[RC5] When RC5 will be finished...
stevenj at acay.com.au
Thu Nov 1 21:45:35 EST 2001
I can see your point with too many projects at once, but the problem is if a new project isn't planned soon we could end up wi5th nothing to do! I admit that probabilistically ( I'm not sure if that's a word, but you know what I mean), RC5 will take until Q3 2002 to finish, but it might just finish
tomorrow. Gee for all I know it might have finished whilst I have been typing this.
If that happens, we will be stuck with nothing but OGR, a project that I for one aren't that keen on. I have problems with its do everything twice philosophy. Yes, I conceed the arguement that it is good science, but it's not good fun. The stats mean that it's hard to keep a handle on how it's going. The
variable size of the nodes, combined with the large time taken to complete a single stats unit leave me feeling that I'm getting nowhere.
I know these are all pretty things, but like everyone else I'm just doing this cause it's cool, (well geek cool), and if it's not fun, I'll find something else for my cycles.
This came out like a whinge. I didn't mean it to be. I too would like to see RC5 completed, and something to come from OGR, but I also think that we should at least be thinking about what's next.
Mike Henry wrote:
> On Wed, 31 Oct 2001 14:25:59 -0500, you wrote:
> >Hi all,
> >When we will finish rc5(we might finish it really soon!), most of us will switch to OGR. I don't know the % of cpu time spent on rc5 vs ogr but request for ogr blocks will at least double(and so on with the speed rate).
> >What I wan't to know is if distributed.net will be able to handle so many request, and also if they are ready to switch to ogr 26 if we complete ogr-25, wich I don't think is the case since they are having trouble to deal with stats with ogr-25. I'm not looking forward to see what it's gonna be with ogr-26!
> >You comments ??
> OGR-24 is still not complete. There is trouble with stats for OGR-24 too
> - but it's been going on so long most people have forgotten about it!
> (it needs a massive amount of work, and a redesign of the database just
> to collate all the results and see if we have finished or not, I
> Personally I think we should halt work on OGR-25 until OGR-24 is
> complete, but that's just my opinion.
> However, to have *3* unfinished projects on the go (OGR-24, OGR-25 and
> OGR-26) after we've finished RC5 would be silly. Let's start 20 projects
> all at the same time and never finish any of them! :-)
> I don't think we should even discuss which projects to do next yet. And
> when we've decided which project is next in line for our CPU time, I
> think we should finish all three current projects completely before
> starting the new one.
> Please - can we just finish OGR-24 and OGR-25 first? Then why not pump
> all resources into RC5 and blitz it?
> I'd prefer that we do not start another OGR project until and unless we
> can sort out the stats problems - working out a way of keeping the %age
> progress on the website up to date, working out a way of validating
> stubs so that we don't need two passes, etc.
> To unsubscribe, send 'unsubscribe rc5' to majordomo at lists.distributed.net
> rc5-digest subscribers replace rc5 with rc5-digest
To unsubscribe, send 'unsubscribe rc5' to majordomo at lists.distributed.net
rc5-digest subscribers replace rc5 with rc5-digest
More information about the rc5