Slashdot Mirror


Intel Putting Wi-Fi into Future Chipsets

Ridgelift writes "Wired's got the story on Intel's plan to incorporate Wi-Fi into the motherboard chipset. "The chipset, however, will not include an actual Wi-Fi radio, so users will still need a wireless add-on card. Intel has said it eventually intends build a Wi-Fi radio into its microprocessors." This would make setting up a wireless network a lot simpler."

216 comments

  1. Is wifi on CPU a good idea? by brejc8 · · Score: 5, Insightful

    It's all very well putting more peripherals onto processors but with the shrinking feature sizes and an increase in cross talk is placing a powerful 2.4GHz source on die really a good idea? You would have to shorten long lines and slow down the processor or suffer random errors. Doesn't inspire me with confidence.

    1. Re:Is wifi on CPU a good idea? by IAR80 · · Score: 3, Interesting

      It is not on the cpu, but the chipset and the RF part is not there. It will just be something like the 802.11 asic dealing with the the protocol not with power up/down conventers or anything like that.

      --
      http://ebgp.net/ccc/
    2. Re:Is wifi on CPU a good idea? by Anonymous Coward · · Score: 0

      OH MY GOD! Call Intel right away and tell their engineers that they're making a huge mistake. I can't believe they didn't think of this!

    3. Re:Is wifi on CPU a good idea? by Anonymous Coward · · Score: 0

      RTFA...

      It's not the radio there, just the logic...

    4. Re:Is wifi on CPU a good idea? by sbma44 · · Score: 1

      As others have mentioned, this is just logic, not the transmitter itself. But even if they do roll out integrated wifi, it'll never be "on die" with the processor. It'll most likely be a motherboard component. If it ever did come as part of the same component as the processor, you can bet there would be a wall of the a faraday cage between the xmitter and the processor.

    5. Re:Is wifi on CPU a good idea? by bhtooefr · · Score: 1

      Hmm, look at the Pentium 4 - there are a LOT of things they didn't think of, like 75W being too fucking much heat for a CPU to be putting out!

      Sign the petition in my sig, and get us some more efficient CPUs.

    6. Re:Is wifi on CPU a good idea? by John+Hurliman · · Score: 2, Insightful

      Putting WiFi technology on the motherboard seems like the next logical leap, but hopefully they keep the RF transmitter hardware outside the box. Otherwise say goodbye to HiFi soundcard quality.

    7. Re:Is wifi on CPU a good idea? by Anonymous Coward · · Score: 0

      Could you at least read the text even if you didn't RTFA?

      Intel has said it eventually intends build a Wi-Fi radio into its microprocessors.

    8. Re:Is wifi on CPU a good idea? by div_2n · · Score: 1

      No, putting more peripherals onboard is not a good idea. Moreover, putting wifi onboard is an extremely bad idea.

      There are many reasons but I can give you at least one--noise. I operate a WISP and one of our customers dropped off our network and it took me at least a half hour to figure out what happened. He put his brand new Centrino laptop into scan mode and it knocked his CPE offline due to side channel noise.

      Just what do you think will happen to your 2.4 ghz cordless phone when every apartment above, beside and below you are blasting 2.4 ghz with their new computers?

      This is an attempt for Intel to cash in on the wireless boom without stopping to think about the consequences because quite frankly they don't have to since 2.4 ghz is unlicensed.

      And now we all get to witness the downside to deregulation of wireless frequency. Don't get me wrong, I love the fact that I don't have to answer the FCC as long as I play within some power requirements. I do not always sleep well knowing I have no protection either from people like Intel.

    9. Re:Is wifi on CPU a good idea? by jonhuang · · Score: 1

      Once again, it's absolutely amazing that the Intel engineers didn't think of this basic point!

    10. Re:Is wifi on CPU a good idea? by darkwhite · · Score: 1

      The only thing hi-fi soundcard quality depends upon is the quality and interference susceptibility of line circuits (op amps and capacitors mostly) on the DAC outputs. Given that DACs are usually given their own ICs and are somewhat spatially isolated from noisy digital components precisely because of the noise they can receive, I don't see how whether the RF amplifier and receiver lines being on chip will affect sound. By design, digital signals will not be corrupted between the components of the chipset.

      Some (non-linear?) RF circuitry will probably still stay off chip, but in any case, I'm not aware of any non-digital data lines on the system chipsets, and wi-fi frequencies and their significant harmonics are far beyond the range of hearing and thus should not affect analog sound circuits, so your worry seems unfounded to me. Now, putting quality sound hardware on notebooks (where wi-fi is most ubiquitous and most closely packed) seems unpopular among notebook manufacturers, but that's another story.

      Also, I don't know about the rest, but on my laptop all major components are shielded by metal sheets and boxes. On the wi-fi card, no less than 5 shields protect the components, while other shields separate it from all other components nearby.

      --

      [an error occurred while processing this directive]
  2. Like Microsoft... by Jake73 · · Score: 4, Insightful

    This is a lot like Microsoft's business plan, but at the hardware level. It leaves ISV's out of the loop. "Centrino" was perhaps the most serious example of this that I've seen.

    1. Re:Like Microsoft... by Anonymous Coward · · Score: 0

      What does this have to do with Independent Software Vendors?

  3. Intel linux support sucks by pe1rxq · · Score: 5, Interesting

    They will probably promise to provide Linux driver like with the centrino chipset and then not even make specs available.
    You will get all kind of lame excuses:
    - We are working on a driver.... (For half a year already)
    - We can't tell you how to operate it because the FCC won't let us (Complete bullshit but sounds nice: 'linux hackers want to interfere with police radio')
    - They might release some binary only modules... (Redhat version bla.bla, kernel version bla.bla and nothing else)

    Jeroen

    --
    Secure messaging: http://quickmsg.vreeken.net/
    1. Re:Intel linux support sucks by noname3 · · Score: 1

      I've had the same problem with D-Link, and it really pisses me off. Both with the Broadcom based PNA2.0 adapters and their asx100 based wifi adapters.

      I wonder how well DriverLoader will work with this.

    2. Re:Intel linux support sucks by Anonymous Coward · · Score: 1, Informative

      Driverloader is working perfectly for my intel 2100 in my laptop. 2 minute install, and then I have another eth* device, use the standard iwconfig and I'm on the 'net. DriverLoader is well worth the 20 bucks.

    3. Re:Intel linux support sucks by Homology · · Score: 5, Informative

      Intel are not forthcoming with documentation about NIC and crypto. Here is what OpenBSD has to say about this
      OpenBSD experience with Intel :


      Much like Intel does for all their networking division components, and completely unlike most other vendors, Intel steadfastly refuses to provide us with documentation. We have talked to about five technical people who are involved in the development of those products. They all want us to have documentation. They commend us on what we have done. But their hands are tied by management who does not perceive a benefit to themselves for providing documentation. Forget about Intel. (If you want to buy gigabit ethernet hardware, we recommend anything else... for the same reason: most drivers we have for Intel networking hardware were written without documentation).

    4. Re:Intel linux support sucks by gmhowell · · Score: 2, Informative

      Not sure if asx100 and acx100 are the same thing, but if so, have you tried this?

      --
      Jesus was all right but his disciples were thick and ordinary. -John Lennon
    5. Re:Intel linux support sucks by drix · · Score: 1

      I wholeheartedl agree with everything you have just said (see my other post to this story.) While we're on the subject, does anyone know anywhere where one can purchase retail miniPCI cards?! They aren't 'supposed' to be available to the general public maybe someone here knows someone who knows someone... that's the answer to all our prayers for those of us chained to the shitty Intel pro/2100 WLAN card...

      --

      I think there is a world market for maybe five personal web logs.
    6. Re:Intel linux support sucks by O · · Score: 2, Informative

      Um, I know Dell sells them. There was a 802.11b MiniPCI card from Dell on sale for $29 the other week. Check Techbargains for it. Also, if you're desperate, you can generally rip apart a consumer AP or "router" and yank out it's MiniPCI card, too.

      --

      1, 1, 2, 3, 5, 8, 13, 21 -- Mathematics is the Language of Nature.
    7. Re:Intel linux support sucks by Aardpig · · Score: 1

      ...and on a similar note, anyone know where I can find a Gemtek micro-USB (internal) wireless card? I've fried the one which came inside my laptop (bad firmware flash), and I'm having real difficulty locating a replacement. Gemtek themselves don't reply to my emails.

      --
      Tubal-Cain smokes the white owl.
    8. Re:Intel linux support sucks by noname3 · · Score: 1

      Oops! Yes, acx100 is what I meant.

      And whoa, unsolicited help. This isn't what I expected. Thanks. :) Right now I'm just using DriverLoader.

    9. Re:Intel linux support sucks by gordyf · · Score: 2, Informative

      I know you said retail, but I bought a lucent orinoco minipci card off Ebay a little while ago. $35 for the minipci card attached to a real pci card - removed two screws, popped it out of the socket and replaced my Intel pro/2100 (piece of crap), and have been happy ever since!

