Results 1 to 25 of 25

Thread: Messy sieve coordination? (designing a better thread)

  1. #1
    Moderator ceselb's Avatar
    Join Date
    Jun 2002
    Location
    Linkoping, Sweden
    Posts
    224
    Originally posted by pixl97
    if anybody reserves before the ranges are cleaned up by the moderators

    Sort of messy right now with the gaps
    Yes. Suggestions?

  2. #2
    Moderator vjs's Avatar
    Join Date
    Apr 2004
    Location
    ARS DC forum
    Posts
    1,331
    Yes...

    Factor density is pretty much uniform from 700T to 1000T so even if they take a while it's no big deal.

    Perhaps we can suggest/require that people reserve a minimum 500G range size or at least a months worth of sieveing.


    The other possibility is a simple color change, font change or a input line as such...
    The main thread could be what's active and the archive thread could contain both reserved and complete.

    Requested Ranges: (in billions)

    359000-360000, 361050-361200, 370000-372000, 374000-374500 alexr
    413300-414000 DarkMan
    418500-420000 expinete
    480000-480500 Moo_the_cow (sieve progress:480155)
    504000-513000 VJS [Complete]
    -------Completed Ranges above this line were archived Jan 1st------
    520000-525000 Stricker (ETA: late april?)
    525000-525200 Silverfish [complete]
    525200-527000 priwo [complete]
    527000-527425 larsivi [complete]
    527425-530000 MaGnatu (ETA: 15 apr 2005)
    530000-540000 engracio [complete]
    540000-541000 larsivi
    541000-542000 Mat67 [complete]
    542000-543000 Troodon
    543000-543300 larsivi [complete]
    543300-543400 maniacken [complete] 5 factors
    543400-544000 Silverfish [complete] 16 factors
    544000-545000 royanee [complete]
    545000-546000 Maddog1 [complete]
    546000-548000 Nuri
    548000-549500 larsivi [complete]
    549500-550000 Rychek [complete]
    550000-550500 Mystwalker [complete]
    550500-553000 Keroberts1
    553000-564000 VJS [Complete] --2^49--
    564000-566000 Nuri
    566000-569000 hc_grove
    569000-570000 priwo [complete]
    570000-570200 torque2k [completed]
    570200-571000 AlanSimpson
    571000-573000 engracio [complete]
    573000-574000 priwo [complete]
    574000-576000 Nuri
    576000-577000 Rychek [complete]
    577000-578000 qbnkiller
    578000-578100 MikeH [complete]
    578100-578150 Keller [complete] 0 factors
    578150-579000 torque2k
    579000-584000 engracio [complete]
    584000-586000 Nuri [complete]
    586000-586300 jaat
    586300-599000 engracio [complete]
    599000-600000 Maddog1 [complete]
    -- Above will be archived Feb 14th --
    600000-606000 MikeH
    606000-608000 Nuri
    608000-609000 shoelace [complete] 26 factors
    609000-610000 Mat67 [complete]
    610000-611000 Rychek [complete]
    611000-612000 MikeH [complete]
    612000-613000 royanee [complete]
    613000-626000 engracio [complete]
    626000-626066 Death [complete] 4 factors
    626066-626600 Frodo42
    626600-626666 Death [complete] 1 factor found
    626666-628000 jimmy
    628000-630000 Nuri [complete]
    630000-630500 Silverfish [complete] - 15 factors found
    630500-630600 chris [complete] 49 factors found
    630600-630650 CedricVonck [complete]
    630650-631000 Grafux [Reserved]
    631000 631050 CedricVonck [Complete] 2 factors
    631050-631500 cjohnsto [complete] 13 factors
    631500-631900 larsivi [complete]
    631900-632500 priwo [complete]
    632500 632510 sslazak [Complete]
    632510-632520 RolanCCC [complete]
    632520-632550 Rum [complete]
    632550-632600 RolanCCC [complete]
    632600-632700 sslazak [Complete]
    632700-633000 Rum [complete]
    633000-634000 Maddog1
    634000-635000 Keroberts1
    635000-665500 engracio (ETA: Late March)
    665500-666000 MaGnatu (ETA: 15 feb 2005)
    666000-666666 Death [complete] 11 factors
    666666-667000 TheCrusher [complete]
    667000-667100 MaGnatu (ETA: 30 ian 2005)
    667100 669000 MikeH
    669000-669066 Death [complete] 1 factor found
    669066-670000 Rychek [ETA late Feb]
    -- Above will be archived Mar 21st --
    670000-670200 sslazak [complete]
    670200-671000 DarkMan
    671000-672000 Rychek
    672000-672100 CedricVonck
    672100-673000 cjohnsto (ETA: End of March)
    673000-673500 shoelace (ETA Mid Feb)
    673500-674000 sslazak
    674000-675000 frodo42
    675000-675300 larsivi
    675300-675500 Rum
    675500-676500 Silverfish (ETA:end of March)
    676500-677500 priwo
    677500-678000 Troodon
    678000-678500 Stromkarl (ETA: end Feb)
    678500-679500 Mat67
    679500-684000 pixl97 [reserved]
    684000-685000 royanee

