Hello,
Every network equipment has a unique MAC address for it. Does the Wiznet NM7010A
ethernet controller also have a MAC address of its own. If no what should the
address be, as the MAC register needs to be set.
Regards,
Vineet.
---------------------------------
Yahoo! Photos
Got holiday prints? See all the ways to get quality prints in your hands ASAP.
MAC address for NM7010A
Started by ●January 12, 2006
Reply by ●January 12, 20062006-01-12
vineet jain wrote:
> Hello,
> Every network equipment has a unique MAC address for it. Does the Wiznet NM7010A ethernet controller also have a MAC address of its own. If no what should the address be, as the MAC register needs to be set.
Hi Vineet,
I don't know this ethernet controller, but if it does not have a
connection for a small EEPROM or something else, you will have to set up
an appropriated MAC address every time when you initialize your
controller. Maybe it will have a default address. But it is not
recommended to sell products with a not-unique default address. You can
order a MAC address block at http://www.ieee.org/ . A block with 16000
address costs about $500, I think.
Sten
--
/************************************************
Do you need a tiny and efficient real time
operating system (RTOS) with a preemtive
multitasking for LPC2000 or AT91SAM7?
http://nanortos.net-attack.de/
Or some open-source tools and code for LPC2000?
http://www.net-attack.de/
************************************************/
> Hello,
> Every network equipment has a unique MAC address for it. Does the Wiznet NM7010A ethernet controller also have a MAC address of its own. If no what should the address be, as the MAC register needs to be set.
Hi Vineet,
I don't know this ethernet controller, but if it does not have a
connection for a small EEPROM or something else, you will have to set up
an appropriated MAC address every time when you initialize your
controller. Maybe it will have a default address. But it is not
recommended to sell products with a not-unique default address. You can
order a MAC address block at http://www.ieee.org/ . A block with 16000
address costs about $500, I think.
Sten
--
/************************************************
Do you need a tiny and efficient real time
operating system (RTOS) with a preemtive
multitasking for LPC2000 or AT91SAM7?
http://nanortos.net-attack.de/
Or some open-source tools and code for LPC2000?
http://www.net-attack.de/
************************************************/
Reply by ●January 12, 20062006-01-12
On 12 Jan 2006 at 8:21, vineet jain wrote:
> Hello,
> Every network equipment has a unique MAC address for it. Does the
> Wiznet NM7010A ethernet controller also have a MAC address of its
> own. If no what should the address be, as the MAC register needs to
> be set.
>
The easiest and cheapest way to get a MAC address which wont cause problems on
a public network, is to take the MAC address of an old broken network card. The MAC
address is normally printed on a label on the card somewhere.
Regards
Anton Erasmus--
A J Erasmus
> Hello,
> Every network equipment has a unique MAC address for it. Does the
> Wiznet NM7010A ethernet controller also have a MAC address of its
> own. If no what should the address be, as the MAC register needs to
> be set.
>
The easiest and cheapest way to get a MAC address which wont cause problems on
a public network, is to take the MAC address of an old broken network card. The MAC
address is normally printed on a label on the card somewhere.
Regards
Anton Erasmus--
A J Erasmus
Reply by ●January 12, 20062006-01-12
Thanks, that'll help.
Cheers,
Vineet.
Anton Erasmus <antone@anto...> wrote: On 12 Jan 2006 at 8:21, vineet jain wrote:
> Hello,
> Every network equipment has a unique MAC address for it. Does the
> Wiznet NM7010A ethernet controller also have a MAC address of its
> own. If no what should the address be, as the MAC register needs to
> be set.
>
The easiest and cheapest way to get a MAC address which wont cause problems on
a public network, is to take the MAC address of an old broken network card. The MAC
address is normally printed on a label on the card somewhere.
Regards
Anton Erasmus--
A J Erasmus
SPONSORED LINKS
Microprocessor Microcontrollers Pic microcontrollers 8051 microprocessor
---------------------------------
YAHOO! GROUPS LINKS ---------------------------------
---------------------------------
Yahoo! Photos
Ring in the New Year with Photo Calendars. Add photos, events, holidays, whatever.
