1)
Message boards :
News :
World Record Woodall Prime!
(Message 116703)
Posted 105 days ago by  Luigi R.
Wow, what a discovery! Congratulations! :)
|
2)
Message boards :
Number crunching :
Year of the Dog Challenge
(Message 116089)
Posted 121 days ago by Luigi R.
Since 321 is only available as CPU apps, that leaves your GPU free to do something else. It won't count for the challenge, but why leave the GPU idle when it can be doing something useful?
Because electricity costs. :D
|
3)
Message boards :
Number crunching :
Year of the Dog Challenge
(Message 116088)
Posted 121 days ago by Luigi R.
I thought the start was today. Well, I'm going to crunch some 321 tasks from now.
|
4)
Message boards :
Number crunching :
Year of the Dog Challenge
(Message 116050)
Posted 122 days ago by Luigi R.
llr321 application is unknown because you have not downloaded it yet or there isn't a file/excecutable named like that. ;)
|
5)
Message boards :
Number crunching :
2018 Tour de Primes
(Message 115520)
Posted 140 days ago by Luigi R.
Thanks for this observation
Talking about PPSE...
total_tasks/primes=128029045/15036=8514.834
That's an average.
I don't know if now it is easier or more difficult finding them. I guess the second one.
Michael said between 10k and 20k.
|
6)
Message boards :
Number crunching :
2018 Tour de Primes
(Message 115511)
Posted 140 days ago by Luigi R.
With a little over 48 hours left of the Tour de Primes challenge it doesn't seem likely that I will end up finding a prime in the PPSE tasks... for PPSE is it still roughly 1 in 8000 tasks to find a prime? I had hoped to get over 10k tasks for this month (and even dished out my old Windows Vista laptop sitting around doing nothing to help run tasks in the middle of the challenge, lol) but I may end up getting a little over 8k, at 7120 now and haven't examined the number much but found I was averaging ~300-400 tasks a day or so.
Don't worry, I ran almost 20000 PPSE this month and I'm pretty sure there is someone that has run more ones. xD
So if I don't pull off a last minute workunit prime I may probably discover one a few weeks into the next month.
Firstly, prime distribution will not remember your past efforts. Next month you will have the same chances of the 1st of February when (maybe) you had run 0 tests.
I go on intuition. Task distribution is random, so discoveries and doublechecks are 'like' independent events. So, if chances are 1/8000tasks, you have roughly 2/8000 chances to be first discover or double checker. This thinking would make our misfortune crazier.
@mathematicians: confirm or reject?
|
7)
Message boards :
Number crunching :
2018 Tour de Primes
(Message 115458)
Posted 142 days ago by Luigi R.
A personal request for those of you who already have earned a TdP badge: please don't complain on the forums about how much better someone else is doing, or how this or that is unfair, or how you've been so unlucky. You've got a badge. Lot's of people do not, despite their best efforts. Please don't rub it in.
Thanks for remembering this.
I don't want to bore anyone about my stats, I guess there has been someone unluckier than me.
This TdP taught me how freaky and fickle fate can be, even when your chances are good.
I was thinking about it yesterday. There are times when you want to be who something happened to and other times when you are lucky not to be the chosen one.
|
8)
Message boards :
Number crunching :
2018 Tour de Primes
(Message 115252)
Posted 147 days ago by Luigi R.
And thanks for reading and commenting...
I read it this morning. I have to study it before commenting. I'm from smartphone, this is my excuse. ;)
|
9)
Message boards :
Number crunching :
2018 Tour de Primes
(Message 115187)
Posted 148 days ago by Luigi R.
I picked these ones.
"I could tell you, but then I'd have to kill you."
It reminds me Stargate SG-1 Season 1 Episode 1 when Colonel Jack O' Neill says: "I’ve thought about it. But then I’d have to shoot anyone that actually read it." LOL
"It's really complicated."
"It works exactly like you think it should."
I was fine with these ones, then the serious answer has completed the picture. Thanks.
|
10)
Message boards :
Number crunching :
2018 Tour de Primes
(Message 115181)
Posted 148 days ago by Luigi R.
On the other hand, by delaying sending out the second task in the workunit, the server allowed you to become the double checker. Otherwise it would have been someone else.
This is true.
It looks the server chose there wouldn't have been any race for that workunit too.
(I'm actually very serious about this -- everyone complains about "getting the task too late". That's nonsense. The timing is up to you, not the server. When you see something like this, all that is happening is you're being given an opportunity to be the double checker -- something that would not have happened had the server sent out the second task at any other instant in time. You never had an opportunity to find that prime because your computer didn't request work at the correct instant.)
No complain, just discuss.
Why does the server not send the second task within 1 minute?
For example we completed 166.8K PPSE tasks in the last 24 hours.
Suppose that 166.8K are first+second tasks only and 83.4K are first tasks.
The server generates almost 1 first task per second.
Why does it not wait for sending all the second tasks before generating new first ones? Or is there a threshold for first/second ratio?
|