      It uses the Orinoco cardbus drivers, since the orinoco minipci card is essentially a PCI card with a TI 1410 PCI/CardBus bridge and a CardBus orinoco card all in a minipci form factor. Cool stuff.

    10. Re:Intel linux support sucks by gilesjuk · · Score: 1

      Thing is you've bought their product, why should you have to buy drivers for it?

    11. Re:Intel linux support sucks by Anonymous Coward · · Score: 0

      I bought a laptop which is designed to run Windows XP. It was not designed to run linux. I cannot fault the manufacturer for choosing not to support a OS used by the minority (especially on laptops).

      If they had stated that it worked in linux, but did not and I was forced to pay extra to get it working, then you may have a valid point, however I was well aware that the wifi did not work in linux at all when I bought it, and I'm quite happy that there is any solution at all to get it working in linux even one which requires an additional (quite small) fee.

    12. Re:Intel linux support sucks by Anonymous Coward · · Score: 0

      Intel released a BSD licenced driver for their gig cards. Either that rant is out-of-date, or it's misguided.

      http://appsr.intel.com/scripts-df/download.asp?u rl =/4105/eng/em-1.7.19.tar.gz

    13. Re:Intel linux support sucks by Homology · · Score: 1

      This is interesting, and it's nice that Intel has changed their mind about some of their producs.

    14. Re:Intel linux support sucks by pe1rxq · · Score: 2, Insightful

      However intel have publicly stated that they would have linux support for their centrino chipset...
      (which the pro wireless 2100 is part off)
      That turned out to be one big lie.

      Jeroen

      --
      Secure messaging: http://quickmsg.vreeken.net/
    15. Re:Intel linux support sucks by Anonymous Coward · · Score: 0

      I can't say for *BSD, but Intel is very active contributing code and specs to Linux, plus tons of seed capital to companies like RedHat, plus they are basically paying Linus through OSDL. That's why the lack of a wireless driver is so confusing.

      Also, unlike most Linux and FreeBSD developers, Theo refuses to sign NDAs, so it's no shock he doesn't get documentation.

    16. Re:Intel linux support sucks by Jeff+DeMaagd · · Score: 1

      I don't understand. How would signing an NDA allow anyone to write an open source driver with closed documentation?

      If the NDA documents are required, and that some information in that protected document gets into the driver, then wouldn't that person be violating the NDA?

      Please clarify.

    17. Re:Intel linux support sucks by Anonymous Coward · · Score: 0

      You will have to ask any of the many Open Source developers that sign these things, but I imagine the NDA says "Don't give away these docs", not "Don't use the information in these docs."

    18. Re:Intel linux support sucks by cheekyboy · · Score: 1

      Hopefully some intel employees would get sacked, then release the specs.

      I would.

      covertly.

      --
      Liberty freedom are no1, not dicks in suits.
    19. Re:Intel linux support sucks by AstroDrabb · · Score: 1

      example only:

      I am an Inetl driver developer, you are a Linux driver developer. I come over your house for dinner and by "accident", I leave the technical specs at your house. You read the specs and write a Linux driver. I signed an NDA with Inetl, you did not. No laws were broken. You did not see ANY of Inetl's source code or "IP". There is nothing Intel can do to stop you from writting a fully-functioning driver since you did not see any of their source or "IP", and you reverse engineered it : )

      Of course, I am not a lawyer so there may be something to stop a situation like this. Though as far as I know, there is nothing wrong with reverse engineering and they cannot prove that you even looked at the Intel specs, etc.

      --
      If Tyranny and Oppression come to this land,
      it will be in the guise of fighting a foreign enemy. -James Madison
    20. Re:Intel linux support sucks by jmauro · · Score: 1

      Except if you knew that your friend only obtained the information under NDA, you would still be sued for releasing information that you knew was a trade secret even though you didn't sign the NDA. I know the whole idea sucks, but that is how the trade secret protection laws are written.

      If you didn't know the information was under NDA, fell happy to know that Intel's legal department will bury you and the project anyway. That and in either case your friend will be spending a lot of quaility time in the court system and talking with lawyers.

    21. Re:Intel linux support sucks by jmauro · · Score: 1

      Then Intel would sue the employee for all their worth and probably then some. Even if you get fired you're still required to protect the trade secrets you had access to while working at a company, especially if you signed something when you started working there.

    22. Re:Intel linux support sucks by Anonymous Coward · · Score: 0

      Exactly. This is why I would not purchase Intel hardware, even though I'm an Intel employee (not in the processors division however).

    23. Re:Intel linux support sucks by gmhowell · · Score: 1

      I've spent the weekend fighting various wireless products for my wife's machine. Finally stumbled onto some supported hardware. Glad it's working for you.

      --
      Jesus was all right but his disciples were thick and ordinary. -John Lennon
    24. Re:Intel linux support sucks by megabeck42 · · Score: 2, Informative
      Frankly, this is taken out of context and is rather misleading. Sometimes I wish for a moderation option entitled "Wrong."

      First of all, the section above is listed under the header called "Intel Ipsec Cards" and more specifically refers to the Intel Encryption Coprocessor on the card.

      Further, Intel has written and released a free, GPL ethernet driver for their EEPro 100, 1000, and 10000 ethernet cards. I shall transcribe for your benefit the top few lines from linux/drivers/net/e100/e100_main.c:

      Copyright(c) 1999 - 2003 Intel Corporation. All rights reserved.

      This program is free software; you can redistribute it and/or modify it
      under the terms of the GNU General Public License as published by the Free
      Software Foundation; either version 2 of the License, or (at your option)
      any later version.

      This program is distributed in the hope that it will be useful, but WITHOUT
      ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or
      FITNESS FOR A PARTICULAR PURPOSE. See the GNU General Public License for
      more details.

      You should have received a copy of the GNU General Public License along with
      this program; if not, write to the Free Software Foundation, Inc., 59
      Temple Place - Suite 330, Boston, MA 02111-1307, USA.

      The full GNU General Public License is included in this distribution in the
      file called LICENSE.

      Contact Information:
      Linux NICS
      Intel Corporation, 5200 N.E. Elam Young Parkway, Hillsboro, OR 97124-6497

      That driver is a GPL implementation meaning that the OpenBSD developers are more than welcome to port it at their leisure.

      Oh, you want real documentation too. Take a look at developer.intel.com:
      Intel 82551ER Fast Ethernet Controller Networking Silicon Datasheet
      Intel 82559ER Fast Ethernet Controller Networking Silicon Datasheet
      Intel 82559ER EEPROM Map and Programming information

      True, Those are for their FastEthernet chipsets, not the Gigabit chipsets that, "Intel steadfastly refuses to provide us with documentation"

      Well, what about these?
      Intel(R) 82541ER Gigabit Ethernet Controller Networking Silicon Datasheet
      Intel(R) 82547GI(EI)/82541ER EEPROM Map and Programming Information Guide

      Ok, so the IPSEC chip on the NIC isn't supported nor is there any data on that chip forthcoming. However, there are a number of papers that show that IPSEC and TCP offloading (not to be confused with TCP fragmentation/checksumming) are not efficient. Specifically, the "hardware" IPSEC is done by firmware downloaded to a small embedded processor on the NIC. This small, embedded NIC is not very fast, in fact, its rather slow.

      Result:
      Processor utilization drops marginally (modern processors can encrypt 10 megabytes/s trivially)
      Latency shoots up (It takes the embedded processor longer to encrypt a packet than the host processor would.)

      There are a number of papers corroborating that latency has a huge effect on maximum bandwidth.

      [I think the paper regarding TCP offloading not being worthwhile is by Mudge. The IPSEC offloading not being worthwhile is my hypothesis, untested, but I feel logically founded.]

      My point is that IPSEC offloading is not an advantage - it probably was in 200MHz K6 days, but it certainly is not in 2.0 ghz K7 days.

      Other notables, for example, the 3Com 3CR990 still doesn't have IPSEC offloading, despite promises from either the openbsd txp driver or the linux typhoon driver.

      Frankly, as far as Gigabit

      --
      fnord.
    25. Re:Intel linux support sucks by gilesjuk · · Score: 1

      No, you bought a laptop with a Centrino chipset. The OEM license deal probably requires a "Designed for Windows XP" to be stuck onto the case.

      Only a few products are designed with Windows in mind, usually cheap modem chipsets or USB devices.

  4. Really? by odyrithm · · Score: 5, Insightful

    This would make setting up a wireless network a lot simpler.

    Woah there, whats so hard with the way its setup now? pcmcia is a matter of plug and go, pci is a matter of modprobe if that.. theres nothing hard about wifi... its a nic with a wireless medium.. thats all.

    Now if intel had some new fangled wep replacement then that would make things simple, no more mac rules on my fw would be nice.. which is unlikely.

    --
    moo
    1. Re:Really? by blackketter · · Score: 1, Flamebait

      How easy or hard it is depends on the software that drives the thing. Setting up an Apple Airport network is extremely simple, and the 802.11 support in XP is pretty darned good.

      modprobe doesn't seem to me to be a simple way to set up anything.

