mersenneforum.org  

Go Back   mersenneforum.org > Great Internet Mersenne Prime Search > Data

Reply
 
Thread Tools
Old 2004-05-02, 14:58   #1
dave_0273
 
dave_0273's Avatar
 
Oct 2003
Australia, Brisbane

2·5·47 Posts
Default Data On Error Codes

First I found out the frequency of each error code in lucas_v.txt. Then I did the same for bad.txt. After doing this, I could then work out the percentage of good results compared to the total amount.

ie %good=[(#of times in lucas_v)/(#of times in lucas + #of times in bad)]*100

The error codes are as followed, ranked in order of % good. Any error code that appeared less than 10 times in total was left out as I felt that there was not enough information to get a decent percentage.

Code:
Error Code   #Lucas_v  #bad	% good
00240000	11	0	100.00%
00130000	35	0	100.00%
00190000	17	0	100.00%
80040000	20	0	100.00%
001A0000	17	0	100.00%
00220000	17	0	100.00%
001E0000	16	0	100.00%
00FF0000	25	0	100.00%
0E000E00	13	0	100.00%
01000001	186	1	99.47%
00090000	109	1	99.09%
03000300	167	2	98.82%
01000100	495	6	98.80%
00000000	356190	5831	98.39%
000E0000	54	1	98.18%
02000200	267	5	98.16%
00100000	52	1	98.11%
000C0000	50	1	98.04%
00030000	570	14	97.60%
04000400	101	3	97.12%
000D0000	67	2	97.10%
00050000	293	9	97.02%
000A0000	95	3	96.94%
00080000	126	4	96.92%
00060000	189	6	96.92%
00150000	31	1	96.88%
00010000	2600	98	96.37%
00020000	950	37	96.25%
80000000	8772	372	95.93%
00070000	165	7	95.93%
02000002	47	2	95.92%
00160000	23	1	95.83%
08000800	22	1	95.65%
00040000	355	18	95.17%
000B0000	78	4	95.12%
00060001	19	1	95.00%
00180000	18	1	94.74%
05000500	69	4	94.52%
000F0000	51	3	94.44%
00110000	31	2	93.94%
0C000C00	15	1	93.75%
00120000	30	2	93.75%
07000700	29	2	93.55%
00200000	13	1	92.86%
001B0000	12	1	92.31%
01000200	24	2	92.31%
000B0001	11	1	91.67%
00210000	11	1	91.67%
000D0001	11	1	91.67%
000A0001	11	1	91.67%
001F0000	10	1	90.91%
09000900	20	2	90.91%
00080001	19	2	90.48%
00090001	18	2	90.00%
06000600	35	4	89.74%
80010000	115	15	88.46%
80020000	38	5	88.37%
00040002	15	2	88.24%
00140000	21	3	87.50%
00020001	77	11	87.50%
81000200	14	2	87.50%
00170000	21	3	87.50%
001D0000	14	2	87.50%
001C0000	20	3	86.96%
00040001	31	5	86.11%
00250000	12	2	85.71%
80050000	11	2	84.62%
00260000	9	2	81.82%
00050001	26	6	81.25%
00070001	26	6	81.25%
80030000	20	5	80.00%
00030001	36	10	78.26%
00010001	146	41	78.07%
000C0001	10	3	76.92%
01000000	123	40	75.46%
00000001	2090	684	75.34%
0D000D00	9	3	75.00%
01000002	11	4	73.33%
00020002	16	6	72.73%
00050002	13	6	68.42%
00030002	10	6	62.50%
02000000	16	10	61.54%
00030003	8	5	61.54%
00020003	11	7	61.11%
00000100	509	336	60.24%
00010004	17	12	58.62%
00030100	7	5	58.33%
00010100	14	10	58.33%
00010005	7	5	58.33%
00000500	12	9	57.14%
80000100	9	7	56.25%
00010003	19	15	55.88%
00010002	42	35	54.55%
80000001	31	26	54.39%
00000002	360	306	54.05%
00000003	157	135	53.77%
03000000	8	7	53.33%
00000010	7	7	50.00%
00000009	16	17	48.48%
00000007	28	35	44.44%
00000011	5	7	41.67%
0000000A	17	25	40.48%
0000000B	12	18	40.00%
00010006	6	9	40.00%
00000101	83	131	38.79%
00000004	75	119	38.66%
00000401	5	8	38.46%
00000006	42	68	38.18%
0000000C	6	10	37.50%
00000005	48	81	37.21%
00000200	59	103	36.42%
00000012	4	7	36.36%
00000008	22	41	34.92%
00000201	23	44	34.33%
0000000F	5	10	33.33%
00000108	4	8	33.33%
00000400	13	28	31.71%
00000103	20	44	31.25%
00000102	35	80	30.43%
80000002	6	14	30.00%
00000300	19	47	28.79%
00000107	4	10	28.57%
00000203	5	13	27.78%
00000202	7	19	26.92%
0000000E	4	11	26.67%
0000010B	3	9	25.00%
00000402	3	9	25.00%
0000000D	4	13	23.53%
00000104	8	26	23.53%
00000302	3	10	23.08%
00000600	3	11	21.43%
00000204	4	17	19.05%
00000700	2	9	18.18%
00000013	2	9	18.18%
00000301	4	20	16.67%
0000010A	2	10	16.67%
00000206	2	11	15.38%
00000106	1	18	5.26%
Therefore, if an error code does not appear in this list, one or more of the following could have happened.
1. It occured in total less than 10 times (ie #in lucas+#in bad<10)
2. It never appeared in lucas_v, so it has never given a correct result

