[QUOTE=Prime95;403844]I'd wait a while to see if the now expired DC assignment on 6-1-15 reports a result.[/QUOTE]

Good idea. I added it to one of my workers but it's pretty deep in there, won't start for a while. I noticed the original assignee hadn't started yet but I'll see how it's doing before I do anything else with it.

Oh, and I didn't know it wouldn't assign the same exponent to the same person for a double-check... that's cool. Good to hear. :smile:

I saw recently that there were some new v4/v5 self verifications done by Curtis' team, but I know those are hard to check against, and given Curtis' large # of results in the past I reckon that's just par for the course.

Which reminds me, I saw recently some discussion about Curtis' systems doing more double-checks... I don't know if that thread has an answer for "why" but maybe just for Curtis' old work we can update the LL results table with his v5 account info, replacing the legacy info on his old stuff? Heck, that might be a "fun" project in general since I think that'd make the v4/v5 stuff more consistent. I notice that many users who migrated have different account names, so it's not always obvious it's the same person.

I've now finished my tests. This will probably be my last contribution to this specific project, unless maybe it drags on past July.

[QUOTE=cuBerBruce;404015]I've now finished my tests. This will probably be my last contribution to this specific project, unless maybe it drags on past July.[/QUOTE]

Thanks for pitching in and helping out. I know it's not glamorous work doing probably-unneeded triple checks, but I appreciate the help.

I did a few double-checks of my own work. If you would be interested in triple checking them. The four exponents I did:

72316429
72779807
72838133
73068419

The reasons I double checked them was checking the overclock on my i5 4690k. I initially had it clocked to 4.5GHz. Failed torture test at 8 hours. Dropped it to 4.4 GHz, and torture test successfully ran for 60 hours. Then did some double-checks that all came back confirmed. Once I started first time tests, I would randomly and rarely awake to the machine being at the log in prompt. I did back down to 4.3GHz then, and did more double-checks; the reboots stopped. I wanted to make sure my first time tests were good, so I double checked them on my i7 4770 that's at stock speeds. Three of the four came back confirmed, but one did not.

I do appreciate the work you have put in to insure integrity of the data.

[QUOTE=endless mike;404065]I did a few double-checks of my own work. If you would be interested in triple checking them. The four exponents I did:

72316429
72779807
72838133
73068419

The reasons I double checked them was checking the overclock on my i5 4690k. I initially had it clocked to 4.5GHz. Failed torture test at 8 hours. Dropped it to 4.4 GHz, and torture test successfully ran for 60 hours. Then did some double-checks that all came back confirmed. Once I started first time tests, I would randomly and rarely awake to the machine being at the log in prompt. I did back down to 4.3GHz then, and did more double-checks; the reboots stopped. I wanted to make sure my first time tests were good, so I double checked them on my i7 4770 that's at stock speeds. Three of the four came back confirmed, but one did not.

I do appreciate the work you have put in to insure integrity of the data.[/QUOTE]

Sure, I already did a triple-check on the last 3 of those. The first one I noticed didn't match between your two runs, and I was kind of waiting to see if you ran it again.

But I'll save you the hassle of doing that (and I'd want to do an independent verification anyway :smile: ) so I'll go ahead and do a check of that first one.

If you felt like returning the favor, I have an exponent where I (accidentally) ran checks on 60027139 multiple times. 4 to be exact. I think I kept adding it to my work to do and because I had the only matching runs, it would show up in my query again. Ran it 4 times total before I realized it. LOL

Anyway, if you wanted to do a verification of 60027139 for me, I'll do a check on 72316429 for you. I mean, I'll do that check of 72316429 anyway, you don't owe me anything. :smile:

Weird one
 

Here's an odd one:
[URL="http://www.mersenne.org/M34531817"]M34531817[/URL]

Odd because I saw it pop up as a self-verified one by LaurV a couple days ago, so I added it to my queue of triple checks.

