Newer milestone thread

[ckdo]

Originally Posted by Brian-E

The completed first-time tests below M(37156667) must include thousands of incorrect results.

Likewise, the completed double-check tests below M(24036583) will certainly include hundreds of fake results. Independent verification of PrimeNet's residues isn't something that actually happens.

[davieddy]

Originally Posted by ckdo

Likewise, the completed double-check tests below M(24036583) will certainly include hundreds of fake results. Independent verification of PrimeNet's residues isn't something that actually happens.

Don't you realize what the probability of two matching
incorrect residues is?

David

[ckdo]

Originally Posted by davieddy

Don't you realize what the probability of two matching incorrect residues is?

If you submit them both, around 100%...

[davieddy]

Originally Posted by ckdo

If you submit them both, around 100%...

Read about the "random shifting" on the Math page.

Or did you, by "you", mean davieddy, arch-rogue extraordinaire?

[davar55]

I think so. Our humor is not an impediment, is it?

[davieddy]

Originally Posted by ckdo

If you submit them both, around 100%...

I don't think my typing accuracy is up to that.

D.

[davieddy]

Originally Posted by davar55

I think so. Our humor is not an impediment, is it?

Mods:
Could this be wafted to one of the following threads:

Hmmm
Darkside of the Moon (a famous detective).
Music
Useless posts from ...... (please append davar55)

David

[davar55]

It is not a "useless" post when its purpose is to
try to restore a bit of sanity and concordance.

[davieddy]

Originally Posted by davar55

It is not a "useless" post when its purpose is to
try to restore a bit of sanity and concordance.

Can't fool me.
There ain't no sanity clause.

David

[davar55]

Originally Posted by davieddy

Can't fool me.
There ain't no sanity clause.

That was from the MARX brothers, as i recall.

[TimSorbet]

Originally Posted by ckdo

If you submit them both, around 100%...

If you only submit results without malice intent, quite low since there are safeguards to prevent having 0 or 1 runnings counted and credited as 2+. While rare, I'd still expect there are some errors in the DB due to accidents.
If you intentionally submit the same result twice (I'm sure there's a way, and I doubt it's too hard or would draw too much attention to yourself), then it depends on whether you ran it once or never: in the case that you ran it once, it depends on your machine's reliability, maybe average 3-4% chance that it's wrong. If you never ran it, (and just guessed the residue) it's around 100% chance that it's wrong. While I should hope there aren't many people intentionally submitting wrong results twice (or even once!), I would be surprised if no numbers are marked as done with DC with incorrect residues due to this.
Put those together, and no, the PrimeNet DB isn't 100% accurate, and no, independent verification of it just isn't done (it would be so much work!), but I'd say as a whole, the DB still has quite high integrity. I'd be surprised if any primes are missed by PrimeNet, though in the future a triple check, with results only allowed through trusted sources, checking large blocks of GIMPS work would probably be worthwhile.