> Another example is Paul Joblings choice for 5 primes for AP
> >I elected to use numbers of the form 2477#*(2^32+k.2^16)+1,
> with k varying. Using
> >his form ensured that each of the numbers generated had 1077 decimal
> digits, and
> >the probability that each was prime was around 1 in 177 (as
> opposed to 1 in 1238 for a random odd 1077 digit number):
> But I wonder, What if Paul had sieved these number for example up to 10,000.
> are these numbers still (after sieving) more likely to be primes than
> randomly selected integers that are factored up to 10,000.
You are quite right, you can either get rid of factors in the expression, or by
sieving. But what I was trying to do there was find an arithmetic progression, so
I really wanted each to be as likely to be prime as possible.
>[Never use the web posting form on onelist.. grrr]
Agreed - I entered a posting there and it came out 3 days later. Which would have
been alright, except that it started "ignore my last posting"....