[normhead]

Interesting post....
A quick analysis, does bring up some points showing why my exif data is what it is.
Note that on the above linked list....


Pentax lenses all have a separate identifier. 0ne number...1-4 followed by a number between 1 and 255. (All us binary dudes understand this.)
Many third party lenses have a period and one more number. So in the case of Aperture, it doesn't pay attention to the third number, while clearly Light Room does. But just from the list posted, it would appear that there is no Pentax glass, using the period one number add on non-enclature. So if it anything it suggest Pentax doesn't use the third number in their own lenses. Whether this is Pentax differentiating between themselves and third party licensers, or third party manufacturers doubling up on the numbers assigned to them, (with or without Pentax's approval) can not be determined.

Every third party manufacture has some lenses with the third number. No Pentax lenses do. The fact that that is true would suggest no third party either supports or is allowed access to the same I.D. system Pentax lenses are. Still a difference worth noting.

To further confuse the issue, some third party lenses do have unique 10 bit identifiers, why some not others? One of the mysteries of the universe.

Did the third party guys all double up on I.D. codes, using an un-athrourized extension , or did Pentax restrict them. Is part of the savings from reverse engineering the lens to buy one I.D. code and then use it on multiple lenses or did Pentax realizing the limitations of a 10 bit code ask them to do it, and then just forgot to mention to Apple when they did Aperture that they had to look at three numbers not 2?

I would find it very difficult to believe that these codes are handed out first come first served in a random order.

Notice the bunch of lenses bunched with a 3 255 code, 255 is the last 8 bit number, did Pentax run out of codes and couldn't make up their minds which way to go next?

Will we ever know for sure? Nice to know this will be cleared up when I'm finally forced to switch to Light Room.

So if you want to reduce this to what we know looking at this data, it would be, sometimes 3rd party vendors use an extension, not currently used by Pentax, although without knowing the I.D for every lens out there, you can't even know that for sure.

There is a difference, we just don't know what causes it, and what effect it might have going forward. I'd love to see a DA 55-300 PLM to try and lift the ID number from the exif, just to see how KAF4 is going to be identified.

Given the above information, my guess is that Pentax charges 3rd party manufacturers for a unique I.D number, one 2 bit digit, one 8 bit digit, but, that's only 2000 lenses that could be identified if you used every available unique digit. My guess is a part of the reverse engineering deal, 3th party manufacturers have put extensions unsupported by Pentax to differentiate different models of lenses, while using one 10 bit code and paying only once.

Someone else got a better guess?

---

Last edited by normhead; 08-12-2016 at 10:12 AM.

[Edgar_in_Indy]

Given the above information, my guess is that Pentax charges 3rd party manufacturers for a unique I.D number, one 2 bit digit, one 8 bit digit, but, that's only 2000 lenses that could be identified if you used every available digit. My guess is apart of the reverse engineering deal, 3th party manufacturers have put extensions unsupported by Pentax to differentiate different models of lenses, while using one 10 bit code and paying only once.

Someone else got a better guess?

Could be that in the early days of digital cameras, when every byte in memory still counted, they were trying to find the most economical way to digitally identify a lens, and didn't think about a day when there would be hundreds of lenses requiring unique identification, and requiring somebody to maintain and coordinate a table of codes across numerous manufacturers and hundreds of lenses.

Going forward, it seems like the most elegant solution would be to allow a lens the additional option to store its name in plain text that gets passed onto the EXIF. That way, the lens makers, the camera manufacturers and the software companies would not have to worry about collaborating on that point.

As a bonus, it would also open up the possibility of hacking into a lens and renaming it whatever you want. So I could have my Sigma 85mm show up as "The Intimidator".

[stevebrot]

Pentax lenses all have a separate identifier. 0ne number...1-4 followed by a number between 1 and 255. (All us binary dudes understand this.)
Many third party lenses have a period and one more number. So in the case of Aperture, it doesn't pay attention to the third number

Ummmmm...a small explanation is in order

The ExifTool tag values in the table are not the code per se. As with all of the maker-specific values parsed from the makernotes, the Pentax-specific tag names and even some aspects of the displayed values are arbitrary and the developer's best guess as to purpose. There is an explanation at the top of the table in regards to the "." values. The number to the right of the "." is arbitrary and not part of the id written into the exif by the camera. For example (from the portion of the table you copied):

