In the old ball park

rjo__ · 2014-04-23 19:35

There is never enough bandwidth or memory. So let's face it, no matter what Chip does, it won't be enough.
However... sometimes we get what we need http://www.youtube.com/watch?v=j7leQB_Oe_k. Chip and Ken do seem to be very facile at discriminating our various needs.
This new chip seems.... wonnnnnnderfulllllll.

After examining the tea leaves, I thought it was time for a sanity check... after all, I'm trying to write code from specs for a camera that I do not yet posses for a processor that won't exist for many moons... If that isn't a reason for a sanity check, I don't know what is. I have some questions... but not for Chip. He is busy and if he answers here, we will know that he is letting you guys down.

This is the first of several, mundane, senile questions that I have:

Is it true that it will take exactly the same amount of time to write a byte to hub memory as it does to write 128 bits to Hub memory?

jmg · 2014-04-23 20:01

rjo__ wrote: »

Is it true that it will take exactly the same amount of time to write a byte to hub memory as it does to write 128 bits to Hub memory?

Chips recent Data had this heading, which answers your question :

New Propeller Chip - 16 April 2014

	200MHz system clock
	16 cogs with 2-clock instructions, hub execution at 50% cog speed
	[B]512KB hub memory with 8/16/32/128 bit cog transfers[/B]
	64 smart I/O pins
	100-pin 14x14mm TQFP with exposed thermal GND pad

Seairth · 2014-04-23 20:03

rjo__ wrote: »

Is it true that it will take exactly the same amount of time to write a byte to hub memory as it does to write 128 bits to Hub memory?

Yes.

rjo__ · 2014-04-23 20:10

Thanks... I think I figured out the answers to my other questions... about cog synchronization.

In the old ball park

Comments