Cheers,
Vineet.
Anton Erasmus <antone@anto...> wrote: On 12 Jan 2006 at 8:21, vineet jain wrote:
> Hello,
> Every network equipment has a unique MAC address for it. Does the
> Wiznet NM7010A ethernet controller also have a MAC address of its
> own. If no what should the address be, as the MAC register needs to
> be set.
>
The easiest and cheapest way to get a MAC address which wont cause problems on
a public network, is to take the MAC address of an old broken network card. The MAC
address is normally printed on a label on the card somewhere.
Regards
Anton Erasmus--
A J Erasmus
SPONSORED LINKS
Microprocessor Microcontrollers Pic microcontrollers 8051 microprocessor
---------------------------------
YAHOO! GROUPS LINKS ---------------------------------
---------------------------------
Yahoo! Photos
Ring in the New Year with Photo Calendars. Add photos, events, holidays, whatever.
Reply by ●January 13, 20062006-01-13
--- In lpc2000@lpc2..., "FreeRTOS Info" <nospam@F...>
wrote:
>
> > Hello,
> > Every network equipment has a unique MAC address for it. Does the
> Wiznet NM7010A ethernet controller also have a >MAC address of its
own. If
> no what should the address be, as the MAC register needs to be set.
>
> If you wish your device to be on a public network or sold
commercially then
> you have to buy a MAC address separately for each device and program
it into
> the NM7010A registers. If you have complete control over the network
> accessibility and the devices connected to the network you can use
any MAC
> address within spec, provided each is unique within the network.
Actually MAC addresses aren't purchased individually, IEEE will assign
you an OUI (the first 3 octets of the MAC address) for a fee that's
around $1250 the last time I checked. Then you're responsible for
assigning MAC addresses within your OUI at your own discretion. This
gives you around 16.7 million MAC addresses. Sadly, IEEE's opinion is
that MAC addresses are not transferrable. One company that purchased
an OUI a while back attempted to sell pieces of it to anyone who
couldn't justify buying a whole OUI block, but IEEE promptly objected.
Although choosing a random MAC address is generally considered bad
form for equipment you've going to sell (and it's probably worth
noting that this isn't a rare occurence, many companies with
low-volume ethernet products actually do this), the odds of ever
having a MAC address collision are exceedingly low.
There are 140737488355327 valid MAC addresses (noting that
00:00:00:00:00:00 is not valid, and the broadcast/group bit must be
clear). Divide by the number of other ethernet hosts you expect to be
on the network. Let's say 1000 hosts, just for kicks. You have a 1
in 140737488355 chance of having a MAC address collision if you choose
a MAC address at random. On the otherhand, the odds of winning the
Powerball jackpot (a nationwide lottery in the US, for non-US people
on the list not familiar with Powerball) is 1 in 146107962. You are
963x more likely to become a multi-millionaire in your lifetime, than
to experience a MAC address collision in your lifetime on a LAN with
1000 ethernet hosts. Or, you'll have hit the smaller $200K Powerball
prize an average of 39493 separate times by the time you've
experienced a MAC address collision.
As i recall, there's a range of MAC addresses set aside for testing
and prototyping, that you can use for internal use however you want.
I can't remember at the moment what OUI's they are.
-Ryan
>
> > Hello,
> > Every network equipment has a unique MAC address for it. Does the
> Wiznet NM7010A ethernet controller also have a >MAC address of its
own. If
> no what should the address be, as the MAC register needs to be set.
>
> If you wish your device to be on a public network or sold
commercially then
> you have to buy a MAC address separately for each device and program
it into
> the NM7010A registers. If you have complete control over the network
> accessibility and the devices connected to the network you can use
any MAC
> address within spec, provided each is unique within the network.
