I am wondering if the older PII / PIII Thinkpads, like the 600 and T20, support hot-inserting of PCMCIA/PC-Cards (Orinoco 802.11 wifi card)? Also, does the trick where you change the cards MAC-ID in the registry and re-insert the card work with these laptops?
Thanks
On 11 Oct 2006 16:50:52 -0700, snipped-for-privacy@mailinator.com wrote in :
They do.
Probably depends on the specific card rather than the host computer.
The correct terminology is "PC Card", not "PCMCIA Card".
In general, all devices that support PC Cards support hot insertion. Also, in general (but less general), removal is ALMOST always supported, although you are supposed to use the system tray icon to "stop" the card before removing it. There are somewhat more issues with removal than with insertion, however, and more still with insertion, removal and RE-insertion. Some drivers (and the issue really is in the drivers) do have problems with removal and/or re-insertion and will lock up the system. But the PC Card specs and PC Card controllers (hardware in the laptop) are universally intended to support hot insertion and then "stopping/removing" the card. Unless the driver screws it up.
The MAC address of a network card is part of the card itself. Noth> I am wondering if the older PII / PIII Thinkpads, like the 600 and T20,
On Wed, 11 Oct 2006 20:57:20 -0400, Barry Watzman wrote in :
-1 point for technical pedantry. ;)
In theory, but not always in practice, older versions of Windoze being a major case in point. -1 point for inaccuracy.
To be equally technical, that's "warm" removal, not "hot" removal. ;)
-1 point for inaccuracy.
Which is the point -- theory is of little comfort in the real world.
Not in the case of a "soft" adapter, like most PC Cards, where much of the functionality is in the host driver. -1 point for incompleteness.
-1 point for inaccuracy. (If you're going to presume to lecture someone else, you should be very careful to get your own facts right.)
See :
SMAC is a powerful, yet an easy-to-use and intuitive Windows MAC Address Modifying Utility (MAC Address spoofing) which allows users to change MAC address for almost any Network Interface Cards (NIC) on the Windows 2000, XP, 2003, and VISTA Server systems, regardless of whether the manufacturers allow this option or not.
SMAC does not change the hardware burned-in MAC addresses. SMAC changes the "software based" MAC addresses, and the new MAC addresses you change will sustain from reboots.
The Registry hack for some other versions of Windoze is at .
Hi, I used Orinoco Gold -11A/b/g card with T21, it's hot pluggable for sure. MAC is like cast in stone. Embedded in the card(only one in the world) I don't think you can play with it. I wonder why you want to change it?
-50 points for being completely incorrect, and -50 further points plus a ruler on the hand for your annoying pseudo-paedagogical style. I can only assume you spent a lot of time around nuns.
While it is usually possible to change the MAC address temporarily or permanently by poking card registers, the MAC address is part of the card; it is generally in an external serial EEPROM and on powerup the MAC [chip] reads it out of EEPROM into on-chip registers. (How else do you think boot-from-LAN is supported?).
There are useful reasons why you might need to set a specific MAC address from the host side - for example, some embedded applications don't have the EEPROM, they use a global flash segment attached to the main microcontroller to store all system configuration data - including the MAC address. Hence all MAC chips that I've seen allow the host to write the MAC address registers directly.
You can also write the EEPROM, if attached, through the MAC chip, and thereby permanently change the MAC address of the device.
Perhaps you might want to consult the IEEE on the issue of how MAC addresses are assigned to hardware.
So he can connect to networks that are locked to specific MAC addresses, and/or so he can run software that's nodelocked to a specific MAC address.
On 12 Oct 2006 04:12:56 -0700, snipped-for-privacy@gmail.com wrote in :
+1 point for essentially confirming what I wrote.-1 point for your own pseudo-pedagogical style.
-1 point for technical accuracy.
-1 point for guessing my religion.
-1 point for courtesy.
-1 point for sense of humor.
I guess Windbag the Sailor deserves one more message...
ObWarning: I design-in this sort of hardware for a living (well, wired Ethernet more than wireless, but at the host side it makes little difference).
The physical part that implements the link layer is referred to as a MAC [chip]. Go look at some datasheets for wired Ethernet parts. Some MACs integrate the PHY and are referred to as MAC+PHY chips. Of course, you'd know this if you had ever designed in one of these parts. The division of labor is slightly different in an 802.11 implementation but the idea is the same. However one doesn't as often hear someone talk about a "wireless MAC chip" as about a [wired] "Ethernet MAC", I grant. The phrase is used, though.
Go to , Communications Network ICs, Wireless LAN ICs, WLAN NIC, IEEE 802.11a/b/g, MAC/BBP. Wow. There you see MAC (Media Access Control) chips and BBPs (Baseband Processors).
Another illustrative link:
Perhaps some elementary electronics courses might help you learn the terminology. Etiquette courses wouldn't hurt either.
Usually EEPROM, as I said. It is external to the MAC chip, generally (though not universally). Yes, it is inside the WiFi card, though this is irrelevant to the discussion. When I said external, I meant external to the MAC.
