mersenneforum.org

mersenneforum.org (https://www.mersenneforum.org/index.php)
-   Aliquot Sequences (https://www.mersenneforum.org/forumdisplay.php?f=90)
-   -   Aliquot sequences that start on the integer powers n^i (https://www.mersenneforum.org/showthread.php?t=23612)

garambois 2022-02-05 18:22

I reserve the base 87 for initialization.

Tomorrow, February 6th, I will do a complete update, unless something unexpected happens.

garambois 2022-02-06 08:34

[QUOTE=garambois;599465]
Tomorrow, February 6th, I will do a complete update, unless something unexpected happens.[/QUOTE]


I'm sorry, but FactorDB's automatic scan fails after a while.
I feel like I no longer have permission to scan all the sequences in the project at once.
So I would have to scan a whole part of the sequences manually as I go along and wait an hour each time to be able to continue.
That would take me a whole day or even longer and I don't have that much time.
So I can't do an update of all the databases.
This time I will just add the new bases and modify the reservations.
[B]Please, has anyone changed the number of sequences we can scan on FactorDB ?[/B]
It becomes difficult to do the work in these conditions !
Sorry if I'm wrong about the cause of my problem to scan FactorDB.

garambois 2022-02-06 09:27

[QUOTE=EdH;598138]I can look at exponents 43 toward 60.

Base 70 table is fully initialized (I think). There are several merges (most already noted):[code]70^3:i23 merges with 1134:i2020
[/code]Working on base 71. . .[/QUOTE]

Please, there must be an error for the 70^3 merger.
I find 70^3:i33 = 1134:i9

Aillas 2022-02-06 09:56

[QUOTE=garambois;599507]I'm sorry, but FactorDB's automatic scan fails after a while.
I feel like I no longer have permission to scan all the sequences in the project at once.
So I would have to scan a whole part of the sequences manually as I go along and wait an hour each time to be able to continue.
That would take me a whole day or even longer and I don't have that much time.
So I can't do an update of all the databases.
This time I will just add the new bases and modify the reservations.
[B]Please, has anyone changed the number of sequences we can scan on FactorDB ?[/B]
It becomes difficult to do the work in these conditions !
Sorry if I'm wrong about the cause of my problem to scan FactorDB.[/QUOTE]

It seems someone is flooding FactorDB, so Markus as lowered the limit. You can see the original post [URL="https://mersenneforum.org/showpost.php?p=599388&postcount=487"]here[/URL]

EdH 2022-02-06 13:03

[QUOTE=garambois;599509]Please, there must be an error for the 70^3 merger.
I find 70^3:i33 = 1134:i9[/QUOTE]I get what you have, too.

Oddly, the 70^3:i23 term is 8720 and the last four digits of the 1134:i2020 term are also 8720. Whatever script I used to detect merges must have made that connection. However, nothing has changed and my script now agrees with your result.

Thanks for keeping me accurate.

EdH 2022-02-06 14:32

[QUOTE=garambois;599507]I'm sorry, but FactorDB's automatic scan fails after a while.
. . .[/QUOTE]How hard are you hitting the db?

I have a full set of all the sequences locally stored and use a couple scripts to update the whole set periodically. I just tested a full update and it ran fine. I think it took less than an hour. (I need to put a time check in the script.) But, I'm not sure what would happen if I needed to start fresh, or even if I was adding several tables.

The first script is base centered and takes the input for the base and last exponent. It then looks at all the local .elfs for that base and does the following:

- if a sequence doesn't exist, it is downloaded
- if a sequence does exist
- - check for prime last term
- - - if prime, move to next sequence
- - - if not prime, get last line from db
- - - - if they match move to next sequence
- - - - if they don't match, download full .elf

All my .elfs are stored in a folder called elfstudy as "elfstudy/base##/base^exponent.elf" e.g. "elfstudy/base8128/8128^34.elf"

The second script is just a list of calls to the first, supplying all the bases and exponents to run. If it was the case that a full update all at once was not doable, pauses could be placed within this script, using "sleep 3600" for an hour pause, or similar.

If you're interested in the scripts, let me know and I'll clean them up a little and post them.

