Mini-drive for k=3010-3200 n=420259-600000

[gamer007]

3017*2^517810-1 is prime.

[gamer007]

517K-518K is complete. 1 prime.

Reserving 518K-519K.

[gamer007]

3125*2^518126-1 is prime.

[gamer007]

518K-519K is complete. 1 prime.

[Svenie25]

Sorry, I am a little late with my range. But I am at hospital since a few days and don't know, how long it will take.

Will finish the range if I am at home again.

MfG
Sven

[gamer007]

Reserving 519K-520K.

[gd_barnes]

Originally Posted by Svenie25

Sorry, I am a little late with my range. But I am at hospital since a few days and don't know, how long it will take.

Will finish the range if I am at home again.

MfG
Sven

I hope you are doing OK Sven. Get well soon!

No problem on your range.

[gd_barnes]

Max and Karsten,

I was just thinking of something here that I think would make sense to do in the future: Whenever anyone completes a manual range, fairly quickly convert it to the format needed by the DB and send it on to Dave for loading into the DB.

This will avoid large jumps in the DB stats. I think the base 5 project did that at some point with a few of the individual-k reservations that they had.

One problem that I see though: Administration. We must keep a detailed log or a graphical listing of some kind of which manual files have been loaded to the DB. It would be quite easy to miss one here and there.

Let me know what you think.


Gary

[mdettweiler]

Originally Posted by gd_barnes

Max and Karsten,

I was just thinking of something here that I think would make sense to do in the future: Whenever anyone completes a manual range, fairly quickly convert it to the format needed by the DB and send it on to Dave for loading into the DB.

This will avoid large jumps in the DB stats. I think the base 5 project did that at some point with a few of the individual-k reservations that they had.

One problem that I see though: Administration. We must keep a detailed log or a graphical listing of some kind of which manual files have been loaded to the DB. It would be quite easy to miss one here and there.

Let me know what you think.


Gary

First of all, I am already keeping a log detailing which manual results have and have not been imported. So that shouldn't be too hard. As for importing manual results as they come, yes, that's pretty much what I had in mind for subsequent manual jobs once the big aggregate bunches are in. For files that small, we don't even have to bug Dave with them--I can just plop them straight into the results folder to be pulled in at the next daily run. (That's what I've already done for some of the drives with very small amounts of manual work.)

My eventual idea is to add a system to the website where people can log in and submit their manual results on their own to be imported. Needless to say that won't be for a little while yet, but I would eventually like it so that the admins don't have to do all this stuff manually.

[kar_bon]

Originally Posted by mdettweiler

My eventual idea is to add a system to the website where people can log in and submit their manual results on their own to be imported. Needless to say that won't be for a little while yet, but I would eventually like it so that the admins don't have to do all this stuff manually.

A point against submitting own maunal results: first one of the admins should verify if exactly those pairs are submitted, which are in the testfile!

I don't know if this was done with all other manual ranges. For the LLRnet/PRPnet ranges i've done (and will continue) this processing and found every missing/double pair at all.

[mdettweiler]

Originally Posted by kar_bon

A point against submitting own maunal results: first one of the admins should verify if exactly those pairs are submitted, which are in the testfile!

My idea was that I would first implement a page on the website where one could select a sieve file, and have the DB automatically check each pair with its contents and report back exactly what's missing, if anything. (This would go along with another page where one could plug in a range of k and n and have the server send back all results in that range in the desired format: together these additions would greatly simplify processing of server results as well.)

I don't know if this was done with all other manual ranges. For the LLRnet/PRPnet ranges i've done (and will continue) this processing and found every missing/double pair at all.

Gary currently verifies all results on all drives in periodic blocks--when he posts in the drive threads saying "all results and primes have been verified up to n=x" that means he's done it for that range.