Strategic Double Clicking - Page 81

Madpoo · 2016-02-23, 03:13

Here's a new list. Some of these are in the "churn" zone... may go fast.

Code:

exponent	Bad	Good	Unk	Sus	Solo	Mis	worktodo
38305409	2	0	1	0	1	0	DoubleCheck=38305409,71,1
38601403	2	1	1	0	1	0	DoubleCheck=38601403,71,1
38941213	3	0	1	0	1	0	DoubleCheck=38941213,71,1
39239509	2	0	1	0	1	0	DoubleCheck=39239509,71,1
39263383	13	6	3	1	2	2	DoubleCheck=39263383,71,1
39273809	3	1	1	0	1	0	DoubleCheck=39273809,71,1
40356571	8	3	6	2	6	2	DoubleCheck=40356571,72,1
40392127	8	4	3	1	2	2	DoubleCheck=40392127,72,1
40568431	18	5	1	4	1	4	DoubleCheck=40568431,72,1
40605407	4	0	5	1	4	2	DoubleCheck=40605407,72,1
46550347	2	1	1	1	1	1	DoubleCheck=46550347,72,1

markr · 2016-02-23, 03:42

A big thank you to Madpoo for triple-checking 35895053 and 36011531! I would have been surprised if the bad results were mine, but you never know. So good to not have a long wait before resolution. They were cat 3's back in August, so the triple-checks would have had to wait at least for the head of the cat 2 queue to move up to them (it's nearly there now) and it probably would have slipped past them anyway, leaving them at the back of the cat 1 queue.

Madpoo · 2016-02-23, 06:11

Quote:

Originally Posted by markr

A big thank you to Madpoo for triple-checking 35895053 and 36011531! I would have been surprised if the bad results were mine, but you never know. So good to not have a long wait before resolution. They were cat 3's back in August, so the triple-checks would have had to wait at least for the head of the cat 2 queue to move up to them (it's nearly there now) and it probably would have slipped past them anyway, leaving them at the back of the cat 1 queue.

Yeah, all part of my (and AirSquirrels too) plan to slog away at some of these triple-checks that are needed.

My ultimate goal with those is to build up the knowledge of bad/good machines, but it's also kind of fun to clear those out anyway.

dragonbud20 · 2016-02-23, 07:01

Quote:

Originally Posted by Madpoo

Here's a new list. Some of these are in the "churn" zone... may go fast.

Code:

exponent	Bad	Good	Unk	Sus	Solo	Mis	worktodo
38305409	2	0	1	0	1	0	DoubleCheck=38305409,71,1
38601403	2	1	1	0	1	0	DoubleCheck=38601403,71,1
38941213	3	0	1	0	1	0	DoubleCheck=38941213,71,1

going to take these for work over the next week or so

bgbeuning · 2016-02-24, 01:23

Quote:

Originally Posted by Madpoo

Here's a new list. Some of these are in the "churn" zone... may go fast.

Code:

exponent	Bad	Good	Unk	Sus	Solo	Mis	worktodo
39239509	2	0	1	0	1	0	DoubleCheck=39239509,71,1
39263383	13	6	3	1	2	2	DoubleCheck=39263383,71,1
39273809	3	1	1	0	1	0	DoubleCheck=39273809,71,1
40356571	8	3	6	2	6	2	DoubleCheck=40356571,72,1
40392127	8	4	3	1	2	2	DoubleCheck=40392127,72,1
40568431	18	5	1	4	1	4	DoubleCheck=40568431,72,1
40605407	4	0	5	1	4	2	DoubleCheck=40605407,72,1
46550347	2	1	1	1	1	1	DoubleCheck=46550347,72,1

Queued up the other 8 of them.

Madpoo · 2016-02-25, 18:47

Here's another handful if anyone's interested. I'll try to just pop a few of these in here now and then.

Overall, I think we've done a really good job of getting the low hanging fruit, and we're getting down to stuff where the CPU has less than a 2;1 bad/good ratio. Still WAY higher than they should be, but just not as certain we'll mismatch on our run.

For those I'll probably start pulling out just the smallest available exponent from those systems so we can test just one more at a time and see which direction we're heading before we pull more from the same thing. I've also setup another query to narrow results down by year and month to try and spot systems that were mostly good for a year or a quarter, but had a bad month... apparently that's a thing, there are machines like that.

Code:

exponent	Bad	Good	Unk	Sus	Solo	Mis	worktodo
39125137	2	1	2	0	2	0	DoubleCheck=39125137,71,1
39239509	2	0	1	0	1	0	DoubleCheck=39239509,71,1
39263383	13	6	3	1	2	2	DoubleCheck=39263383,71,1
39273809	3	1	1	0	1	0	DoubleCheck=39273809,71,1
39285991	2	0	3	0	3	0	DoubleCheck=39285991,71,1
46550347	2	1	1	1	1	1	DoubleCheck=46550347,72,1
52276529	25	3	1	5	1	5	DoubleCheck=52276529,73,1
63462271	23	11	2	5	2	5	DoubleCheck=63462271,74,1

Prime95 · 2016-02-25, 18:56

I took these:

Code:

39125137	2	1	2	0	2	0	DoubleCheck=39125137,71,1
39239509	2	0	1	0	1	0	DoubleCheck=39239509,71,1
39263383	13	6	3	1	2	2	DoubleCheck=39263383,71,1
39273809	3	1	1	0	1	0	DoubleCheck=39273809,71,1
39285991	2	0	3	0	3	0	DoubleCheck=39285991,71,1

That 52M one with 25 bad to 3 good looks very tasty....