    ------- No Gaps in reservation exist above this line ------
    ------- Only reserve ranges below this line -----------



    690000-697000 MikeH
    697000-699000 Nuri
    699000-700000 Death (ETA: March)
    700000-700300 Moo_the_cow
    700300-700320 Guilherme [complete] - no factors found
    700320-700400 Guilherme (ETA: Feb/04/2005)
    700400-700450 Guilherme (ETA: Feb/09/2005)

    Ranges over 750T:
    777000-778000 pixl97 [complete]
    Last edited by vjs; 02-08-2005 at 09:43 PM.

  3. #3
    Another idea could be to ask people to reserve the first free range SEQUENTIALLY and not accept any reservations leaving gaps behind them! (although this would have been far more effective if it was implemented from the start of the sieving project...)
    If you don't want to do that, VJS' plan sounds reasonable to me.

  4. #4
    Moderator vjs's Avatar
    Join Date
    Apr 2004
    Location
    ARS DC forum
    Posts
    1,331
    The gaps ahead don't really bother me, often times people are testing for factor density, or speed at a certain point. Want to sieve around a range where they found a factor or some other valid reason. I know for certain JoeO for example does alot of this type of testing.

    Also those ranges above 1120T are only sieveable currently with slow sieve clients.

    In the past I also use to reserve 10000G or more 100T ahead of the main effort. It's always been suggested that people don't MAKE holes, for example...

    899500-900000 MikeH [complete]
    999000-999900 Joe_O [complete}

    Reserving:
    900000-910000 username [3 months]
    or
    990000-999000 username [3 months]
    for example

    Is a perfectly good reserve, if not appreciated especially if you can't stay on top of things, complete quickly, or have the time etc... pixl97 for example is doing a great job up there.

    905000-905050 username [3days]

    Is probably not appreciated by most people here.


    One thing I'd like to say however, regardless of what's reserved where and by whom. Everything sieved really helps the project and we should do more to encourage sieve than discourage sieve. Especially in our home teams that prp etc. If you look at my condensed list you can acutally see that not very many people are acutally sieving compared to prp. Also there are even fewer people factoring than sieve.
    Last edited by vjs; 02-08-2005 at 06:46 PM.

  5. #5
    Moderator ceselb's Avatar
    Join Date
    Jun 2002
    Location
    Linkoping, Sweden
    Posts
    224
    Originally posted by vjs
    Factor density is pretty much uniform from 700T to 1000T so even if they take a while it's no big deal.
    Agreed.
    Perhaps we can suggest/require that people reserve a minimum 500G range size or at least a months worth of sieveing.
    Forcing a minimum range will drive (some) people away.

