Gosia2 brute force error estimation

[hayes]

Hi, Christopher,

A side note:

...I define the experiments such, that I throw out the angles around that point and define 1 experiment before and 2 after the maximum scattering angle, that looks fine.

I keep forgetting exactly how you are fitting. If you are doing it with OP,MINI, you need to make your own corrected yields, because Gosia doesn't sum two experiments to compare to yields from the full range of the two. This is probably clear to you.

Now, Rutherford cross-section and integrated yields look rather the same no matter which angles I define. But then, I had another surprise: The corrected yields are different! Shouldn't they be similar at least...

You should get different corrected yields whenever you change the point-scattering angle in the EXPT lines. Since

Code: Select all

-22,48,360.,19.61,3,1,0,0,360,0,1


and

Code: Select all

-22,48,360.,-19.61,3,1,0,0,360,0,1


(with the minus flag vs. without it, or equivalently a 19.6 deg projectile angle vs. a 19.6 deg target angle) in EXPT correspond to different c.o.m. scattering angles, the point cross sections will be different, but the integrated yields in OP,CORR don't use the point angle in EXPT for anything, so they will not change. Then, obviously, the ratio (correction factor) will be different. Maybe I could have said that more simply:

The OP,MINI command compares the calculated point yields to the corrected yields, so the latter have to represent point yields also (on some scale) at the scattering angles in EXPT.

There is a huge difference in the point yields between projectile and target detection, is this correct? Here are also the 2 EXPT inputs, where the input files differ:

Target detection:

Code: Select all

EXPT
3,60,140
-22,48,360.,-19.61,3,1,0,0,360,0,1
-22,48,360.,-19.33,3,1,0,0,360,1,1
-22,48,360.,-17.86,3,1,0,0,360,1,1


Projectile detection:

Code: Select all

EXPT
3,60,140
-22,48,360.,19.61,3,1,0,0,360,0,1
-22,48,360.,19.33,3,1,0,0,360,1,1
-22,48,360.,17.86,3,1,0,0,360,1,1


I am a little confused about these differences, don't know which calculations one can trust so.

This has the same explanation: They should be different in the two cases, since a target angle of 19.6 degrees has a different c.o.m. angle than a projectile angle of 19.6 deg. I know this is obvious when I say it; the obvious gets lost in the coding. This also tells you that the point cross sections can be very misleading, since they can change rapidly with the point angle.

One other side note--you use 3 magnetic substates. Since your input probably uses little memory, you can probably use the maximum of 8 without any problem. It shouldn't make a noticeable difference, but I always use the most accuracy to avoid doing other checks later.

Best,
Adam

[LPGaffney]

Hi Adam,

Hi Liam and/or Christopher,

I am just about to finally fix this in the "GUI." Maybe you said it before, but does this error happen in Gosia 1 also?
Adam

Yes, this is the behaviour I see in gosia 1.

That seems really unusual. I'm a tiny bit suspicious that something is wrong, because OP,MAP usually takes a fraction of a second for me, even with many matrix elements. Maybe your case is just different and more complicated for Gosia.

I can see that the time it takes OP,MAP to complete is in some way related to the complexity of the input. It takes me ~15 seconds for a fairly complicated definition of ~15 free and ~80 total matrix elements and 10 experiments. This becomes ~150 seconds when I increase the matrix elements to ~120.

Cheers,
Liam

[C.Bauer]

You should get different corrected yields whenever you change the point-scattering angle in the EXPT lines. Since

Code: Select all

-22,48,360.,19.61,3,1,0,0,360,0,1


and

Code: Select all

-22,48,360.,-19.61,3,1,0,0,360,0,1


(with the minus flag vs. without it, or equivalently a 19.6 deg projectile angle vs. a 19.6 deg target angle) in EXPT correspond to different c.o.m. scattering angles, the point cross sections will be different, but the integrated yields in OP,CORR don't use the point angle in EXPT for anything, so they will not change. Then, obviously, the ratio (correction factor) will be different.

I agree to all your comments, except the one above. To my understanding the minus sign shouldn't change anything. I just checked the manual:

Code: Select all

The mean scattering angle must be positive if the projectile is detected. A negative sign should
be given with the true value of the projectile scattering angle if the recoiling target nucleus has been
detected to ensure the selection of the proper kinematics.


So I always give the true projectile angle in the LAB system, no matter what I detect (w/ or w/o minus sign) shouldn't I get the same point yields?

[hayes]

You should get different corrected yields whenever you change the point-scattering angle in the EXPT lines. Since

Code: Select all

-22,48,360.,19.61,3,1,0,0,360,0,1


and

Code: Select all

-22,48,360.,-19.61,3,1,0,0,360,0,1


(with the minus flag vs. without it, or equivalently a 19.6 deg projectile angle vs. a 19.6 deg target angle) in EXPT correspond to different c.o.m. scattering angles, the point cross sections will be different, but the integrated yields in OP,CORR don't use the point angle in EXPT for anything, so they will not change. Then, obviously, the ratio (correction factor) will be different.

I agree to all your comments, except the one above. To my understanding the minus sign shouldn't change anything. I just checked the manual:

Code: Select all

The mean scattering angle must be positive if the projectile is detected. A negative sign should
be given with the true value of the projectile scattering angle if the recoiling target nucleus has been
detected to ensure the selection of the proper kinematics.


So I always give the true projectile angle in the LAB system, no matter what I detect (w/ or w/o minus sign) shouldn't I get the same point yields?

Oh! Of course! My mistake. Maybe I should look at your full input again. Can you send me the latest one? Might take a few days, but I want to see if there is a problem here. The point cross sections aren't necessarily vital in fitting, because they are used only to quickly calculate the equivalent integrated yields.

I think it could be a bug, since it could go unnoticed, because it won't affect the final calculations in the vast majority of cases. If this is a serious bug, we need to look into it quickly.

I'm suspicious because you have obviously surpassed my Gosia skills!

Edit: I just realized that changing to target detection will make some difference related to the Q-value you have chosen (or the default). If this is the case, we can see how big a difference to expect, but your rough description "huge" seems unlikely at first thought.

Adam

[hayes]

Regarding the MAP, MINI errors--

Yes, this is the behaviour I see in gosia 1.

Thanks for the help! I have to make this a high priority. (Still not sure why I rarely get this problem. It must also be related to the machine or compiler, but it's no use in wondering.)

Adam