332.2M - 333.9M (aka 100M digit range) - Page 14

chalsall · 2009-07-20, 15:08

Quote:

Originally Posted by Primeinator

For example, if 330 - 331M has 20,000 exponents that have been trial factored, and there are two colors on this column (one color representing 0 to 5,000 and the second 5,000 to 20,000) then I assume this means that 5,000 have been factored at the first depth and 15,000 (20 - 5) at the other? Sorry for my run on sentence. It is late and my grammar seems to be deteriorating from inadequate sleep...

That is correct.

Each column is the total number of exponents which have not yet been factored (or at least, that will be the case shortly -- there are a few which have been factored which my database doesn't know about; these generally appear as low factor levels).

Within each column, each colour represents the total number of exponents which have been trial factored to that depth.

So, for example, on the 332M to 333M graph, the column for 332.5 (to 332.6) shows maybe 40 which have been trial factored to 2^64 (in fact these are where a factor was found), approximately 800 trial factored to 2^65, and the rest trial factored to 2^66.

I'm working on a simple web interface, so people will be able to view any range they want. I'll let everyone know when it's online.

Primeinator · 2009-07-20, 15:17

Excellent, that would be most helpful.

I think I have decided to assist this project once my current batch of LLs finish on my laptop. I assume the time required to trial factor to 2^66 or so is trivial on a 2.0Ghz 32bit processor?

Primeinator · 2009-07-20, 20:20

I would like to go ahead and reserve fifty or so exponents that are below 62 bits. Could someone point to where I can find available exponents to pick through? Thanks.

petrw1 · 2009-07-20, 20:34

Quote:

Originally Posted by Primeinator

I would like to go ahead and reserve fifty or so exponents that are below 62 bits. Could someone point to where I can find available exponents to pick through? Thanks.

http://www.mersenne.org/report_facto...=1&B1=Get+Data

Primeinator · 2009-07-20, 20:39

Quote:

Originally Posted by petrw1

http://www.mersenne.org/report_facto...=1&B1=Get+Data

Thank you. Now how do I reserve a certain group and then put them in my worktodo? Note- I will not start on it until Aug 2 or so, which is when my laptop will be free. But before that happens, I need to have it switch work types.

Uncwilly · 2009-07-20, 20:44

Quote:

Originally Posted by Primeinator

Thank you. Now how do I reserve a certain group and then put them in my worktodo? Note- I will not start on it until Aug 2 or so, which is when my laptop will be free. But before that happens, I need to have it switch work types.

http://www.mersenneforum.org/showthread.php?t=11308
It's a sticky.

chalsall · 2009-07-22, 14:44

Quote:

Originally Posted by chalsall

I'm working on a simple web interface, so people will be able to view any range they want. I'll let everyone know when it's online.

Hey all.

Well, it's not "painted" yet (read: no pretty CSS), and the navigation isn't yet as easy as I plan (read: no clickmaps), but the web interface is now online.

Please see http://www.wabbit.com/trialfactored.cgi and let me know if you have any comments. If anyone can suggest a better pallet for the colors, that would be great (I'm *so* not a graphic artist).

I'm attaching an example of a rendering of the entire range currently in Primenet.

Note that I've moved away from the "3D Bars", as while they look cool, they actually make it harder to determine values.

10metreh · 2009-07-22, 15:33

Quote:

Originally Posted by chalsall

Hey all.

Well, it's not "painted" yet (read: no pretty CSS), and the navigation isn't yet as easy as I plan (read: no clickmaps), but the web interface is now online.

Please see http://www.wabbit.com/trialfactored.cgi and let me know if you have any comments. If anyone can suggest a better pallet for the colors, that would be great (I'm *so* not a graphic artist).

I'm attaching an example of a rendering of the entire range currently in Primenet.

Note that I've moved away from the "3D Bars", as while they look cool, they actually make it harder to determine values.

Just curious, who has been taking large numbers of exponents in the 700M range to 63 bits?

chalsall · 2009-07-22, 15:57

Quote:

Originally Posted by 10metreh

Just curious, who has been taking large numbers of exponents in the 700M range to 63 bits?

Appears to be "monst".

http://www.mersenne.org/report_expon...&B1=Get+status

cheesehead · 2009-07-22, 17:11

Quote:

Originally Posted by chalsall

Please see http://www.wabbit.com/trialfactored.cgi and let me know if you have any comments.

Looks great!

Now, on behalf of those of us who are color-vision-impaired, could you add varied cross-hatching or something to help us see the differences between the bit-levels denoted by colors we confuse? For instance, to me the colors for levels 65, 67 and 71 are practically identical (maybe they really are!).

Here are two web sites that might help:

http://webstandards.psu.edu/accessib...lor/colorblind

http://webdesign.about.com/od/access...a/aa062804.htm

Let me add that this wouldn't be essential for distinguishing the bars on the graph, just a nice feature (and that's why I've put this in small font). From the alternations of colors I can distinguish, tracing back to the bit-level color legend takes only a second or two.

10metreh · 2009-07-22, 17:46

Quote:

Originally Posted by cheesehead

For instance, to me the colors for levels 65, 67 and 71 are practically identical (maybe they really are!).

They aren't: 65 is lighter orange, 67 is darkish green, 71 is dark orange. But those can look similar if you're colour-blind.

2009-07-20, 15:17	#145
Primeinator "Kyle" Feb 2005 Somewhere near M52.. 3×5×61 Posts	Excellent, that would be most helpful. I think I have decided to assist this project once my current batch of LLs finish on my laptop. I assume the time required to trial factor to 2^66 or so is trivial on a 2.0Ghz 32bit processor? Last fiddled with by Primeinator on 2009-07-20 at 15:17

Similar Threads
Thread	Thread Starter	Forum	Replies	Last Post
GPU72 / MISFIT use for 100M digit range?	Uncwilly	GPU to 72	64	2013-03-31 02:45
I want a 100M digit Mersenne that....	JuanTutors	PrimeNet	8	2012-12-06 13:47
How far along are you in your 100M digit LL test?	JuanTutors	Lounge	6	2012-02-21 07:36
100M-digit n/k pairs	__HRB__	Riesel Prime Search	0	2010-05-22 01:17
100M digit prime	Unregistered	Information & Answers	10	2010-03-24 20:16

2009-07-20, 20:20	#146
Primeinator "Kyle" Feb 2005 Somewhere near M52.. 3·5·61 Posts	I would like to go ahead and reserve fifty or so exponents that are below 62 bits. Could someone point to where I can find available exponents to pick through? Thanks.