[Halld-online] Adoption of front-end board data format

[C. Cuevas]

Hi Elliott,

Sure.  We touched on this topic at the Trigger meeting last Friday and 
should be able to iron out the details this Friday.

-Chris
~~~~~~~~~~
On 3/5/2013 3:18 PM, Elliott Wolin wrote:
> Hi Ben,
>
> The whole topic is a little beyond me, since I don't understand the 
> ramifications in all the boards.  In any case it must be discussed in 
> an Electronics meeting.
>
> Chris...can you discuss this at the next meeting and get back to me?
>
> Thanks,
>
>
>
> On 03/05/2013 02:35 PM, Benjamin Raydo wrote:
>> Hi Elliott,
>>
>> That document is in need of an amendment before I'll agree to conform 
>> to it. The lacking ability of the protocol to support a 32bit 
>> quantity for event readout is pretty ridiculous. With that said I've 
>> suggested that a "User defined" data type should be allowed to 
>> include in it the number of words to follow that should not be 
>> interpreted by the protocol, as currently defined. I understand that 
>> it is likely desired to define a protocol that doesn't depend on 
>> "user defined" values to figure out how to decode where data for a 
>> module starts and stops, but in reality you must understand the data 
>> format of the modules anyway so this shouldn't a crazy request.
>>
>> Alternatively I suppose a special firmware for Hall D where 32bit 
>> scalers are made 31bits could be done (or with no change a penalty 
>> for allowing scalers to count beyond 31bits could be in the form of 
>> event builder corruption :) ).
>>
>> Can we get that document to reflect this? The discriminator, fADC250, 
>> as well as the SSP (probably many other Hall D boards and the other 
>> boards Hall B boards I've done that probably you're not worried 
>> about) will be using 32bit scalers (unless there's a big push to cut 
>> in half the time it takes to overflow these by enforcing 31bit scalers).
>>
>> Ben
>>
>> ----- Original Message -----
>> From: "Elliott Wolin"<wolin at jlab.org>
>> To: halld-online at jlab.org, "Chris Cuevas"<cuevas at jlab.org>, "Ed 
>> Jastrzembski"<jastrzem at jlab.org>, "Hai Dong"<hdong at jlab.org>, 
>> "William Gu"<jgu at jlab.org>, "Ben Raydo"<braydo at jlab.org>, "Scott 
>> Kaneta"<skaneta at jlab.org>, "Fernando Barbosa"<barbosa at jlab.org>, 
>> "jeff wilson"<wilson at jlab.org>, "Cody Dickover"<dickover at jlab.org>
>> Cc: "nick Nganga"<nganga at jlab.org>
>> Sent: Tuesday, March 5, 2013 2:03:33 PM
>> Subject: Adoption of front-end board data format
>>
>> Hi,
>>
>> Having 1) sent out my original note (see below), 2) talked with many 
>> of the relevant parties, and 3) having received no feedback to the 
>> contrary, it is my understanding that all involved agree:
>>
>> a) To implement Dave and Ed's VME readout format in their modules.
>>
>> b) The modules will be ready by the time Hall D needs to read them 
>> out (start this summer).
>>
>> If you do NOT agree with both of these please get back to me ASAP, as 
>> we are proceeding with online and offline programming as if this is 
>> the case.
>>
>> I note that this VME format may differ from that used by some 
>> previously released modules (e.g. discriminators), but that having a 
>> uniform format across all modules is more important.
>>
>> Thanks,
>>
>>
>> On 03/01/2013 03:07 PM, Elliott Wolin wrote:
>>
>> Hi,
>>
>> Attached is Dave A's and Ed's proposal for a consistent data format 
>> for all JLab front-end modules. At the last Online meeting we were 
>> led to believe all JLab modules that get read out over the VME 
>> backplane would implement this format (FADC25, FADC125, F1TDC, TI, 
>> TS, etc). We are proceeding as if this is the case.
>>
>> Further, we expect to begin reading out detectors this summer, 
>> probably BCAL and FCAL first (check with Fernando for our crate 
>> installation schedule). We would very much like this format to be 
>> implemented by the time the crates are installed.
>>
>> FPGA programmers: please let me know ASAP if you are NOT planning to 
>> implement this format, or if you won't be able to implement it by the 
>> time your modules get installed in Hall D and we begin reading them 
>> out. We very much do not want to have to deal with multiple formats 
>> for the same module (e.g. the June 2013 format, the Mar 2015 format, 
>> whatever).
>>
>> Finally, we are developing simulation, readout and analysis code that 
>> expects this format, so we need to know ASAP if it will change.
>>
>> Thanks,
>>
>>
>>
>> _______________________________________________
>> Halld-online mailing list Halld-online at jlab.org 
>> https://mailman.jlab.org/mailman/listinfo/halld-online
>