Always learning wrote:
I always thought, mistakenly, IP6 was 6 segments, because it was IP6. IP4 had 4 segments. However IP6 is >actually IP version 6 and it has 8 segments.
I don't think I ever heard IP6, but always IPv6. Counting segments might not be as meaningful. IPv6 has twice (8) segments compared to IPv4 however each segment is 2 octets making IPv6 address space 4 times (128 bits) compared to IPv4 (32 bits).
Thanks Sheraz Sent from my Verizon Wireless BlackBerry
On Sat, 2011-02-26 at 20:58 +0000, sheraznaz@yahoo.com wrote:
IPv6 has twice (8) segments compared to IPv4 however each segment is 2 octets making IPv6 address space 4 times (128 bits) compared to IPv4 (32 bits).
Oct... means 8.
Each segment of an IP6 segment can contain 4 hexadecimal digits. Hexadecimal means 0 to F.
Are you sure 'octets' is correct?
2 hex digits is 1 octet (or byte).
On Feb 26, 2011, at 3:04 PM, Always Learning wrote:
On Sat, 2011-02-26 at 20:58 +0000, sheraznaz@yahoo.com wrote:
IPv6 has twice (8) segments compared to IPv4 however each segment is 2 octets making IPv6 address space 4 times (128 bits) compared to IPv4 (32 bits).
Oct... means 8.
Each segment of an IP6 segment can contain 4 hexadecimal digits. Hexadecimal means 0 to F.
Are you sure 'octets' is correct?
--
With best regards,
Paul. England, EU.
CentOS mailing list CentOS@centos.org http://lists.centos.org/mailman/listinfo/centos
IPv6 has twice (8) segments compared to IPv4 however each segment is 2 octets making IPv6 address space 4 times (128 bits) compared to IPv4 (32 bits).
Oct... means 8.
Each segment of an IP6 segment can contain 4 hexadecimal digits. Hexadecimal means 0 to F.
Are you sure 'octets' is correct?
4 hex digits == 16 bits. 2 octets == 2 * 8 = 16 bits.
Octets
Thanks for pointing-out my misunderstanding.
I'll remember 2 octets are really 2 characters (IBM's bytes) = 2 digits, 4 octal numbers or 4 hexadecimal numbers.
On 02/26/11 9:46 PM, Always Learning wrote:
Octets
Thanks for pointing-out my misunderstanding.
I'll remember 2 octets are really 2 characters (IBM's bytes) = 2 digits, 4 octal numbers or 4 hexadecimal numbers.
4 octal (base 8) digits only represents 12 bits. byte oriented computers really don't fit into octal at all well
On 27/02/11 06:46, Always Learning wrote:
Octets
Thanks for pointing-out my misunderstanding.
I'll remember 2 octets are really 2 characters (IBM's bytes) = 2 digits, 4 octal numbers or 4 hexadecimal numbers.
This is a confusing summary.
3 bits = 1 octal number (values 0-7) 4 bits = 1 nibble (values 0-15 or in hex 0x0-0xF) 8 bits = 2 nibbles = 1 byte or 1 octet (values 0-255 or in hex 0x00-0xFF)
Don't mix in octal numbers, as that's a completely different numeric system which is very seldom used nowadays. Octal numbers are smaller than nibbles, which is usually the smallest unit referred to in today's computers.
IPv4 uses 32 bits addresses, hence 4 bytes (4 bytes * 8 bits per byte = 32 bits). Organised into 4 "group", separated by dot. Each "group" contains 1 byte, where user interfaces uses decimal notation, with values from 0 to 255
IPv6 uses 128 bits addresses, hence 16 bytes (16 bytes * 8 bits per byte = 128 bits). Organised into 8 "groups" separated by colon. Each "group" contains of 2 bytes, where user interfaces uses hexadecimal notation, with values 0x0000 to 0xFFFF.
That's basically it.
kind regards,
David Sommerseth
--On Saturday, February 26, 2011 9:04 PM +0000 Always Learning centos@g7.u22.net wrote:
Are you sure 'octets' is correct?
https://secure.wikimedia.org/wikipedia/en/wiki/Octet_%28computing%29
Those of us who've used older mainframes (such as the PDP-10) remember "byte" being a synonym for "bit field" and a byte could be any number of bits, typically from 1 to 36 (on a 36-bit-wide machine). 7-bit and 9-bit bytes were quite common on such machines.
The PDP-11 and microcomputers used 8-bit bytes, and their popularity meant most people using computers at home or in small businesses assumed that that was the only size a byte could be.
On Sun, 2011-02-27 at 04:12 -0800, Kenneth Porter wrote:
Those of us who've used older mainframes (such as the PDP-10) remember "byte" being a synonym for "bit field" and a byte could be any number of bits, typically from 1 to 36 (on a 36-bit-wide machine). 7-bit and 9-bit bytes were quite common on such machines.
PDP being a 'main franme'? Baby mainframe perhaps when compared to Honeywell's (later Bull's) Level 66? Level 66 had 36 bit words which could be used as 6 BCD characters or 4 ASCII characters.
The PDP-11 and microcomputers used 8-bit bytes, and their popularity meant most people using computers at home or in small businesses assumed that that was the only size a byte could be.
Those *were* the days.
With best regards,
Paul. England, EU.
On 02/27/11 5:32 AM, Always Learning wrote:
On Sun, 2011-02-27 at 04:12 -0800, Kenneth Porter wrote:
Those of us who've used older mainframes (such as the PDP-10) remember "byte" being a synonym for "bit field" and a byte could be any number of bits, typically from 1 to 36 (on a 36-bit-wide machine). 7-bit and 9-bit bytes were quite common on such machines.
