Re: Is short-circuit evaluation possible with parfor() ?
[email protected] Tue, 10 Feb 2026 17:13:53 +0100
| Newsgroups | gmane.comp.mathematics.pari.user |
|---|---|
| Message-ID | <[email protected]> |
On 2026-02-10 14:51, Ruud H.G. van Tol wrote:
>
> Sure, but cleaning up for 192 threads, just takes a bit more time than
> for half of that, etc.
>
but running with have the number of threads takes double the time, but I
see your point.
>
>> Proving BIG numbers prime can earn you 150,000 USD price money ;-)
>> https://www.eff.org/awards/coop
>
> Good luck!
>
I used isok3() to plot the primes (#s) of prime(n)#/prime(i)+1 with GP.
The last row #s correspond to primes of 100 digits. The 100K digit
primes I mentioned are a factor 1,000 bigger. And another factor 1,000
on those give the 150,000 USD ;-)
In case not proving prime mathematically I bought 8x AMD Instinct MI50
GPUs with a measured total peak of 50 TFLOPs FP64 (around 100€ used on
ebay). One of them can test a 100 million decimal digit Mersenne number
for being prime in less than 10 days ...
Here the GP primorial prime art:
?
pdivisors(n)={L=List();;my(p=prime(n+1)#);parfor(i=2,n+1,ispseudoprime(1+p/prime(i)),r,if(r,listput(~L,i)););L};
?
for(j=1,54,S=Set(pdivisors(j));print1(".");for(i=2,j+1,print1(if(vecsearch(S,i)!=0,"#
"," ")));print("."));
.# .
.# # .
.# # # .
. # # # .
. # # # .
. # # # .
.# # .
. # # .
.# # # .
. # # # # .
. # # # # # .
. # # .
. # .
. # # # .
. # # .
. # # # .
. # .
. .
. # .
. # .
. # # # # .
. # # # # # .
. .
. .
. # # .
. # .
. # .
. # # # .
. # # .
. # .
. # # # .
. .
. # # # .
. # .
. # .
. # #
.
. # #
.
. # # # # #
# .
. # #
.
. # # # #
.
. #
.
. # # # #
.
. #
# # # .
. # #
# .
. # # #
# .
. # #
# .
. # #
.
. # # #
# .
. #
.
.
.
.
# # # .
. # # #
.
. # #
# .
. # # # #
.
?
Regards,
Hermann.