[BUG] Lens detection

Bugs & Suggestions
Post Reply
RobiWan
Posts: 5
Joined: Wed Mar 14, 2012 2:06 am
operating_system: Mac
System_Drive: N/A
32bit or 64bit: 64 Bit

[BUG] Lens detection

Post by RobiWan »

Hi,

the very old Bibble BUG is still present :(
Some Sigma and Canon Lenses are recognized not correctly. Sigma 50/1.4 on Canon 7D. Sigma 85/1.4 on Canon 7D as example.

Has it somebody posted to https://www.surveymonkey.com/s/ASPFeedback ?

Regards Robert
grubernd
Posts: 248
Joined: Thu Jan 12, 2012 8:17 pm
operating_system: Linux
System_Drive: N/A
Location: Graz, Austria

Re: [BUG] Lens detection

Post by grubernd »

just write down what exactly happens - "wrong" is not enough - and feed it to the monkey.
since the monkey is a one-way-street with no direct feedback unless there are questions, and we dont know what's in there already, just submit the issues you stumble upon.
Bibble since 2004. Aftershot until 2020. From then on darktable.
RobiWan
Posts: 5
Joined: Wed Mar 14, 2012 2:06 am
operating_system: Mac
System_Drive: N/A
32bit or 64bit: 64 Bit

Re: [BUG] Lens detection

Post by RobiWan »

Hello,

now some examples:

Sigma 85/1.4 HSM - >ASP mean "Carl Zeiss Planar T* 85mm f=1.4 ZF .2"
Sigma 50/1.4 HSM -> ASP mean "Canon EF 50mm f/1.4 USM"
EF 100mm f/2.8 L Macro IS USM - >ASP mean "Canon EF 100mm f/2.8 Macro" (NOT L Version)

On my Canon 5D MKII are all three Lenses correctly detected.
There are only three examples. There are some another wrong detections on Canon, Minolta (Dynax 7D), Sony (A700).

Regards Robert
jp_wieseth
Posts: 6
Joined: Wed Apr 11, 2012 8:29 pm
operating_system: Windows 7 Professional
System_Drive: C
32bit or 64bit: 64 Bit
motherboard: ASRock 890GX Pro3
processor: AMD X6 1100T
ram: 16GB
Video Card: ATI HD6670
sound_card: onBoard
Hard_Drive_Capacity: 2 TB
Monitor/Display Make & Model: Samsung SyncMaster S27A850D 27"

[BUG] Lens detection

Post by jp_wieseth »

Hello,

now new example:

Sigma 70-200 mm F2,8 EX DG OS HSM - >ASP mean "Nikkor 70-200 f/2.8G AF-S ED VR II"

On my Nikon D90 all other Lenses are correctly detected.


jp_wieseth
gregglee
Posts: 55
Joined: Mon Jan 16, 2012 9:54 am
operating_system: Windows 7 Home Premium
System_Drive: C
32bit or 64bit: 64 Bit
motherboard: Asus Sabertooth 990FX R2.0
processor: AMD FX-8350 8core 4600 MHz
ram: 16 gb
Video Card: nvidia GTX 660ti
sound_card: on board
Hard_Drive_Capacity: 10 TB
Monitor/Display Make & Model: viewsonic vb2365WB + LG 1933TR
Location: Somervell County, Texas, USA

Re: [BUG] Lens detection

Post by gregglee »

RobiWan wrote:Hi,

the very old Bibble BUG is still present :(
Some Sigma and Canon Lenses are recognized not correctly. Sigma 50/1.4 on Canon 7D. Sigma 85/1.4 on Canon 7D as example.
Regards Robert
This is not so much a ASP/Bibble bug as a "bug" in the lens ids used by lens and camera manufacturers. Sony/Minolta and Canon occasionally use duplicate lens id's, but more of a problem, Sigma and Tamron frequently use duplicate lens id's, dupicating both other mfg ids and their own.

Lens id is a numeric field in metadata (in makernotes, not exif) that all lens mfgs have included for decades.

There is hope for more recent lenses. Sony (and so I imagine other manufacturers) since 2007 lenses have provided additional information to the camera besides lens id. Also Sony has added a table to their cameras to translate the lens id of older sony and minolta lenses into a more informative (still numerical) "lens spec" field that includes min and max aperture at min and max focal length, whether full or crop frame, and several identifiers. Will translate other brand lenses if they provide the data.

