Join PrimeGrid
Returning Participants
Community
Leader Boards
Results
Other
drummers-lowrise
|
Message boards :
Sophie Germain Prime Search :
Increase n to make it to T5K?
| Author |
Message |
|---|
 Bur Volunteer tester
Send message
Joined: 25 Feb 20
Posts: 515
ID: 1241833
Credit: 414,278,186
RAC: 40,609
|
|
I know the main goal of SGS is not finding record primes, but finding twins and SG primes.
Still, it would be a nice bonus if a prime found in SGS that is neither, would at least qualify for T5K. That would require approximately 100000 additional digits. Which would mean going from n=129000 to around n=470000, I think.
I have no idea how that would impact calculation times. Maybe it's worth it?
Though I just realized the sieve is probably not done in that area? | |
|
Crun-chiVolunteer tester
Send message
Joined: 25 Nov 09
Posts: 3208
ID: 50683
Credit: 135,132,479
RAC: 57,320
|
|
With current know software this kind of sieve can be made in home in two days, and you can find prime in next 30 days. Today software is fast, MT capable....
But as I know PG has no plan to stop current or make new sieve...
____________
92*10^1439761-1 NEAR-REPDIGIT PRIME :) :) :)
4 * 650^498101-1 CRUS PRIME
2022202116^131072+1 GENERALIZED FERMAT
Proud member of team Aggie The Pew. Go Aggie! | |
|
robishVolunteer moderator
Volunteer tester
Send message
Joined: 7 Jan 12
Posts: 2197
ID: 126266
Credit: 7,323,343,059
RAC: 3,133,740
|
|
I've been sieveing SG at 1M digits since lockdown started. It's at 1.7P and 67M candidates left. Its still removing a k every 2 seconds.
If there's any interest in it as a new project just let me know ;)
____________
My lucky numbers 10590941048576+1 and 224584605939537911+81292139*23#*n for n=0..26 | |
|
Michael GoetzVolunteer moderator
Project administrator
Send message
Joined: 21 Jan 10
Posts: 13956
ID: 53948
Credit: 393,166,793
RAC: 184,751
|
I know the main goal of SGS is not finding record primes, but finding twins and SG primes.
Still, it would be a nice bonus if a prime found in SGS that is neither, would at least qualify for T5K. That would require approximately 100000 additional digits. Which would mean going from n=129000 to around n=470000, I think.
I have no idea how that would impact calculation times. Maybe it's worth it?
Though I just realized the sieve is probably not done in that area?
We certainly could make SGS larger so it's once again eligible for T5K. The question is do we want to do so? The answer is no, and there's a good reason.
To explain why, let's start with a different question: "Why isn't SGS large enough to be eligible for T5K?" The answer is that it used to be, but there's now too many larger primes on the list, so SGS is no longer big enough.
Where did those other primes comes from? They came from PrimeGrid, of course. The vast majority of primes on the T5K list are found by PrimeGrid. We, ourselves, are the ones that knocked SGS off the list.
We have several projects that produce primes at a prodigious rate: SGS, PPS, GFN-15, and GFN16. All four of those used to be on the list, but only two are today. The problem is that no matter what we do, one of our subprojects will always be pushing some other subproject off the T5K list. It's foolish to try to keep everything on the list; it's not possible.
If I were to increase the size of SGS such that it was on the list again, but was smaller than PPSE, it would not take long for PPSE to once again push SGS off the list. If I increased the size of SGS so that it's bigger than PPSE, then SGS would push PPSE off the list. But PPSE grows faster, so it would eventually get back on the list, grow larger than SGS, and once again push SSGS off the list. In both scenarios, however, the increased size would mean that fewer SGS primes are being found, which in turn means it's harder to find Sophie Germain and Twin primes.
For this reason, we're going to keep running SGS at its current size until we exhaust the current sieve file. At that time we'll make a decision about what comes next.
____________
My lucky number is 75898524288+1 | |
|
|
|
For this reason, we're going to keep running SGS at its current size until we exhaust the current sieve file. At that time we'll make a decision about what comes next.
Yes, during the SGS challenge around 1% of the sieve file was exahusted. And we're currently at 55.2%.
____________
My lucky number is 6219*2^3374198+1
 | |
|
BurVolunteer tester
Send message
Joined: 25 Feb 20
Posts: 515
ID: 1241833
Credit: 414,278,186
RAC: 40,609
|
|
Danke Michael, makes sense.
So, it's either looking for a record size prime or looking for a special prime.
Some special ones are also record but conincidentally. | |
|
BurVolunteer tester
Send message
Joined: 25 Feb 20
Posts: 515
ID: 1241833
Credit: 414,278,186
RAC: 40,609
|
Which would mean going from n=129000 to around n=470000, I think.
Doesn't change anything, but still: I just saw it's actually n = 1 290 000, and that would need to increase to approximately n = 1 620 000. | |
|
|
|
Which would mean going from n=129000 to around n=470000, I think.
Doesn't change anything, but still: I just saw it's actually n = 1 290 000, and that would need to increase to approximately n = 1 620 000.
1.29M to 1.59M, which is PPSE now
But PPSE is sure to pass 1.7M in 2020.
____________
My lucky number is 6219*2^3374198+1
| |
|
BurVolunteer tester
Send message
Joined: 25 Feb 20
Posts: 515
ID: 1241833
Credit: 414,278,186
RAC: 40,609
|
1.29M to 1.59M, which is PPSE now Not to sound like a smart ass, just to make sure I understand correctly.
I thought to increase by 100 000 digits we need to increase n by ln(10)/ln(2) * 300 000 which is approximately 330 000? | |
|
|
|
1.29M to 1.59M, which is PPSE now Not to sound like a smart ass, just to make sure I understand correctly.
I thought to increase by 100 000 digits we need to increase n by ln(10)/ln(2) * 300 000 which is approximately 330 000?
You're absolutely correct, but 100000 digits is "too much", more than PPSE quite a lot.
____________
My lucky number is 6219*2^3374198+1
| |
|
|
|
|
Why not keep SGS "as is" and create a SGS-MEGA.
There's PPSE, PPS and MEGA.
There's GFN "regular" and MEGA. | |
|
JaySend message
Joined: 27 Feb 10
Posts: 132
ID: 56067
Credit: 63,631,031
RAC: 13,097
|
Why not keep SGS "as is" and create a SGS-MEGA.
If finding a prime that is listed in T5K is important, then choose one of the many subprojects that offer that. There are projects that will land you in the top 10, or even first place, if that's what you desire.
| |
|
|
|
Why not keep SGS "as is" and create a SGS-MEGA.
There's PPSE, PPS and MEGA.
There's GFN "regular" and MEGA.
SGS has a specific purpose for running low numbers. Higher numbers won't help.
____________
My lucky number is 6219*2^3374198+1
| |
|
Message boards :
Sophie Germain Prime Search :
Increase n to make it to T5K? |