One thing to consider when looking at the 00000000 error code. Version WUx had a bug that didn't count errors properly, so some of the 'bad' results with 00000000 error code could have had errors.

Unfortunately, none of the results from hrf3.txt can be used as they have not been confirmed good or bad.

Last fiddled with by dave_0273 on 2004-05-02 at 15:09
dave_0273 is offline   Reply With Quote
Old 2004-05-02, 15:08   #2
dave_0273
 
dave_0273's Avatar
 
Oct 2003
Australia, Brisbane

2×5×47 Posts
Default

I then sorted each of the error codes into 4 main catagorys..

1. Error codes of the form 00XX0000 which are relatively harmless
2. Error codes of the form XY00XY00 which are also relatively harmless
3. Error codes of the form 8XXXXXXX. Error codes that start with an 8 mean that at least some of the test was run on a program that did not count errors. Therefore, not much can be gathered from this error code as we don't know how many (and of what type) error occured (if any).
4. The rest

Code:
Error Code   #Lucas_v  #bad	% good
00000000	356190	5831	98.39%

Of the form 00XX0000
Error Code   #Lucas_v  #bad	% good
00240000	11	0	100.00%
00130000	35	0	100.00%
00190000	17	0	100.00%
001A0000	17	0	100.00%
00220000	17	0	100.00%
001E0000	16	0	100.00%
00FF0000	25	0	100.00%
00150000	31	1	96.88%
00010000	2600	98	96.37%
000D0000	67	2	97.10%
00050000	293	9	97.02%
000A0000	95	3	96.94%
00080000	126	4	96.92%
00060000	189	6	96.92%
00160000	23	1	95.83%
00040000	355	18	95.17%
000B0000	78	4	95.12%
00180000	18	1	94.74%
000F0000	51	3	94.44%
00110000	31	2	93.94%
00120000	30	2	93.75%
001B0000	12	1	92.31%
00210000	11	1	91.67%
001F0000	10	1	90.91%
00140000	21	3	87.50%
00170000	21	3	87.50%
001D0000	14	2	87.50%
001C0000	20	3	86.96%
00250000	12	2	85.71%
00260000	9	2	81.82%

Of the form XY00XY00
Error Code   #Lucas_v  #bad	% good
0E000E00	13	0	100.00%
03000300	167	2	98.82%
01000100	495	6	98.80%
02000200	267	5	98.16%
04000400	101	3	97.12%
08000800	22	1	95.65%
05000500	69	4	94.52%
0C000C00	15	1	93.75%
07000700	29	2	93.55%
09000900	20	2	90.91%
06000600	35	4	89.74%
0D000D00	9	3	75.00%

Of the form 8XXXXXXX
Error Code   #Lucas_v  #bad	% good
80040000	20	0	100.00%
80000000	8772	372	95.93%
80010000	115	15	88.46%
80020000	38	5	88.37%
81000200	14	2	87.50%
80050000	11	2	84.62%
80030000	20	5	80.00%
80000100	9	7	56.25%
80000001	31	26	54.39%
80000002	6	14	30.00%

