Author Topic: 7.3728Mhz  (Read 4628 times)

0 Members and 1 Guest are viewing this topic.

Offline JVRTopic starter

  • Regular Contributor
  • *
  • Posts: 201
  • Country: be
7.3728Mhz
« on: October 14, 2011, 06:58:45 pm »
Hi gents/ladies.

Been reading the forum for a while and I am stuck with a question that I dont think can be answered anywhere else.

Why would you clock a DSP at 7.3728Mhz?  I have a dsPICdem board with this on, and I need to generate a 50us timer interrupt, but it seems to not be possible. 

I'll work around the problem, but would love to know the logic behind using such a random (to me) clock speed

 

Offline Bored@Work

  • Super Contributor
  • ***
  • Posts: 3932
  • Country: 00
Re: 7.3728Mhz
« Reply #1 on: October 14, 2011, 07:36:36 pm »
Why would you clock a DSP at 7.3728Mhz?  I have a dsPICdem board with this on, and I need to generate a 50us timer interrupt, but it seems to not be possible. 

I'll work around the problem, but would love to know the logic behind using such a random (to me) clock speed

It is not random, it is a nice frequency. It is easily divided into clock frequencies for some common "high speed" baud rates:

7372800 / 16 = 460800
7372800 / 32 = 230400
7372800 / 64 = 115200
7372800 / 128 = 57600
7372800 / 256 = 28800
7372800 / 512 = 14400

and a bunch more, which aren't too common, and aren't "high speed" (7200, 3600, ...).
I delete PMs unread. If you have something to say, say it in public.
For all else: Profile->[Modify Profile]Buddies/Ignore List->Edit Ignore List
 

Offline JVRTopic starter

  • Regular Contributor
  • *
  • Posts: 201
  • Country: be
Re: 7.3728Mhz
« Reply #2 on: October 14, 2011, 08:02:37 pm »
That makes some sense.  First time I've worked with it though. 

It does seem a bit daft clocking the whole micro to mach baud rates, as sampling and generation frequencies seems to take a massive accuracy knock



PS:  Do we need to enter the Captha for every post? Or is it just in the beginning >x posts?
 

Offline Bored@Work

  • Super Contributor
  • ***
  • Posts: 3932
  • Country: 00
Re: 7.3728Mhz
« Reply #3 on: October 14, 2011, 08:32:01 pm »
It does seem a bit daft clocking the whole micro to mach baud rates, as sampling and generation frequencies seems to take a massive accuracy knock

Tell us about the alternatives. How many independent asynchronous clock sources can your device take?

In other words, that is just what is happening with such digital devices. Everything comes in steps, and something is always screwed with a single clock. Live with it. Do you think it is better to screw communication, or frequency generation? It is your job as a designer to find and use the least problematic frequency to satisfy all of your requirements.
I delete PMs unread. If you have something to say, say it in public.
For all else: Profile->[Modify Profile]Buddies/Ignore List->Edit Ignore List
 

Offline JVRTopic starter

  • Regular Contributor
  • *
  • Posts: 201
  • Country: be
Re: 7.3728Mhz
« Reply #4 on: October 14, 2011, 09:54:28 pm »

 How many independent asynchronous clock sources can your device take?

Two :(

I guess if it wasn't 23:50 here I could just solder a 32KHz crystal in and use another timer for freq generation.  Uni projects FTL

Thanks for clearing the 7Mhz bit up though.
 

Offline Mechatrommer

  • Super Contributor
  • ***
  • Posts: 11713
  • Country: my
  • reassessing directives...
Re: 7.3728Mhz
« Reply #5 on: October 15, 2011, 01:45:17 am »
Why would you clock a DSP at 7.3728Mhz? would love to know the logic behind using such a random (to me) clock speed
It is not random, it is a nice frequency. It is easily divided into clock frequencies for some common "high speed" baud rates:
7372800 / 16 = 460800
7372800 / 32 = 230400
7372800 / 64 = 115200
7372800 / 128 = 57600
7372800 / 256 = 28800
7372800 / 512 = 14400
and a bunch more, which aren't too common, and aren't "high speed" (7200, 3600, ...).
if its up to me, 6.5536MHz will be nicer. :D there should be a better reason than that.
Nature: Evolution and the Illusion of Randomness (Stephen L. Talbott): Its now indisputable that... organisms “expertise” contextualizes its genome, and its nonsense to say that these powers are under the control of the genome being contextualized - Barbara McClintock
 


Share me

Digg  Facebook  SlashDot  Delicious  Technorati  Twitter  Google  Yahoo
Smf