Actually MAC addresses aren't purchased individually, IEEE will assign
you an OUI (the first 3 octets of the MAC address) for a fee that's
around $1250 the last time I checked. Then you're responsible for
assigning MAC addresses within your OUI at your own discretion. This
gives you around 16.7 million MAC addresses. Sadly, IEEE's opinion is
that MAC addresses are not transferrable. One company that purchased
an OUI a while back attempted to sell pieces of it to anyone who
couldn't justify buying a whole OUI block, but IEEE promptly objected.
Although choosing a random MAC address is generally considered bad
form for equipment you've going to sell (and it's probably worth
noting that this isn't a rare occurence, many companies with
low-volume ethernet products actually do this), the odds of ever
having a MAC address collision are exceedingly low.
There are 140737488355327 valid MAC addresses (noting that
00:00:00:00:00:00 is not valid, and the broadcast/group bit must be
clear). Divide by the number of other ethernet hosts you expect to be
on the network. Let's say 1000 hosts, just for kicks. You have a 1
in 140737488355 chance of having a MAC address collision if you choose
a MAC address at random. On the otherhand, the odds of winning the
Powerball jackpot (a nationwide lottery in the US, for non-US people
on the list not familiar with Powerball) is 1 in 146107962. You are
963x more likely to become a multi-millionaire in your lifetime, than
to experience a MAC address collision in your lifetime on a LAN with
1000 ethernet hosts. Or, you'll have hit the smaller $200K Powerball
prize an average of 39493 separate times by the time you've
experienced a MAC address collision.
As i recall, there's a range of MAC addresses set aside for testing
and prototyping, that you can use for internal use however you want.
I can't remember at the moment what OUI's they are.
-Ryan
Reply by ●January 14, 20062006-01-14
Thanks for the input.
Actually there are 2^48 MAC addresses existent in the world.The 1st three octets are assigned by the IEEE. Look up the link for details:
http://anonsvn.ethereal.com/ethereal/trunk/manuf
After which i feel, the network product the companies sell can configure the last 5 octets. If the production doesn't suffice, they need to buy more first three octets.
Anyways, the best way to set the address, an appropriate one, would be to take one from an already bought network card.
Cheers,
Vineet. Ryan Niemi <ryan@ryan...> wrote: --- In lpc2000@lpc2..., "FreeRTOS Info" <nospam@F...> wrote:
>
> > Hello,
> > Every network equipment has a unique MAC address for it. Does the
> Wiznet NM7010A ethernet controller also have a >MAC address of its
own. If
> no what should the address be, as the MAC register needs to be set.
>
> If you wish your device to be on a public network or sold
commercially then
> you have to buy a MAC address separately for each device and program
it into
> the NM7010A registers. If you have complete control over the network
> accessibility and the devices connected to the network you can use
any MAC
> address within spec, provided each is unique within the network.
Actually MAC addresses aren't purchased individually, IEEE will assign
you an OUI (the first 3 octets of the MAC address) for a fee that's
around $1250 the last time I checked. Then you're responsible for
assigning MAC addresses within your OUI at your own discretion. This
gives you around 16.7 million MAC addresses. Sadly, IEEE's opinion is
that MAC addresses are not transferrable. One company that purchased
an OUI a while back attempted to sell pieces of it to anyone who
couldn't justify buying a whole OUI block, but IEEE promptly objected.
Although choosing a random MAC address is generally considered bad
form for equipment you've going to sell (and it's probably worth
noting that this isn't a rare occurence, many companies with
low-volume ethernet products actually do this), the odds of ever
having a MAC address collision are exceedingly low.
There are 140737488355327 valid MAC addresses (noting that
00:00:00:00:00:00 is not valid, and the broadcast/group bit must be
clear). Divide by the number of other ethernet hosts you expect to be
on the network. Let's say 1000 hosts, just for kicks. You have a 1
in 140737488355 chance of having a MAC address collision if you choose
a MAC address at random. On the otherhand, the odds of winning the
Powerball jackpot (a nationwide lottery in the US, for non-US people
on the list not familiar with Powerball) is 1 in 146107962. You are
963x more likely to become a multi-millionaire in your lifetime, than
to experience a MAC address collision in your lifetime on a LAN with
1000 ethernet hosts. Or, you'll have hit the smaller $200K Powerball
prize an average of 39493 separate times by the time you've
experienced a MAC address collision.
