"Discussion on initial conditions, minijet survival, quadrupole vs minijet systematics"
Lanny Ray · 2011
Thorsten: b = 0 exists in the theorist's mind, not in [reality].Tom: In 0-5% [central collisions] v 2 /ε from hydro should be quite significant [nonzero].These observations [David's measurements] contradict that expectation by a lot.Yuri: Because of fluctuations [in eccentricity]?Tom: No, because that's where you are in centrality [b].Let's forget fluctuations.0-5% corresponds to a significant impact parameter.It's like 3.5 fm.Raimond: I would not take as a fact that for 0-5% v 2 is zero.There's one estimate of v 2 which is zero.Uli: There's lots of published data that are nonzero.Raimond: That's right.And experimentally it's extremely hard to get a number there.So, maybe the most conservative thing to do would be to just draw a smooth curve from all the centralities through zero and then estimate what the number should be.I don't think any of our measurements would be able to do much better at it.It is measured with a large error bar at that centrality.That's my take on it.Uli: I would say, somebody asked how you can falsify hydrodynamics.If you find that in the 5% most-central collisions there is really zero v 2 then I think we have a problem with hydrodynamics.Because those collisions have, just by fluctuations, they don't have perfectly round [zero eccentricity] initial conditions.So, if the system thermalizes it will have an anisotropic flow developing, and we will see the [nonzero] v 2 .Any [hydro] calculation will predict this.And if you don't measure it, it's a problem.But, that's why I'm somewhat hesitant to believe that value that has been thrown out here today of around zero.Because I have seen so many other values that are already in the published literature that contradict this data.Note added (Tom): What is being contrasted is two fundamentally different measurement techniques -a) nongraphical numerical methods denoted v 2 {method} applied to azimuth correlations (published) and b) model fits to 2D angular correlations on azimuth and pseudorapidity denoted v 2 {2D} (not yet published).Most of the methods a) produce very significant non-zero v 2 for 0-5% Au-Au; method b) is consistent with zero and establishes a small upper limit which strongly contradicts methods a).It is not justified a priori to describe a) and b) in terms of a common systematic uncertainty.Derek: Why would you mess around with this 0-5% bin where you can't measure?You go just a tiny little bit away [to less-central bins] and then you have a beautiful measurement, and it's large.And it's the right size predicted by hydro.Mike: I have another way to falsify hydro.Uli predicts at LHC v 2 /ε goes up 15%.Wit Busza predicts it goes up 60%.If Uli is right hydro is not falsified.If Wit is right, I assume it's falsified.Is that true?Uli: If they confirm elliptic flow [v 2 ] that's a factor two bigger at LHC than at RHIC then [Derek: it's really tough.]Yes.Derek: There's one caveat here, which is [the possibility that] eccentricity changes with energy.Uli: But a factor of two is hard.