cancel
Showing results for 
Search instead for 
Did you mean: 

MT9M001 Camera and DCMI

simfake
Associate II
Posted on September 01, 2016 at 02:08

 

 

The original post was too long to process during our migration. Please click on the attachment to read the original post.
3 REPLIES 3
simfake
Associate II
Posted on September 03, 2016 at 06:09

So I had a look with my logic analyzer. After fixing the bug where I was allowing too many DMA transfers per picture (there is 2 pixels per DMA transfer - wasn't do that), I now have similar problems to before.

HS = Horizontal sync. Every line/row of pixels.

VS = Vsync. Every entire picture

STATUS2: DCMI Line interrupt

STATUS1: DMA interrupt.

First, the 164ms delay seen in the zoomed out picture. I'm enabling DCMI in the DCMI interrupt, and somehow there is a 164ms delay? That would ideally be exactly on the right time.

Second, the data doesn't seem to be valid still. I would expect that the results are just a little bit shifted/the picture is distorted. But no, it's just nor working nicely.

0690X000006057RQAQ.png

0690X00000605RfQAI.png

Any ideas? Any ideas for things that I could check?

megahercas6
Senior
Posted on September 03, 2016 at 08:17

The original post was too long to process during our migration. Please click on the provided URL to read the original post. https://st--c.eu10.content.force.com/sfc/dist/version/download/?oid=00Db0000000YtG6&ids=0680X000006I6jc&d=%2Fa%2F0X0000000buE%2Fg7fA3NSU9BwoEj3zz40sz_p0EdD7FauZruyvukdhHOY&asPdf=false
simfake
Associate II
Posted on September 04, 2016 at 04:05

Thanks! On first glance, there's only a little different in the DMA setup, and a lot different in synchronisation. I'll have a look.