mersenneforum.org The Appallingly Blue Page
 Register FAQ Search Today's Posts Mark Forums Read

 2012-07-12, 17:50 #12 Batalov     "Serge" Mar 2008 Phi(4,2^7658614+1)/2 2×4,591 Posts ...I was just looking at the magic page too: it still has 99240 in the old state, so... March explains it! EDIT: after checking a few latest recorded terminations we can date this webpage between #428106 and #438966, i.e. late Jan .. early Feb. Last fiddled with by Batalov on 2012-07-12 at 17:54
 2012-07-13, 03:42 #13 Dubslow Basketry That Evening!     "Bunslow the Bold" Jun 2011 40
 2012-07-13, 17:56 #14 Greebley     May 2009 Dedham Massachusetts USA 15138 Posts 2^2*7 makes sense as the most common, I think. 2^2 is fairly common to hit and it isn't hard to get a factor of 7. 2*3 is harder because the 1st power of 2 is harder to get, and 2^4*31 is harder because of the 31. I was surprised that 2^4*31 was as common as it was though. I would have thought that one would occur less with the 31 being uncommon. There is also a strong bias towards having a driver, because sequences without one will usually be pursued first, until they get too big, pickup up a driver, terminate, or merge. Only the first option leaves them on the list as a guide
2012-07-13, 22:16   #15
fivemack
(loop (#_fork))

Feb 2006
Cambridge, England

23×797 Posts

Quote:
 Originally Posted by Dubslow I added the driver counts to the stats page, and 2^2*7 is by far the most prevalent driver, followed by 2*3 and 2^3*3*5.
Maybe I should leave off the 180-digit sequences and apply some serious smiting to the 114-digit ones; my last batch of smiting produced the visible spike at 125 ...

But the Big Machine isn't Internet-connected at the moment.

 2012-07-22, 07:27 #16 Dubslow Basketry That Evening!     "Bunslow the Bold" Jun 2011 40
2012-07-22, 15:22   #17
henryzz
Just call me Henry

"David"
Sep 2007
Cambridge (GMT/BST)

34·71 Posts

Quote:
 Originally Posted by Dubslow PS Should I add a class column? The reason I ask is because while 2^2 * 7 is class -1, 2^2*7^2 is class 2, even though they technically have the same driver.
Sounds a good idea here. Can I also suggest checkboxes that hide sequences with drivers/3s?

2012-07-22, 16:51   #18
Dubslow

"Bunslow the Bold"
Jun 2011
40<A<43 -89<O<-88

3·29·83 Posts

Quote:
 Originally Posted by henryzz Can I also suggest checkboxes that hide sequences with drivers/3s?
Umm... what? You mean hide sequences with drivers, leaving only sequences with guides (not drivers) that don't include a 3?

If I interpreted that correctly, well... I'm not sure how to do that. You can try looking through here, and I'll try and think of a possible way to do it...

By the way, I've had reports that sorting by Last Updated does an alphabetical sort, rather than sort by date. It works fine for me (sorts by date) with Linux/Chrome/AdBlock, but I've had reports of failure (sort by letters of day) on Linux/Firefox/No-blocking and on Mac/Firefox/Safari/ad-and-js-blockers. How many of you see this?

 2012-07-22, 18:03 #19 henryzz Just call me Henry     "David" Sep 2007 Cambridge (GMT/BST) 34×71 Posts I see I have thoroughly confused you. I meant have a filter for sequences with drivers. This could be done using regular expressions or other filters. Maybe you could use a hidden column although I am not certain how you would allow us to use the feature. This could also be done for removing sequences with 3 as a factor as they rise and it can be hard to lose the 3. Thinking this might be harder to do depending on how you implement this. This post is a bit brainstormy but should hopefully give you some clues.
2012-07-22, 18:26   #20
Dubslow

"Bunslow the Bold"
Jun 2011
40<A<43 -89<O<-88

3·29·83 Posts

Quote:
 Originally Posted by henryzz I see I have thoroughly confused you.
Yep
Quote:
 Originally Posted by henryzz I meant have a filter for sequences with drivers.
Still lost.