As i recall, there's a range of MAC addresses set aside for testing
and prototyping, that you can use for internal use however you want.
I can't remember at the moment what OUI's they are.
-Ryan
SPONSORED LINKS
Microprocessor Microcontrollers Pic microcontrollers 8051 microprocessor
---------------------------------
YAHOO! GROUPS LINKS --------------------------------- __________________________________________________
Actually there are 2^48 MAC addresses existent in the world.The 1st three octets are assigned by the IEEE. Look up the link for details:
http://anonsvn.ethereal.com/ethereal/trunk/manuf
After which i feel, the network product the companies sell can configure the last 5 octets. If the production doesn't suffice, they need to buy more first three octets.
Anyways, the best way to set the address, an appropriate one, would be to take one from an already bought network card.
Cheers,
Vineet. Ryan Niemi <ryan@ryan...> wrote: --- In lpc2000@lpc2..., "FreeRTOS Info" <nospam@F...> wrote:
>
> > Hello,
> > Every network equipment has a unique MAC address for it. Does the
> Wiznet NM7010A ethernet controller also have a >MAC address of its
own. If
> no what should the address be, as the MAC register needs to be set.
>
> If you wish your device to be on a public network or sold
commercially then
> you have to buy a MAC address separately for each device and program
it into
> the NM7010A registers. If you have complete control over the network
> accessibility and the devices connected to the network you can use
any MAC
> address within spec, provided each is unique within the network.
Actually MAC addresses aren't purchased individually, IEEE will assign
you an OUI (the first 3 octets of the MAC address) for a fee that's
around $1250 the last time I checked. Then you're responsible for
assigning MAC addresses within your OUI at your own discretion. This
gives you around 16.7 million MAC addresses. Sadly, IEEE's opinion is
that MAC addresses are not transferrable. One company that purchased
an OUI a while back attempted to sell pieces of it to anyone who
couldn't justify buying a whole OUI block, but IEEE promptly objected.
Although choosing a random MAC address is generally considered bad
form for equipment you've going to sell (and it's probably worth
noting that this isn't a rare occurence, many companies with
low-volume ethernet products actually do this), the odds of ever
having a MAC address collision are exceedingly low.
There are 140737488355327 valid MAC addresses (noting that
00:00:00:00:00:00 is not valid, and the broadcast/group bit must be
clear). Divide by the number of other ethernet hosts you expect to be
on the network. Let's say 1000 hosts, just for kicks. You have a 1
in 140737488355 chance of having a MAC address collision if you choose
a MAC address at random. On the otherhand, the odds of winning the
Powerball jackpot (a nationwide lottery in the US, for non-US people
on the list not familiar with Powerball) is 1 in 146107962. You are
963x more likely to become a multi-millionaire in your lifetime, than
to experience a MAC address collision in your lifetime on a LAN with
1000 ethernet hosts. Or, you'll have hit the smaller $200K Powerball
prize an average of 39493 separate times by the time you've
experienced a MAC address collision.
As i recall, there's a range of MAC addresses set aside for testing
and prototyping, that you can use for internal use however you want.
I can't remember at the moment what OUI's they are.