    2. Re:Really? by Artifex · · Score: 2, Interesting
      Woah there, whats so hard with the way its setup now? pcmcia is a matter of plug and go, pci is a matter of modprobe if that.. theres nothing hard about wifi... its a nic with a wireless medium.. thats all.


      You guys are not looking at the big picture, here. Adding this on board allows OEMs to specify a motherboard platform with this, and then buy the cards at cheaper prices than full implementations on cards would cost. It's just following the trend of AMR slots and onboard video, and, in the last couple of years, wired NICs.

      Early hard drive and floppy controllers used to only be on cards, too.

      --
      Get off my launchpad!
    3. Re:Really? by odyrithm · · Score: 1

      modprobe doesn't seem to me to be a simple way to set up anything.

      Id take modprobe over windows drivers anyday ;)

      I know it depends on the software, any hardware depends on software at the end of the day.

      --
      moo
    4. Re:Really? by odyrithm · · Score: 1

      and then buy the cards at cheaper prices than full implementations on cards would cost.

      Im not trying to be picky, but that made no sense, what you mean is.. this can be seen as the winmodem of wifi?

      Early hard drive and floppy controllers used to only be on cards, too.

      My raid controller is on a pci card... its not very early though ;)

      --
      moo
    5. Re:Really? by Tim+C · · Score: 1

      Id take modprobe over windows drivers anyday ;)

      Why's that? With Windows, I just run an executable, and 99% of the time the drivers install correctly and work fine (in my experience). Of course, that 1% is the rub - then, you're reduced to completely uninstalling them, rebooting, reinstalling from scratch, sacrificing chickens, etc.

      In my experience of linux, though, if you actually have to start loading kernel modules or hacking config files yourself, that figure goes to more like 80%, *and* you're going to be searching for and reading HOW-TOs and other docs, that weren't always written with the end user in mind.

      Oh, and to stop the useless "Linux newbie! M$ sux0rz, Linux rox0rz!!!" replies, note that my first Linux install was Slackware 3 off floppies 5 or so years ago. I know what I'm doing, mostly, and it's still not easy if there isn't an install script or similar *for your exact distro*.

    6. Re:Really? by Anonymous Coward · · Score: 0

      The hardest thing is getting the sodding closed-intel-crap working on linux...

    7. Re:Really? by odyrithm · · Score: 1

      my first Linux install was Slackware 3 off floppies 5 or so years ago. I know what I'm doing

      I hate to be the one to break this to you... but you really dont.

      Just because you installed slack3 does not mean in anyway you know what your doing, I hate it when people say "oh I installed version N" and for some crazy reason believe that makes them knowledgable of GNU/Linux... its just madness.. I started on windows 3.11, and have had to follow the windows all the way to XP and 2003, have you followed GNU/Linux to this date? and used it day in day out? proberly not.

      --
      moo
    8. Re:Really? by odyrithm · · Score: 1

      thats why you gotta love the intelligent folk who know how to driver!!!

      omg I crack myself up sometimes. ;)

      --
      moo
    9. Re:Really? by be-fan · · Score: 2, Interesting

      I'd say the number is higher than 99%, but its exactly that few percent that are the kicker. With modprobe, I can type "modprobe i810_audio" and the thing just starts working. If something goes wrong, I can check the dmesg output, and ahve a range of debugging options. With Windows, if something goes wrong, you're screwed. I've been trying to debug persistent wonkiness in the wireless card of one of my Windows machines. After months, I've just given up and take to reinstalling the drivers every month. It doesn't help that there are two competing wireless managers (the one that came with the card, and the one that comes with Windows) and I really don't think I can reproduce the exact steps I took to disable one and keep the two from interfering with each other.

      Of course, that's largely a moot point --- current distros use something like hotplug or kudzu, so you don't have to use modprobe anyway. But the advanced troubleshooting options are still there if you need them.

      --
      A deep unwavering belief is a sure sign you're missing something...
    10. Re:Really? by Anonymous Coward · · Score: 0

      Id take modprobe over windows drivers anyday ;)

      I use a Mac.

      What's a "driver?"

    11. Re:Really? by Anonymous Coward · · Score: 0

      I started on windows 3.11 Really? I started on 1.02!

    12. Re:Really? by Anonymous Coward · · Score: 0

      yes, the big picture is that this would position intell to set certan aspects of market standards. this isn't neccesarily a bad thing but i remeber a ordeal with the centrino chips that if you use a different _non-intel_ add on wifi card you would lose some of the benefits of the centrino platform (that was more than the wifi). i wonder if adding the wifi supoport to the cipset could (un_intentionaly) lock out competitors or force them to pay for compatability?

      would what apears to be a benefit end up locking consumers into a sub standard experience because other product won't work or they will disable advantages that makes the platform attractive in the first place.? As we have seen with some intell product they are locking non-mainstream operating systems out while claiming suport, is it because the driver that can be released for **nix style os's would have the key to allow the limitation in the centrino platfrom to be corected?

      of course i have no problem with a company that has a tuned set of hardware that works best when using "thier tuned product" but when the tuning apears to lockout competition, i'm not sure if that is a benefit at all.

    13. Re:Really? by SCHecklerX · · Score: 1

      I sure hope you don't think you are really secure by simply enabling mac filtering. At least use WEP too. Even the windoze drivers for most cards allow lusers to easily change their card's mac address. It's trivial to capture valid MAC addresses using tools like kismet.

    14. Re:Really? by Rob+Simpson · · Score: 1

      Yeah, this sounds exactly like an AMR slot. Has anyone actually used one of those? Or even seen an AMR modem? If this was actually totally integrated and functional, like a built-in NIC, it might actually be useful.

  5. What's the point? by bobthemuse · · Score: 5, Insightful

    The chipset, however, will not include an actual Wi-Fi radio, so users will still need a wireless add-on card. Intel has said it eventually intends build a Wi-Fi radio into its microprocessors.

    Why put in a chipset without a radio? Maybe one could argue an attempt to get market share by making their add-on card cheaper than the others (just radio, no chipset), but this card will have a more limited market, since it wouldn't be compatible with older or non-intel mobos.

    Now if the were to put a software radio on board, *that* would be cool! Think of upgrading to future standards with just a flash rom upgrade...

    1. Re:What's the point? by odyrithm · · Score: 1

      Why put in a chipset without a radio?

      As mad as it seems it does make sense, this way you can bolt on anykind of radio/antenna you wanted, should give plenty of us somthing todo on a slow sunday, wifi over packet radio anyone?? ;)

      My ideas are proberly floored I know, sorry, but its sunday and its not only /. thats slow right now ;p

      --
      moo
    2. Re:What's the point? by Zocalo · · Score: 1

      Maybe to avoid more allegations of Intel being a monopoly? This is kind of like Microsoft embedding an HTML rendering engine in Windows instead of the entire browser, or Intel's own onboard graphics set; it's up to the vendors what to do with it. They could ignore it and provide their own (presumably higher spec) system, (like all those onboard Intel boards with an ATI/Nvidia AGP card fitted). Alternatively, they can use it and either provide a cheap radio with short range, or go for it and bundle a radio that can communicate with nearby star systems.

      --
      UNIX? They're not even circumcised! Savages!
    3. Re:What's the point? by Hoser+McMoose · · Score: 1

      Initially it will probably be targeted at OEMs of the Dell and HPaq variety, so compatiability with older boards (or future boards for that matter) will be a non-issue. For these guys, saving a few dollars on parts is a fairly big deal, even if it means that users have a hard time getting replacement parts on their own (in fact, this may be seen as a benefit to the OEMs).

      As for the future, I'd imagine that Intel will try to put as much of a software radio on board as they possibly can. Anything that can bring more of the total system dollars into Intel's pocket is a good thing in their books. It used to be that Intel ONLY made the CPU in most computers. Now, Intel is responsible for the CPU, the memory controller, hard drive controller, extra I/O controller (USB, serial, keyboard, etc.), sound, modem, NIC, etc. etc. Heck, Intel is even competing with nVidia to be the #1 supplier of graphics processors in the world now. The more things that they can integrate, the larger chunk of the total cost of a computer goes into Intel's pockets. And if the integration is of the software variety and needs a little extra processor umph, that's an extra bonus for Intel since it gives users more reason to upgrade to the latest and greatest processor.

  6. users will still need a wireless add-on card by rokzy · · Score: 4, Insightful

    I don't follow tech too closely but this sounds like those worthless AMR slots or whatever they were called - the ones that were like having a built-in sound card or modem (in the sense that the board cost more and had less space available) except didn't actually do anything...?

    1. Re:users will still need a wireless add-on card by RLiegh · · Score: 1

      I don't know what an AMR slot is, but after fighting with gateway I got "as a present" back in 2001 which has an intel soundcard and modem built into the mobo, I think I'm DEFINATELY going to wait and see what the competition is.

      On a side note, is it just me, or is EVERYTHING that intel makes in the way of peripherials (mice, webcams, etc) total shit? Why don't they stick to CPU's? I can't imagine them making any $ off of their shitty, non-functioning web cameras.