    The other possibility is a simple color change, font change or a input line as such...
    Sure, but not as the only method, we'd still have a mess.

    The main thread could be what's active and the archive thread could contain both reserved and complete.
    It seems like a lot more work (for me). Also you'd still have to have the upper reserves visable to prevent cross reserves.

    Why not go all out and only state free ranges like this?:
    Code:
    EXAMPLE***EXAMLE
    685000-690000 FREE
    700450-777000 FREE
    778000-875000 FREE
    880000-899500 FREE
    900000-999000 FREE
    1000050-1100000 FREE
    1100050-1111110 FREE
    1111150-1120000 FREE
    EXAMPLE***EXAMLE

  6. #6
    Moderator vjs's Avatar
    Join Date
    Apr 2004
    Location
    ARS DC forum
    Posts
    1,331
    celseb,

    Agreed with everything you said except for the all out. Also I don't really like the idea of >500G either, lets drop this idea altogether.

    Why not simply try something simple first...

    1.

    Still use my idea of the yellow and green lines

    ------- No Gaps in reservation exist above this line ------
    ------- Only reserve ranges below this line -----------
    684000- <--- Forget this as well too confusing

    2.

    Everything above the green reservation line could be small

    3.

    You could also try changing

    -- Completed ranges above this line were archived Jan 1st --

    This would be the same as the lowest active range.



    Note...

    My post was edited to reflect these changes.

  7. #7
    why not simply give more people the ability to edit the thread so that it can be cleaned more often?

  8. #8
    Senior Member
    Join Date
    Jan 2003
    Location
    UK
    Posts
    479
    Don't we have a place for discussions like this?

    My opinion. It's been working OK for some years now, carry on as usual

  9. #9
    TeamRetro Siever
    Join Date
    Oct 2004
    Location
    Lebanon, NH
    Posts
    39
    FYI, from a noob's point of view, reserving a range is confusing. Here is a bit of what I went through to get a range reserved.

    I started by reading the first message through. It wasn't immediately obvious where to reserve a range, so I checked each range in the whole message for gaps. Once I found a gap (not quite near the end), I wrote down the numbers I wanted to reserve. Then I went to the end of the thread to reserve the range. I started to write a reply to the thread when I noticed that someone else had already reserved the same range. So, I went back and read all the remaining replies to the first message, adjusted my range, then reserved the range.

    Some of the things that would improve this immensely are:
    Originally posted by ceselb

    Why not go all out and only state free ranges like this?:
    Code:
    EXAMPLE***EXAMPLE
    685000-690000 FREE
    700450-777000 FREE
    778000-875000 FREE
    880000-899500 FREE
    900000-999000 FREE
    1000050-1100000 FREE
    1100050-1111110 FREE
    1111150-1120000 FREE
    EXAMPLE***EXAMPLE
    IMHO, Free ranges should be ADDED to the current list.


    Originally posted by vjs

    Why not simply try something simple first...
    1.
    ------- No Gaps in reservation exist above this line ------
    3.
    -- Completed ranges above this line were archived Jan 1st --
    I also think that these two items could be added to clear things up a bit. Just change the month and day the files were archived. Keep the warning date for future archiving. In addition, I think a message to warn sievers to check the message replies for reserved ranges not yet incorporated into the main list would help.

    Just thought you might want see what this thread looks like from noob's point of view. And yes, I know that there is another thread for this, but I didn't want to split up the discussion.

    Stromkarl

  10. #10
    Moderator vjs's Avatar
    Join Date
    Apr 2004
    Location
    ARS DC forum
    Posts
    1,331
    Stromkarl,

    I see your point but strongly disagree with the marking of free ranges

    Problem is if it's marked as free people may stop there and just think "well it's free" even if it has been reserved further down the page, also it may promote small reservations and holes etc.

    Mike is correct of course, in the last year or so there has maybe been 3-5 times where people have reserved the same range twice. Generally we go by the rule of first come first serve of course.

