search

hnordquist / INCC6

INCC6
4 stars 8 forks source link

Acquisition parameters using list mode #26

Open jflongo opened 8 years ago

jflongo commented 8 years ago

Issue by MartynSw Wed May 27 14:12:44 2015 Originally opened as https://github.com/hnordquist/INCC6-WithHistory/issues/88

If I use list mode acquisition with "multiplicity with fast accidentals", the gate width and predelay seems to come from the "measurement parameters", rather than those I have set up on the list mode acquisition screen.

I wonder if there should only be one set of parameters, but they can be changed in either place, so that they are always the same?

jflongo commented 8 years ago

Comment by jflongo Wed May 27 16:35:57 2015

Sounds like a bug, but ...there is a lot to discuss about this. Right now, the code uses all the parameters from the "measurement parameters", and uses the predelay and gatewidth values from the LM wizard page for the "multiplicity with fast accidentals", Similarly for "conventional". That is how it is supposed to work now It is probably buggy though (e.g. #87) The bigger issue is how do we want it to actually work?

Related, Suppose no LM analyzers are defined but the user does a rates-only analysis on an LM file or instrument. I then create a new Fast Acc analyzer based on the current "measurement parameters", using the default values for predelay, gatewidth and the long delay. So the user gets SDT and mult bins from the LM stream for her rates-only acquire after all.

jflongo commented 8 years ago

Comment by jflongo Wed May 27 16:37:32 2015

Supper time

jflongo commented 8 years ago

Comment by MartynSw Wed May 27 18:57:08 2015

I found a quote; quote2

jflongo commented 8 years ago

Comment by jflongo Wed May 27 19:31:15 2015

Ok, right. Nice. The INCC DB is designed in this spirit. There is one place for each parameter. Parameters that are related to a detector exist, as do parameters related to a unique set of multiplicity parameters ,(the SR params, and a few more for the virtual shift register) The possibility for multiple virtual shift registers exists. I have constrained it via the UI to 2. The DB permits as many as you want, defined by the predelay, gate width, long delay and FA type. Not to mention the other types of analyzers (Feynman etc), and we have not broached the topic of assigning channel subsets to virtual shift registers. Or masking off subsets of channels. There is much to do but it is not well- defined. We have a good foundation to move forward with. Good evening.

jflongo commented 8 years ago

Comment by hnordquist Mon Jun 15 16:24:29 2015

What are we doing with this?

jflongo commented 8 years ago

Comment by jflongo Wed Jun 17 18:14:03 2015

chatting about it so far

jflongo commented 8 years ago

Comment by henzlova Tue Jun 23 17:39:30 2015

I think my comment #112 is related to this. Also it is a little bit misleading if 'fast acc.' is selected in 'measurement parameter setup' window, but the analysis is still performed with 1024us gate separation. Fast accidentals would be more like 1MHz rate of gate generation. So it seems there should be two options - either 'fast acc.' is selected, in which case the default gate separation is say 1us or 'fast acc.' is NOT selected and then the default long delay of 1024 us is used. In either case, the information would be printed in the .txt/.csv file.

jflongo commented 8 years ago

Comment by hnordquist Tue Jun 23 18:23:13 2015

Joe, this could also be related to the part of your code that I commented out because all the files weren't present. I have a newer build I can have Daniela try.

On Tue, Jun 23, 2015 at 11:39 AM, henzlova notifications@github.com wrote:

I think my comment #112 https://github.com/hnordquist/INCC6/issues/112 is related to this. Also it is a little bit misleading if 'fast acc.' is selected in 'measurement parameter setup' window, but the analysis is still performed with 1024us gate separation. Fast accidentals would be more like 1MHz rate of gate generation. So it seems there should be two options - either 'fast acc.' is selected, in which case the default gate separation is say 1us or 'fast acc.' is NOT selected and then the default long delay of 1024 us is used. In either case, the information would be printed in the .txt/.csv file.

— Reply to this email directly or view it on GitHub https://github.com/hnordquist/INCC6/issues/88#issuecomment-114584047.

jflongo commented 8 years ago

Comment by jflongo Tue Jun 23 18:33:00 2015

Sure, we can make different defaults for the two options.