Primenet web design - Page 51

kladner · 2015-11-03, 22:26

Quote:

Originally Posted by chalsall

Stupid...

... (Programmer Error)....

Yours, or his?

LaurV · 2015-11-04, 05:36

Quote:

Originally Posted by Madpoo

EDIT again: I've kind of confirmed... it's the type of work you get when you select "trial factoring to low limits" from the client. Is/was that a part of the Lone Mersenne Hunters thing? I'm trying to make sense of some comments in the code.

It was (and is) a type of TF reserved for very old CPUs that can not do LL and have too less memory to do P-1 or ECM. It is what we would call today "bitlevel first" (or "breadth first", as opposite to "deep first"), where one guy takes a lot of assignments of very high exponents to very low bitlevels. The invention of mfaktX rendered it obsolete (the low limit bitlevel raised, even for "not so high" exponents, where TF is slower, the GPUs are much faster, etc)

Madpoo · 2015-11-04, 18:57

Quote:

Originally Posted by LaurV

It was (and is) a type of TF reserved for very old CPUs that can not do LL and have too less memory to do P-1 or ECM. It is what we would call today "bitlevel first" (or "breadth first", as opposite to "deep first"), where one guy takes a lot of assignments of very high exponents to very low bitlevels. The invention of mfaktX rendered it obsolete (the low limit bitlevel raised, even for "not so high" exponents, where TF is slower, the GPUs are much faster, etc)

I guess that would explain why I haven't really encountered them or paid attention. There are precisely 10 active assignments of that kind.

snme2pm1 · 2016-09-10, 01:44

http://www.mersenne.org/report_factors
Would it be feasible to have an optional tickbox filter to show just first found factors?
That could diminish noise and reduce the need to probe statistics of an exponent deeper.
Edit: I seek to mask subsequently found factors for a particular time range, perhaps the last month, which are somehow less exciting.
PS. An alternative semantic might be to choose just the lowest factor.

petrw1 · 2016-09-10, 02:46

Quote:

Originally Posted by snme2pm1

http://www.mersenne.org/report_factors
Would it be feasible to have an optional tickbox filter to show just first found factors?
That could diminish noise and reduce the need to probe statistics of an exponent deeper.
Edit: I seek to mask subsequently found factors for a particular time range, perhaps the last month, which are somehow less exciting.
PS. An alternative semantic might be to choose just the lowest factor.

+1

For example I've seen the same few factors reported hundreds of time for ECM-F for 4096.
Granted it is still a results and the contributor is still getting credit for it but I'd agree to have an option to NOT see duplicates.

snme2pm1 · 2017-06-03, 07:13

It seems a long time since anybody paid attention to this thread.
Also a while since I've visited; I've been extremely overworked in recent months.

Is that to suggest that everything is as best as it should be?
Well I suspect, largely so...
I was the instigator of a thrust to revise date formats to more ISO provisions, first at mersenne.org then later at mersenneforum.org.
It is far more joyous to see dates these days.
Mostly. I hope the thrust is sustained and not backsliding.
Many thanks for Aaron on this subject already, for his persistence and partial calming of the presented material, including some older content...
There remain several places at mersenne.org where a cocktail assortment of date formats might challenge the senses of a viewer, including: https://www.mersenne.org/report_milestones/
At the time there was a tacit acknowledgment that pages with editorial styled comment might venture to retain a US date style, though that stance is obviously strained for the page mentioned, with different date formats exhibited.
The use of "Third-Quarter" perhaps provides for a further aspect of discussion, and the ISO provisions I suspect provide a suitable remedy, if deemed tolerable to the masses?
I believe there are further such places that could be subject to similar comment.
G
Ed: Just looked up ISO 9601, but did not find any mention of quarters! Bother!

Madpoo · 2017-06-04, 20:37

Quote:

Originally Posted by snme2pm1

There remain several places at mersenne.org where a cocktail assortment of date formats might challenge the senses of a viewer, including: https://www.mersenne.org/report_milestones/
At the time there was a tacit acknowledgment that pages with editorial styled comment might venture to retain a US date style, though that stance is obviously strained for the page mentioned, with different date formats exhibited.
The use of "Third-Quarter" perhaps provides for a further aspect of discussion, and the ISO provisions I suspect provide a suitable remedy, if deemed tolerable to the masses?
I believe there are further such places that could be subject to similar comment.

Yeah, the milestone page has long dates (e.g. "January 3rd, 2011") for what you might call editorial content, but the same aggregated info for all such milestones is presented in the lists using ISO 8601 (YYYY-MM-DD). Either way, I don't think there's any ambiguity.

