Strategic Double Clicking

[Madpoo]

Originally Posted by Madpoo

...
One of my side projects, that is never-ending, is doing daily checks for any exponents that need a triple-check and getting it done. Besides the ones where I've already done one of the checks, I think there are only about 65 or so outstanding. I've got something like 28 of them queued up and the other 37 are assigned to someone else.
...

Here's a list of available exponents that need a triple-check ... I can't do 'em because I already ran of of the previous tests. The highest 3 of these were false positives so you'll only see one result in the report page (the false "is prime" result is hidden).

41363341
41396569
41464483
41582521
41845787
41869493
41910059
41992429
60358603
78355799
78605777
80051219

[storm5510]

Originally Posted by Madpoo

Yeah, that's the point of this little project, is finding the exponents only tested once and try to predict if that solo test was good or bad.
...:

That was a lot to write and I thank you for the effort.

I often peruse the 'recently cleared' list to see if anything looks odd. Of course, the one thing that jumps out is a LL and DC reported on the same day, at the same time, and by the same user. This, in my mind, would immediately fall into the triple-check category.

I also look at the 'manual testing' entries which indicate a composite. I do not recall seeing anything that indicates if these are being ran on Prime95, a Linux process, or a GPU program like CUDALucas.

Has any user ever been contacted and informed that they may have a questionable device?

[Madpoo]

Originally Posted by storm5510

Of course, the one thing that jumps out is a LL and DC reported on the same day, at the same time, and by the same user. This, in my mind, would immediately fall into the triple-check category.

Any exponent that was tested and double-checked by the same person has been getting an independent triple-check. It happens, and some users do it on purpose because they don't trust their systems and would hate to miss a potential prime because their computer messed up. We had a big side project not too long ago where a bunch of nice people helped work through a huge backlog of those. I forget how many there were in total, but it was several thousand, but as a team effort we got through them and now I just do the odd one that shows up sometimes.

Another category I looked at were what I call the "temporally proximal" results, where it was from 2 different users but within a short time period of each other. Those seemed weird to me and not really by chance, so I did independent checks on those to be sure.

I also look at the 'manual testing' entries which indicate a composite. I do not recall seeing anything that indicates if these are being ran on Prime95, a Linux process, or a GPU program like CUDALucas.

In the raw result text sent to the server, it indicates the software name and version which is helpful to be able to refer to if there was some kind of problem. I suppose it might be nice to include that info somewhere on the exponent report page. Hmm... food for thought.

Has any user ever been contacted and informed that they may have a questionable device?

Since I started working with the Primenet server, sometimes either I or George have contacted a user that was sending in some strange results, like the false positives that were coming in from one user. In that case we tried to see if we could figure out what the issue was but we never were able to nail anything down.

[ATH]

Originally Posted by Madpoo

Any exponent that was tested and double-checked by the same person has been getting an independent triple-check. It happens, and some users do it on purpose because they don't trust their systems and would hate to miss a potential prime because their computer messed up. We had a big side project not too long ago where a bunch of nice people helped work through a huge backlog of those. I forget how many there were in total, but it was several thousand, but as a team effort we got through them and now I just do the odd one that shows up sometimes.

It's funny I started that "triple check" thread back in Aug 2012 but there was no interest at all, no one thought it was very important so I just left it. Then 2.5 years later in March 2015 VictordeHolland made a post in the thread (bumped it) which I guess made you notice it, and then with your fast resources and database sleuthing it took off from there I think you did at least 80% of the triple check work by yourself, and after that was done, you came up with this "Strategic Double Checking" thread.

[Madpoo]

Here's a fun one that needs a quad check:
M48704479

[Madpoo]

Originally Posted by ATH

It's funny I started that "triple check" thread back in Aug 2012 but there was no interest at all, no one thought it was very important so I just left it. Then 2.5 years later in March 2015 VictordeHolland made a post in the thread (bumped it) which I guess made you notice it, and then with your fast resources and database sleuthing it took off from there I think you did at least 80% of the triple check work by yourself, and after that was done, you came up with this "Strategic Double Checking" thread.

It was certainly great to get all the help from other people.

Speaking of, I mentioned this in another obscure thread, but I wanted to mention it here too...

I'm losing all of my hardware... yup, the cloud has reared it's ugly head into my world, and after my current assignments finish (just a handful) that'll be it. :unhappy:

I'll still generate lists of exponents of various interesting things, I just won't be able to join in the fun firsthand, only vicariously.

[Mark Rose]

Originally Posted by Madpoo

Speaking of, I mentioned this in another obscure thread, but I wanted to mention it here too...

I'm losing all of my hardware... yup, the cloud has reared it's ugly head into my world, and after my current assignments finish (just a handful) that'll be it. :unhappy:

I'll still generate lists of exponents of various interesting things, I just won't be able to join in the fun firsthand, only vicariously.

Time for a home lab ;)

[storm5510]

Originally Posted by Madpoo

It was certainly great to get all the help from other people.

Speaking of, I mentioned this in another obscure thread, but I wanted to mention it here too...

I'm losing all of my hardware... yup, the cloud has reared it's ugly head into my world, and after my current assignments finish (just a handful) that'll be it. :unhappy:

I'll still generate lists of exponents of various interesting things, I just won't be able to join in the fun firsthand, only vicariously.

.
This sucks, badly!

Originally Posted by Mark Rose

Time for a home lab ;)

I agree. There are more than a few people here, including me, who do not want to see you riding off into the sunset.

[moebius]

The exponent 45863893 needs a triple-check. Probably my Residue is wrong because I currently give a try to other DRAM timings in the bios and this is my first Double Check
since Years.

[Madpoo]

Originally Posted by moebius

The exponent 45863893 needs a triple-check. Probably my Residue is wrong because I currently give a try to other DRAM timings in the bios and this is my first Double Check
since Years.

I did pick that one up. I have a few more hours of checking power available.

[rudi_m]

Originally Posted by Madpoo

I'm losing all of my hardware... yup, the cloud has reared it's ugly head into my world, and after my current assignments finish (just a handful) that'll be it. :unhappy:

Maybe the cloud will be cheaper for your company because your local machines used too much electrical energy for mprime. ;)