RING BUFFER 18LF4620 Oshonsoft

Status
Not open for further replies.
Hi R,
I think you're correct, but I wan't sure, well done. Now the INTERRUPT is as mentioned, 80uS, so much smaller.

I've been using what I've learnt plus the smaller INTERRUPT to repair the original LINEAR BUFFER, which I'll try first. At the moment I'm stuck on CSV's, but may be able to figure it out.

I'll have to re-read this post, later, as you appear to suggest that a RING BUFFER can be smaller that a LINEAR one, and be better.
Thanks,
C.
 
Imagine two people each doing different tasks on a production line:

The first completes and hands on the item on an exact fixed schedule; the second can usually do his part in less time, but occasionally more.

Passing the items direct one to the other, the first guy occasionally drops the item to where the second guys hand should! be to catch it, but he's not ready and the item gets lost.

That's your existing situation, with data at a fixed interval from the serial port, and occasionally a delay while you parse the linear buffer content.

The small ring buffer will acting as a FIFO, a first-in, first-out store, between the two.

It's like putting a small ramp shute between the first guy to the second - the items can stack occasionally until the second guy catches up, so nothing lost when the two different timings do not match up.

The two existing parts, interrupt input and linear buffer processing, still do exactly what they did before, just without any critical timing between them.
 
Hi R,
I follow the idea, but I am slow at programming, I'll come back to RING BUFFERS later.
Thanks.
C.
 
Hi,
I been trying to make a LINEAR INTERRUPT?PARSE program first, rather frustratingly! I had 3x goes so far, and hope I'm improving each time.
Here is the latest:

With 'correct' $Sentences, it appears to be ok, but I've had faults, with corrupt $Sentences. I'm not sure why?
EDIT: I've changed it slightly, and think it is working now. I would eppreciate any feedback for other simulators please.
C.
 

Attachments

  • 18LF4620 8MHz XTL BASE PCB4 300520 1700.txt
    9.7 KB · Views: 311
Last edited:
Hi,
#44 program slightly changed and may be woking now.

If it is, thanks. I hope to try to make a RING next.
C.
 
Status
Not open for further replies.
Cookies are required to use this site. You must accept them to continue using the site. Learn more…