The use of "quarter" for some of the milestones is unfortunate since the exact date is unknown and we only had the approximate timeframe.

If you find other instances of dates in the wrong format, just holler. I think we've managed to get them all standardized but it's possible we missed something (press releases aside, which probably have the long format).

LaurV · 2017-06-06, 04:17

Quote:

Originally Posted by Madpoo

but the same aggregated info for all such milestones is presented in the lists using ISO 8601 (YYYY-MM-DD)

potonono · 2017-06-15, 22:34

The latest Firefox version (v54.0 64-bit) isn't handling the 'blue' theme right on pages like mersenne.org/results or mersenne.org/workload. Text, tables, and most elements are fine, but the blue color is being applied on top of images (like Marin's picture) until it is hovered over, then it pops in front of the blue background.

James Heinrich · 2017-06-15, 23:52

I can confirm it's a FF54 issue, the page looked perfectly normal in FF53, I applied the update and now it's... wonky.

Madpoo · 2017-06-18, 18:35

Quote:

Originally Posted by James Heinrich

I can confirm it's a FF54 issue, the page looked perfectly normal in FF53, I applied the update and now it's... wonky.

Well that's weird. It's been a while since I set all that up and as far as I can remember it's just making CSS changes (through Javascript).

I can't think of anything off the top of my head that should have made an update in Firefox cause such a drastic difference.

I might have to dust off my install of Firefox (if I even have it installed anymore) and see what's what.

2016-09-10, 01:44	#554
snme2pm1 "Graham uses ISO 8601" Mar 2014 AU, Sydney 11110011₂ Posts	Known factors http://www.mersenne.org/report_factors Would it be feasible to have an optional tickbox filter to show just first found factors? That could diminish noise and reduce the need to probe statistics of an exponent deeper. Edit: I seek to mask subsequently found factors for a particular time range, perhaps the last month, which are somehow less exciting. PS. An alternative semantic might be to choose just the lowest factor. Last fiddled with by snme2pm1 on 2016-09-10 at 02:21

2017-06-03, 07:13	#556
snme2pm1 "Graham uses ISO 8601" Mar 2014 AU, Sydney 11110011₂ Posts	Dates again... It seems a long time since anybody paid attention to this thread. Also a while since I've visited; I've been extremely overworked in recent months. Is that to suggest that everything is as best as it should be? Well I suspect, largely so... I was the instigator of a thrust to revise date formats to more ISO provisions, first at mersenne.org then later at mersenneforum.org. It is far more joyous to see dates these days. Mostly. I hope the thrust is sustained and not backsliding. Many thanks for Aaron on this subject already, for his persistence and partial calming of the presented material, including some older content... There remain several places at mersenne.org where a cocktail assortment of date formats might challenge the senses of a viewer, including: https://www.mersenne.org/report_milestones/ At the time there was a tacit acknowledgment that pages with editorial styled comment might venture to retain a US date style, though that stance is obviously strained for the page mentioned, with different date formats exhibited. The use of "Third-Quarter" perhaps provides for a further aspect of discussion, and the ISO provisions I suspect provide a suitable remedy, if deemed tolerable to the masses? I believe there are further such places that could be subject to similar comment. G Ed: Just looked up ISO 9601, but did not find any mention of quarters! Bother! Last fiddled with by snme2pm1 on 2017-06-03 at 07:32 Reason: Maybe ISO 9601 doesn't help with quarters

Similar Threads
Thread	Thread Starter	Forum	Replies	Last Post
Database design	xilman	Astronomy	1	2017-04-30 22:25
Theoretical Experiment Design	c10ck3r	Homework Help	7	2015-02-03 08:54
Digital Logic Design	henryzz	Puzzles	9	2014-12-04 20:56
new intel design	tha	Hardware	5	2007-04-19 11:38
design factoring algorithms	koders333	Factoring	14	2006-01-25 14:08

2017-06-15, 22:34	#559
potonono Jun 2005 USA, IL 193 Posts	The latest Firefox version (v54.0 64-bit) isn't handling the 'blue' theme right on pages like mersenne.org/results or mersenne.org/workload. Text, tables, and most elements are fine, but the blue color is being applied on top of images (like Marin's picture) until it is hovered over, then it pops in front of the blue background.

2017-06-15, 23:52	#560
James Heinrich "James Heinrich" May 2004 ex-Northern Ontario 11×311 Posts	I can confirm it's a FF54 issue, the page looked perfectly normal in FF53, I applied the update and now it's... wonky.