nvidia rtx-2080 - Page 12

LaurV · 2019-02-17, 12:56

Anybody tested if this misses factors or not? I mean, excuse my grumpiness, but we had x% improvements in the past, many times, and usually they were missing at least x% of the factors

Chuck · 2019-02-17, 13:58

Quote:

Originally Posted by LaurV

Anybody tested if this misses factors or not? I mean, excuse my grumpiness, but we had x% improvements in the past, many times, and usually they were missing at least x% of the factors

I'll put one of my known previous factors through this version.

Chuck · 2019-02-17, 14:13

Quote:

Originally Posted by Chuck

I'll put one of my known previous factors through this version.

It successfully re-found the factor for 59825761.

LaurV · 2019-02-17, 14:41

Well, one factor is not much, I am going to give it a spin on a file I have with many exponents that have easy-to-find factors. But this will not be today (here already almost 10PM Sunday night and tomorrow I will have a quite busy day, so I will turn towards the bed soon). Probably next week, and it will take some time.

nomead · 2019-02-17, 15:07

The self-tests pass, of course this doesn't tell much yet, only that nothing major is wrong. In the past weeks I've been running it mostly on >1000M exponents from mersenne.ca and factors found per TF span have statistically been pretty much as expected in the long run. Of course this still tells nothing about missing a few % of them somewhere... so yes, agreed, further tests are needed.

Chuck · 2019-02-17, 15:56

Quote:

Originally Posted by LaurV

Well, one factor is not much, I am going to give it a spin on a file I have with many exponents that have easy-to-find factors. But this will not be today (here already almost 10PM Sunday night and tomorrow I will have a quite busy day, so I will turn towards the bed soon). Probably next week, and it will take some time.

Good idea. Looking forward to the results.

nomead · 2019-02-17, 16:59

Okay so I extracted some known factors from the mersenne.ca database, also some exponents with multiple factors in different TF ranges, but no run was really long (only up to about two minutes per exp). Single range, multiple range, different exponent sizes, 2-64 bit range, highest was 73-74 bits (for 1000M-size exponents, though). 612 factors found of expected 611! No expected factors were missed, though... The one extra was M1000030483 - the database already had 1428663559409788789639 (found in 2016) but in the same 70-71 bit range there was also 2021118137372188931399. Still, more tests naturally can't hurt.

Chuck · 2019-02-17, 20:00

We could retest every factor ever found by MFAKTC and this version still MIGHT miss one in the future. We can't ever prove that this version is completely correct. When is enough enough? I am satisfied.

LaurV · 2019-02-18, 13:55

Quote:

Originally Posted by nomead

The one extra was M1000030483 - the database already had 1428663559409788789639 (found in 2016) but in the same 70-71 bit range there was also 2021118137372188931399.

nomead · 2019-03-03, 22:41

Quote:

Originally Posted by kriesel

Have RTX20xx become reliable yet? (Reduced, low probability of return for repair within months, weeks, or days) The initial going seemed to be sort of dismal, with some users indicating 2 of 3 or 2 of 2 early failures, including replacements failing.

Two months on the RTX 2080 at work now, with various work loads, no problems.

One month on the RTX 2060 at home now, running pretty much 100% 24h a day with mfaktc, no problems with that one either.

However... My older kid (20 years but still living at home, meh) bought some upgrade parts for his computer, including an RTX 2060 card. This week, he started having some random blue screens when playing games, and earlier today, we went through the drivers and BIOS and everything is up to date now... but no help. Later in the evening, after one more BSOD the system stayed "on" with a black screen but wouldn't reboot anymore after that. More troubleshooting will need to be done tomorrow but signs point to display card failure. I have a couple older spare cards to try but they're not at home, so that's why it's tomorrow. And it shouldn't be because of too much stress on the card - he's not even playing the latest titles that stress the GPU to the max, and definitely not 24h a day. Under two weeks in use...

kriesel · 2019-03-04, 02:11

Quote:

Originally Posted by nomead

Two months on the RTX 2080 at work now, with various work loads, no problems.

One month on the RTX 2060 at home now, running pretty much 100% 24h a day with mfaktc, no problems with that one either.

However... My older kid (20 years but still living at home, meh) bought some upgrade parts for his computer, including an RTX 2060 card. This week, he started having some random blue screens when playing games, and earlier today, we went through the drivers and BIOS and everything is up to date now... but no help. Later in the evening, after one more BSOD the system stayed "on" with a black screen but wouldn't reboot anymore after that. More troubleshooting will need to be done tomorrow but signs point to display card failure. I have a couple older spare cards to try but they're not at home, so that's why it's tomorrow. And it shouldn't be because of too much stress on the card - he's not even playing the latest titles that stress the GPU to the max, and definitely not 24h a day. Under two weeks in use...

Who installed your son's gpu? With careful antistatic precautions?
(Way to get our hopes up and then dash them, by the way...)

Similar Threads
Thread	Thread Starter	Forum	Replies	Last Post
Nvidia GTX 745 4GB ???	petrw1	GPU Computing	3	2016-08-02 15:23
Nvidia Pascal, a third of DP	firejuggler	GPU Computing	12	2016-02-23 06:55
AMD + Nvidia	TheMawn	GPU Computing	7	2013-07-01 14:08
Nvidia Kepler	Brain	GPU Computing	149	2013-02-17 08:05
What can I do with my nvidia GPU?	Surge	Software	4	2010-09-29 11:36

2019-02-17, 12:56	#122
LaurV Romulan Interpreter "name field" Jun 2011 Thailand 41×251 Posts	Anybody tested if this misses factors or not? I mean, excuse my grumpiness, but we had x% improvements in the past, many times, and usually they were missing at least x% of the factors

2019-02-17, 14:41	#125
LaurV Romulan Interpreter "name field" Jun 2011 Thailand 41×251 Posts	Well, one factor is not much, I am going to give it a spin on a file I have with many exponents that have easy-to-find factors. But this will not be today (here already almost 10PM Sunday night and tomorrow I will have a quite busy day, so I will turn towards the bed soon). Probably next week, and it will take some time.

2019-02-17, 15:07	#126
nomead "Sam Laur" Dec 2018 Turku, Finland 317 Posts	The self-tests pass, of course this doesn't tell much yet, only that nothing major is wrong. In the past weeks I've been running it mostly on >1000M exponents from mersenne.ca and factors found per TF span have statistically been pretty much as expected in the long run. Of course this still tells nothing about missing a few % of them somewhere... so yes, agreed, further tests are needed.

2019-02-17, 16:59	#128
nomead "Sam Laur" Dec 2018 Turku, Finland 317 Posts	Okay so I extracted some known factors from the mersenne.ca database, also some exponents with multiple factors in different TF ranges, but no run was really long (only up to about two minutes per exp). Single range, multiple range, different exponent sizes, 2-64 bit range, highest was 73-74 bits (for 1000M-size exponents, though). 612 factors found of expected 611! No expected factors were missed, though... The one extra was M1000030483 - the database already had 1428663559409788789639 (found in 2016) but in the same 70-71 bit range there was also 2021118137372188931399. Still, more tests naturally can't hurt.

2019-02-17, 20:00	#129
Chuck May 2011 Orange Park, FL 2⁵×29 Posts	We could retest every factor ever found by MFAKTC and this version still MIGHT miss one in the future. We can't ever prove that this version is completely correct. When is enough enough? I am satisfied.