    I think, however that my suggestion of reserve below this line "no gaps in reserveation, etc above this line". May help people out, also making the upper ones smaller would also help.

    I wouldn't make drastic changes either, the reserve below this line works in P-1.

    My appologies to Mike about the discussion in this thread hopefully everything will be moved momentarily.

  11. #11
    Forgotten Member
    Join Date
    Dec 2003
    Location
    US
    Posts
    64
    The biggest problem I see here is that were using a message board to reserve ranges, its not very effective, and makes a lot of work for the mods. Like trying to pull a livestock trailer with a Yugo

    If it were me doing this I'd whip out a page with a php script on it that had a few options on it like

    Id like to reserve the lowest avalible range of (selectable) size,
    Id like to reserve (typehere) to (typehere) - then put a limit of 10G on each selection, so one dumb user could not lock up 200T or so by accident.
    or maybe for fun, Id like to select a random range of (selectable) size.

    have php check this against a mysql backend so no overlapping ranges can be selected.

    With a setup like this clients could write there own scripts to autoreserve/get blocks for proth to do. We would be well on our way to automated sieveing!

    This is not something that needs done right away, but with 2^50 comming closer every day, or sieveing over 20M, a more automated reserve system would be welcome.

    Mods Imagine no more populating the sieve statistic pages my hand, just have the scripts do it for you. Im not sure why noone has done this yet, you must enjoy the punishment

    edit: after thinking about it, if the moderators of this forum forced the programmers of the software to manage the sieve by hand, this would have been corrected a long time ago. Give a programmer the choies of 1. make a script to do it 2. do it by hand or 3. slave labor, most programmers choose 3.

    Feel free to move this to a new thread, its getting way too long in here.
    Last edited by pixl97; 02-09-2005 at 04:58 PM.

  12. #12
    Moderator ceselb's Avatar
    Join Date
    Jun 2002
    Location
    Linkoping, Sweden
    Posts
    224
    Originally posted by MikeH
    Don't we have a place for discussions like this?
    Yes, moved to a brand new thread

    Originally posted by pixl97
    The biggest problem I see here is that were using a message board to reserve ranges, its not very effective, and makes a lot of work for the mods.
    I agree and yes, it's a lot of work.

    Originally posted by pixl97
    If it were me doing this I'd whip out a page with a php script
    Great, I'd love something like that.

    Originally posted by pixl97
    after thinking about it, if the moderators of this forum forced the programmers of the software to manage the sieve by hand, this would have been corrected a long time ago. Give a programmer the choies of 1. make a script to do it 2. do it by hand or 3. slave labor, most programmers choose 3.
    Absolutely. What do you suggest I do about it? Go on strike?

    Originally posted by Stromkarl
    FYI, from a noob's point of view, reserving a range is confusing. [...] It wasn't immediately obvious where to reserve a range. [...] Just change the month and day the files were archived. Keep the warning date for future archiving. In addition, I think a message to warn sievers to check the message replies for reserved ranges not yet incorporated into the main list would help.

    Just thought you might want see what this thread looks like from noob's point of view.
    Thanks. I've made the suggested changes.


    (To everyone) Comments on the 'new' look?

  13. #13
    Senior Member engracio's Avatar
    Join Date
    Jun 2004
    Location
    Illinois
    Posts
    237
    ceselb,


    Looks good to me, THANKS!!!!



    e


    BTW, This range [635000-665500 engracio (ETA: Late March)]
    will probably be done ETA: Mid-April. Thanks again for doing a GREAT job on a Thankless job.



  14. #14
    Moderator vjs's Avatar
    Join Date
    Apr 2004
    Location
    ARS DC forum
    Posts
    1,331
    Looks better celseb,

    I still suggest making the portion with no gaps a smaller case like the high ranges.

    Also the red is hard on my eyes, I'd really make that green and the archive dates yellow.

