![]() | ![]() | ![]() |
| | |||||||
| Micro Controllers Discuss all aspects of micro controllers - building them, coding them, etc. All controllers are welcome - PIC, BASIC, Z8 Encore!, etc. |
| | LinkBack | Thread Tools | Display Modes |
| | (permalink) |
| I'm curious how does AVR acheive a one clock per instruction speed? Is it a larger pipeline prefetch? Execution on both clock edges? | |
| |
| | (permalink) |
| Here are a couple of pages from the ATTiny13 datasheet. Might give you some insight. Never really bothered with trying to understand it myself. | |
| |
| | (permalink) | |
| Quote:
Upon decoding of the instructions there are enables placed on all elements needed to perform the specific instruction with just a single clock pulse, and yes I would assume that both the leading and trailing edges are needed to make that all happen in one cycle. It's a classic tradeoff of how much logic to expense Vs speed of operation. Also in retrospect, IF a 4 cycle per instruction processor can be clocked 4X the clock speed of a single clock per instruction processor, does the user of the device really care? Lefty
__________________ Measurement changes behavior | ||
| |
| | (permalink) | |
| Quote:
If a 4 cycle per instruction processor can be clocked at 4X the clock speed of a single clock per instruction processor while consuming no increase in overall power, does the user of the device really care?
__________________ L.Chung | ||
| |
| | (permalink) | |
| Quote:
I recall one in-house mini-computer design I worked on in the 70s that only had one real hardware register, the ALU accumulator, the others (program counter, variable registers, etc) were just reserved addresses in the main core memory, less logic required, slower speed but fast enough to do the purpose task of the in house application. Logic design of computers is a fascinating subject. Even a product series like the famous DEC PDP-11 series of the 70s & 80s had tremendous differences in implementing the many models all with the standard PDP-11 instruction set. Lefty
__________________ Measurement changes behavior Last edited by Leftyretro; 9th December 2007 at 02:07 AM. | ||
| |
| | (permalink) |
| Yes. Respect! And without the hard work of so many people involved in earlier pioneering work in computing, we would not have what we have today.
__________________ L.Chung | |
| |
| Bookmarks |
| Thread Tools | |
| Display Modes | |
| |
| | ||||
| Title | Starter | Forum | Replies | Latest |
| Problem with LCD code? | spondootre | Micro Controllers | 15 | 29th April 2008 10:15 AM |
| Hardcode a PIC by hand? (or 16F74 + par.port.prog = problem) | smilen | Micro Controllers | 15 | 21st November 2005 10:23 AM |
| counter clock | 111 | General Electronics Chat | 7 | 3rd July 2005 01:51 AM |
| Help me on basketball's game clock and shot clock! | syaw_007 | Electronic Projects Design/Ideas/Reviews | 1 | 6th September 2004 10:50 PM |
| Processor Instruction Format | mus3na | Micro Controllers | 1 | 19th January 2004 07:13 PM |