PDP being a 'main franme'? Baby mainframe perhaps when compared to Honeywell's (later Bull's) Level 66? Level 66 had 36 bit words which could be used as 6 BCD characters or 4 ASCII characters.
the PDP-10 was in fact considered a mainframe in the 1960s. They were more commonly called DECsystem-10, or KA10, KL10. the CPU was multiple cabinets, the KL10 supported up to 4 megawords of ram (where a word was 36 bits). They were commonly used as timesharing systems which was relatively uncommon in the late 1960s
--On Sunday, February 27, 2011 10:48 AM -0800 John R Pierce pierce@hogranch.com wrote:
the PDP-10 was in fact considered a mainframe in the 1960s. They were more commonly called DECsystem-10, or KA10, KL10. the CPU was multiple cabinets, the KL10 supported up to 4 megawords of ram (where a word was 36 bits). They were commonly used as timesharing systems which was relatively uncommon in the late 1960s
https://secure.wikimedia.org/wikipedia/en/wiki/PDP-10
I used them at MIT in the early 80's and also at Systems Concepts, which designed a clone.
On Sun, 2011-02-27 at 10:48 -0800, John R Pierce wrote:
the PDP-10 was in fact considered a mainframe in the 1960s. They were more commonly called DECsystem-10, or KA10, KL10. the CPU was multiple cabinets, the KL10 supported up to 4 megawords of ram (where a word was 36 bits). They were commonly used as timesharing systems which was relatively uncommon in the late 1960s
What type of memory did it have?
At my second computer job in 1967 on a Honeywell H-120 (a baby machine with 3 tapes which took 1 hour to do a Cobol compilation ... and then another hour for a recompile to correct the 400 errors the Punch Room had mysteriously added to 'verified' coding sheets) the memory was magnetic cores using 3 wires physically through each hollow core or ring. The memory total was, I think, octal 37777.
I can still read punch cards held upto the light to see where the holes are :-)
On Sun, 27 Feb 2011 13:32:34 +0000 Always Learning centos@g7.u22.net wrote:
On Sun, 2011-02-27 at 04:12 -0800, Kenneth Porter wrote:
Those of us who've used older mainframes (such as the PDP-10) remember "byte" being a synonym for "bit field" and a byte could be any number of bits, typically from 1 to 36 (on a 36-bit-wide machine). 7-bit and 9-bit bytes were quite common on such machines.
PDP being a 'main franme'? Baby mainframe perhaps when compared to Honeywell's (later Bull's) Level 66? Level 66 had 36 bit words which could be used as 6 BCD characters or 4 ASCII characters.
Baby? http://en.wikipedia.org/wiki/PDP-10
BR, Bob
On Sun, 2011-03-06 at 14:36 +0100, Bob Marcan wrote:
On Sun, 27 Feb 2011 13:32:34 +0000 Always Learning centos@g7.u22.net wrote:
PDP being a 'main franme'? Baby mainframe perhaps when compared to Honeywell's (later Bull's) Level 66? Level 66 had 36 bit words which could be used as 6 BCD characters or 4 ASCII characters.
I never saw any DEC installtion :-(
Working exclusively on Honeywell for over 30 years I was a bit biased.
Saw the Amstelveen (NL) computer centre of KLM. It had over 400 hard disk drives!
I also saw Honeywell upgrading a L66 machine so it would run faster. The engineer pulled-out a PCB and took it away. That 'upgrade' cost over 1 million NLG (Dutch guilders).
With best regards,
Paul. England, EU.
On 06/03/2011 13:44, Always Learning wrote:
I also saw Honeywell upgrading a L66 machine so it would run faster. The engineer pulled-out a PCB and took it away. That 'upgrade' cost over 1 million NLG (Dutch guilders).
Very annoying those big iron companies. We had two banks of ICL Eagle drives (10GB in five full height filing cabinet sized boxes). We upgraded to Albatrosses (20GB) for a mill or so (don't know the actual price). All the engineer did was swap a couple of jumpers and told us to reformat in M2FM instead of MFM. Definitely worth the money.
The other one, much later was a 3 x 1GB upgrade for a GA mini. £3k we were quoted when we could buy the drives for about £250 each. The supplier said 'fine but we're still charging £3k for the authorisation code'
Now the nearest to specialised hardware we use are Dell servers so we can't be held hostage.
On 06/03/2011 13:44, Always Learning wrote:
I also saw Honeywell upgrading a L66 machine so it would run faster. The engineer pulled-out a PCB and took it away. That 'upgrade' cost over 1 million NLG (Dutch guilders).
Very annoying those big iron companies. We had two banks of ICL Eagle drives (10GB in five full height filing cabinet sized boxes). We upgraded to Albatrosses (20GB) for a mill or so (don't know the actual price). All the engineer did was swap a couple of jumpers and told us to reformat in M2FM instead of MFM. Definitely worth the money.
In case of NC machines it was quite common that the amount of memory usable was just a configuration setting. After you paid a horrible amount of money a service engineer came, entered a special code, reconfigured the amount of memory and that was it. With a modem connection it could even be done remotely :)
Simon
-
Always Learning -
Bob Marcan -
David Sommerseth -
John R Pierce -
Kenneth Porter -
Kevin K -
Kevin Thorpe -
sheraznaz＠yahoo.com -
Simon Matter