Even more important the new exif standard in 2010 includes exif fields for more lens data. This means that they are the same for all brands, unlike fields in makernotes. One field the defined same as the makernotes lensspec field I mentioned. Another is the full name of the lens in an alphanumeric field. This should allow reliable identification of all newer lenses.
There are several others. Not all mfgs will write all fields, but these fields always have the same exif address and specification.

The challenge is to get Corel/ASP and other image editors to add the new information to their lens id logic. Phil Harvey and his contributors to EXIFTool have already figured all this out, so it's just a matter of adding it.

Refer to the exiftool website for the list http://owl.phy.queensu.ca/~phil/exiftoo ... index.html for examples. click your brand then scroll down to "lens type values." For industry standard tags click exif tags (note the tag definitions and addresses are standard, but not all mfgs use all tags.
display the list then search for any field containing "lens." )

The exifgtool forum may also be informative http://u88.n24.queensu.ca/exiftool/foru ... ?board=4.0
Last edited by gregglee on Wed Jun 06, 2012 2:43 pm, edited 1 time in total.
Gregg Lee
12 miles from Glen Rose, Texas
camera equipment
gregglee
Posts: 55
Joined: Mon Jan 16, 2012 9:54 am
operating_system: Windows 7 Home Premium
System_Drive: C
32bit or 64bit: 64 Bit
motherboard: Asus Sabertooth 990FX R2.0
processor: AMD FX-8350 8core 4600 MHz
ram: 16 gb
Video Card: nvidia GTX 660ti
sound_card: on board
Hard_Drive_Capacity: 10 TB
Monitor/Display Make & Model: viewsonic vb2365WB + LG 1933TR
Location: Somervell County, Texas, USA

Re: [BUG] Lens detection

Post by gregglee »

You can also use exiftool or exiftoolgui to write the new exif fields to images taken by older lenses, which will enable editing software to identify them when/if they add logic to read the 2010 lens info exif fields aka exif tags.
Gregg Lee
12 miles from Glen Rose, Texas
camera equipment
RobiWan
Posts: 5
Joined: Wed Mar 14, 2012 2:06 am
operating_system: Mac
System_Drive: N/A
32bit or 64bit: 64 Bit

Re: [BUG] Lens detection

Post by RobiWan »

gregglee wrote:This is not so much a ASP/Bibble bug as a "bug" in the lens ids used by lens and camera manufacturers. Sony/Minolta and Canon occasionally use duplicate lens id's, but more of a problem, Sigma and Tamron frequently use duplicate lens id's, dupicating both other mfg ids and their own.

Lens id is a numeric field in metadata (in makernotes, not exif) that all lens mfgs have included for decades.
I'm sorry do you have read this?
RobiWan wrote:EF 100mm f/2.8 L Macro IS USM - >ASP mean "Canon EF 100mm f/2.8 Macro" (NOT L Version)

What are you trying to tell me straight? That's not a ASP bug. Of course, there is one. Aperture and Lightroom do not have these problems. There are ways and means to identify the correct lens, you just have to want and to deal with the issues apart.

Regards Robert
gregglee
Posts: 55
Joined: Mon Jan 16, 2012 9:54 am
operating_system: Windows 7 Home Premium
System_Drive: C
32bit or 64bit: 64 Bit
motherboard: Asus Sabertooth 990FX R2.0
processor: AMD FX-8350 8core 4600 MHz
ram: 16 gb
Video Card: nvidia GTX 660ti
sound_card: on board
Hard_Drive_Capacity: 10 TB
Monitor/Display Make & Model: viewsonic vb2365WB + LG 1933TR
Location: Somervell County, Texas, USA

Re: [BUG] Lens detection

Post by gregglee »

RobiWan wrote:There are ways and means to identify the correct lens, you just have to want and to deal with the issues apart.

Not always. If two (or more ) lenses report the same information to be recorded in metadata, then ASP cannot read your mind to determine which one it was. Sigma has used some lens code 10 times. The Sigma 50 and 85 for Canon have the same lens code for example.

You can of course override the ASP choice, and create a presets to make it easy.

You can also rearrange ASP lens tables to change which duplicate it picks first. Solves the problem unless you own more than one of the duplicates. How to do this so has been covered in other threads.
Gregg Lee
12 miles from Glen Rose, Texas
camera equipment
Post Reply