Some lenses seem to not work

[aslidegh]

Hey, great project, I hope it is still alive!
I was wondering, if you had ever tested this with EF 100 Macro, EF 24-70 2.8 II or 16-35 f4 lenses. While other lenses seem to work fine, I keep getting 0xFF when busy polling above mentioned lenses.
Any ideas?

[Jana-Marie]

Hey, sorry for the late reply. Glad that you replicated and like the project :3

I'm still active, but I also haven't touched this project in a while. Is it correct that all the lenses that do not work are rather modern "red-ring" lenses? Are all working lenses "non-red-ring" lenses?

[aslidegh]

Hey,
mmh yeah kind of. My tests are as follows:

• 50mm f1.8 STM ; works (I think this version of the lens is also kind of "new")
• 70-200 f4 red-ring ; works (but it's very old)
• 24-70 f2.8 II red-ring ; not working
• 100 f2.8 macro red-ring ; not working
• 16-35 f4 red-ring ; not working

Are you observing similar issue?

[jeanthom]

I came across this: http://oliford.co.uk/phys/canon-lens-protocol/canon-135mm-switchOn-x3 (Canon 135mm L, old lens but red ring). It seems like the init sequence goes with more than just 0x0A on the body side for these lens. It legacy from the analog EF era?

This is sort of confirmed in http://jp79dsfr.free.fr/_Docs%20et%20infos/Photo%20Tech%20_%20Canon%20EOS-EF%20Protocol.pdf:

Commande 0x0A : Demande disponibilité objectif

Cette commande est généralement la première rencontrée lors de l'émission d'une trame. Elle
est alors répétée jusqu'à obtention d'une réponse 0xAA de la part de l'objectif, d'autres commandes
pouvant alors suivre. Cette méthode de fonctionnement est typique d'un test de réponse-
acquittement comme celui utilisé en communication série Ascii avec ENQ/ACK (ou TcpIp avec
Syn/Ack mais nettement moins d'actualité lors de la conception du protocole en 1987).

On retrouve ce principe dans l'essai précédent sans objectif ou une suite massive de commande
0x0A était émise avant abandon ou tentative ultérieure. Avec l'Eos 5000, ces suites de commande
0x0A sont suivies des commandes 0X09, 0E, 0F ce qui peut laisser supposer que celles-ci
correspondent a des ordres d'initialisation ou de reset. Il est à noter que 0x0E ne sera rencontré
ultérieurement qu'avec des objectifs réglés en autofocus manuel.

...or in English:

0x0A command

This command is generally the first when emitting a frame. It is repeated until getting a 0xAA reply from the lens, after which other commands can be sent. [Author goes into a long explanation about why it looks like a SYN/ACK mechanism to them]

With the EOS 5000 [an older EF analog body], the 0x0A commands are followed by 0x09 [is it a typo?], 0x0E, 0x0F which could lead us to think that they are part of the init or reset process. It is important to consider that 0x0E will only be observed afterwards when the lens is set to manual focus.

[aslidegh]

I did already modify the init procedure doing similar, but seems like it doesn't change anything for me at least.

[Jana-Marie]

Hey, mmh yeah kind of. My tests are as follows:

* 50mm f1.8 STM ; works (I think this version of the lens is also kind of "new")

* 70-200 f4 red-ring ; works (but it's very old)

* 24-70 f2.8 II red-ring ; not working

* 100 f2.8 macro red-ring ; not working

* 16-35 f4 red-ring ; not working

Are you observing similar issue?

Thanks! I have not experienced similar issues, but I also do not own a red-ring lens. Thanks for the test!

[Jana-Marie]

I came across this: http://oliford.co.uk/phys/canon-lens-protocol/canon-135mm-switchOn-x3 (Canon 135mm L, old lens but red ring). It seems like the init sequence goes with more than just 0x0A on the body side for these lens. It legacy from the analog EF era?

This is sort of confirmed in http://jp79dsfr.free.fr/_Docs%20et%20infos/Photo%20Tech%20_%20Canon%20EOS-EF%20Protocol.pdf:

Commande 0x0A : Demande disponibilité objectif
Cette commande est généralement la première rencontrée lors de l'émission d'une trame. Elle
est alors répétée jusqu'à obtention d'une réponse 0xAA de la part de l'objectif, d'autres commandes
pouvant alors suivre. Cette méthode de fonctionnement est typique d'un test de réponse-
acquittement comme celui utilisé en communication série Ascii avec ENQ/ACK (ou TcpIp avec
Syn/Ack mais nettement moins d'actualité lors de la conception du protocole en 1987).
On retrouve ce principe dans l'essai précédent sans objectif ou une suite massive de commande
0x0A était émise avant abandon ou tentative ultérieure. Avec l'Eos 5000, ces suites de commande
0x0A sont suivies des commandes 0X09, 0E, 0F ce qui peut laisser supposer que celles-ci
correspondent a des ordres d'initialisation ou de reset. Il est à noter que 0x0E ne sera rencontré
ultérieurement qu'avec des objectifs réglés en autofocus manuel.

...or in English:

0x0A command
This command is generally the first when emitting a frame. It is repeated until getting a 0xAA reply from the lens, after which other commands can be sent. [Author goes into a long explanation about why it looks like a SYN/ACK mechanism to them]
With the EOS 5000 [an older EF analog body], the 0x0A commands are followed by 0x09 [is it a typo?], 0x0E, 0x0F which could lead us to think that they are part of the init or reset process. It is important to consider that 0x0E will only be observed afterwards when the lens is set to manual focus.

Thanks to you too! I will look into it next week, maybe I can find something

[aslidegh]

I digged a bit more but couldn't find any useful information related to newer red ring lenses unfortunately.

[whoigaurav]

can anyone add a bin file or hex file for the stm32 programmer?