    2. Re:users will still need a wireless add-on card by bhtooefr · · Score: 1

      Take the whole family (ACR, AMR, CNR), okay? BTW, ACR/AMR is SHIT.

      ACR Audio Communications Riser, a slot that is only used by large OEMs to add cheap devices to their computers. Since ACR devices are software based, they often times sap a lot of CPU resources.

      AMR Audio Modem Riser, like the ACR, AMR's are extremely low cost software devices that can do a host of functions. Since they are software based, they use quite a bit of CPU resources.

      CNR Communications Network Riser - Small slot next to the PCI slots which a modem or sound riser card plugs into.

  7. Dumb Question by Anonymous Coward · · Score: 2, Funny

    Customer:If I can have a 3 Ghz processor why cant I have WiFi at 3 Ghz
    Radio Shack Assistant: Mam the 3Ghz speed refers to CPU speed while Wi Fi works at 2.4 Ghz
    Customer:Dont get technical on me . I know this Ghz speed keeps increasing all the time . It cant be fixed at 2.4 Ghz

  8. Hospitals and Airlines beware by ericspinder · · Score: 2, Interesting

    Building more functionality into the motherboard is an ongoing trend, but adding a radio cannot be a good thing. Due to potential interferance, you cannot go into a hospital or airplane without being told to turn off your cellphone.

    --
    The grass is only greener, if you don't take care of your own lawn.
    1. Re:Hospitals and Airlines beware by gmhowell · · Score: 1

      Both of those industries, IIRC, are basing things on spurious evidence at best. However, given the possibly life threatening nature, better to be safe than sorry.

      But, I should also point out that healthcare uses a TREMENDOUS amount of wireless technology. Nurses with wireless pads to enter vitals, monitors sending info via wireless, etc. Not sure if they use 2.4 or some other frequency, but they exist.

      --
      Jesus was all right but his disciples were thick and ordinary. -John Lennon
    2. Re:Hospitals and Airlines beware by iantri · · Score: 1

      This is not a problem because, if you read the blurb above, you will see that the radio is NOT included. If you want to enable the Wi-Fi features you need to add this.

    3. Re:Hospitals and Airlines beware by ericspinder · · Score: 1
      Really, I did read both the article and the summary. Which blurb in the summary were you talking about, was it the one right before "Intel has said it eventually intends build a Wi-Fi radio into its microprocessors."

      I am sorry to be a bit of a jerk about it, you might have noticed your error after you posted. Sometimes, it's a bit of a problem that you can't edit after you commited a comment.

      --
      The grass is only greener, if you don't take care of your own lawn.
  9. Wouldn't this compromise security? by twiztidlojik · · Score: 4, Insightful

    I think that if an admin of a corporate network had several thousand wireless clients, a simple airsnort would compromise thousands of computers, without requiring the packet sniffer actually access any sort of ethernet cabling.

    This could have a grave impact on the sales of wireless-based chipsets in the corporate market.

    --
    I will now redundantly add my name to the end of my post. You know, in case you forgot me or something.
    1. Re:Wouldn't this compromise security? by Bishop · · Score: 1

      It is already a pain in the ass. Many laptops now ship with built in wireless. So far it has been possible to disconnect the antennas. But it has been impossible in most cases to disable or remove the WiFi completely as the wireless card is also the ethernet and modem card.

    2. Re:Wouldn't this compromise security? by Anonymous Coward · · Score: 0

      I was wondering too. Plus, has anyone done any real tests with these so-called antennas being removed? I know on my cell phone, I've removed the external antenna, but if I'm next to a tower, I still get a signal.

      This could be a huge problem when installing unified chipset drivers that are closed source. Hopefully, Intel is smart enough to have the software disregard WiFi by default or at the very least allow the user to turn that aspect of the driver off.

  10. how convenient by vinsci · · Score: 4, Insightful

    So when Big Brother wants to know what you're up to, they don't need you to be online to the public internet. How convenient. The "trusted" BIOS can always let them bypass your firewall, as the BIOS is going to handle the net connection too.

    --

    Trusted Computing FAQ | Free Dawit Isaak!
    1. Re:how convenient by SectorNotFound · · Score: 1

      My first thought. Backdoor-equipped for Big Brother. Intel for Intel.

    2. Re:how convenient by maximilln · · Score: 1

      All too true. Unless you happen to be a computer engineer with the specs for all of the chips on the mobo in front of you along with the BIOS source code then who's to say we're not all backdoored already?

      --
      +++ATHZ 99:5:80
    3. Re:how convenient by drinkypoo · · Score: 1

      The BIOS would have to know everyone's network card, since they all have different drivers, or it would have to be called at some point after initialization, which doesn't happen; the BIOS is neglected on 32 bit operating systems once you have booted.

      --
      "You're right," Fisheye says. "I should have set it on 'whip' or 'chop.'"
    4. Re:how convenient by maximilln · · Score: 1

      That's what everyone keeps telling me but I'm skeptical. I think about it in this fashion:

      -Get an electrical signal to a network card using the TCP/IP protocol.
      -Most network card architectures are going to be similar for passing that signal to the PCI/ISA bus.
      -Find a flaw in the target OS which allows signals from the PCI bus to begin filling memory registers (maybe simple flooding will cause a voltage overload which "spills" into subsequent registers).
      -Using a list of common possible memory configurations start filling up stack space and inserting backdoor code.

      Sure it sounds sketchy but I won't be the least bit surprised if, in six months or less, we hear that someone's doing it. I was flamed mercilessly on #LFS for suggesting that routers could be compromised at the hardware BIOS level. Two days later HP was releasing a patch for their hardware routers.

      Go figure.

      --
      +++ATHZ 99:5:80
    5. Re:how convenient by vinsci · · Score: 1
      Please read up on the Trusted Computing FAQ, which I can tell you haven't read yet. Then read:
      --

      Trusted Computing FAQ | Free Dawit Isaak!
  11. Unintentional network compromises through bridging by AgTiger · · Score: 5, Insightful

    I sincerely hope they're going to have these things configured in one of the following manners:

    1. Disable the Wi-Fi by default
    2. If not disabled, seed the encryption key with a pseudorandom number before the user specifically configures it.

    You don't want new computers forming unintended bridges or access points between the untrusted network/airspace and your trusted internal network between when they're first powered up and when the overworked sysadmin has a chance to configure them properly. So much for your company's firewalls having a chance to do their job.

  12. Huh? by Aneurysm9 · · Score: 2, Insightful

    Wi-Fi without a radio? Sounds kinda like a soundcard without a DAC or a videocard without a RAMDAC. What's the point of including functionality while not including necessary pieces except, perhaps, to rachet up the marketingspeek and pressure out other manufacturers?

    --
    There was Cowboy Neal at the wheel of a bus to never-ever land.
    1. Re:Huh? by maximilln · · Score: 1

      Perhaps it's the same premise as the winmodem. The entire push, from the hardware level on, was to foil functionality under competing operating systems by leaving the functional part off of the hardware board. The necessary functionality is added from the software side and it's only written for the Windows platform.

      --
      +++ATHZ 99:5:80
    2. Re:Huh? by Anonymous Coward · · Score: 0

      Actually it's more like a soundcard with a DAC converter without the port to hook the speaker in, or a video card without the video port to hook the monitor in. It's like alot of motherboards have onboard lan in the chipset but some manufacturers choose not to have a lan port on the motherboard.

    3. Re:Huh? by Anonymous Coward · · Score: 0

      Err, software radio can be done, but it requires an expensive (four figures) high-speed A/D converter.

      Basically everything except the radio is easy to do in software.

      However, motherboard chipsets are in many ways more similar to software than hardware in their capabilities (pure digital logic), so it is sort of natural for Intel to include that part.

  13. Should wireless really be in the chipset?? by Anonymous Coward · · Score: 3, Interesting

    I think it should be integrated into the motherboard (like regular ethernet) but not the chipset.

    Or is intel upset that sales of centrino are so poor? To be an "official" centrino laptop, you have to offer intel's 802.11b wireless. Not surprisingly, many people want faster (802.11b/g or 802.11a/b/g) wireless cards.

    Broadcom has been eating intel's lunch in the oem ethernet (wireless & wired) card market. Sounds like anticompetitive monopolist activity to me.

  14. WiFi Security by Chalybeous · · Score: 4, Interesting

    What about security issues?
    I like my router (at home, I share a cable internet connection between two desktop PCs and, occasionally, my laptop). It has an inbuilt firewall according to the manufacturer, and I know if I ever do have a serious problem which I suspect originates from the internet, I can physically disconnect it. Sure, cables are archaic, but they're cheaper and more secure than wireless networks - especially for the novice (like myself).
    But if you enable a CPU to act as a wireless hub - or, eventually, if WiFi comes as a full onboard feature (rather like many motherboards now have onboard sound and graphics) - would that not open up your PC and network to security issues? My parents would not be best pleased if someone warchalked on the fence, but since they have little idea of technology or computer security, I think if they bought a new machine enabled with this kind of tech, every l33t hax0r in a two mile radius would be camping out to leech their access.

    Any other thoughts, opinions or predictions?

