PrimeGrid
Please visit donation page to help the project cover running costs for this month
1) Message boards : Number crunching : Lennart Vogel Honorary Challenge (Message 131612)
Posted 348 days ago by Profile composite
Did anyone ever figure out what happened to him?

It looks to me like he's chosen to retire from computers and ham radio to spend more time enjoying nature with his family.
Let's raise a toast and respect the man's privacy (there's so little left these days).
2) Message boards : Problems and Help : How to only download one WU at a time? (Message 131607)
Posted 348 days ago by Profile composite
AND it ran the wu's using all 28 threads too?


Yep. Here is a task with bionic prefs set to 100%:

https://www.primegrid.com/result.php?resultid=1015618473

Here is a task set to 3.57%:

https://www.primegrid.com/result.php?resultid=1015629384

Both similar run times. Both show 28 threads in the Stderr output.

Also, both were reported first. ;)

28 threads is excessive: 79353.39 seconds of total CPU time for a runtime of 3374 seconds, so you have an effective CPU efficiency of 4.25%.
Using 28 threads on one task wastes over 95% of your CPU power.
You could use a lot less threads and have about the same runtime on the PPS task,
and use the other threads to run more PPS tasks simultaneously. Your throughput will be much higher.

With 4 threads per task for PPS they each should still run in around 3400 seconds.
Except that you can run 7 of them simultaneously with -t 4 if you set your %CPU to just over 1/7 => 15%
and then your CPU efficiency will be closer to 50%, and you will have many more tasks "first" each day, while you are racking up credit 7 times faster.
3) Message boards : General discussion : PrimeGrid Motto (Message 131080)
Posted 366 days ago by Profile composite
Isaac Newton in 1666 wrote:
I am ashamed to tell you to how many figures I carried these computations, having no other business at the time.
about his calculation of Pi to 15 digits.

I am sure this timeless phrase applies equally well to our quest for prime numbers.
4) Message boards : Sierpinski/Riesel Base 5 Problem : Welcome to the Sierpinski / Riesel Base 5 Project (Message 130730)
Posted 382 days ago by Profile composite
According to the 2019 Mega Primes forum thread, now (June 2019) we found the primes 88444 * 5^2799269 - 1 and 322498 * 5^2800819 - 1 (with almost 2 million digits each), so we got rid of k=88444 and k=322498 of the Riesel "part" of the SR5 conjecture, right ?

You can always see live data at http://www.primegrid.com/stats_sr5_llr.php, including all the primes found.

In case anyone missed it, the live data page is one of the Range Statistics links on the Subproject status page in the PrimeGrid menu at the top/left sidebar of this page.
5) Message boards : General discussion : Duh, what am I missing here? (Message 130679)
Posted 384 days ago by Profile composite
... and the C program computed in around 2 minutes what the Perl program did in 32 days. That's the difference between using Perl arrays and hashes, versus bit string operations in C, and being able to re-use some intermediate computations in the C version. This includes recompiling the C version with -O3, which accounted for a 4.9 times speedup over the unoptimized version, and a modest speedup of 1.3 times by switching from linear search to bisection search (which was penalized by using insertion sort for keys). I think I could do better time-wise and storage-wise by building a tree.

The benefit of having a Perl implementation for reference is that it was easy to determine the existence of program bugs in the C version, especially in memory addressing. It took a few hours to write the Perl version and over 2 weeks (with sporadic attention) to debug the C version. And I even noted a minor issue in output from the Perl version (index value off by 1, correctible after the fact).

Alas, all it did was confirm a pattern I had previously observed for k up to 250 thousand, extending that range to 30 million in 842 minutes. The arrays gobbled up 10.2 GB of RAM. I could make it more efficient memory-wise but there's no point to repeating or extending the computation. It's an interesting result. The hard part is figuring out why I'm seeing this, and if it has any consequence to what I am trying to prove.
6) Message boards : Generalized Fermat Prime Search : 271643232^131072 + 1 ? (Message 130676)
Posted 384 days ago by Profile composite
It means that some day PG will be doing a double-check of Crun-chi's previously searched range. This is not as important as the double-check for SoB's range, as in the latter case we are trying to prove a conjecture, and a missed prime there has consequences for future computing energy expenditures.
7) Message boards : Sieving : MSR Sieving - P / watt for GPUs (Message 130673)
Posted 384 days ago by Profile composite
Precisely, "power" is not a measure of energy, it is a measure of energy per unit time. We pay for electicity by energy usage, hence the billing in kilowatthours. That applies in France as well as North America.
8) Message boards : Sieving : MSR Sieving - P / watt for GPUs (Message 130628)
Posted 385 days ago by Profile composite
That's not right - calculating P per watt. Watt is a unit of measure of instantaneous power, so you need to integrate power with respect to time to make an apples-to-apples comparison for P/day and watt-hours per day, usually scaled to kilowatt hours per day. Then it would make sense to talk about energy efficiency in terms of watt-hour per P, or it's reciprocal P per watt-hour. To get this value, divide the numbers quoted in this forum by 24.
9) Message boards : General discussion : Update your Firefox browser now (Message 130627)
Posted 385 days ago by Profile composite
I'm not altogether crazy about Firefox phoning home when it wants to restore tabs after you had a browser crash (perhaps by turning off the computer). Isn't restoring tabs a local computer function? The browser contacts IPv4 addresses registered to Amazon, apparently EC2 instances. I'm thinking about categorically dropping all outgoing connections to Amazon addresses with the firewall.
10) Message boards : General discussion : i9 9900k causes noises when running Primegrid (Message 130564)
Posted 387 days ago by Profile composite
Clearly the problem is that your fans and hard disk drives are not loud enough to drown out the CPU sounds. ;)


Next 10 posts
[Return to PrimeGrid main page]
DNS Powered by DNSEXIT.COM
Copyright © 2005 - 2020 Rytis Slatkevičius (contact) and PrimeGrid community. Server load 0.02, 0.01, 0.00
Generated 14 Jul 2020 | 17:11:14 UTC