mersenneforum.org

mersenneforum.org (https://www.mersenneforum.org/index.php)
-   PrimeNet (https://www.mersenneforum.org/forumdisplay.php?f=11)
-   -   Trial Factor assignments squatting in Cat 0 and Cat 1 DC range (https://www.mersenneforum.org/showthread.php?t=26001)

Uncwilly 2020-09-23 17:05

Trial Factor assignments squatting in Cat 0 and Cat 1 DC range
 
As of this moment there are 68 TF assignments in the DC Cat 0 and 1 range below 54M. [B][U]All were assigned in the last month.[/U][/B]:bob:
And they have not been showing progress like they should if they were using a GPU.
[url]https://www.mersenne.org/assignments/?exp_lo=53310629&exp_hi=54000000&execm=1&exdchk=1&exfirst=1&exp1=1[/url]

Why is this allowed? These are keeping other users from completing the DC on them. The assignments were (I am guessing) gotten from the manual GPU assignment page. It lists the bottom of the available DC-TF range as the lowest exponent yet to get a DC. That should be moved up into Cat 3. The issue is the same with the FTC-TF range.

It is getting sorely tempting to just do the DC to kill them off. They are not even slow, they are stopped And with this many exponents not real likely to find a single factor.

De Wandelaar 2020-09-23 17:28

A majority of those TF's were assigned between 2020-09-14 and 2020-09-22.
Poaching those assigments after a couple of days or weeks (while the participants are not at fault) could imho discourage them definitively from contrinuting especially if they are relatively new participants.

storm5510 2020-09-23 17:59

Curious: Why TF's on exponents which have a LL test needing verified? I looked at the detail for the first three. :confused:

Uncwilly 2020-09-23 18:08

[QUOTE=De Wandelaar;557657]A majority of those TF's were assigned between 2020-09-14 and 2020-09-22.
Poaching those assigments after a couple of days or weeks (while the participants are not at fault) could imho discourage them definitively from contrinuting especially if they are relatively new participants.[/QUOTE]
They should not be open to TF assignments. It should be DC only. If these are newbies (all of them are ranked less than 100 on the TF top producers list), they may not understand that they should not be working down there and holding on to the exponents. All the more reason to not allow them to take the exponents. They have moved up a bunch in the 90 day range. The highest, assuming a dead start 90 days ago is producing 708 GHz-days/day (which is less than a single GeForce GTX 1070).

Uncwilly 2020-09-23 18:14

[QUOTE=storm5510;557659]Curious: Why TF's on exponents which have a LL test needing verified? I looked at the detail for the first three. :confused:[/QUOTE]
If they had too little TF already That is not the case. Or they hope to find a factor and eliminate the need for a DC. Current DC TF range according to GPU72 should be in the 67M range. If someone wanted to hammer away at the 58M range, fine, that is way out in the Cat 3 area. Just take 2 weeks worth of work at a time and turn it in.

ATH 2020-09-23 18:25

I sent a PM to Madpoo about this thread, lets see if he can fix it.

It is curious these cat0 and cat1 TF assignments are stuck so often with no progress. It might be deliberate attempts to stall / sabotage.

De Wandelaar 2020-09-23 18:30

[QUOTE=Uncwilly;557660]They should not be open to TF assignments. It should be DC only. If these are newbies (all of them are ranked less than 100 on the TF top producers list), they may not understand that they should not be working down there and holding on to the exponents. All the more reason to not allow them to take the exponents. They have moved up a bunch in the 90 day range. The highest, assuming a dead start 90 days ago is producing 708 GHz-days/day (which is less than a single GeForce GTX 1070).[/QUOTE]

Completely agree with your statement, those exponents should not be open to TF assignments.
Assignment algorithmm is not 100 % waterproof but again, participants are absolutely not at fault.

Do you think it's better to poach those exponents or to wait (worst case) for the 60 days expiry date ?

Edit : @ATH, I'm not convinced of deliberate attempts to stall but it's not impossible (??)

firejuggler 2020-09-23 19:14

Or maybe do a short PM1 on those and hope for the best.

S485122 2020-09-23 19:46

[QUOTE=De Wandelaar;557665]...
Do you think it's better to poach those exponents or to wait (worst case) for the 60 days expiry date ?
...
[/QUOTE]Those TF assignments have no expiry date. But poaching is a lack of respect of the other users. In the past that kind of TF assignments were completed ... after a time.
[QUOTE=firejuggler;557667]Or maybe do a short PM1 on those and hope for the best.[/QUOTE]Once again this would be poaching.
[QUOTE=ATH;557663]I sent a PM to Madpoo about this thread, lets see if he can fix it.
...[/QUOTE]I did the same yesterday : PM to Aaron, James. James forwarded to George.

Jacob

It is curious these cat0 and cat1 TF assignments are stuck so often with no progress. It might be deliberate attempts to stall / sabotage.

kruoli 2020-09-23 20:40

[QUOTE=S485122;557669][...] poaching is a lack of respect of the other users. [...][/QUOTE]

Yes, that's especially true if the assignees did not do this on purpose, which I cannot determine.

On the other hand, if someone is trying to hinder progress knowingly, I do not think we need to respect them doing this. :beatdown:

storm5510 2020-09-23 23:03

[QUOTE=ATH;557663]I sent a PM to Madpoo about this thread, lets see if he can fix it.

It is curious these cat0 and cat1 TF assignments are stuck so often with no progress. It might be deliberate attempts to stall / sabotage.[/QUOTE]

It seems to me that if anyone were going to stall / sabotage, they would do it with a lot more than 68. If [U]Madpoo[/U] cannot do a fix, he may be able to give a clearer reason as to how this happened.


All times are UTC. The time now is 13:50.

Powered by vBulletin® Version 3.8.11
Copyright ©2000 - 2021, Jelsoft Enterprises Ltd.