The Appallingly Blue Page - Page 21

EdH · 2018-04-03, 15:37

Is it just me, or did you dim the cell background color a bit? I like the color of the previous image (post #212) better than the new one (post #218). Maybe it's just the capture? The blue border looks the same.

Column formatting is better in the latter.

Thanks for all the work.

Dubslow · 2018-04-03, 18:44

Hmm, no I didn't change anything but it looks like my screenshot-taking method somehow did something different, even though I didn't. Pretty sure the former is "actual" correct version. None of the background blues have changed, just the in-table link color.

EdH · 2018-04-04, 03:58

I think it's all good then...

gd_barnes · 2018-04-06, 00:02

On the statistics page here: https://www.rechenkraft.net//aliquot/statistics.html

There are a few inaccuracies as a result of project progression:

1. "This page contains some simple statistics based on the overall status of all un-terminated Aliquot sequences < 1M." <--- Change to 2M.
2. "First, a table giving the counts of how many sequences are at what size. (Sizes < ~110 are probably FactorDB sequence errors.)" <--- Statement in parenthesis could be changed or removed.
3. "Third, a table giving the counts of how many sequences are at what length. (Lengths < 400 are definitely FDB errors.)" <--- Statement in parenthesis could be changed or removed.

Dubslow · 2018-04-06, 00:12

That's exactly why I've been working on getting everything to 400 ~~digits~~ terms since the extension

Probably just going to change it anyways. Ah well. Thanks for the other heads ups as well.

LaurV · 2018-04-06, 04:29

Quote:

Originally Posted by Dubslow

That's exactly why I've been working on getting everything to 400 digits...

Wow... You must have a couple hundreds of peta-flops under the hood there.... hehe...

On topic: the link colors look much better than the original version. Could you try zooming in/out in Firefox (or what browser you use) and see how the new fonts behave? (this is because some of us keep the zoom level to 80-90 and not all fonts scale well, also, are they pure text that can be changed from the FF's options/general/language/fonts&colors tab? - very useful for some pages especially with lots of text and with large tables, to show my font, not the embedded font). It is not requirement, just question. Don't fix the font in nails...

Dubslow · 2018-04-06, 04:41

Quote:

Originally Posted by LaurV

Wow... You must have a couple hundreds of peta-flops under the hood there.... hehe...

Oooops.... certainly is handy to be able to edit posts for more than an hour!

Quote:

Originally Posted by LaurV

On topic: the link colors look much better than the original version. Could you try zooming in/out in Firefox (or what browser you use) and see how the new fonts behave? (this is because some of us keep the zoom level to 80-90 and not all fonts scale well, also, are they pure text that can be changed from the FF's options/general/language/fonts&colors tab? - very useful for some pages especially with lots of text and with large tables, to show my font, not the embedded font). It is not requirement, just question. Don't fix the font in nails...

No fonts have changed. In fact I think none are specified, you can probably fiddle with yourself in your browser.

LaurV · 2018-04-06, 08:51

Quote:

Originally Posted by Dubslow

No fonts have changed. In fact I think none are specified, you can probably fiddle with yourself in your browser.

As I already see them different it means that it is possible, and the browser is showing my own font already, for the actual page. So, everything is well here. Except maybe about fiddling with myself, but about that I don't do in the browser, and I don't like to discuss in public...

Dubslow · 2018-04-06, 18:49

Oh jeez

Keep in mind that my screenshots have the font used by my browser/OS and are not representative of what someone else may see.

Dubslow · 2018-05-16, 03:54

Quote:

Originally Posted by ChristianB

There is more progress to report for the progress page (pun intended). It is now fully automated and running hands-off. Sequences that are determined as terminated are checked using a recent download of the ELF file from factordb using aliqueit to make sure the sequence has no broken indices in factordb. This will make the next extension of the list much less painful than the last one as mergers and terminations are automatically purged from the list. Things are also written and set up that, should the current server ever fail, we can spin up the list on a new server within hours.

