[Hadstruct] GEVP

[Joseph Karpie]

I have not studied these so closely, but Colin has data on many ops. I just blindly ran reconfit on these to get a number for Lorenzo. Kostas, if it's easy enough for you to run  with a new input, then you should take a look at the same momentum with whatever number of evecs and configs Colin used on the larger basis. The adat file is found at

/qcd/work/JLabLQCD/cegerer/nucleon_2pt/cl21_32_64_b6p3_m0p2350_m0p2050/t0_avg/momXYZ.0.0.5/EDB/cl21_32_64_b6p3_m0p2350_m0p2050.nuc_p005.n64.t0_avg.edb

which I now realize is a link which maybe is why previous backup attempt failed.... Anyway the true file is

/qcd/work/JLabLQCD/cegerer/nucleon_2pt/cl21_32_64_b6p3_m0p2350_m0p2050/phased/d001_2.00/t0_avg/momXYZ.0.0.5/EDB/cl21_32_64_b6p3_m0p2350_m0p2050.nuc_p005.n64.t0_avg.edb

Many other momenta can be found in nearby directories by changing the momXYZ.?.?.? in the path.
________________________________
From: Hadstruct <hadstruct-bounces at jlab.org> on behalf of Dutrieux, Herve via Hadstruct <hadstruct at jlab.org>
Sent: Wednesday, March 26, 2025 12:11 PM
To: Orginos, Konstantinos <knorgi at wm.edu>; Joseph Karpie via Hadstruct <hadstruct at jlab.org>
Subject: [Hadstruct] [EXTERNAL] Re: GEVP

Hmmm... so does it mean we have to turn to the 9 ops? I thought I heard you say it did not make a big difference compared to the 3 ops (for the ground state?).
Is there any chance anyone has an extraction of the Z factors and energies of this group of 4 excited states for the phased D0J0? Even if it does not come with the proper cfg by cfg statistical fluctuation, that would be an interesting quantity.

Cheers,
Hervé

________________________________
From: Hadstruct <hadstruct-bounces at jlab.org> on behalf of Orginos, Konstantinos via Hadstruct <hadstruct at jlab.org>
Sent: Wednesday, March 26, 2025 11:50:22 AM
To: Joseph Karpie via Hadstruct <hadstruct at jlab.org>
Subject: [Hadstruct] [EXTERNAL] GEVP

Hi everybody,

 I finally resolved the issue I had. My ordering of the operators caused me to compare the ground state principal correlator effective mass with that of a non-D0J0 ops. As a result the single operator was showing stronger excited state contamination. These are the plots that Herve saw sometime ago that got us excited about a possible improvement of our spectrum and matrix elements calculations at large momentum.

After fixing this problem my results are in line with Chris’ (and the rumored Robert’s results). The 3 operator GEVP gives no advantage over the single (D0J0) operator.

As I said on the call, given that the excited state nucleon spectrum has several  (N>2) nearly degenerate states close by a 3 operator basis is not expected to solve any problems.
This is the P=0,0,5 plot

Cheers,

K.

[cid:415c1b08-93ea-4ffc-93b8-1b31d8a92d76 at namprd05.prod.outlook.com]
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <https://mailman.jlab.org/pipermail/hadstruct/attachments/20250326/31bae76d/attachment-0001.html>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: Figure_1.png
Type: image/png
Size: 33786 bytes
Desc: Figure_1.png
URL: <https://mailman.jlab.org/pipermail/hadstruct/attachments/20250326/31bae76d/attachment-0001.png>