ERROR: SUM(INPUTS) != SUM(OUTPUTS)

flava · 2003-12-25, 19:55

Hi,

I got this error today while running a LL test on a 10M digits number (1792K FFT).

ERROR: SUM(INPUTS) != SUM(OUTPUTS), 3.430791490553064e+048 != 3.430791490553064e+048

The computer is not overclocked, it passed a night of torture test, it passes memtest OK and it doesn't run very hot (CPU 48°C, MB 37°C). It doesn't crash on me except when the DSL modem's drivers go crazy while connecting or diconnecting internet.
The funny thing is the two numbers in this message appear equal from what is displayed.
I ran a 1792K inplace FFT test for a few hours and all seems OK.
Shuld I toss away some weeks of work (the test got to about 30%) or should I blame it on a driver, windoze, etc...?

Prime95 · 2003-12-26, 21:02

This thread http://www.mersenneforum.org/showthr...7968#post17968 shows that you have a 23.1% chance of returning a bad result if no other errors crop up.

Your percentage *might* be less in that you were diligent in running a torture test and memtest when the error occured. Others were probably not as diligent.

Your call on whether you want to restart.

flava · 2003-12-27, 20:33

Thank you for that link! I'll let it run for now. I save a file every 1.000.000 iterations so I know the expected interim residues. I'll run the first 10M iterations on a fast Xeon at work to make a partial "doublecheck". If they are OK I'll let my computer finish, if not I'll let the Xeon finish this number.

flava · 2004-01-19, 17:52

I did the "partial doublecheck" and it is good!

2003-12-25, 19:55	#1
flava Feb 2003 2×59 Posts	ERROR: SUM(INPUTS) != SUM(OUTPUTS) Hi, I got this error today while running a LL test on a 10M digits number (1792K FFT). ERROR: SUM(INPUTS) != SUM(OUTPUTS), 3.430791490553064e+048 != 3.430791490553064e+048 The computer is not overclocked, it passed a night of torture test, it passes memtest OK and it doesn't run very hot (CPU 48°C, MB 37°C). It doesn't crash on me except when the DSL modem's drivers go crazy while connecting or diconnecting internet. The funny thing is the two numbers in this message appear equal from what is displayed. I ran a 1792K inplace FFT test for a few hours and all seems OK. Shuld I toss away some weeks of work (the test got to about 30%) or should I blame it on a driver, windoze, etc...?

2003-12-26, 21:02	#2
Prime95 P90 years forever! Aug 2002 Yeehaw, FL 1110101100110₂ Posts	This thread http://www.mersenneforum.org/showthr...7968#post17968 shows that you have a 23.1% chance of returning a bad result if no other errors crop up. Your percentage might be less in that you were diligent in running a torture test and memtest when the error occured. Others were probably not as diligent. Your call on whether you want to restart. Last fiddled with by Prime95 on 2003-12-26 at 21:04

Similar Threads
Thread	Thread Starter	Forum	Replies	Last Post
ERROR: SUM(INPUTS) != SUM(OUTPUTS)	AES	PrimeNet	6	2007-10-05 18:48
LLR SUM(INPUTS) != SUM(OUTPUTS) error	jbristow	Software	4	2007-08-14 04:07
v. 2.13: SUM(INPUTS) != SUM(OUTPUTS)	Kaiw	Software	7	2005-10-26 14:49
SUM(INPUTS) != SUM(OUTPUTS)	rudi_m	Software	3	2005-08-22 15:42
ERROR: SUM(INPUTS) != SUM(OUTPUTS)	ebx	Software	5	2004-01-02 22:25

2003-12-27, 20:33	#3
flava Feb 2003 2×59 Posts	Thank you for that link! I'll let it run for now. I save a file every 1.000.000 iterations so I know the expected interim residues. I'll run the first 10M iterations on a fast Xeon at work to make a partial "doublecheck". If they are OK I'll let my computer finish, if not I'll let the Xeon finish this number.

2004-01-19, 17:52	#4
flava Feb 2003 2·59 Posts	I did the "partial doublecheck" and it is good!