Initial Setup (only needs to be performed once):
Write Cycle:
These sequences are fairly close to those of the OV511/OV511+, except
that the control register is now 0x47 instead of 0x40. There is also a
conspicuous lack of status checking, which indicates to me that the control
register no longer acts as a status register at all. It looks like the
only way to determine if the slave device is responding is to read from
a subadress that has a known value (such as the manufacturer IDs in registers
0x1C and 0x1D).
Here is a summary of the I2C registers:
Reg | OV511/OV511+ function | OV518+ function |
40h | Control bits and status bits | [unknown] |
41h | I2C write slave ID | [unchanged] |
42h | Subaddress for write cycles | [unchanged] |
43h | Subaddress for read cycles | [unchanged] |
44h | I2C read slave ID | [unchanged] |
45h | Data port | [unchanged] |
46h | I2C clock prescaler | Apparently not used anymore. |
47h | Time out counter | Control bits are the same as in OV511 reg 40h;
Probably not a status register any more |
48h | Subaddress for snapshot write cycles | Probably no longer used |
49h | Data port for snapshot write cycles | Probably no longer used |
NOTE: It appears that all of the snapshot-specific registers (19h-1fh, 48h, 49h) are no longer used. The snapshot button seems to only be used for launching a vendor's application.
Version History:
Date | Version | Change |
2001/01/15 | 1.0 | Initial Release |