    --

    "It is dark. You are likely to be eaten by a grue." -- Zork

    1. Re:WiFi Security by gmhowell · · Score: 1

      Thoughts? Yeah, go into the BIOS and turn it off, like you can onboard IDE controllers, ethernet ports, serial ports, parallel ports, etc, etc, etc.

      Oh, or get your head out of your ass and reread the article where it says this is just the chipset, and doesn't include a radio.

      --
      Jesus was all right but his disciples were thick and ordinary. -John Lennon
    2. Re:WiFi Security by 1337_h4x0r · · Score: 1

      I won't leech their access, I promise. :)

    3. Re:WiFi Security by Chalybeous · · Score: 1

      I read the article... it just occurred to me that, with a WiFi card installed or if motherboards ever feature full WiFi, as they have come to do with graphics and sound, there could be security issues.
      You and I are lucky. We know about turning off onboard stuff via the BIOS, or putting up a firewall, or disabling network connections. But not everyone does. That's the point I was making; my parents have a PC, and they sure as heck don't know about internet security. Three years, no antivirus software, no patches, no maintenance - I spent three days with various tools, bringing their Win98 up to date and getting rid of Gator and other stuff that they installed purely because they didn't know any better.

      So, to sum up: my opinion is it's good tech if you can run your OS of choice on it and if you can deal with the security issues should you have a WiFi card installed. Otherwise, it's possibly a bad choice.

      --

      "It is dark. You are likely to be eaten by a grue." -- Zork

    4. Re:WiFi Security by Chalybeous · · Score: 1

      *grin*
      Thanks, man. I've been writing a paper on Jane Eyre (no tattooist jokes, please) and researching another on double negatives this afternoon, and I needed a laugh.
      Two days here and already /.ers are making me smile, without having to resort to photoshopping the goatse guy. ;-)

      --

      "It is dark. You are likely to be eaten by a grue." -- Zork

    5. Re:WiFi Security by Dun+Malg · · Score: 1
      But if you enable a CPU to act as a wireless hub - or, eventually, if WiFi comes as a full onboard feature (rather like many motherboards now have onboard sound and graphics) - would that not open up your PC and network to security issues? My parents would not be best pleased if someone warchalked on the fence, but since they have little idea of technology or computer security, I think if they bought a new machine enabled with this kind of tech, every l33t hax0r in a two mile radius would be camping out to leech their access.

      Cripes, man, you don't even have to RTFA to figure this out-- it's in the /. blurb:

      "The chipset, however, will not include an actual Wi-Fi radio, so users will still need a wireless add-on card"
      Can we end the insanity now? No one will be hax0ring your box through the wireless chipset if there's no freekin radio transceiver connected to it.
      --
      If a job's not worth doing, it's not worth doing right.
    6. Re:WiFi Security by gmhowell · · Score: 1

      You're absolutely right. A fact I was reminded of yesterday morning when my mother said "Do you want to come down for dinner tomorrow? BTW, my computer isn't acting right..."

      --
      Jesus was all right but his disciples were thick and ordinary. -John Lennon
  15. Might make it easier to find by man_of_mr_e · · Score: 1

    When a server gets "lost", you can track it by it's radio signature.

  16. Why do companies expand into other est business by Bruha · · Score: 1

    I think it's silly how Intel is now embracing a trend to include features already provided by another market. This is the same situation that MS caused by intergrating IE into the system. So now people will quit buying wireless AP's and routers becuase it came with the motherboard.

    Intel will probalby do this witout adding more than 20 employees and in turn drive about 10,000 people out of jobs due to their companies going out of business.

    1. Re:Why do companies expand into other est business by Artifex · · Score: 5, Insightful
      I think it's silly how Intel is now embracing a trend to include features already provided by another market. Intel will probalby do this witout adding more than 20 employees and in turn drive about 10,000 people out of jobs due to their companies going out of business.


      Oh, yes, it's Intel's fault if 20 of their employees can make a product better suited to the market than 10,000 other peoples' efforts.

      By the same token, buses and trains and taxis have all taken jobs away from the good hard-working people with horse-drawn taxis. And we really should go back to rooms full of seamstresses making clothing by hand, like before those evil industrial looms were created. Oh, and the cotton for the clothes (synthetics put farmers out of work) should be picked by immigrant laborers.

      It's not society's obligation to prop up inefficient methods of production; quite the contrary. There's a word us old-timers use sometimes. It's called "progress." Might want to study it.

      --
      Get off my launchpad!
    2. Re:Why do companies expand into other est business by King_TJ · · Score: 1

      I might be wrong, but I don't think the original poster was trying to argue that companies have no business ever developing products superior to what's already on the market!?

      I think his point was, companies are wise to stick with their strengths, and not to venture into areas that over-extend their reach.

      With the flood of complete wi-fi solutions out there, does it really make sense for Intel to start offering a partial wi-fi solution like this, integrated into their chipsets?

      For starters, many people like the idea of keeping the wi-fi router physically seperate from the PC. It's a good bet that almost none of these independent wi-fi hubs/routers/residential gateways will start using full-blown Intel chipset-based motherboards inside of them. Therefore, Intel's not going to capture any of that market.

      They're probably hoping to make some money at the expense of vendors selling PCI or USB cards to add into workstations, though. Will it work? Probably so, to an extent. But IMHO, this has potential to be the wireless version of the software-based "winmodem". It's a cheaper solution, but also an inferior one. People selling radios that integrate with Intel's wi-fi chipset are locked into whatever technology Intel has. Meanwhile, competitors are free to build complete solutions that out-perform Intel's setup, because they control the entire wi-fi package inside their cards. (Already, some vendors have wi-fi setups that adhere to the standards with competitor's products, but talk at higher speeds with their own products.)

    3. Re:Why do companies expand into other est business by Anonymous Coward · · Score: 0
      Except progress ain't always a good thing. Progress is supposed to make the lives of future generations better, remember? That's kind of the whole point. Are our lives better than our parents? grandparents? great-grandparents? To my knowledge of history the answer is a qualified maybe. Hot and cold running water are awesome, but the only reason that wasn't present is that christians destroyed civilization 1500 years ago. Water doens't really take high technology. Modern medicine is great, except the diseases are evolving immunity to our treatments, so our grandchildren may very well be subject to smallpox, polio, etc. etc.

      And we've lost a hell of a lot. Transportation has actually gone down the shitter in the states in my grandparents life-time, from being pretty good to abjectly horrible. From being able to go anywhere cheaply by comfortable train travel to being crammed into an uncomfortable airplane seat, with lousy service, and a big possibility of getting strip searched and probed before you leave. Progress is great, ne?

    4. Re:Why do companies expand into other est business by Bruha · · Score: 0

      Sure those 20 people may make something better.. but do they have the buying power of 10,000 people? Each time we have a business that does it better with fewer people you tighten the job fields and increase the strain on the markets.

      What do we do when everyone's been replaced by a robot but they dont have money to spend on anything.

      Capitialism is a symbiant circle if you dont keep it in balance with money flowing through the circle you get chaos.

      A computer could easily do my job but then instead of me pumping nearly 22k into my local economy on things I buy thoughout the year into a burden of 1200 dollars per month in unemployment. Coupled with food stamps and whatever other public assistance I could muster I could actually drain more out of the local area then I was putting in.

      And I make nearly 55k a year.

      Not that all of those 10k people were getting paid that much but most times people can put equal drains on public assistance reguardless of how much they made before. Either way it's better for something to be crappy and people to be employeed than have a superior product and put thousands out of business. If they're not paying taxes then you're paying it for them.

    5. Re:Why do companies expand into other est business by Anonymous Coward · · Score: 0

      It's not silly. It's what market economies are all about.

      Intel can make a PC wireless capable more efficiently than their competitors. Intel's stockholders, employees, and consumers benefit.

      Those who can not currently compete, either get smarter and produce product better for the consumer than Intel or perish.

      If you think this is the wrong approach, please send me a donation to keep my sealing wax company afloat. We are having trouble making our payroll this month.

      Please think of all my employees' children!

    6. Re:Why do companies expand into other est business by Anonymous Coward · · Score: 0

      Yeah, sure, capitalism uber alles, but remember capitalism is about *competition*, not *monopolies*.

      When a big company wipes out the small ones, it's competition that suffers, and likewise progress. If intel corners the market on wi-fi enabled motherboards, their rate of innovation will slow. Most people will use the wi-fi on-bard, instead of paying extra for another one, even if it is better.

      You have to keep in mind there is a balance between progress and efficiency (as measured by profit margins). They might use the profits for research, or they might not. But the little guy always has to be innovating.

    7. Re:Why do companies expand into other est business by cpeterso · · Score: 1


      yeah, your grandparents could use those trains to travel cross-country in just a few hours or anywhere in the world in just half a day. Today's planes can't do THAT. And I bet today's plane tickets are (comparitively) cheaper.