Quote:

Originally Posted by Dubslow

To extend this, yesterday CB officially pushed to production new code that does the same for mergers, specifically it verifies the entire elf file of both the sequence to be dropped and the sequence it merged into, and then emails us with the common index:

As a result of this second batch of work in this vein, all sequences dropped from the table are verified via aliqueit for correctness, so that any errors in the Factor Database will never (again??) result in a faulty drop of a sequence. (We can't guarantee the non-existence of errors in sequences listed in the table, but we can guarantee that any such errors will never lead to an incorrect unlisting of a sequence.

In this vein, using my recently made python+pexpect+yafu clone of R. Gerbicz's "aliquot siever", I made a list of all sequences below 1e6 which remain open after factoring up to 10^40, then compared to those currently in the master blue page list, and found these 1133 seqs to be verified once more:

Code:

I'm downloading ELFs from the FDB at a rate of roughly 5 per hour to be checked my aliqueit, so a bit over a week to check the lot. This will ensure that no seqs are missing from the table either, in addition to preventing future missing seqs.

The original 10277 seqs that survived sieving to 10^40 are attached. If it's easy for you to independently verify, I would appreciate verification of this list.

LaurV · 2018-05-16, 13:56

Quote:

Originally Posted by Dubslow

I'm downloading ELFs from the FDB at a rate of roughly 5 per hour to be checked my aliqueit

No need. I have my DB independent of factorDB and there is no conflict compared with the list on the blue page.

Edit: In fact, you can just add this list to the blue page, and see they will resolve by themselves in a week or so, anyhow do not need to dld them separately.

2018-04-06, 00:12	#225
Dubslow Basketry That Evening! "Bunslow the Bold" Jun 2011 40<A<43 -89<O<-88 3·29·83 Posts	That's exactly why I've been working on getting everything to 400 ~~digits~~ terms since the extension Probably just going to change it anyways. Ah well. Thanks for the other heads ups as well. Last fiddled with by Dubslow on 2018-04-06 at 04:39

Similar Threads
Thread	Thread Starter	Forum	Replies	Last Post
Russian Blue	xilman	Science & Technology	24	2015-10-16 00:03
How do you explain the colour blue to someone that was born blond?	retina	Lounge	32	2015-03-28 13:08
Blue-eyed Islanders	fetofs	Puzzles	28	2005-11-03 15:50
Blue Gene\L	ixfd64	Hardware	9	2005-11-01 11:34
Blue Screen Of Death	dave_0273	Hardware	17	2005-05-19 14:33

2018-04-03, 15:37	#221
EdH "Ed Hall" Dec 2009 Adirondack Mtns 3,539 Posts	Is it just me, or did you dim the cell background color a bit? I like the color of the previous image (post #212) better than the new one (post #218). Maybe it's just the capture? The blue border looks the same. Column formatting is better in the latter. Thanks for all the work.

2018-04-03, 18:44	#222
Dubslow Basketry That Evening! "Bunslow the Bold" Jun 2011 40<A<43 -89<O<-88 3×29×83 Posts	Hmm, no I didn't change anything but it looks like my screenshot-taking method somehow did something different, even though I didn't. Pretty sure the former is "actual" correct version. None of the background blues have changed, just the in-table link color.

2018-04-04, 03:58	#223
EdH "Ed Hall" Dec 2009 Adirondack Mtns 3,539 Posts	I think it's all good then...

2018-04-06, 00:02	#224
gd_barnes May 2007 Kansas; USA 19×541 Posts	On the statistics page here: https://www.rechenkraft.net//aliquot/statistics.html There are a few inaccuracies as a result of project progression: 1. "This page contains some simple statistics based on the overall status of all un-terminated Aliquot sequences < 1M." <--- Change to 2M. 2. "First, a table giving the counts of how many sequences are at what size. (Sizes < ~110 are probably FactorDB sequence errors.)" <--- Statement in parenthesis could be changed or removed. 3. "Third, a table giving the counts of how many sequences are at what length. (Lengths < 400 are definitely FDB errors.)" <--- Statement in parenthesis could be changed or removed.

2018-04-06, 18:49	#229
Dubslow Basketry That Evening! "Bunslow the Bold" Jun 2011 40<A<43 -89<O<-88 3·29·83 Posts	Oh jeez Keep in mind that my screenshots have the font used by my browser/OS and are not representative of what someone else may see.