[QUOTE]Unverified LL A4209DE258C211__ by "Flavius"[/QUOTE]

I has a first time LL done, a verified one is called a doublecheck, so you should doublecheck it instead, since it need a shifted bitrate to count as a doublecheck.

I *think* you can just right:

[Worker #1]
Doublecheck=18000343,67,1

but not sure, can't remember format for doublecheck.

[quote=ATH;147850]I has a first time LL done, a verified one is called a doublecheck, so you should doublecheck it instead, since it need a shifted bitrate to count as a doublecheck.

I *think* you can just right:

[Worker #1]
Doublecheck=18000343,67,1

but not sure, can't remember format for doublecheck.[/quote]

Yes, that's right, except I don't know if worktodo is case sensitive - you might have to put in [B]D[/B]ouble[B]C[/B]heck=xxxxxxxx,xx,x.

Also note that if you see "Suspect LL" for an LL residue in an exponent's history, then someone has returned a first-time LL result, but with error codes. So, if you wanted to run such an exponent, you would do so as a first time LL test, adding Test=xxxxxxxx,xx,x to worktodo.

[quote=petrw1;147833]Okay ... I took a chance.

I looked at the Exponent Status report and found:
[code]18000343 No factors below 2^67
P-1 B1=100000, B2=1850000
Unverified LL A4209DE258C211__ by "Flavius" [/code]

This appears to be a test that does NOT have a first time LL test done (at least not a verified one).

So I created a worktodo.add file and inserted the lines:
[code][Worker #1]
Test=18000343,67,1[/code]

About 2 seconds after saving it (and without me doing anything else) my status window showed the Worker stopping and starting and the worktodo.txt now has the line in it without any warnings or errors.

I'll keep you posted.

Update: I refresehed my screen on the Exponent Status report but I am not showing up as Assigned. Maybe this won't happen until my test is active?

Can I manually edit the worktodo.txt file and move this new assignment right below the currently active one to see sooner what will happen?

Update 2: It does NOT show up on my Account Assignment report yet.[/quote]

You need to force a Manual Communication with the PrimeNet server. Get worktodo set up the way you want, and then stop Prime95. Then continue Prime95, and your added exponents will be picked up. Next go to Advanced -> Manual Communication. Check the box for "Send new expected completion dates to server" and then hit OK. You should see your exponents get updated one by one in the Communication Thread window (with registration for assignment keys, if necessary). Once the communication with the server is done, your new assignment(s) should appear in your account report. Note, however, that if you happen to be inadvertently taking an already assigned exponent, you will get a "No Assignment - exponent already assigned" message from PrimeNet during the communication. This means that your exponent more than likely belongs to someone else, and should *not* be picked up at this time. (If in doubt, you can always run an Exponent Status query on the exponent(s) in question.)

Yes, Manual Communication made it show up and it accepted it as a first time LL test.

And with the new KEY

[QUOTE=starrynte;147777]if it is assigned the exponent report will show:

332272879No factors below2^61
AssignedTrial factoring to "starrynte" on 2008-11-04[/QUOTE]

So the Project Summary shows 1 available LL in the 18M range. I could do the Exponent Status report and manually look for it but the report is limited to 100 Exponents. This could take a LONG time. Is there not an easier way to find this 1 Available LL test?

Mind you this contradicts the Miletone reports that says all exponents up to 22M have been tested once.

[QUOTE=petrw1;147867]Yes, Manual Communication made it show up and it accepted it as a first time LL test.

And with the new KEY[/QUOTE]

Why are you doing a new first time test, when this "Flavius" allready did it?

[QUOTE=ATH;147895]Why are you doing a new first time test, when this "Flavius" allready did it?[/QUOTE]

I guess I misunderstood the meaning of "Unverified Test" to mean no valid LL test was done yet. I will fix my assignment

For future reference, if you're looking at an exponent's history, and you see:

"Unverified LL" - this means that the first LL has been completed, and no error codes were returned. What you could do: Double-check this exponent (or, I suppose, if a really low level of P-1 was done, you could re-P1 the exponent in hopes of finding a factor).

"Verified LL" - this means that *both* the first LL and the double-check have been completed, without error codes, and both residues agreed. What you could do: Try P-1 if you really wanted to look for a factor, and the previous P-1 efforts were on the low side.

"Suspect LL" - this means that the first LL has been completed, and harmful error codes were returned. There's about a 50-55% chance the listed residue is incorrect. What you could do: P-1 if it hasn't been already done to a decent level, or run a *first* LL. I have had my exponents poached before and turned in with suspect LLs - when I ran them myself, I *always* got different residues.

"Bad LL" - this means that two (or more) LLs have been returned and one (or more) of their residues do not agree. What you could do: P-1 if it's worthwhile, then run a first check, or double (or n-tuple) check as appropriate.

As I mentioned above, keep in mind that residues having error codes attached are about 55% likely to be wrong. Thus an exponent with a suspect LL, in my opinion, is just as interesting/exciting to test as a first-time, never-been-tested exponent.

[quote=petrw1;147868]So the Project Summary shows 1 available LL in the 18M range. I could do the Exponent Status report and manually look for it but the report is limited to 100 Exponents. This could take a LONG time. Is there not an easier way to find this 1 Available LL test?

Mind you this contradicts the Miletone reports that says all exponents up to 22M have been tested once.[/quote]

Right, and I'm inclined to believe the milestone report, as even on the v4 server, we've been beyond 18M for a long time now. I wonder if this is a case of a "stuck" exponent, or if there is just an error in the database.

I agree we need a better way of tracking down lone assignments, though. Looking through a range of 1M, 100 exponents at a time, is soul-destroying. Maybe this is something George will be working on when he gets back from vacation.

tracking lost exponents
 

To track exponents you do not have to use "Exponent Status" with its limit of 100 results, just use "LL Results", excluding the verified results and the exponents for which a factor was found.
The only thing you could miss there is an untested exponent. To track those is more complicated : make a list of primes in the range you want to explore, tick of those for which factors where found, the queries are limited to 10 000 rows, so query ranges of 200 000 at a time (e.g. : 17000000-17200000), tick of the LL results, queries are limited to 10 000 rows for each category (verified, unverified, factored and bad.)

With the list of leftover exponents you can now use the Exponent status query to check if they are assigned to a user by PRIMENET. Be careful !!! I think some exponents have been assigned to some users without internet access and are not tracked in the database : they would appear unreserved while they are assigned to someone by George. Meaning a bad case of poaching !!! There are also a lot of exponents stuck on the v4 server for different reasons... It is better, IMO, to wait for such queries until George is back from holidays and give him and Scott the time to stabilise the situation. Then you could send the query result to George and ask for his comments.

Jacob

Yes I think that exponent may be manually assigned to someone by George and this may be a case of poaching. Also some computers from v4 still haven't communicated with the server and given that there used to be a 60 day period before expiry we should give them 60 days before accepting v5's assertion that an exponent is unassigned.