You ought to look at the datasheets, it's all in there. Generally a WiFi card contains a micro with a small bootloader only. The host uploads the main bulk of the firmware to the card when the driver is loaded, in such instances. Implementations vary; some devices have the complete radio firmware in flash and hit the ground running as soon as power is applied. But the realtime stuff is all run on the micro in the card, in any case.
You're not talking about the same thing. I'm talking about _booting_ an appliance from the LAN, i.e. a diskless workstation. Of course you'd know what I was talking about if you ever set up a network more complex than plugging in a bit of 10baseT and opening Network Neighborhood.
Irrelevant.
John Navas hath wroth:
The MAC address is stored in a protected portion of the NVRAM on the wireless card. This section also contains the boostrap loader and sometimes a TFTP server for making post production changes, especially if the card is earmarked for use in a wireless router. Some PCMCIA cards also include a JTAG port for companies that like to ship their prototypes.
I'll ignore the steps in the boot process as they don't have much to do with the original question.
When Windoze loads the wireless card driver, it reads the MAC address from the card and stores it temporarily in the registry. This value can be changed by various utilities and registry tweakers thus making MAC spoofing possible. In some drivers, it can even be changed in the Windoze device properties.
The current question revolves around whether Windoze re-reads the MAC address from the PCMCIA card when hot swapping, or if it continues to use the MAC address in the registry. From what little tinkering I've done, it seems to follow the device driver logic. If the driver is smart enough to know that a different card has been inserted, then the MAC address will be read from the new card. If it's lazy and doesn't check for a card swap, then it uses the old MAC address from the registry.
In other words, it's not the hardware (i.e. older Thinkpad) but rather the card driver. As I vaguely recall, the early Orinocco drivers were not NDIS drivers and did some rather interesting things (such as promiscuous mode support). Later Proxim drivers are NDIS 5.1 based, which eliminates these interesting things, and fixed the card swap "problem".
+1 point to me for explaining how it works.On 12 Oct 2006 09:19:13 -0700, snipped-for-privacy@gmail.com wrote in :
-1 point for relevance.
-1 point for technical accuracy.
-1 point for courtesy.
Since your response wasn't relevant to the issue at hand (much less accurate), and since you're becoming ever more rude, the "discussion" is obviously over. Have a nice day.
On Thu, 12 Oct 2006 09:22:00 -0700, Jeff Liebermann wrote in :
Ummm... Wasn't the question at hand how to do it? ;)
(...)
That's not the way I read the question. I thought he was asking if it worked on the older IBM laptops. My comments were basically that the laptop doesn't really matter. It's the driver that either supports hot swap (with or without retaining the MAC address).
On Fri, 13 Oct 2006 03:28:04 GMT, Jeff Liebermann wrote in :
Isn't that what I said way back at the beginning? ;)
No, it's perfectly valid to call it a PCMCIA card. PC card's just for the marketing droids that can't spell PCMCIA.
Yes, cards do have a default MAC address programmed into them from the factory. Unless you have a specific need to change it you're better off leaving it alone. But there are a number of perfectly legitimate situations where you'd need to change the MAC address. Most folks are fortunate to not have to change it.
This is entirely incorrect. Nearly all cards support changing their hardware MAC address. Pull up the driver's control panel and it's usually a value you can set manually. Now, whether or not this is "in windows" is merely arguing semantics.
True, some old-school cards do have a separate utility that'll let you permanently alter the on-card MAC address. But why bother since the OS lets you do it quite easily. Again, some folks might need it permanently changed, but most won't.
Err, no. MAC addresses are only local to the networks nearby. Anything upstream to the ISP is going to use TCP/IP to route the packets.
Often not, as the manufacturer will disable the ability to change the MAC address in EEPROM. There uses do be some protection circuitry that you could get around with the proper key, but I doubt that the 88¢ PCI Ethernet cards are including these extra components.
Fortunately, boot from LAN seems to becoming more rare. Even on diskless thin clients, they are now booting a reduced size OS from internal flash, rather than booting over the network. When I was designing network cards, it was always a pain in the butt having to put that big DIP socket on the network card for boot from LAN.
I'll chip in for those etiquette and network classes, but alas, I don't think there's much chance of him attending!
If that's your best defense, for the raft of errors in your post, well, that's rather weak.
Having done PCMCIA card development I'm aware of the marketing name change.
Actually I understood it to originally mean:
"Peripheral Component Microchannel Interconnect Architecture"
but later changed to
"Personal Computer Memory Card International Association"
Although some circles claims it means:
"Personal Computer Manufacturers Can't Invent Acronyms" or my favorite:
"People Can't Memorize Computer Industry Acronyms"
Although the acronym PCMCIA is still correct for PCMCIA Type I cards. As these should *not* be called PC Cards ever! Only PC Cards Type II and III.
Worse is the fact that PCMCIA 2.1 or later should not even be called PC Cards at all, but rather a CardBus. If you are not confused yet, a JEIDA
4.2 or later is also called a CardBus.Wow and ready for this? There is a PC Card Type IV (4) too. Which should never be called PCMCIA at all. As Toshiba created this baby and was never officially sanctioned by the PCMCIA.
Cabling-Design.com Forums website is not affiliated with any of the manufacturers or service providers discussed here. All logos and trade names are the property of their respective owners.