Edit: I did some experimenting and adding a full table did cause a limit crash during a full update, so maybe my scripts would need the pauses added in. I might play further, but unfortunately, the lockouts are affecting other things I have running against the db currently.

garambois 2022-02-06 15:34

[QUOTE=Aillas;599510]It seems someone is flooding FactorDB, so Markus as lowered the limit. You can see the original post [URL="https://mersenneforum.org/showpost.php?p=599388&postcount=487"]here[/URL][/QUOTE]

Thank you very much, I did not have this information.
And do you know if it would be possible to ask for the possibility to have an extension of this limit for some IPs, for example mine ?
(But I guess everyone would want such an extension !)


[QUOTE=EdH;599516]I get what you have, too.

Oddly, the 70^3:i23 term is 8720 and the last four digits of the 1134:i2020 term are also 8720. Whatever script I used to detect merges must have made that connection. However, nothing has changed and my script now agrees with your result.

Thanks for keeping me accurate.[/QUOTE]

Everything looks good now !


[QUOTE=EdH;599519]How hard are you hitting the db?

I have a full set of all the sequences locally stored and use a couple scripts to update the whole set periodically. I just tested a full update and it ran fine. I think it took less than an hour. (I need to put a time check in the script.) But, I'm not sure what would happen if I needed to start fresh, or even if I was adding several tables.

The first script is base centered and takes the input for the base and last exponent. It then looks at all the local .elfs for that base and does the following:

- if a sequence doesn't exist, it is downloaded
- if a sequence does exist
- - check for prime last term
- - - if prime, move to next sequence
- - - if not prime, get last line from db
- - - - if they match move to next sequence
- - - - if they don't match, download full .elf

All my .elfs are stored in a folder called elfstudy as "elfstudy/base##/base^exponent.elf" e.g. "elfstudy/base8128/8128^34.elf"

The second script is just a list of calls to the first, supplying all the bases and exponents to run. If it was the case that a full update all at once was not doable, pauses could be placed within this script, using "sleep 3600" for an hour pause, or similar.

If you're interested in the scripts, let me know and I'll clean them up a little and post them.

Edit: I did some experimenting and adding a full table did cause a limit crash during a full update, so maybe my scripts would need the pauses added in. I might play further, but unfortunately, the lockouts are affecting other things I have running against the db currently.[/QUOTE]

Thank you very much Edwin for this proposal !
This will really be my last resort and it's still a solution in case everything else fails.
But I hope that the "classic" method will work again very soon.
I'm going to try it again later today or tomorrow and hopefully it will work properly !
I'll keep you posted...

garambois 2022-02-06 15:53

Page updated.
Many thanks to all for your help !

[B]Added bases : 70, 71, 77, 88, 95, 96, 97, 99, 800, 968.[/B]
[B]Update of bases 50, 79, 385 and [/B][B]200560490130 with reservations.[/B]

Our project now has 123 bases and 11282 sequences.
Can you please check the items that apply to you and notify me of any errors ?

I added the bases 77 and 88 whose initialization is not finished for the Open-End sequences.
But I have reserved them and I will finish the calculations up to 125 digits for all the sequences, so it will be more than an initialization.

I release from my reservation the bases 200560490130 and 385, except for the sequence 385^22 that I am calculating.

RichD 2022-02-07 11:15

99^61 & 99^63 terminates.

EdH 2022-02-07 14:05

Base 968 table is fully "primed."

Moving haphazardly to working on some of the high end of base 98 among other things. . .

Aillas 2022-02-08 07:51

[QUOTE=garambois;599522]Thank you very much, I did not have this information.
And do you know if it would be possible to ask for the possibility to have an extension of this limit for some IPs, for example mine ?
(But I guess everyone would want such an extension !)
[/QUOTE]

I think so, but it will not be by IP. I think you must create an account on factordb and use this account when connecting. The limit should be set by account not by IP. If I well remember.

Maybe we should ask this privilege directly for a list of people (from here) instead of one by one individually ?


All times are UTC. The time now is 05:34.

Powered by vBulletin® Version 3.8.11
Copyright ©2000 - 2022, Jelsoft Enterprises Ltd.