Edge Rewrite
// HTMLRewriter · presentation

This page was redesigned at the edge.

Cloudflare fetched the original article and streamed it through HTMLRewriter to apply an entirely new visual system without rebuilding the source page.

// request.cf · coarse context

A page that knows where it met you.

Only coarse request metadata is shown. This demo does not display or persist visitor IP addresses.

Country
US
Cloudflare location
CMH
Connection
HTTP/2
Language
Not provided

Ray ID: a24936334aa9879c

Jump to content

Talk:4B5B

Page contents not supported in other languages.
Add topic
From Wikipedia, the free encyclopedia
Latest comment: 8 years ago by Kvng in topic Line code or block code

Category

[edit]

This article should be listed in http://en.wikipedia.org/wiki/Category:Line_codes.

Done. --68.0.120.35 19:08, 7 December 2006 (UTC)Reply


5b6b code

[edit]

There's also a 5b6b code, used by some IEEE 802 class standard. A 5-bit input block makes a 6-bit , DC-balanced output block.

Output blocks are either DC-balanced (overall disparity 0), or are assigned in pairs, (overall disparity ±2). The latter codes are assigned in pairs, and the member of the pair chosen that will drive the running disparity toward 0.

Because input blocks have an odd number of bits, they always have an odd disparity. Blocks with a disparity of ±1 are output by appending one more bit, chosen to make the overall disparity 0.

Blocks with disparities ±3 and ±5 are assigned pairs of codes, as follows. Note that the pairs are bitwise complements, so the encoder only needs to store one value, and optionally invert it.

InputOutput
00000010111101000
00001100111011000
00010011011100100
00100101011010100
01000110011001100
10000011101100010
11110110110001001
11101101110010001
11011111001000110
10111110101001010
01111101101010010
11111111010000101

Why (0,3) RLL, when GCR gives (0,2) RLL?

[edit]

One thing that's puzzled me is why FDDI chose a code with up to three zeros between ones (sync edges), when codes were already known (GCR) that limited. Does it have something to do with DC bias? But the code isn't DC-balanced... 71.41.210.146 (talk) 14:01, 13 March 2009 (UTC)Reply

FDDI control codes

[edit]

FDDI uses TT as Ending Delimiter for Tokens

FDDI uses TRRR as minimum End of Frame Sequence for Data Frames

Repeaters may change an R to an S

A received Data Frame is invalid if the first two symbols of the End of Frame Sequence are not TR

FDDI-II defines other control code sequences for Cycle Control and for Tokens and Data within the Packet Channel

71.117.221.89 (talk) 20:16, 11 June 2010 (UTC)Reply


Question re the command codes

[edit]

Looking at the QQ code -- all zeroes -- therefore no transistions similarly QH and HQ -- only one transtion ?don't these "break" the idea of having two transistions Feldercarb (talk) 15:05, 11 September 2016 (UTC)Reply

Line code or block code

[edit]

I have edited to avoid calling 4b5b a block code because Block code is something with error correction. Line code doesn't quite hit me right either because MLT-3 encoding is definitely a line code and 4b5b is use in conjunction with MLT-3 in Fast Ethernet. Well, at least now we're giving both 4b5b and 8b10b the same description. Any further improvements or discussion are welcomed. ~Kvng (talk) 21:39, 11 September 2017 (UTC)Reply