Thoughts on future NPLB searches

[nuggetprime]

But what about the smilie he posted?

[masser]

Originally Posted by nuggetprime

But what about the smilie he posted?

What about it? If I take you behind the woodshed and beat the tar out of you, but then smile about it, does my smile make you feel better?

[nuggetprime]

No,but why should he have posted this smilie?
Something unusual must have been happened.

Anyway, I think he has confused something here.

--nuggetprime

[masser]

Oh... you were referring to the smilie in ltd's original post. I thought you were referring to henryzzz's smilies.

Perhaps ltd was just surprised that for a heavy-k sequence like that he didn't find any primes in the range he completed up to n = 132 k. If he started at n = 110 k and tested up to n = 132 k, he probably did many tests without finding a prime.

[henryzz]

Originally Posted by masser

Umm... I followed the links to the original post by ltd - first of all, the quote is taken out of the context of the original thread it was in - was this a thread that was occasionally purged of older content? was it a thread where the original readers and posters knew the background info about ranges previously searched? Second, I believe ltd is a non-native english speaker, so his saying "no primes found" might easily be considered/interpreted as "no new primes found" It's a little harsh to call that post (missing one adjective that when taken out of context can look erroneous) the Most Erroneous post. Certainly there are some good candidates for that title in the Miscellaneous threads, and I would venture that there are some good candidates in the Soapbox thread as well.

because kar_bon had misunderstanded ltd there were 85 errors on his page
that was what i was pointing out
we dont want that

[kar_bon]

Originally Posted by henryzz

because kar_bon had misunderstanded ltd there were 85 errors on his page
that was what i was pointing out
we dont want that

those days i extracted all data i could find for those pages; obviously i found no other information about that k (or not yet) in the threads for RPS- and 15k-project.

if anyone can find some other hints for this k telling something different, let me know!

[gd_barnes]

Originally Posted by masser

Umm... I followed the links to the original post by ltd - first of all, the quote is taken out of the context of the original thread it was in - was this a thread that was occasionally purged of older content? was it a thread where the original readers and posters knew the background info about ranges previously searched? Second, I believe ltd is a non-native english speaker, so his saying "no primes found" might easily be considered/interpreted as "no new primes found" It's a little harsh to call that post (missing one adjective that when taken out of context can look erroneous) the Most Erroneous post. Certainly there are some good candidates for that title in the Miscellaneous threads, and I would venture that there are some good candidates in the Soapbox thread as well.

Sorry if my statement about it being the most erroneous post was taken a little harshly. I actually found the statement of "no primes found" on an extremely high-weight k funny and wasn't intending to insult anyone. I knew the 15k project didn't search entire ranges like RPS or NPLB attempt to do so the "85 errors" was meant somewhat in fun but to also point out how things can easily be misinterpreted if specifics are not given.

That reminds me of a funny story that I'll tell here that points out language barriers at times. I had a girlfriend in my mid-20's who had been a foreign-exchange student to Belgium for a year in high school. She was quite fluent in 3 other languages besides English when I dated her several years after that. She told me a story about how she was trying to tell her Belgian "parents" that she had had enough to eat for supper. She did this by attempting to directly translate "I'm full" to Dutch. I believe that was the language they spoke.

As the story goes, "I'm full" in English directly transaltes to "I'm pregnant" in Dutch. :surprised (Or at least the way she mistranslated it did.) She said it took her 15 minutes to calm her parents down and communicate what she really meant.

20 years later that's still one of my favorite miscommunication stories to tell.

(My apologies if I have the language incorrect. It's been 20+ years since I heard the story.)


Gary

[MyDogBuster]

On the other hand searching one or more fixed k for a large n range can be quite interesting too ...

I like MrOzzy's idea also. Put a k or 2 into a server and let er' rip. Fill in some of those gaps that Gary loves so much.

I can't believe Gary suggested CREATING gaps. LOL

[gd_barnes]

Originally Posted by MyDogBuster

I like MrOzzy's idea also. Put a k or 2 into a server and let er' rip. Fill in some of those gaps that Gary loves so much.

I can't believe Gary suggested CREATING gaps. LOL

He-he. Well, they would be CONTROLLED gaps unlike the hideously uncontrolled gaps that currently exist in k=1005-2000. If people searched a few k=300-400 for n=900K-1M, I'd pretty quickly come up behind them and fill in n=600K-900K and suggest that others help with it. (lol)

As for putting a k or 2 into a server and letting 'er rip, we could do what amounts to likely a double-check on k=289 up to n=1M with a file that is already sieved to P=26T. I personally have it reserved and the 4th drive had already searched it to n=680K. I figure I'll take it to n=~750K-800K after the 1st drive is over but if some people want to help take it higher, that would be great.

Obviously k=289 is a very boring effort to n=1.1M, which is around where the highest prime was found. Perhaps for something more interesting, we could take our lowest "official" k, k=301, sieve it for n=1M-2M, and then load it up in a server. It's pretty high weight so we should get at least a prime or two out of it. Another possibility is to do the same for k=289. If we double-check it to n=1M (technically likely n=1.1M), I'm confident it would not create any controversy to take it to n=2M even though it is k<300.

Although these are interesting efforts, keep in mind that we gain huge sieving efficiency by sieving 10's and 100's of k's at once, which is why we've moved up the project primes and scores list so fast at the top-5000 site. Therefore, I'd like to keep them somewhat limited to perhaps 3-4 k's at the most at any one time.


Gary

[em99010pepe]

Forget about the mini-drive using C443 server.
I'll pick up random 4-5 k's to C443 and take them up to 1M with help of the usual clients on the llrnet server. I really prefer to clean k by k all k's between 300 and 400 way up to 1M. It's a nice job to be done.

[gd_barnes]

Originally Posted by em99010pepe

Forget about the mini-drive using C443 server.
I'll pick up random 4-5 k's to C443 and take them up to 1M with help of the usual clients on the llrnet server. I really prefer to clean k by k all k's between 300 and 400 way up to 1M. It's a nice job to be done.

lol. You contradicted yourself. On one line, you say forget about the mini-drive using C443 server. On the next you say you'll pick 4-5 random k's to put in the C443 server. Wouldn't 4-5 k's be the same as a mini-drive? lol

So is what you're proposing to do is to take the 5 lowest unreserved k's (or reserved by you) and load them into the server? If so, are you talking about running them by n-range? That would be my preference instead of one k at a time.

We can be very flexible on this drive to keep it as interesting as possible for as many people as possible. Of course I would like to avoid leaving gaps on any k's for any extended period.

Probably my main preference would be that k=315 be run by a server. It is so heavy weight that it would be nice to share all of the testing on it. But you've reserved it. If you can put several cores on it so that it doesn't take an extended period to test, then that's fine.


Gary