSLR has to assume as you say all lens are perfect and correct for WB assuming all lens attached are perfect, when we know they are not.

I guess this is the cause of color cast between lens? You can make whites white, between two different lens, but in doing so you alter every single pixels color in the pic for bad or good; probably a mix of both.

Could the next step for in camera lens aberration correction be built in software that corrects the color balance, measured for each lens used on a camera. Or the lens come with the data already in it. Measured by an international standard.
08-11-2011, 04:37 AM   #27
Site Supporter




Join Date: Jul 2008
Location: Detroit MI, USA
Posts: 502
QuoteOriginally posted by rawr Quote
FYI.

Just came across a recent article on Lenstip.com that attempts to explain some of the issues around colour rendering in lenses and binoculars. They mention Pentax several times - eg as the pioneer of lens multicoating and even have a light transmission chart for the DA40 as an illustration of the good work modern multilayer coatings can do.

Colour rendering in binoculars and lenses - Colours and transmission - Lenstip.com

An interesting read.
BTW I just thought Lenstip is using the same tool they measured UV filters with. The best to mid range UV filters all had relative to lens very flat response and little light degradation. This shows using the best tested UV filters, Hoya PRO 1 Digital and a few other Hoyas at worse you can expect a 1%-5% loss at the ver edges of the visible light spectrum. I doubt considering the scale unless a function shows other wise that this would ever be visible to the human eye.

Article shows +1 for filters...

Or does it show that adding a filter degrades less than perfect lens even more, how much does a -5% loss at 690nm make? I say very little!


now shoot and as I drop down low to hide from the bullets of UV filter non-believers. LOL

It also shows there is much more to a good lens then the resolution charts we really like looking at; and put a lot of bias on, show at most review sites. Do most know because we are very sensitive to contrast in the 30 LW/PH, (my research) that in this range we see sharpness best, that reviewers have shown some proof manufactors know this and when one of their lens are put on may boost the contrast in this range to make the lens appear more sharp! (Amateur Photographer, 2005 lens reviews; review upon request).

Tests such as those done at Photozone can miss key aspects of a lens performance because it does not show a scale rather a snap shot at the most demanding region of resolution that depending on the lens can mean very little! Again my well liked DA55-300mm, read ephotozine review on it. Note where it performs well (in LW/PH), now Photozone shows results at 50 LW/PH. The lens could be superb from 40 LH/PH down and this will make a sharp lens appear soft in Photozones tests. Klaus (of Photozone) however to his credit will note when a lens does show good to strong contrast and good field performance (as in the DA55-300mm) or if it drops in contrast!.

People including me are at times of judging a lens (such as I did with the DA18-135mm WR) with simple data and not looking at the whole picture (pun, yes ). Because there is not enough data in any one test that I know of to judge a lens. The closes is the one people hate the most PopularPhotographys SQF method. Because it looks at the whole picture (sans color) and rates on prints. Also SLRgears blur chart will closely show you what your eyes see in terms of sharpness (dpr as well).

Last edited by jamesm007; 08-11-2011 at 04:08 PM. Reason: ephotozine tested the DA55-300mm, not IR
08-11-2011, 04:59 AM   #28
Pentaxian
bdery's Avatar

Join Date: Mar 2008
Location: Quebec city, Canada
Photos: Gallery
Posts: 5,654
QuoteOriginally posted by jamesm007 Quote
Your job sounds fun now that I am into lens
It is a great deal of fun. However, I'm not a photographic lens designer. I actually design fiber optics test and measurement tools for telecomms. So when Verizon fixes your broadband connexion, they're probably using stuff we design and build. But I had to study physics in general, and optics in particular, to get there

QuoteOriginally posted by jamesm007 Quote
However it seems good optical lens design as far as light transmission in the visible spectrum is still somewhat of an art
Not only in the visible spectrum. Lens design in general is an art once you understand the "basic" science. It's something you have to practice to become good at (and I won't pretend I'm particularly good at designing lenses).

QuoteOriginally posted by jamesm007 Quote
I don't agree with DPR or any other entity saying WB can be made to match between two different lens, say the above lens. If you made the white balance the same for both the lens every single color/pixel will be effected and that's not a good thing. In fact that may be the very problem. Some P/S cameras seem to have a magic with their color response (look at the Kodak P880; has a cult following). Kodak could tune the colors/hues in PP for that lens. It does not need to worry about any other lens. a dSLR has to assume as you say all lens are perfect and correct for WB assuming all lens attached are perfect, when we know they are not.

I guess this is the cause of color cast between lens? You can make whites white, between two different lens, but in doing so you alter every single pixels color in the pic for bad or good; probably a mix of both.
It's a way of putting it, yes. Preset WB settings only assume your scene is a "blackbody" (ask wikipedia for it, maybe) with a given temperature, thus a given colour cast. But your scene is not really a blackbody, and your camera does not capture light evenly. WB processing tries to account for that, but there are too many variations for a small processor to handle them all.

QuoteOriginally posted by jamesm007 Quote
Could the next step for in camera lens aberration correction be built in software that corrects the color balance, measured for each lens used on a camera. Or the lens come with the data already in it. Measured by an international standard.
We could imagine this, but I doubt it's worth it.
08-11-2011, 10:33 AM   #29
Inactive Account




Join Date: Dec 2008
Location: Ames, Iowa, USA
Photos: Albums
Posts: 2,965
QuoteOriginally posted by newarts Quote
If manual white balance adjusts the R, G, and B channels independently it ought to do a good job I think... not perfect of course and it can never correct certain light sources adequately but it ought to do real well on old yellowish cast lenses; ie on coloration which is smooth like due to single or no coatings rather than color casts with sharp peaks like mercury vapor lamps etc.
Here's an extreme example of correcting tinting by white balance. A white and yellow panel were placed side by side on my laptop screen and photographed first using the white panel for white balance then using the yellow panel for white balance.


As expected, white balancing on the yellow panel turned it (almost) white (and removed the equivalent yellow from the originally white panel).

This clearly demonstrates that manual (and I suspect auto-) white balancing is different from color temperature shifts.

Next I'll try to demonstrate it with a yellow filter placed over a lens.
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!
article, binoculars, colour, k-mount, lenses, pentax lens, slr lens, transmission
Thread Tools Search this Thread
Search this Thread:

Advanced Search


Similar Threads
Thread Thread Starter Forum Replies Last Post
The recent "Fast Sports Zoom" article Entropy Pentax SLR Lens Discussion 6 02-14-2011 02:20 PM
Pentax Shooter uses 6x7 for recent Magazine Article desertscape Pentax Medium Format 8 01-28-2011 01:13 PM
article about recent QC issues with Sigma rfortson Pentax SLR Lens Discussion 7 10-22-2008 01:48 PM
3D-like Rendering and XYZ Lenses? Frank Fletcher Pentax DSLR Discussion 6 07-03-2008 11:15 PM
question about lenses rendering depth/3d jake123 Pentax SLR Lens Discussion 16 05-11-2008 07:51 AM



All times are GMT -7. The time now is 07:54 PM. | See also: NikonForums.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