search

jlingema / uGMTScripts

Script collection for upgraded GMT
1 stars 4 forks source link

Add output muons to testbench patterns #10

Closed jlingema closed 9 years ago

jlingema commented 9 years ago

@dinyar the output format will be: OUT N PT PHI ETA SIGN SIGNVALID QUALITY SORTRANK 0 ISO where N is the rank of the muon. For the intermediate results the same structure as for the inputs will be used.

Btw: Should SIGN+SIGNVALID be one quantity or is this OK?

dinyar commented 9 years ago

Yes, that looks good. Will ISO be represented as an int (i.e. 0-3) or as two bits (i.e. RISO AISO)?

SIGN and SIGNVALID are fine in the current way, I'm still planning on separating these later on in both the interface and internally.

On Fri Nov 07 2014 at 10:12:26 AM Joschka Lingemann < notifications@github.com> wrote:

@dinyar https://github.com/dinyar the output format will be: OUT N PT PHI ETA SIGN SIGNVALID QUALITY SORTRANK 0 ISO where N is the rank of the muon. For the intermediate results the same structure as for the inputs will be used.

Btw: Should SIGN+SIGNVALID be one quantity or is this OK?

— Reply to this email directly or view it on GitHub https://github.com/jlingema/uGMTScripts/issues/10#issuecomment-62116452.

jlingema commented 9 years ago

Along the same line of reasoning as for the sign/signvalid I would stick to one int for iso (as it is planned to be encoded in one int not two bits)?

jlingema commented 9 years ago

Testpatterns were pushed, please check and let me know if they are OK. Then I'll also push the script and close the ticket.

dinyar commented 9 years ago

Looks good. One thing though: Could you add information which sorter the intermediate came from to the intermediates? (e.g. IMDB, IMDO and IMDF)

dinyar commented 9 years ago

Just to clarify: The intermediates are not split into positive and negative at my outputs. The negative side is index 0 to 3 while the positive side is index 4 to 7.

jlingema commented 9 years ago

I don't understand fully what you mean with the last comment. Currently the order is: imd_barrel, imd_ovl_pos, imd_ovl_neg, imd_fwd_neg, imd_fwd_pos... I didn't pay much attention to this as it was only debugging output, the order can of course be changed: What would you like to see?

dinyar commented 9 years ago

Ok, the ordering is great. I wasn't sure if negative would be placed before positive.

jlingema commented 9 years ago

Well currently it's a mess as for fwd it's the opposite w.r.t. ovl ;).. Will be changed in the next version.

dinyar commented 9 years ago

Ok, awesome. Thanks!

jlingema commented 9 years ago

I guess that closes the issue. If there are any other issues, please let me know.