Mark Rose · 2016-02-25, 19:04

I took these two:

Code:

exponent	Bad	Good	Unk	Sus	Solo	Mis	worktodo
46550347	2	1	1	1	1	1	DoubleCheck=46550347,72,1
52276529	25	3	1	5	1	5	DoubleCheck=52276529,73,1

fivemack · 2016-02-26, 20:36

My first batch of strategic DCs in the 50M range is over.

Could I request triple-checks for 50803771, 53998279, 50967491 ?

bgbeuning · 2016-02-27, 03:04

Quote:

Originally Posted by fivemack

Could I request triple-checks for 50803771, 53998279, 50967491 ?

They are running.

bgbeuning · 2016-02-28, 01:44

Quote:

Originally Posted by Madpoo

Code:

exponent	Bad	Good	Unk	Sus	Solo	Mis	worktodo
39125137	2	1	2	0	2	0	DoubleCheck=39125137,71,1
39239509	2	0	1	0	1	0	DoubleCheck=39239509,71,1
39263383	13	6	3	1	2	2	DoubleCheck=39263383,71,1
39273809	3	1	1	0	1	0	DoubleCheck=39273809,71,1
39285991	2	0	3	0	3	0	DoubleCheck=39285991,71,1
46550347	2	1	1	1	1	1	DoubleCheck=46550347,72,1
52276529	25	3	1	5	1	5	DoubleCheck=52276529,73,1
63462271	23	11	2	5	2	5	DoubleCheck=63462271,74,1

This list and the one from 22 Feb seem to have a couple of duplicates.
I guess we are quad checking those.

2016-02-23, 03:42	#882
markr "Mark" Feb 2003 Sydney 1075₈ Posts	Thanks! A big thank you to Madpoo for triple-checking 35895053 and 36011531! I would have been surprised if the bad results were mine, but you never know. So good to not have a long wait before resolution. They were cat 3's back in August, so the triple-checks would have had to wait at least for the head of the cat 2 queue to move up to them (it's nearly there now) and it probably would have slipped past them anyway, leaving them at the back of the cat 1 queue.

2016-02-25, 18:47	#886
Madpoo Serpentine Vermin Jar Jul 2014 7·11·43 Posts	Here's another handful if anyone's interested. I'll try to just pop a few of these in here now and then. Overall, I think we've done a really good job of getting the low hanging fruit, and we're getting down to stuff where the CPU has less than a 2;1 bad/good ratio. Still WAY higher than they should be, but just not as certain we'll mismatch on our run. For those I'll probably start pulling out just the smallest available exponent from those systems so we can test just one more at a time and see which direction we're heading before we pull more from the same thing. I've also setup another query to narrow results down by year and month to try and spot systems that were mostly good for a year or a quarter, but had a bad month... apparently that's a thing, there are machines like that. Code: exponent Bad Good Unk Sus Solo Mis worktodo 39125137 2 1 2 0 2 0 DoubleCheck=39125137,71,1 39239509 2 0 1 0 1 0 DoubleCheck=39239509,71,1 39263383 13 6 3 1 2 2 DoubleCheck=39263383,71,1 39273809 3 1 1 0 1 0 DoubleCheck=39273809,71,1 39285991 2 0 3 0 3 0 DoubleCheck=39285991,71,1 46550347 2 1 1 1 1 1 DoubleCheck=46550347,72,1 52276529 25 3 1 5 1 5 DoubleCheck=52276529,73,1 63462271 23 11 2 5 2 5 DoubleCheck=63462271,74,1

2016-02-25, 19:04	#888
Mark Rose "/X\(‘-‘)/X\" Jan 2013 B74₁₆ Posts	I took these two: Code: exponent Bad Good Unk Sus Solo Mis worktodo 46550347 2 1 1 1 1 1 DoubleCheck=46550347,72,1 52276529 25 3 1 5 1 5 DoubleCheck=52276529,73,1

2016-02-26, 20:36	#889
fivemack (loop (#_fork)) Feb 2006 Cambridge, England 6419₁₀ Posts	My first batch of strategic DCs in the 50M range is over. Could I request triple-checks for 50803771, 53998279, 50967491 ? Last fiddled with by fivemack on 2016-02-26 at 20:38

Similar Threads
Thread	Thread Starter	Forum	Replies	Last Post
Double-Double Arithmetic	Mysticial	Software	52	2021-04-23 06:51
Clicking an exponent leads to 404 page	marigonzes	Information & Answers	2	2017-02-14 16:56
x.265 half the size, double the computation; so if you double again? 1/4th?	jasong	jasong	7	2015-08-17 10:56
What about double-checking TF/P-1?	137ben	PrimeNet	6	2012-03-13 04:01
Double the area, Double the volume.	Uncwilly	Puzzles	8	2006-07-03 16:02