  17. Yeh but look at Centrino by Majin+Viper · · Score: 1

    Yeh but look at Centrino, they only had support for Windows, what about linux users we had to try and reverse engineer all the stuff and it still barely works. why can't people just add support for the main OS's used on PC's, Doze, Linux and MacOS

  18. Not complete bullshit by mindstrm · · Score: 3, Insightful

    The FCC DOES require that ISM band consumer devices opearate within specifications. They are certified for a given antenna configuration, among other things. Even the anntenna connectors are non-standard, for this very reason (so the consumer doesn't think you are supposed to hook it up to an amp).

    It's nto that you aren't allowed to modify it legally, of course you are, as long as you operate within spec... but that the company has to make it so.

    It's not a big stretch for them to feel releasing driver code is a risk, as anyone who gets it from them could easily use it to operate outside of spec.

    1. Re:Not complete bullshit by timeOday · · Score: 4, Interesting
      It's not a big stretch for them to feel releasing driver code is a risk, as anyone who gets it from them could easily use it to operate outside of spec.
      It's no different than selling cars that can exceed the speed limit. My compliance with FCC regulations is not Intel's concern.
    2. Re:Not complete bullshit by shepd · · Score: 2, Insightful

      They might be able to exceed the speed limit, but how many can go over 250 km/h without being modified? :-)

      --
      If you could be told what you can see or read, then it follows that you could be told what to say or think - BoC
    3. Re:Not complete bullshit by pe1rxq · · Score: 1

      That would be a good analogy if these cards were able to produce just about any frequency and any power.... Thats not true.
      The differences are in the specific channels available for various regions and even then it is a difference of only a few channels at most. The power is also limited by the hardware, you can't just push it way over the limits just by software.
      The amount of interference you could produce is comparable with a car that can go slightly faster then the speed limit, not some suppercharged hot rod.btw since when are cars modified to go over 250 km/h outlawed? (Assuming the modifications don't involve removing stuff required for safety)

      Jeroen

      --
      Secure messaging: http://quickmsg.vreeken.net/
    4. Re:Not complete bullshit by shepd · · Score: 1

      >btw since when are cars modified to go over 250 km/h outlawed

      Never outlawed, just that most companies limit them this way so that fast cars don't *get* outlawed. :)

      That way only people smart enough to tinker with their cars (not me) are able to drive them at speeds fast enough to kill dozens, maybe hundreds of people.

      --
      If you could be told what you can see or read, then it follows that you could be told what to say or think - BoC
    5. Re:Not complete bullshit by Anonymous Coward · · Score: 0

      More then you think... I've touched about 260km/h (indicated), in my Honda (Stock) S2000...

    6. Re:Not complete bullshit by Anonymous Coward · · Score: 0

      10 miles an hour, (rufly 25km/h) should be enough to kill dozens of people. in the rite situations maybe even more. maybe there is a little difference here.
      (not that you said there wasn't or anything).

    7. Re:Not complete bullshit by mindstrm · · Score: 1

      Yes, it's very different, I"m sorry.

      Cars are not required to be limited by the manufacturer to beneath the posted limit. an ISM band defice is REQUIRED to behave within limits... and if it came with abutton that said "press here to ignore the limits", it would NOT get approved for sale, even though, as you said, once you buy it, it's your responsbility, not theirs.

      (You realize you need FCC certification in order to sell these things commercially, right? Without passing FCC, you get nowhere)

      Your compliance is not their concern.. but that's not the point. If they distribute the code, it's like selling a box with a button on it that lets you violate FCC. As a consumer mass market ISM item, it's not allowed to do that. They cannot be percieved as selling a product that operates outside those limits.

      YES, it's a stupid argument, and it's stretching things a bit too far... however, it's not total bullshit, and that was the point. It's rooted in real law.

      This is why they have non-standard antenna connectors.. not because you aren't ALLOWED to hook aynthing else up, but so that there is some kind of barrier to entry, so you don't just grab some antenna from ratshack and say "Hey it fits.". So the consumer who goes out and does something weird with it can't say "Well, it came that way, I thought it was okay"

  19. Re:Unintentional network compromises through bridg by lanalyst · · Score: 1

    Considering no vendor seems to slipstream service packs or security fixes onto new XP systems (probably never will - too expensive): just what we need - another attack vector.

  20. There are other problems by plusser · · Score: 5, Informative

    The reason why Intel probably don't want to integrate the Micoprocessor with the actual WiFi transmitter and receiver is quite simple. If they add it inside the IC, they will have to go through radio use approval for every different potential market in the world, before they sell a single component. By letting the motherboard/add on card manufacturer's do this instead, they can concentrate on developing better microprocessors.

    As somebody in the know, I do worry that these new WiFi enabled equipment could be the next mobile phone when it come to interference of avionic systems; especially as many modern microprocessors are prone to soft faults at altitude due to the effects of the upper atmosphere.

  21. To clarify: by HoldmyCauls · · Score: 2, Informative

    This is to make it so that an average desktop computer can function as a router for WiFi traffic in the home or office. The card is needed NMW, in order to grab that traffic. A poster above mentioned using a software radio, but it seems that that would only be useful if things were reversed: the software radio *interprets* the signal, and *generates* one to return to the WiFi device in question, but ultimately it is a radio device which transmits that signal into the air. The problem Intel will face is explaining this in terms that a PHB who *signs* the check to buy this stuff.

    --
    Emacs: for people who just never know when to :q!
  22. Re:Here is a picture of wi-fi chip... by Majin+Viper · · Score: 1

    oh grow up

  23. Re:Here is a picture of wi-fi chip... by Chalybeous · · Score: 1

    Just be grateful it wasn't the goatse guy again - saw that one for the first time yesterday and almost lost my lunch. Jeez, but these things are turning into the /. equivalent of Fark photoshop contest cliches...

    --

    "It is dark. You are likely to be eaten by a grue." -- Zork

  24. Bluetooth? by ybmug · · Score: 5, Insightful

    802.11a/b/g on board would be nice.. but, I would really like to see more motherboards coming with bluetooth onboard. This would seem to make sense with things like keyboard, mice, headsets, and cellphones that are bluetooth enabled coming to market.

    1. Re:Bluetooth? by Doctor+Faustus · · Score: 1

      And Bluetooth PDAs. I imagine the wireless synching would be nice.

      Bluetooth PDAs don't really get interesting, though, until they're paired with Bluetooth cell phones. Then they can use the cell phone for internet access, while the phone is still in your pocket. Unfortunately, there are only a few such phones so far, and they all seem to be GSM.

  25. Real estate isn't free, something's got to go by Anonymous Coward · · Score: 0

    Transistors formerly responsible for generating floating-point errors will now be repurposed towards wi-fi.

  26. I'm going to have to RTFA by AndroidCat · · Score: 1
    They're putting WiFi into the chipset -- but you still need the radio card.

    Isn't that like buying batteries (batteries not included)?

    --
    One line blog. I hear that they're called Twitters now.
  27. DRM implementation by 4volt · · Score: 3, Informative

    I'm all for WiFi everywhere, but it sounds like a pretty big backdoor to me, I don't think I'd want to have a WiFi connection built onto my board that I couldn't disable with anything mroe then software. Next thing you would know Microsooft is using it to send DRM related information or usage stastics without you knowing.

    I realize that it would probably able to be disabled in BIOS, but it wouldn't take much that if M$ wanted to take control they could do it with a few sentences in the EULA.

    Improbable, but possible.

    1. Re:DRM implementation by Dun+Malg · · Score: 1
      I'm all for WiFi everywhere, but it sounds like a pretty big backdoor to me, I don't think I'd want to have a WiFi connection built onto my board that I couldn't disable with anything mroe then software.

      RTFA-- no, wait-- just read the damn /. blurb, it's only a few lines long. There's no radio transceiver. It's just the chipset. If you don't plug in the add-on part with the radio, you've got nothing to worry about.

      Mods, stop marking knee-jerk shrieks of "all intel mobos will have a wireless backdoor!" as "informative"

      --
      If a job's not worth doing, it's not worth doing right.
    2. Re:DRM implementation by 4volt · · Score: 1

      I was speaking of a more towards the trend of having on-board wifi, I know that this specific application dosn't have the transcever. I would like to see (if they do make mobo's with complete wifi) a good way to lock it out if necessary.

  28. Easier? by toupsie · · Score: 4, Insightful
    This would make setting up a wireless network a lot simpler."

    How freaking simple can it be now?
    1) Insert Airport Extreme Card into PowerBook 12"
    2) Turn on PowerBook 12" 3) Select Network from Airport menu & Enter WEP if needed
    4) Wirelessly communicat

    --
    Strange women lying in ponds distributing swords is no basis for a system of government.
    1. Re:Easier? by codeonezero · · Score: 1

      And best of all there's no step 3)

      Sounds like an iMac commercial with voice over by Jeff Goldblum ;-)

      --

      ....
      int main (void) { ... }

    2. Re:Easier? by Anonymous Coward · · Score: 0

      Well this will save the PC folks from the pain of step #1 (but unfortunately they still have to deal with their steps 2-34, and if they are mixing products from different manufacturers, steps 35-67).

