RE: [certi-dev] Re: encode and decode

[Gotthard, Petr]

Other reason might be that big endian is prevailing endianness used in network 
protocols. Think of TCP/IP suite, for example.

Just my 2 cents....


Petr

-----Original Message-----
From: address@hidden [mailto:address@hidden On Behalf Of Yaomin Fu
Sent: Monday, February 22, 2010 10:01 PM
To: Eric Noulard
Cc: CERTI development discussions
Subject: [certi-dev] Re: encode and decode

As you guessed rtis is owned by JFCOM/DMSO, and may be coded by 
Lookhead martin.
I am not exactly clear about its current status, only thing I know is it is
alive, serves as a fallback rti for JSAF, and it implements SISO DLC 1.3
interface, that's the reason I used it.

In HLA specification, there is nothing says about what endian should be 
encoded.
I have found from rtis, it uses big endian. I guess the reason could be that
rtis is for RPR FOM federations that follows DIS to use big endian.

It's your choice.

Quoting Eric Noulard <address@hidden>:

> 2010/2/21 Yaomin Fu <address@hidden>:
>> I am not sure whether this is a bug, but certainly can submit it to your bug
>> repository.
>
> This is definitely a bug even from a CERTI only perspectuve
> because it does not tale into account possible heterogeneity
> between federate.
>
> Even for CERTI fed1 to CERTI fed2 the encode-->send-->receive->decode
> will fail if fed1 and fed2 do not have the same endianess ...
>
> Please file a  bug.
>
> Explain what you expect.
> You said that you expected BigEndian encoding which is off course
> an option but, would you tell me if this is a RTI-s choice or whether
> if it is coming from the HLA specs?
>
> I did search the spec without founding information about HOW to encode?
>
> By the way, is RTI-s the RTI implementation from JFCOM?
> I tried to go to
> http://dss.ll.mit.edu/dss.web/stow.html
> (found from 
> http://en.wikipedia.org/wiki/Run-Time_Infrastructure_%28simulation%29)
> but the site seems to be unreachable?
>
> What is the status of RTI-s?
>
> --
> Erk
> Membre de l'April - < promouvoir et défendre le logiciel libre > -
> http://www.april.org
>
>





-- 
CERTI-Devel mailing list
address@hidden
http://lists.nongnu.org/mailman/listinfo/certi-devel