[RC5] OGR-28

Roman Trunov proxyma at tula.net
Thu Nov 29 04:10:25 EST 2012


Unfortunately, these estimations are not quite correct. We're at the end 
of stubspace, where stubs are very, very short. So why absolute numbers 
in stub counts are so high. As soon as we'll recycle (within a week or 
two), stub counts will return back to usual values.

We could assume that a third recycle pass will be required to verify 
unreturned and mismatched stubs. This is not an exact value, because 
we'll complete 70-72% before recycle and actual amount of job required 
could be less then 1/3, but we could use this as starting point. Early, 
we had:

1st pass - 729 days
2nd pass - 520 days (estimated)

So, very, very, very optimistic estimation for third pass, considering 
increasing computing power, could be 1 year.

As for OGR-28, it's ready to run, we could switch to it almost 
transparently. Probably no client update will be necessary. 
Even-numbered OGRs are usually shorter then odd ones, so OGR-28 is 
expected to be same or even shorter then OGR-27.


Jarod McClintock wrote:

> If we consider the project throughput of the last days, 
> http://stats.distributed.net/project/ogr_status.php?project_id=27 than 
> the project will finish in August 2013. I can't say yet what will happen 
> afterwards. So you have time until August to catch some OGR badges.
>  So...what planning is out there for the next OGR contest? I ask because 
> as of tonight, we're 5-7 months away from being done with OGR-27...

More information about the rc5 mailing list