  29. call me paranoid by ciaran_o_riordan · · Score: 2, Insightful

    I'm no Wi-Fi expert, but couldn't a wi-fi-enabled cpu transmit data without your permission?

    Unique cpu ids? Treacherous Computing Group data?

    1. Re:call me paranoid by gl4ss · · Score: 1

      why wouldn't any wifi enabled anything be able to do it?

      it's just integrating, saving costs in manufacturing(and needing to add new features to be better than the old product and really i don't think this is that big of a deal to incorporate)..

      --
      world was created 5 seconds before this post as it is.
  30. Re:(none) by Anonymous Coward · · Score: 0

    and while you dummies are doing
    dumb, someone is planing on puting
    a pyramid on earth ...
    -
    why abcdefghijklmnopqrstuvwxyz
    doesnt cut it anymore ...
    -
    we're lost def. lost. /me feels
    soo sad that all your stinking brains
    are just gonna rot away in worm
    stomaches... (your stinking comment here)

  31. real transcript by rootofevil · · Score: 2, Funny

    Customer:If I can have a 3 Ghz processor why cant I have WiFi at 3 Ghz
    Radio Shack Assistant: Oh, i dont know. I bet it does.
    Customer: Ok. good.

    --
    turn up the jukebox and tell me a lie
  32. Will make tracking easier too by nurb432 · · Score: 1

    If every cpu transmits its presence to the world ( and unique MAC address ) it will make monitoring ones whereabouts that much easier..

    Oh, and make the spectrum a total mess with all that noise....

    ( and just for the record, they were talking about this a year ago, but to discuss it you had to have an NDA )

    --
    ---- Booth was a patriot ----
  33. Monopolistic BS by drix · · Score: 2, Informative

    This just is another step in Intel's ploy to rule the wireless market through cheap and underhanded business practices. Not many people know this, or at least I didn't till I started shopping for a laptop 2 days ago, but all new laptops carrying the Centrino designation have to come with an Intel miniPCI WLAN card preinstalled or they cannot be called Centrino. Which is great except that Intel refuses to support Linux on their stinkin' card. (Yes I could go elsewhere, but for the price, speed, and power consumption, Centrino is far and away the best on the market right now.) If you want to monopolize an entire hardware sector, fine--good luck trying. But don't chain me to a stupid Wintel platform because of it. If Intel had their way they'd be the only supplier of WiFi cards within a few short years--then WTF do we do if we're not on Windows?

    --

    I think there is a world market for maybe five personal web logs.
    1. Re:Monopolistic BS by RdsArts · · Score: 1

      Intel "owning" the WiFi market would be great.

      Why? Because having only one chip-set makes it easier to figure out which target to choose to try and make your own drivers.

      The big problem right now is everyone switching chip sets left and right on their WiFi cards/devices. Sometimes the same card will have a different chip set based on the time of day you buy it, because each run will have a different, more software-based chipset and driver. It's a big old moving target.

      So if Intel can "own" the WiFi market legally, more power to them.

    2. Re:Monopolistic BS by akuma(x86) · · Score: 1

      Ummm...There are plenty of Wi-Fi alternatives. Intel by far does not have a monopoly on the Wi-Fi market. They only have a monopoly on the name Centrino.

  34. Ack by FrostedWheat · · Score: 4, Interesting

    But why!

    Onboard soundcards (chips?) are rubbish, onboard NICs are quite often crap (not always), onboard modems are a joke and onboard video is nasty. Apart from some specific cases (VIA's mini-itx stuff) I think manufacturers should be moving away from this onboard-everything obsession.

    PCI was invented for a reason! Customisability is what set the PC apart from the Amiga or similar machines!

    1. Re:Ack by antiMStroll · · Score: 2, Informative
      We use on-board ACL 650 sound for commercial use and it's fine, much better than being forced to load Creative's nasty software (circa Audigy 1, haven't tried the 2.) The motherboards we select use Intel integrated NICs, vastly superior to the typical Realtek-level junk found in so many machines. Modems, you have a point, integrated video however has its uses. The desktop demands in a corporate environment are typically meagre. Integrated handles the task fine and has the added benefits of reduced noise, heat and cost, plus increased reliability (no GPU fan) and generally better stability since the GPU and MB were designed as one.

      If AutoCAD or UT2003 aren't part of the design spec, motherboard integration makes a great deal of sense.

    2. Re:Ack by be-fan · · Score: 1

      Actually, that's not really true. The audio and NIC chips on NVIDIA's nForce is pretty good, and its not like I need anything more than the GeForce-class graphics processor on the nForce to run my xTerms :)

      --
      A deep unwavering belief is a sure sign you're missing something...
    3. Re:Ack by SharpFang · · Score: 1

      Think of this as of IrDA interface. Most of the chipset is on-board, and what is missing is one small chip, one resistor, some wire and optionally a box and a couple of capacitors. Nobody seems to be complaining though.

      --
      45 5F E1 04 22 CA 29 C4 93 3F 95 05 2B 79 2A B2
    4. Re:Ack by FrostedWheat · · Score: 1

      Think of this as of IrDA interface. Most of the chipset is on-board, and what is missing is one small chip, one resistor, some wire and optionally a box and a couple of capacitors. Nobody seems to be complaining though.

      Bluetooth would make more sense here, rather than 802.11.

    5. Re:Ack by man_ls · · Score: 1

      That realtek "junk" has a more versatile driver than pretty much anything else I've seen...Case in point:

      At a school (TSTC) we were given a single ethernet drop. Problem: It's keyed to a Widows 98 machine's MAC address that doesn't exist with us. We know the MAC that needs to be cloned.

      My laptop's Broadcom chipset can't do it.
      Another laptop's Intel chipset can't do it.
      A third laptop's 3Com can't do it. Neither could his Linksys.

      A fourth person's crappy Realtek chipset NIC was the only one that could do MAC cloning out of the batch.

    6. Re:Ack by CyberSlugGump · · Score: 1

      You can spoof the MAC address on Windows 2k/XP/2003address with the program SMAC:

      SMAC does not change the hardware burned-in MAC addresses. It is not necessary. SMAC changes the "software based" MAC addresses on the Windows 2000, XP, and 2003 Server systems, and the new MAC addresses you change will sustain from reboots.

      http://www.klcconsulting.net/smac/

      I've also seen a way to do it for Windows 98. I imagine it's even easier on Linux.

    7. Re:Ack by megabeck42 · · Score: 2, Interesting

      Frankly, that's a product of the driver not a result of chip design or flaws. Further, all Windows (NDIS?) drivers I've had to work with have had the option called "Locally Administered Address" where you can force a custom MAC address.

      If you take a look at most ethernet drivers, you'll notice that most drivers need to read a MAC address out of a serial eeprom. This address is then configured into the ethernet chip so it can receive appropriate unicast frames and is used by the driver when constructing ethernet frame headers for transmission.

      Further, regarding just how "junky" the Realtek stuff is, I refer you to this page: http://www.fefe.de/linuxeth/realtek.txt which states quite clearly and specifically why the realtek card is "junk."

      For the record, I can spoof the MAC address on my 3Com 3C905's, Intel EEPro100 and Broadcom Tigon 3.

      --
      fnord.
  35. Re:Unintentional network compromises through bridg by Anonymous Coward · · Score: 0

    "Trusted internal network" ... good joke, thanks, I needed a laugh.

  36. Re: integration by King_TJ · · Score: 4, Interesting

    Personally, I'm not really convinced all these integrated parts do us any favors.

    Just last week, for example, I installed a new Pentium 4 motherboard and CPU in a standard ATX case that was formerly running a PII system. This was done for a law firm, and was upgraded on-site, because they couldn't afford to have much downtime.

    Well, as luck would have it, the integrated EIDE controller was faulty. I kept getting "data corrupt" type messages when it tried to boot Win2K on the drive that just worked in the other system. I tried a different hard drive with a fresh format, and had the same issue. Even the secondary channel had problems.

    If it hadn't been intergated, I could have simply swapped a $15 or $20 controller card and gotten everything back up and running for them.

    The more devices Intel can integrate into motherboards using their chipsets, the more often they get to sell people an entire new board when they only need one small part.

    On-board video has been a disaster since day 1, for both PC and Mac users. What seems "high end" when a machine is new turns into "mediocre" within a year or two. Then come all the conflicts trying to get the on-board video disabled when you add a new, add-in video card. (I'm sure many long-time Mac users can remember the dislike for the "Performa" towers like the 6400/6500, largely due to the on-board video only allowing up to 2MB of video RAM.)

    Integrated NICs may work fine when they work, but again - I've seen many a blown NIC card due to power surges/spikes. I'd rather swap a card and have a fully functional machine again than have a dead port permanently soldered onto the back of my computer....

  37. apparently by the_2nd_coming · · Score: 1

    Intel is still a bit worried about the allegations of the Lone Gunmen.

    --



    I am the Alpha and the Omega-3
  38. security by spoonist · · Score: 3, Interesting

    I, for one, do not like this trend of integrating wireless into everything.

    As a security conscious individual, I want to be able to physically choose whether or not I want wireless when I want wireless.