When you say filter, do you mean activate the filter and the sequences with drivers disappear from the table?
Quote:
 Originally Posted by henryzz Maybe you could use a hidden column although I am not certain how you would allow us to use the feature.
Actually, I already use two hidden columns. One is a version of the date in integer format, which is supposed to be what's sorted on when sorting the dates, not the dates themselves (but I guess this isn't working for some browsers). (The other hidden column contains the FDB-ID of the last line, used for merge detecting.)

Code:
// visible in my page's source code
"aoColumnDefs": [
{ "sClass": "factorcol", "aTargets": [4, 5] },
{ "bSearchable": false, "bVisible": false, "aTargets": [3, 6] }, // id and sorttime are invisible
{ "aDataSort": [7], "aTargets": [6] }
// Sort column 7 by the invisible column 6
]

Last fiddled with by Dubslow on 2012-07-22 at 18:32 Reason: [code]

2012-07-22, 22:35   #21
henryzz
Just call me Henry

"David"
Sep 2007
Cambridge (GMT/BST)

575110 Posts

Quote:
 Originally Posted by Dubslow Yep Still lost. When you say filter, do you mean activate the filter and the sequences with drivers disappear from the table?
Yes. Currently you can sort by driver/guide which is useful but finding the easy sequences is still not trivial. A class column would help but sequences with hard factors to lose(mainly 3) that aren't part of the driver won't be classed as harder.

Quote:
 Actually, I already use two hidden columns. One is a version of the date in integer format, which is supposed to be what's sorted on when sorting the dates, not the dates themselves (but I guess this isn't working for some browsers). (The other hidden column contains the FDB-ID of the last line, used for merge detecting.) Code: // visible in my page's source code "aoColumnDefs": [ { "sClass": "factorcol", "aTargets": [4, 5] }, { "bSearchable": false, "bVisible": false, "aTargets": [3, 6] }, // id and sorttime are invisible { "aDataSort": [7], "aTargets": [6] } // Sort column 7 by the invisible column 6 ]
I noticed looking at the source. I didn't find it very easy to read I hope you understand it better than I do. Don't try too hard just to please me. Assuming you have worked out the language/api enough you should understand it better than me.

2012-07-22, 22:48   #22
Dubslow

"Bunslow the Bold"
Jun 2011
40<A<43 -89<O<-88

3×29×83 Posts

Quote:
 Originally Posted by henryzz Yes. Currently you can sort by driver/guide which is useful but finding the easy sequences is still not trivial. A class column would help but sequences with hard factors to lose(mainly 3) that aren't part of the driver won't be classed as harder.
What I do is sort by driver, then sub-sort by factors columns so all those with 3s come first, then 5s, then 7s, etc., within each driver.

E.g., search for 2^2, sort the driver column to put the 2^2 on top, then shift-click on factors. You can see all the 2^2s * 1-- or *2---, scroll past the 3s and see the rest.

tl;dr: get creative with sub sorting

Nonetheless, I'll look into a checkbox to filter out drivers or 3s. It might take a few days though. (And the other changes I mentioned would be done today will wait until then, because adding/removing columns is something of a pain.)

 Similar Threads Thread Thread Starter Forum Replies Last Post xilman Science & Technology 24 2015-10-16 00:03 retina Lounge 32 2015-03-28 13:08 fetofs Puzzles 28 2005-11-03 15:50 ixfd64 Hardware 9 2005-11-01 11:34 dave_0273 Hardware 17 2005-05-19 14:33

All times are UTC. The time now is 15:50.

Tue Dec 1 15:50:16 UTC 2020 up 82 days, 13:01, 3 users, load averages: 3.12, 2.67, 2.21