    If you need a hand editing posted etc just ask for help. I'd help out, I'm sure a few others would as well, MikeH should definetly have the ability.

  15. #15
    Moderator vjs's Avatar
    Join Date
    Apr 2004
    Location
    ARS DC forum
    Posts
    1,331
    The large text for the red is much better

    You may also want to consense both our posts into the main post, client link your tutorial etc. You could even make it a short paragraph to save the number of lines

  16. #16
    Moderator vjs's Avatar
    Join Date
    Apr 2004
    Location
    ARS DC forum
    Posts
    1,331
    Example,

    To start sieving first visit MikeH's Site then download the cmov client for non-P4's or the sse2 for P4's. Extract the new dat into the same folder as the client. A tutorial for rieselsieve may help out. For references purposes the old dat, a clear text version, and A very small zip file, indicating the next couple of days of main and DC PRP efforts and the daily shrinkage of the new sob.dat file.
    A Excel spreadsheet, to help you calculate, g/day based upon speed, time to complete a range. Also Sobistator, a great addon tool!!!

    No need to write out the code, as a moderator I can edit all posts. /ceselb
    Last edited by ceselb; 02-10-2005 at 12:38 PM.

  17. #17
    Moderator vjs's Avatar
    Join Date
    Apr 2004
    Location
    ARS DC forum
    Posts
    1,331
    Wasn't sure about the code... just tring to help out.

    Looks alot better now IMHO, hopefully it hasn't changed so much that people become confused. I really can't tell the difference between the orginal and now. But then again after a while who looks closely.

    I'll get some fresh eyes...

  18. #18
    Just thought i'd drop a post to say i hope to have an automated seiveing allocation thingy running shortly.. hopefully by teh weekend.. but it maybe monday instead.

    i'll keep y'all posted

  19. #19
    TeamRetro Siever
    Join Date
    Oct 2004
    Location
    Lebanon, NH
    Posts
    39
    ceselb,
    Could you please shorten the reminder about duplicate reserves so it doesn't wrap around to the next line?

    Thanks,
    Stromkarl

  20. #20
    Moderator ceselb's Avatar
    Join Date
    Jun 2002
    Location
    Linkoping, Sweden
    Posts
    224
    Originally posted by Stromkarl
    so it doesn't wrap around to the next line?
    Wow, you on 800x600 or something? Better now?

  21. #21
    TeamRetro Siever
    Join Date
    Oct 2004
    Location
    Lebanon, NH
    Posts
    39
    Originally posted by ceselb
    Wow, you on 800x600 or something? Better now?
    Actually, at home I am on 1024 x 768. At work it looked ok with 1280 x 1024. I think it has to do with the font size you are using more than anything else.

    It looks great now! Thanks,

    Stromkarl

  22. #22
    Moderator vjs's Avatar
    Join Date
    Apr 2004
    Location
    ARS DC forum
    Posts
    1,331
    Your webbrowser is probably set to largest fornt size.

  23. #23
    TeamRetro Siever
    Join Date
    Oct 2004
    Location
    Lebanon, NH
    Posts
    39
    Originally posted by vjs
    Your webbrowser is probably set to largest fornt size.
    Thanks! That was it. I knew that you can set font size in HTML, but I wasn't aware that the brower over-rode that. Is that something new, or just something I overlooked all these years?

    Stromkarl

  24. #24
    It's been there for a long time, you may have accidently changed the font size perhaps by holding control and using the mouse wheel?



  25. #25
    Senior Member
    Join Date
    Apr 2004
    Location
    Florianopolis - Santa Catarina - Brazil
    Posts
    114
    I can help ceselb on the maintenance of the threads "Sieve coordination (1M - 20M)" and "Archived sieve ranges (1M - 20M)", if you guys wish.

Posting Permissions

  • You may not post new threads
  • You may not post replies
  • You may not post attachments
  • You may not edit your posts
  •