Overclocking AVR features/instructions list

Go To Last Post
9 posts / 0 new
Author
Message
#1
  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

Is there any list available for each AVR feature and instructions when overclocking?
By watching this video the AVR (ATmega8) pins can be overclocked to 51MHz.

http://www.youtube.com/watch?v=g...

E.g. If a program/system need to do lots of calculations this overclocking (on certain parts in the program) would save a lot of time.
Anyone tested instructions on high speeds?

RES

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

Quote:

Is there any list available for each AVR feature and instructions when overclocking?

I'll bite: Where would you expect such a list to come from? Atmel? Why in the world would anyone spend any time characterizing features beyond specified limits? Now, how long a list do you want? If there are 100 AVR models and 100 instructions each and 100 "features" each, that's a million entries. I look forward to seeing your table.

Once you have that done, better check each variant of each instruction. That is about 50000 times the 100 models.

You can put lipstick on a pig, but it is still a pig.

I've never met a pig I didn't like, as long as you have some salt and pepper.

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

I'd also recommend testing 50 of each chip (all from different batches) so multiply Lee's numbers by another 50.

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

Quote:

If there are 100 AVR models and 100 instructions each and 100 "features" each, that's a million entries. I look forward to seeing your table.

Once you have that done, better check each variant of each instruction. That is about 50000 times the 100 models.


Then take into onsideration "silicon revisions". (Or perhaps even manufacturing week?)

OP: If you need that kind of performance, then get something that promises that kind of performance. E.g. somne of the Cortex-M3's.

Overclocking an AVR by at least 2.5 times, perhaps more than 4 times.. Ridicolous for anything but singular hobby actiities, if even that. And there the solution is simple: Overclock. Test specimen. Works? OK, keep. Works not? Crap, throw away.

If you are interested in the general attitude towards overclocking here at AVRfreaks, then do a search for "overclocking" and the like. (Since you're "RES", a member here close to 6 years, you should know this. OTOH, since you're "RES", nothing surprises us... :evil:)

Unless this thread leads to something very revolutionary new I suggest a mod locks this thread in a day or so. If requested, I can locate a few of the previous threads on the subject and post here before locking.

As of January 15, 2018, Site fix-up work has begun! Now do your part and report any bugs or deficiencies here

No guarantees, but if we don't report problems they won't get much of  a chance to be fixed! Details/discussions at link given just above.

 

"Some questions have no answers."[C Baird] "There comes a point where the spoon-feeding has to stop and the independent thinking has to start." [C Lawson] "There are always ways to disagree, without being disagreeable."[E Weddington] "Words represent concepts. Use the wrong words, communicate the wrong concept." [J Morin] "Persistence only goes so far if you set yourself up for failure." [Kartman]

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

I thin k the mega will do perfect when clocking at 50MHz. It might jump an instruction or two for each instruction fetched, but it will do something. If you make sure after each command there are at least 4 nops, you should be just fine in making the processor run at 50MHz.

bit dont be surprised to burn your fingers and for un expected code behaviour. the nops might just not do what you expect them to do......

ow and indeed as others siad this is much discussed, so a simple search on the topic should yield enough information to keep you reading for a while.

have fun in playing

ow and don't forget arm cortex chips are so passe, an over clocked AVR is the buzz word now a days.

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

Why would the failure mode necessarily be the Program Counter skipping? Failure could come in one of a trillion ways.

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

It has been observed (by more than one person) that EEPROM writes will fail when over-clocking at temperature extremes.

But, how could lack of any "features" be acceptable? And, what special "features" would you expect from overclocking (that is, other than being unpredictably unusable - hey, what a great "feature"!)?

Jim

 

Until Black Lives Matter, we do not have "All Lives Matter"!

 

 

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

Quote:
Failure could come in one of a trillion ways.
M$... :) :) :)
Quote:
hey, what a great "feature"!)?
BSOD... :) :) :)
Give me a Kernel Panic any day ;)
Overclockers are nuts!

--greg
Still learning, don't shout at me, educate me.
Starting the fire is easy; the hardest part is learning how to keep the flame!

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

Quote:

Why would the failure mode necessarily be the Program Counter skipping? Failure could come in one of a trillion ways.


I thought the post was tongue-in-cheek. I especially liked
Quote:

If you make sure after each command there are at least 4 nops, you should be just fine in making the processor run at 50MHz.

Still laughing.

You can put lipstick on a pig, but it is still a pig.

I've never met a pig I didn't like, as long as you have some salt and pepper.