Printable View
may i am too stupid to understand the words of big-endian and little-
endian?
my understand is that big-endian means "int" data type in memory
layout from high bits to low bits ,and little-endian means "int" data
type in memory layout from low bits to high bits.
so big-endian should be named big-start ,little-endian should be name
little-start.
if so ,it is easy to understand thes terminologys literally.
any one agree me ?
[email]zhangyefei.yefei@gmail.com[/email] schrieb:
[color=blue]
> any one agree me ?[/color]
I disagree.
Endianness is about the *byte* order, not the *bit* order. The bit order
is usually not determinable by conventional means (i.e. without a logic
analyzer).
See [url]http://en.wikipedia.org/wiki/Endianness[/url]
Regards,
Johannes
On Sep 11, 3:52*am, Johannes Bauer <dfnsonfsdu...@gmx.de> wrote:[color=blue]
> zhangyefei.ye...@gmail.com schrieb:
>[color=green]
> > any one agree me ?[/color]
>
> I disagree.
>
> Endianness is about the *byte* order, not the *bit* order. The bit order
> is usually not determinable by conventional means (i.e. without a logic
> analyzer).
>
> Seehttp://en.wikipedia.org/wiki/Endianness[/color]
Actually, it refers to both bits and bytes.
Some L1 schemes send the LSbit first, others send the LSbit last, in
every byte transferred. For instance, the L1 of FDDI and ATM send the
LSbit last, MSbit first, byte by byte. But Ethernets send LSbit first.
That is also referred to as endianness.
At the byte level, RFC 791 Appendix B says that IP is big endian.
And although this always causes great controversy, the headers of
Ethernet, FDDI, and ATM frames are pretty clearly meant to be big
endian byte order also, if you look at any field of the header where
byte order matters. Such as, the "length" field, for example.
Bert
On Sep 11, 1:02*am, "zhangyefei.ye...@gmail.com"
<zhangyefei.ye...@gmail.com> wrote:[color=blue]
> may i am *too stupid *to understand the words of big-endian and little-
> endian?
>
> my understand is that big-endian means *"int" data type in memory
> layout from high bits to low bits ,and little-endian means *"int" data
> type in memory *layout from low bits to high *bits.
>
> so big-endian should be named big-start ,little-endian should be name
> little-start.
>
> if so ,it is easy to understand *thes terminologys literally.[/color]
Basically you're correct, although people often discuss endianness in
terms of units larger than bits (typically bytes), particularly in
programs (at the hardware level, for example, within the stream of
bits on an Ethernet cable, we almost always discuss it at the bit
level).
Your confusion is most likely do to the terms themselves, which
actually come from Jonathan Swift's, "Gulliver’s Travels" where there
were two rival kingdoms, one where they ate eggs starting at the
little end (aka the "little endians") and the other in which it was
correct to eat the egg from the big end. So it's really a statement
of which end you start the number (or egg) from.
On Sep 12, 8:16*am, "robertwess...@yahoo.com"
<robertwess...@yahoo.com> wrote:[color=blue]
> On Sep 11, 1:02*am, "zhangyefei.ye...@gmail.com"
>
> <zhangyefei.ye...@gmail.com> wrote:[color=green]
> > may i am *too stupid *to understand the words of big-endian and little-
> > endian?[/color]
>[color=green]
> > my understand is that big-endian means *"int" data type in memory
> > layout from high bits to low bits ,and little-endian means *"int" data
> > type in memory *layout from low bits to high *bits.[/color]
>[color=green]
> > so big-endian should be named big-start ,little-endian should be name
> > little-start.[/color]
>[color=green]
> > if so ,it is easy to understand *thes terminologys literally.[/color]
>
> Basically you're correct, although people often discuss endianness in
> terms of units larger than bits (typically bytes), particularly in
> programs (at the hardware level, for example, within the stream of
> bits on an Ethernet cable, we almost always discuss it at the bit
> level).
>
> Your confusion is most likely do to the terms themselves, which
> actually come from Jonathan Swift's, "Gulliver’s Travels" where there
> were two rival kingdoms, one where they ate eggs starting at the
> little end (aka the "little endians") and the other in which it was
> correct to eat the egg from the big end. *So it's really a statement
> of which end you start the number (or egg) from.[/color]
Oh ,i see now.
My confusion comes from my poor english ,and your saying " they ate
eggs starting at the little end " make me understand the actual
meaning of big-endian and little-endian.
Thank you.
Thanks all.
On 2008-09-11 20:16:42 -0400, "robertwessel2@yahoo.com"
<robertwessel2@yahoo.com> said:
[color=blue]
> Your confusion is most likely do to the terms themselves, which
> actually come from Jonathan Swift's, "Gulliver’s Travels" where there
> were two rival kingdoms, one where they ate eggs starting at the
> little end (aka the "little endians") and the other in which it was
> correct to eat the egg from the big end. So it's really a statement
> of which end you start the number (or egg) from.[/color]
Side note? That is a damn cool piece of trivia I did not know until
today - thanks!
/dmfh
--
_ __ _
__| |_ __ / _| |_ 01100100 01101101
/ _` | ' \| _| ' \ 01100110 01101000
\__,_|_|_|_|_| |_||_| dmfh(-2)dmfh.cx
Johannes Bauer wrote:
[color=blue]
> Endianness is about the *byte* order, not the *bit* order. The bit order
> is usually not determinable by conventional means (i.e. without a logic
> analyzer).[/color]
It can be either, or both. Sometimes bit and byte order are different.
There are few bit addressable processors now, though some have
instructions that do bitwise operations. Otherwise, bits tend
to be numbered in processor documentation, and not always
consistent with the byte order.
Also, note the byte order in VAX floating point formats.
(Fixed point is little-endian.) The byte addressable VAX
stores floating point in little endian 16 bit words, which
are in big endian order.
For ethernet, the bit order on the wire is normally not visible.
It is only the length field if IEEE 802.3 frames that byte order
is significant to ethernet. (Not counting higher layer protocols.)
Ethernet MAC addresses are bit patterns, and don't have most and
least significant bits. (Though in some cases one may consider
that they do.)
-- glen
<zhangyefei.yefei@gmail.com> wrote in message
news:07bb1588-106c-41b7-a8bc-d346278de9d9@i24g2000prf.googlegroups.com...[color=blue]
> may i am too stupid to understand the words of big-endian and little-
> endian?
>
> my understand is that big-endian means "int" data type in memory
> layout from high bits to low bits ,and little-endian means "int" data
> type in memory layout from low bits to high bits.
>
> so big-endian should be named big-start ,little-endian should be name
> little-start.
>
> if so ,it is easy to understand thes terminologys literally.
>
> any one agree me ?[/color]
Lack of agreement on endianness has hindered the industry greatly. Little
endian is the way to go and the internet should be switched over to little
endian before we turn the clocks away from daylight savings time.
Tony ;)