Then I just checked it in and saw "For Research" had checked in a result yesterday. Strange.

Did LaurV poach the assignment "For Reasearch" got for a double check? :smile:

[QUOTE=Madpoo;404102]Odd because I saw it pop up as a self-verified one by LaurV a couple days ago, so I added it to my queue of triple checks. Then I just checked it in and saw "For Research" had checked in a result yesterday. Strange. Did LaurV poach the assignment "For Reasearch" got for a double check? :smile:[/QUOTE]

All I can say is this was assigned to one of my (slower) machines by Primenet 2015-06-08 07:11:32 (UTC), and it completed 2015-06-14 22:16:59.

[QUOTE=Madpoo;404083]Sure, I already did a triple-check on the last 3 of those. The first one I noticed didn't match between your two runs, and I was kind of waiting to see if you ran it again.

But I'll save you the hassle of doing that (and I'd want to do an independent verification anyway :smile: ) so I'll go ahead and do a check of that first one.

If you felt like returning the favor, I have an exponent where I (accidentally) ran checks on 60027139 multiple times. 4 to be exact. I think I kept adding it to my work to do and because I had the only matching runs, it would show up in my query again. Ran it 4 times total before I realized it. LOL

Anyway, if you wanted to do a verification of 60027139 for me, I'll do a check on 72316429 for you. I mean, I'll do that check of 72316429 anyway, you don't owe me anything. :smile:[/QUOTE]

Added to my worktodo.txt. Looks like two weeks to finish.

[QUOTE=endless mike;404108]Added to my worktodo.txt. Looks like two weeks to finish.[/QUOTE]

Thanks, that will help. I didn't want to leave any self-verified work dangling, and I [I]think[/I] that's the only one where I'm the one who self-verified, by mistake. :smile:

I have yours behind one other largish (62M) assignment on a 10-core worker that will finish in 21 more hours, and then it will take something like 50-55 hours... something like that.

[QUOTE=Madpoo;404115]Thanks, that will help. I didn't want to leave any self-verified work dangling, and I [I]think[/I] that's the only one where I'm the one who self-verified, by mistake. :smile:

I have yours behind one other largish (62M) assignment on a 10-core worker that will finish in 21 more hours, and then it will take something like 50-55 hours... something like that.[/QUOTE]

And done...your second check was the correct one:
[URL="http://www.mersenne.org/M72316429"]M72316429[/URL]

Updated triple-check list
 

I thought I'd update with the latest list. 452 to go.

I've got everything from 33M up still. I'm planning to add the 32M range when the other stuff is done (probably the first week of July, give or take).

The unclaimed stuff is 28.5M - 33M (92 in that range) and 27.32M - 28M (32 exponents in that range).

Every now and then a new one shows up...Curtis has been checking in some self-checked stuff in the 60M range lately which I've been adding in to be done first, so that might slow me up here and there. :smile:

My progress on my other list of multiple self-verified work (3+ checks by the same person) and "self-verified by anonymous" users is nearly done. 31 left which should be done in the next couple days. I think it started out around 500 or something. Whew! They were mostly on the small side (20M-40M) which helped me get through faster.

One final mini-list I've been keeping up with are exponents where a v5 user self verified their v4 account's work. I did have them done, but lately Curtis (and some by LaurV) have shown up. Curtis has like 3 more assigned that will check in soon where I'll do triple-checks, but no recent assignments so maybe George kept that from happening from now on? Plus, I think sometimes Curtis' system will put the same work on multiple machines, without an extra assignment. They check in within days of each other... not sure if that's on purpose to verify a suspect system, or accidental (moving work around between systems)?

Good news is that once all this backlog is done I shouldn't have a problem keeping up with new arrivals here and there. People might still accidentally get a double-check for something they did under v4, but sounds like users shouldn't be getting DC work for their own v5 submissions.

That'll be nice... maybe I can just do regular double-checking from that point on.