The Rest
Error Code   #Lucas_v  #bad	% good
01000001	186	1	99.47%
00090000	109	1	99.09%
000E0000	54	1	98.18%
00100000	52	1	98.11%
000C0000	50	1	98.04%
00030000	570	14	97.60%
00020000	950	37	96.25%
00070000	165	7	95.93%
02000002	47	2	95.92%
00060001	19	1	95.00%
00200000	13	1	92.86%
01000200	24	2	92.31%
000B0001	11	1	91.67%
000D0001	11	1	91.67%
000A0001	11	1	91.67%
00080001	19	2	90.48%
00090001	18	2	90.00%
00040002	15	2	88.24%
00020001	77	11	87.50%
00040001	31	5	86.11%
00050001	26	6	81.25%
00070001	26	6	81.25%
00030001	36	10	78.26%
00010001	146	41	78.07%
000C0001	10	3	76.92%
01000000	123	40	75.46%
00000001	2090	684	75.34%
01000002	11	4	73.33%
00020002	16	6	72.73%
00050002	13	6	68.42%
00030002	10	6	62.50%
02000000	16	10	61.54%
00030003	8	5	61.54%
00020003	11	7	61.11%
00000100	509	336	60.24%
00010004	17	12	58.62%
00030100	7	5	58.33%
00010100	14	10	58.33%
00010005	7	5	58.33%
00000500	12	9	57.14%
00010003	19	15	55.88%
00010002	42	35	54.55%
00000002	360	306	54.05%
00000003	157	135	53.77%
03000000	8	7	53.33%
00000010	7	7	50.00%
00000009	16	17	48.48%
00000007	28	35	44.44%
00000011	5	7	41.67%
0000000A	17	25	40.48%
0000000B	12	18	40.00%
00010006	6	9	40.00%
00000101	83	131	38.79%
00000004	75	119	38.66%
00000401	5	8	38.46%
00000006	42	68	38.18%
0000000C	6	10	37.50%
00000005	48	81	37.21%
00000200	59	103	36.42%
00000012	4	7	36.36%
00000008	22	41	34.92%
00000201	23	44	34.33%
0000000F	5	10	33.33%
00000108	4	8	33.33%
00000400	13	28	31.71%
00000103	20	44	31.25%
00000102	35	80	30.43%
00000300	19	47	28.79%
00000107	4	10	28.57%
00000203	5	13	27.78%
00000202	7	19	26.92%
0000000E	4	11	26.67%
0000010B	3	9	25.00%
00000402	3	9	25.00%
0000000D	4	13	23.53%
00000104	8	26	23.53%
00000302	3	10	23.08%
00000600	3	11	21.43%
00000204	4	17	19.05%
00000700	2	9	18.18%
00000013	2	9	18.18%
00000301	4	20	16.67%
0000010A	2	10	16.67%
00000206	2	11	15.38%
00000106	1	18	5.26%
Once again, just remember that the "percentage good" could technically be a little higher due to the bug in version WUx that didn't count errors properly, so some of the 'bad' 00000000 results could have had errors in them.

Last fiddled with by dave_0273 on 2004-05-02 at 15:11
dave_0273 is offline   Reply With Quote
Reply

Thread Tools


Similar Threads
Thread Thread Starter Forum Replies Last Post
Error Codes? Antonio Information & Answers 6 2011-09-27 13:20
Alt codes : should shift = +32 in them ? science_man_88 Programming 13 2011-06-24 06:52
Very odd error 7, send benchmark data. Uncwilly PrimeNet 2 2008-12-29 04:48
Prime Generators and/or Source Codes Erasmus Factoring 3 2004-05-21 13:55
Most popular error codes GP2 Data 14 2003-12-28 22:02

All times are UTC. The time now is 08:30.

Wed Nov 25 08:30:39 UTC 2020 up 76 days, 5:41, 4 users, load averages: 1.17, 1.36, 1.38

Powered by vBulletin® Version 3.8.11
Copyright ©2000 - 2020, Jelsoft Enterprises Ltd.

This forum has received and complied with 0 (zero) government requests for information.

Permission is granted to copy, distribute and/or modify this document under the terms of the GNU Free Documentation License, Version 1.2 or any later version published by the Free Software Foundation.
A copy of the license is included in the FAQ.