-Ryan
SPONSORED LINKS
Microprocessor Microcontrollers Pic microcontrollers 8051 microprocessor
---------------------------------
YAHOO! GROUPS LINKS --------------------------------- __________________________________________________
Reply by ●January 14, 20062006-01-14
---- Original Message ----
From: "Ryan Niemi" <ryan@ryan...>
To: <lpc2000@lpc2...>
Sent: Saturday, January 14, 2006 12:38 AM
Subject: [lpc2000] Re: MAC address for NM7010A
> Actually MAC addresses aren't purchased individually, IEEE will assign
> you an OUI (the first 3 octets of the MAC address) for a fee that's
> around $1250 the last time I checked. Then you're responsible for
> assigning MAC addresses within your OUI at your own discretion. This
> gives you around 16.7 million MAC addresses. Sadly, IEEE's opinion is
> that MAC addresses are not transferrable. One company that purchased
> an OUI a while back attempted to sell pieces of it to anyone who
> couldn't justify buying a whole OUI block, but IEEE promptly objected.
You can also buy a block of 4096 addresses, called an IAB, for USD 550.
http://standards.ieee.org/regauth/index.html
Karl Olsen
From: "Ryan Niemi" <ryan@ryan...>
To: <lpc2000@lpc2...>
Sent: Saturday, January 14, 2006 12:38 AM
Subject: [lpc2000] Re: MAC address for NM7010A
> Actually MAC addresses aren't purchased individually, IEEE will assign
> you an OUI (the first 3 octets of the MAC address) for a fee that's
> around $1250 the last time I checked. Then you're responsible for
> assigning MAC addresses within your OUI at your own discretion. This
> gives you around 16.7 million MAC addresses. Sadly, IEEE's opinion is
> that MAC addresses are not transferrable. One company that purchased
> an OUI a while back attempted to sell pieces of it to anyone who
> couldn't justify buying a whole OUI block, but IEEE promptly objected.
You can also buy a block of 4096 addresses, called an IAB, for USD 550.
http://standards.ieee.org/regauth/index.html
Karl Olsen
Reply by ●January 14, 20062006-01-14
--- In lpc2000@lpc2..., vineet jain <vineetrvce@y...> wrote:
>
> Thanks for the input.
>
> Actually there are 2^48 MAC addresses existent in the world.
> The 1st three octets are assigned by the IEEE. Look up the
> link for details:
No, the LSB of the first octet must be clear on a valid MAC address.
This is the broadcast / group (multicast) bit. Thus, there are 2^47 -
1 *valid* MAC addresses (since 00:00:00:00:00:00 is not valid), which
yields 140737488355327 valid MAC addresses.
> http://anonsvn.ethereal.com/ethereal/trunk/manuf
>
> After which i feel, the network product the companies sell
> can configure the last 5 octets. If the production doesn't
> suffice, they need to buy more first three octets.
A MAC address contains 6 octets, not 8 as you suggest. The first
three are assigned by IEEE to a specific manufacturer, and is known as
the OUI. Look at the link you provided, which is Ethereal's
hopelessly outdated copy of the IEEE OUI list, and lists the OUI's
(first 3 octets) of the manufacturers. The manufacturer may configure
the last 3 octets, not 5. If there were 5 configurable octets, there
could only be 256 ethernet manufacturers in the world. Rather, there
are 3 octets for the OUI, and 3 octets manufacturer-assigned.
Here's the MAC address breakdown:
yy:yy:yy:zz:zz:zz
yy:yy:yy = IEEE-assigned OUI
zz:zz:zz = Manufacturer-assigned
Additionally, the first octet of the OUI may not have the LSB set.
There are 2^23 possible OUI's, and 2^24 possible manufacturer-assigned
addresses within an OUI.
-Ryan
>
> Thanks for the input.
>
> Actually there are 2^48 MAC addresses existent in the world.
> The 1st three octets are assigned by the IEEE. Look up the
> link for details:
No, the LSB of the first octet must be clear on a valid MAC address.
This is the broadcast / group (multicast) bit. Thus, there are 2^47 -
1 *valid* MAC addresses (since 00:00:00:00:00:00 is not valid), which
yields 140737488355327 valid MAC addresses.
> http://anonsvn.ethereal.com/ethereal/trunk/manuf
>
> After which i feel, the network product the companies sell
> can configure the last 5 octets. If the production doesn't
> suffice, they need to buy more first three octets.
