From: Ian Bearden (bearden@nbi.dk)
Date: Fri Jun 27 2003 - 15:39:43 EDT
Here are a (very) few things I can add to the discussion. These are responses to Dana and JHs questions from earlier today. Dana asks...in a font which does almost not show up in my mail program: "The STAR PP data -is it published or should we refer to it as preliminary star data. Maybe we want to check with star." I am pretty sure that it has been published...let me check...not published yet, the reference I have is nucl-ex/0305015. This will find it way into the draft. Dana also asks: We will check more on the wiggles.The pt=4 point is in the middle panel, but off scale. We should try to include the error bar (at least!) on the plot. "Figure 2 Rcp has some low Pt wiggles in it, both eta=0 and 2. Are we sure these are okay? The middle eta=2 panel is missing the Pt=4 point so I do not understand how you get a pt=4 point in the bottom panel." The pt=4 point is in the middle panel, but off scale. We should try to include the error bar (at least!) on the plot. Dana notes that there are low pt points missing in fig. 3. I am not sure about this (so Claus will have to respond on Sunday) but I think that this is at least partially due to the fact that he has rebinned the data to be able to compare MRS and FS. Other than that, though, I think you are right, and it looks to me as if R(eta=2.2)/R(eta=0) is largest at the lowest pt(around 400MeV??) and falls to 1 at 800MeV. Is there any reason in particular to worry about this? Dana asks: "Will there be any systematic error bands on this plot? It appears that this is beyond statisctics so the remark similar or even superior seems a bit weak unless the sytematic error bands rule this out. Has anyone estimated the chisq/df this is from flat." Yes, the systematic errors will be shown. Exactly how they will look is still being discussed. I think that we do not want to 'oversell' the case of more suppression at eta=2.2. We have not estimated chisq compared to flat, but we certainly can do that. I agree with Dana's last comment that it is critical to do this measurement correctly next time (much higher statistics) and to do similar measurements at higher rapidities. I almost lost JH's mail, since my spam filter read the 'kinky feature' and thought it was about something else entirely! :-) As for these kinky features, I am not sure how significant they are ( I have only the printed version of the paper with very tiny figures!) but I think we should probably look into this. I don't see (at least not now...) how point-to-point uncertainties would not cancel in the ratio Rcp, since the acceptance is the same (and is corrected for in the same way). I think that the efficiencies are also quite similar, but that should probably be checked on Monday. As for comments 1 and 2, I think you answered your own question. Yield increases, and correction increases effect stays the same (almost). Not comforting, but it seems to be true. Also the Rcp is not effected by these problems and remains 'nice' Thanks for the comments, keep 'em coming! Cheers, Ian
This archive was generated by hypermail 2.1.5 : Fri Jun 27 2003 - 15:41:13 EDT