Potential conflict .ca vs .org

[snme2pm1]

Hi friends, I have not visited often in the last year or two, but still been diligent.
I also am aware that where is a channel for James, but I just don't recall.
Suggest move this to that channel.
If I tap https://www.mersenne.ca/status/tf/0/0/4/11800, there are reported 4 exponents suffering below 76 bits of exploration, yet from .org only two exponents are reported.
Am I missing something?

[snme2pm1]

Originally Posted by snme2pm1

Hi friends, I have not visited often in the last year or two, but still been diligent.
I also am aware that where is a channel for James, but I just don't recall.
Suggest move this to that channel.
If I tap https://www.mersenne.ca/status/tf/0/0/4/11800, there are reported 4 exponents suffering below 76 bits of exploration, yet from .org only two exponents are reported.
Am I missing something?

I'm unimpressed that discrepancies would arise at all, I always imagined the arrangements to be relatively safe.

[axn]

Thread moved to correct forum

[James Heinrich]

Originally Posted by snme2pm1

If I tap https://www.mersenne.ca/status/tf/0/0/4/11800, there are reported 4 exponents suffering below 76 bits of exploration, yet from .org only two exponents are reported.

The two exponents that are really still at 74 are 118690421 and 118690987 -- both have been assigned for TF to "CHM" since 2021-04-28 with no progress (I don't know why Primenet hasn't released the assignment yet).

The other two exponents were 118088989 and 118715687 which both showed as being at 75 instead of 76. In both cases the effort for 74-75 and 75-76 was reported by the same user at the exact same time. I suspect for whatever reason the results were presented for import out of order (since they have the same timestamp) and the system won't update the TF for new results if there's a gap -- if it saw the 75-76 result before the 74-75 result it would ignore it, then update the exponent to 75 when it (very soon after) sees the 74-75 result. However the database is somewhat self-correcting, simply looking at the exponent page on mersenne.ca was sufficient for it to realize that the TF gap had been filled and so both exponents are now automatically at 76 (as will be shown on tomorrow's TF status report).

[snme2pm1]

Originally Posted by James Heinrich

exponent to 75 when it (very soon after) sees the 74-75 result. However the database is somewhat self-correcting, simply looking at the exponent page on mersenne.ca was sufficient for it to realize that the TF gap had been filled .

Don't misunderstand, I love your work.
Good that any discrepancy has been resolved.
I have habits of sorting incoming material, how about that idea?
Sorry if such messages wake you in a different (opposing) time zone.

[snme2pm1]

Originally Posted by snme2pm1

Don't misunderstand, I love your work.
Good that any discrepancy has been resolved.
I have habits of sorting incoming material, how about that idea?
Sorry if such messages wake you in a different (opposing) time zone.

I would be prepared to forward some C code that I regularly use to sort file records (Memory + File Heapsort).

[Zhangrc]

Originally Posted by James Heinrich

The two exponents that are really still at 74 are 118690421 and 118690987

I'd like to manually reserve them and TF them to 2^77, they are:
Factor=118690421,74,77
Factor=118690987,74,77

[kriesel]

Originally Posted by Zhangrc

I'd like to manually reserve them and TF them to 2^77

Then do so. For 1G<p<2³², read and use the upper left portion of https://www.mersenne.ca/tf1G/

[James Heinrich]

Originally Posted by kriesel

Then do so. For 1G<p<2³², read and use the upper left portion of https://www.mersenne.ca/tf1G/

But the two exponents in question are solidly in GIMPS range, 118M. Zhangrc is going to have to "poach" them since PrimeNet has (for whatever reason) not released the 6-month-old assignments (maybe it will release it at the 180-day mark = Oct 21?).

[snme2pm1]

Originally Posted by James Heinrich

Zhangrc is going to have to "poach" them since PrimeNet has (for whatever reason) not released the 6-month-old assignments

I suspect that ancient assignments are not a rarity.
Yet imagine that the best practice would be that they be expired in a useful manner.
Yet also I am aware that some assignments become completed long after expiry.

Even in my region of play I can see 118002503 being 760 days out of expectation.
Not sure why it is helpful that such be retained.
I guess that GW isn't bothered about some such, yet.

[Zhangrc]

Originally Posted by snme2pm1

Even in my region of play I can see 118002503 being 760 days out of expectation.

Doesn't matter. The rule of Category 4 says "Assignments are recycled if assignment is not started within 180 (unless manual testing) days or when the exponent moves midway into the first category and the assignment is more than 360 days old." The 118M exponent is now in Cat 4 and it'll take another year before it falls in Cat 1 (unless someone like Ben Delo increases the computation power)

Also I'm holding the exponent 141145141 (yes, with a "114514" in it ), which runs on an old machine and would certainly take years to complete. If it gets expired, I would be very disappointed.