The dangers of working on a live system. I'm trying to wedge in a new feature -- please bear with me.

[QUOTE=Prime95;506675]The dangers of working on a live system. I'm trying to wedge in a new feature -- please bear with me.[/QUOTE]

When unreserving an exponent the server will not give you the same exponent right back. OK, not a new feature more of a bug fix.

Hm, might this be related to something I just noticed when submitting the results for [URL="https://www.mersenne.org/report_exponent/?exp_lo=96312169&exp_hi=&full=1&ecmhist=1"]M96312169[/URL]?
I don't know what exactly happened there as I can't see the assignment history.
For me this looks like the exponent has been reassigned to me... or I messed up the worktodo file.

Crazy amount of GHz-Days???
 

Looks like someone is submitting a crazy amount of GHz-Days of work on the website using trial factoring. and the GHz-Days number on the server for today is just going crazy.

[QUOTE=xx005fs;506935]Looks like someone is submitting a crazy amount of GHz-Days of work on the website using trial factoring. and the GHz-Days number on the server for today is just going crazy.[/QUOTE]

Hmmm...one of the fastest GPUs at 4,000 GhzDays/Day would take about 1800 years to complete that much factoring on that exponent; 900 years on the last bit level alone. And since as far as I know TF can't be parallelized even 100 such GPUs can't save time.

Unless there was a Delorean involved.

[QUOTE=xx005fs;506935]Looks like someone is submitting a crazy amount of GHz-Days of work on the website using trial factoring. and the GHz-Days number on the server for today is just going crazy.[/QUOTE]

All of the Anonymous results from 2^72 onwards should be deleted.

[QUOTE=petrw1;506937]And since as far as I know TF can't be parallelized even 100 such GPUs can't save time.[/QUOTE]Everyone take one TF class, or a small portion of the search space. Should be no problem to parallelise. I'm not saying that happened here though.

Could be some new QC that we are unaware of yet. :surprised:

The GHz-Days values are all fixed but the PrimeNet Graph is still not fixed

I didn't even know [url=https://www.mersenne.org/primenet/graphs.php]those graphs[/url] existed. :redface:

[QUOTE=James Heinrich;506948]I didn't even know [url=https://www.mersenne.org/primenet/graphs.php]those graphs[/url] existed. :redface:[/QUOTE]

Surprise! I added those at some point when I was having way too much fun with the Google graph API.

The data comes from a table of info generated daily and I'll have to manually re-run it for that day to clear out the bogus stuff.

[QUOTE=Madpoo;506950]Surprise! I added those at some point when I was having way too much fun with the Google graph API.

The data comes from a table of info generated daily and I'll have to manually re-run it for that day to clear out the bogus stuff.[/QUOTE]
All but one graph cover 180 days. What's the reason for that one covering only 30 days?