'3 255' = Sigma Lens (3 255)
'3 255.1' = Sigma 18-200mm F3.5-6.3 DC
'3 255.2' = Sigma DL-II 35-80mm F4-5.6
'3 255.3' = Sigma DL Zoom 75-300mm F4-5.6
'3 255.4' = Sigma DF EX Aspherical 28-70mm F2.8
'3 255.5' = Sigma AF Tele 400mm F5.6 Multi-coated
'3 255.6' = Sigma 24-60mm F2.8 EX DG
'3 255.7' = Sigma 70-300mm F4-5.6 Macro
'3 255.8' = Sigma 55-200mm F4-5.6 DC
'3 255.9' = Sigma 18-50mm F2.8 EX DC

All lenses on the list pass '3 255' to the camera resulting in a huge ambiguity. If ExifTool is able to narrow the selection by other means (e.g. focal length and/or aperture) a specific id may be applied for display purposes as part of the composite LensID tag. The original (read-only) code in the makernotes remains intact. It is important to remember that even if ExifTool is able to determine an accurate ID from the makernotes, that is no guarantee that Lightroom will be able to do the same. This is true even for non-ambiguous ids. It all depends on the completeness of Lightroom's lookup tables.

Case in point might be my Sigma 17-70/2.8-4 (C). ExifTool correctly identifies the lens as "Sigma 17-70mm F2.8-4 DC Macro HSM Contemporary" based on the non-ambiguous code '8 30'. Lightroom 5.x, on the other hand, does not detect my lens on import despite having a profile file. Go figure.*


Steve

* My workaround is to have ExifTool write its mapped "friendly name" into the standard exif "LensModel" and "LensInfo" tags. These will be used by Lightroom for both display and as searchable metadata. It is an extra step, but worth it.

[anu l]

Could be that in the early days of digital cameras, when every byte in memory still counted, they were trying to find the most economical way to digitally identify a lens, and didn't think about a day when there would be hundreds of lenses requiring unique identification, and requiring somebody to maintain and coordinate a table of codes across numerous manufacturers and hundreds of lenses.

Going forward, it seems like the most elegant solution would be to allow a lens the additional option to store its name in plain text that gets passed onto the EXIF. That way, the lens makers, the camera manufacturers and the software companies would not have to worry about collaborating on that point.

As a bonus, it would also open up the possibility of hacking into a lens and renaming it whatever you want. So I could have my Sigma 85mm show up as "The Intimidator".

Then Pentax would prefer to mostly hard-code the names in ROM.....

[gelokrol]

anyone knows if the problem occurs with the sigma 17-50 mm f/2.8 ex dc os hsm lens?

[stevebrot]

anyone knows if the problem occurs with the sigma 17-50 mm f/2.8 ex dc os hsm lens?

The list is here:

Use of Pentax mount SIGMA interchangeable lenses when attached to the ?Pentax K-70? - SIGMA CORPORATION


Steve

[Kuzma]

Pentax does everything to make me give up on purchasing it´s new cameras( 2 of my most used lenses are in this list.

---

Last edited by BigMackCam; 08-14-2016 at 02:39 AM. Reason: Removed vulgarity

[Not a Number]

Pentax does everything to make me give up on purchasing it´s new cameras( 2 of my most used lenses are in this list.

And how does "Pentax" do that?

1. "Pentax" as far as digital cameras goes exists as a brand name and not a corporation
2. Your copies of the listed lenses may not have a problem at all.
3. The lenses can modified at no cost by Sigma
4. If the "scratch" doesn't bother you, you can use the lenses as is.

[D1N0]

In the mean time I haven't seen said scratch yet

[monochrome]

In the mean time I haven't seen said scratch yet

The lack of evidence of a scratch proves the existence of a scratch.

[BigMackCam]

The lack of evidence of a scratch proves the existence of a scratch.

Until we actually see the scratch, it simultaneously exists and does not exist. This is "Schrödinger's Scratch"....

[D1N0]

Until we actually see the scratch, it simultaneously exists and does not exist. This is "Schrödinger's Scratch"....

well don't look then, this topic can last forever. To scratch or not to scratch. That is the question.

[stevebrot]

Until we actually see the scratch, it simultaneously exists and does not exist. This is "Schrödinger's Scratch"....

Camera must be placed in a box...


Steve

[Edgar_in_Indy]

Camera must be placed in a box...

With a Sigma lens that may, or may not be, on the list of known scratchers...

[rawr]

Pentax does everything to make me give up on purchasing it´s new cameras( 2 of my most used lenses are in this list.

Having experience with two of the Sigma 'scratcher' lenses when mounted on the K-1 (35 f1.4 A and 70-200 HSM II - both of which are also on the K-70 list), I can say that the issue is unfortunate, but it should not discourage you from buying a K-70 (or K-1 ). There was no impact on the actual operation of both my lenses, even though a tiny scratch was evident after use. And Sigma world-wide seem more than happy to perform the lens mount plate upgrade for free.