[ibis-interconn] Re: Draft of TSIRD 2.1 for review
- From: "Angulo, John" <john_angulo@xxxxxxxxxx>
- To: "Mirmak, Michael" <michael.mirmak@xxxxxxxxx>, "IBIS-Interconnect" <ibis-interconn@xxxxxxxxxxxxx>
- Date: Wed, 14 Jul 2010 14:22:13 -0700
Michael,
My apologies: in the various hex examples I have sent so far, I have
exercised almost every available degree of freedom to provide wrong
information :). In the example text incorporated into the latest draft
of TSIRD2.1, I provided the little-endian double-precision
representation of 1.0 instead of 10.0.
Read in ascending address order, the text "40 24 00 00 00 00 00 00" is
the big-endian double-precision representation of 10.0 in hex.
So, in the lastest draft, we must replace the eight bytes of hex
following the initial zero byte with "00 00 00 00 00 00 24 40" to
provide the little-endian representation.
The remainder of the hex example in the lastest draft is indeed the
little-endian single-precision representation of the ASCII text later in
the text.
Best regards,
John
________________________________
From: ibis-interconn-bounce@xxxxxxxxxxxxx
[mailto:ibis-interconn-bounce@xxxxxxxxxxxxx] On Behalf Of Mirmak,
Michael
Sent: Wednesday, July 14, 2010 10:01 AM
To: IBIS-Interconnect
Subject: [ibis-interconn] Re: Draft of TSIRD 2.1 for review
I am a tad confused as to our hexadecimal example, as the hex
calculators on the Wiki page shown below don't quite agree.
If I want to represent "10" (decimal) in hex format, assuming 64-bit,
the tools indicate I should see this:
40 24 00 00 00 00 00 00
The 32-bit version would look like this
41 20 00 00
Is this right? The tools seem to assume big-endian (I think), so I may
be comparing apples and oranges.
Note that our example uses 10 as the frequency data, in MHz; the format
will not convert from MHz to Hz for us.
* MM
_____________________________________________
From: Mirmak, Michael
Sent: Wednesday, July 14, 2010 8:43 AM
To: 'IBIS-Interconnect'
Subject: Draft of TSIRD 2.1 for review
Enclosed please find a draft of TSIRD2.1, incorporating the minor
changes discussed at the last meeting. This assumes Walter's proposals
will be addressed through separate TSIRDs.
You can find information on the IEEE floating point standard here (at
the usual Source of All Knowledge and Wisdom):
http://en.wikipedia.org/wiki/IEEE_754-2008
<http://en.wikipedia.org/wiki/IEEE_754-2008>
<< File: tsird2.1_draft1_markup.pdf >> << File: tsird2.1_draft1.pdf >>
* MM
*
Other related posts:
