Hi Yuri,

On Mar 7, 2013, at 6:54 PM, Yuri Gershtein <[log in to unmask]> wrote:
> 
> ;) no harm done.

Good.

>>> Again, sorry,  I'm just trying to come up to speed with the computing model, 
>>> may be I can google those kind of questions as well…

I thought about joking that you should also try googling "how do I find dark force carriers" in response to this, but I decided I don't know you and your sense of humor well enough yet!

> 
>> 
>> I think all of your questions are good ones that we don't have answers to yet, and are the types of things we are grappling with in these discussions.
> 
> I guess I propose we figure out the boundary conditions resource-wise that we're dealing with first,
> and then figure out what kind of event content we can afford.

We've done this in various ways at various times, and it has always been my own conclusion that we will need a L3 trigger.  We've thought a little in the past about doing fast tracking with a histogram technique for this, and I think that is a viable approach that is within our means.

> The question of ROOT vs LCIO here is probably separate. ROOT is pretty good at packing the data,
> but then again I guess LCIO is as well, so size-wize this should not matter, correct?

I think LCIO is pretty well optimized, but I don't have any figures to back that up relative to ROOT.

> And if I can write a C++ program that reads in  a file and writes out a root tree, this is just my analysis job. 
> I.e. even in CMS, where all data, even RAW, are root files, I make another root tree for my own analyses purposes,
> and I actually see advantages in having LCIO microDSTs. An Ntuplizer that has TClonesArrays of
> some simple classes is pretty straightforward to write once we know the content, so I do not see how 
> this is worth an argument at this point...

Yes, I think for anyone used to dealing with much more complicated and opaque frameworks (ATLAS in my case) either LCIO or ROOT access to HPS data will be child's play.

Tim
########################################################################
Use REPLY-ALL to reply to list

To unsubscribe from the HPS-SOFTWARE list, click the following link:
https://listserv.slac.stanford.edu/cgi-bin/wa?SUBED1=HPS-SOFTWARE&A=1