A MAC address contains 6 octets, not 8 as you suggest. The first
three are assigned by IEEE to a specific manufacturer, and is known as
the OUI. Look at the link you provided, which is Ethereal's
hopelessly outdated copy of the IEEE OUI list, and lists the OUI's
(first 3 octets) of the manufacturers. The manufacturer may configure
the last 3 octets, not 5. If there were 5 configurable octets, there
could only be 256 ethernet manufacturers in the world. Rather, there
are 3 octets for the OUI, and 3 octets manufacturer-assigned.
Here's the MAC address breakdown:
yy:yy:yy:zz:zz:zz
yy:yy:yy = IEEE-assigned OUI
zz:zz:zz = Manufacturer-assigned
Additionally, the first octet of the OUI may not have the LSB set.
There are 2^23 possible OUI's, and 2^24 possible manufacturer-assigned
addresses within an OUI.
-Ryan
Reply by ●January 14, 20062006-01-14
One thing I have done in the past, when I needed just a few MAC
addresses, was to go to a computer disposal or recycling store and buy
a box of old (generally ISA bus) Ethernet controllers. Cost about $1
each. Every one has a serial eeprom containing a MAC address.
Sometimes it's simply written on the board, other times you need to
power the board up, and at worst you need to desolder the eeprom chip
and read it in a programmer. Then destroy the boards, so the addresses
you got will be guaranteed to be unique. A cheap solution for a few
addresses.
Of course, if you need lots of addresses then the IEEE pricing won't
worry you...
Adrian.
--- In lpc2000@lpc2..., "Ryan Niemi" <ryan@f...> wrote:
> Actually MAC addresses aren't purchased individually, IEEE will assign
> you an OUI (the first 3 octets of the MAC address) for a fee that's
> around $1250 the last time I checked. Then you're responsible for
> assigning MAC addresses within your OUI at your own discretion. This
> gives you around 16.7 million MAC addresses.
addresses, was to go to a computer disposal or recycling store and buy
a box of old (generally ISA bus) Ethernet controllers. Cost about $1
each. Every one has a serial eeprom containing a MAC address.
Sometimes it's simply written on the board, other times you need to
power the board up, and at worst you need to desolder the eeprom chip
and read it in a programmer. Then destroy the boards, so the addresses
you got will be guaranteed to be unique. A cheap solution for a few
addresses.
Of course, if you need lots of addresses then the IEEE pricing won't
worry you...
Adrian.
--- In lpc2000@lpc2..., "Ryan Niemi" <ryan@f...> wrote:
> Actually MAC addresses aren't purchased individually, IEEE will assign
> you an OUI (the first 3 octets of the MAC address) for a fee that's
> around $1250 the last time I checked. Then you're responsible for
> assigning MAC addresses within your OUI at your own discretion. This
> gives you around 16.7 million MAC addresses.
Reply by ●January 14, 20062006-01-14
--- In lpc2000@lpc2..., "Karl Olsen" <kro@p...> wrote:
> You can also buy a block of 4096 addresses, called an IAB, for USD 550.
> http://standards.ieee.org/regauth/index.html
>
> Karl Olsen
Hi Karl,
Ah, so it seems! This appears to be a new IEEE offering since the
last time I dealt with OUI registration. I see IEEE also jacked the
cost of an OUI up from $1250 to $1650 in the meantime. I remember
back when it was still just $1000. Actually, it wasn't so long ago
either, just a few years..!
-Ryan
> You can also buy a block of 4096 addresses, called an IAB, for USD 550.
> http://standards.ieee.org/regauth/index.html
>
> Karl Olsen
Hi Karl,
Ah, so it seems! This appears to be a new IEEE offering since the
last time I dealt with OUI registration. I see IEEE also jacked the
cost of an OUI up from $1250 to $1650 in the meantime. I remember
back when it was still just $1000. Actually, it wasn't so long ago
either, just a few years..!
-Ryan