    I like to be able to physically pull out the wireless card in my laptop because then I know I can't be h4x0r3d via my WLAN card.

    Fine. Call me paranoid. I don't mind.

    (Yeah, I know they said the RF part would still be an add-on... I'm just talking in general that I want add-ons and not fully integrated wireless stuff that I can't pull out without desoldering chips.)

  39. Centrino is NOT the price/performance/powr leader! by MarcQuadra · · Score: 1

    I totally disagree that Centrino is the 'best' mobile chipset. Transmeta's offerings are way too slow for my taste, AMD CPUs draw gobs of power, and Centrino is 'black boxed' to the Linux community.

    The 'best' mobile CPU in my opinion is available from Apple, as either a PPC750 or a PPC74xx. You really can't beat the iBook line in terms of price/performance/quality. Sure, the clock speed is a bit low, but even the G3 has tons of horsepower. A 900MHz iBook running Linux feels about as fast to me as a 2GHz Centrino.

    --
    "Sometimes, I think Trent just needs a cup of hot chocolate and a blankie." -Tori Amos on Nine Inch Nails
  40. You shoulda seen Otellini by Anonymous Coward · · Score: 0

    Intel has said it eventually intends build a Wi-Fi radio into its microprocessors.

    You should have seen Paul Otellini when he explained how easy this would be (April/May of last year? can't recall exactly). He looked like he was about to wet his pants.

  41. Dear God, would you people please think? by gmhowell · · Score: 3, Interesting

    Would you people please think before posting?

    This includes a chipset, not a radio. Therefore, it won't be sending out your world control schemes to everyone in existence. Yes, Intel will at some point in the nebulous future include a radio. As will many manufacturers. At that point, we go to the next paragraph:

    Every integrated soundcard/videocard/ethernet controller/serial port/etc. I've ever seen has a setting in the BIOS. If you don't want the location of your laser embedded sharks known to the black helicopter people, switch it off.

    Finally, when is the last time your built in ethernet card just randomly spewed data out the port to the CIA? Oh, last week? Then you have more problems than just a wireless AP built into your motherboard.

    --
    Jesus was all right but his disciples were thick and ordinary. -John Lennon
  42. Risk. by SharpFang · · Score: 1

    This would make setting up a wireless network a lot simpler."

    As in "The cops are at the doors! This is a set-up!" ?

    Say, I want a box that's remotely uncrackable. Nothing simpler, remove all network cards. But Intel is known from claiming its CPU 'features' can't be re-enabled without reboot, while they can. So I have a potentially harmful piece of hardware in my box...

    --
    45 5F E1 04 22 CA 29 C4 93 3F 95 05 2B 79 2A B2
  43. Re:If it's actually on the motherboard by YouHaveSnail · · Score: 1

    Will it not need some kind of cooling to prevent it overheating?

    Since they can't seem to prevent their processors from emitting lots of energy, they might as well emit it at a useful frequency. Once they add the radio to the chip, it will actually function as a cooling device itself. When the chip gets too hot, you just transmit more packets. ;-)

  44. Re:Here is a picture of wi-fi chip... by Anonymous Coward · · Score: 0

    First time? Damn, [whistles] you must be new around here ;)

  45. you mean like airport from day 1? by jpellino · · Score: 1

    just add the radio card to any os9/X than can (slot loading iMacs on up, iBooks on up) and press 'share'
    do all the protective stuff of course...

    --
    "Win treats sysadmins better than users. Mac treats users better than sysadmins. Linux treats everyone like sysadmins."
  46. This could be fun... by AetherBurner · · Score: 0

    I wonder what would happen when you give the chipset a specific channel to use and it decides to use the old pentium math section and approximates the transmit frequency - all the computers start talking on a Fed's channel...too funny. Now I guess Intel wants to implement planned obsolescence in the Wi-Fi market. My Orinoco cards work just fine thank you. One other idea comes to mind: Microsoft integrates web browser into Windows -> Intel integrates WiFi into their chips. I smell a parallel here...

  47. Wintel are both right. by Anonymous Coward · · Score: 0

    Check out a Computer Structure and organization book. Anything in hardware can be done in softwear and vise versa.

    Do you lament the fact that all modern chips hae an FPU? Builtin memory management? Altivec?

    Intel will eventually want to put the graphics card, and sound card in the CPU too. Think of the peak bandwidth!

    1. Re:Wintel are both right. by bhtooefr · · Score: 2, Informative

      http://slashdot.org/articles/99/11/25/094246.shtml

      Actually, they already tried that. However, their Israel design team had it almost finished when they pulled the plug, and they were PISSED. Ah, well, they were happy when they got the Pentium M project - what the P4 shoulda been!

      BTW, have you played with a Cyrix MediaGX? If so, then you should be modded funny. If not, they need to make a mod for stupid - Cyrix's implementation SUCKED - 44-50MHz CPU speed was lost, and the video was worse than i810.

    2. Re:Wintel are both right. by Anonymous Coward · · Score: 0

      Obviouly if it can't be done now it can never be done, and will never be done. Thank you for your insight.

      I for one welcome trolling our new luddite overloards.

    3. Re:Wintel are both right. by bhtooefr · · Score: 1

      I'm just saying that it sucked the first time someone tried, and the second time Intel burned itself. BTW, Timna was also Intel's first CPU with an on-board memory controller in a LONG time. Their official stance is now to never have one, which leaves AMD (I don't think VIA will repeat Cyrix's mistake - not with the direction they're headed) to screw up again - any bets that they won't?

    4. Re:Wintel are both right. by Anonymous Coward · · Score: 0

      As the pressure to increase clockrates grows, the distances on the motherboards get longer and longer. As features get ever smaller, and silicon waffers ever larger it's cheaper and cheaper to put more eggs in one basket. It's the nature of the beast.

      The first couple of large scale experiments failed, more or less, and so far. But just as chipmakers know they'll eventually be leaving even extreme UV lithography completely in the dust, the too know that chips are only going to pick up more functions. It's just too cheap.

      The bind there in, everyone is in, isn't just finding the right time to make those specific leaps, but to make sure they're not passing up better more subtle fuctions to integrate in the nearer term. Will every CPU come with an embedded GPU in three years? No. Almost certainly none will, aside from niche markets. Five years, still probably not. Ten? Maybe, maybe not. Twenty? I'd say we're approaching metaphysical certainy.

      Look at the strides Motorola and Intel have made in just the past decade, just what's in stores. Forget distributed clocks, their high end chips, what's in their labs, and even the slow but impressive gains in quantum computing. Hell, how fast was the cache on the computer you were using in 1992, and did you have more that 256k?

      Nearly every success was, is, and will be preceeded by failure (occasionally spectacular) I see no reason this incremental process should be considered somehow special. After all, the Neanderthals might have been great apes that walked upright, had opposable thumps, language, trade, and could throw spears, but clearly their performance isn't an accurate indicator of our success.

    5. Re:Wintel are both right. by bhtooefr · · Score: 1

      What cache? I was using a WDC 65C02 @ 1MHz w/o a cache, you insensitive clod! (I didn't get my first PC until 1994, and it was a 386SX/16MHz w/o a 387. 1997 was when I got both a 386DX/33 w/o 387 and a Cyrix MediaGX/166)

  48. Re:Centrino is NOT the price/performance/powr lead by O · · Score: 1

    And there's still no support for AiportExtreme on Linux, greatly reducing the utility of Linux on that iBook.

    --

    1, 1, 2, 3, 5, 8, 13, 21 -- Mathematics is the Language of Nature.
  49. Why wireless? by wideBlueSkies · · Score: 1

    I don't get it.

    I'm happy with my hardwired LAN. Why do I need to think about wireless hardware being in my setup by default? I'm assuming that in this situation I'd have to do something to disable it so I can plug in a good old fashined NIC card. Right?

    This is troubling. OK, maybe putting it on the MOBO isn't so bad, but on the CPU? Why? This is good in what way?

    wbs.

    --
    Huh?
  50. Re:Hospitals by Jarrik · · Score: 3, Interesting

    I work in a fairly large hospital (200+ beds) and I can tell you from alot of experance that this is no longer the case for hospitals with modern equipment. The hospital I work for just had Nextel come in and install signal multiple repeaters on every floor and building of our entire campus and put a cell tower on top of our main building that houses all patent rooms / OR's with no problems what so ever. There are certain areas where we are advised not to have our Cell phones on but that is due more to respect of patent famlies and not becuase of any interferance issues.

  51. How about AM/FM on a PC? by tjstork · · Score: 1, Interesting

    You can talk to someone in Bangladesh but can't get the football game commentary on WIP? What's up with that!

    --
    This is my sig.
  52. but... by sbma44 · · Score: 2, Informative
    you probably *could* have swapped in an ide controller board and just turned off the faulty integrated controller in the bios.

    I agree that we should not sacrifice modularity for all-in-one disposability, but for all the applications you list (IDE, NIC, video) you can put in a modular card and override the integrated stuff. Personally, I think ubiquitous integrated mobo NICs are one of the handiest hardware improvements of the last five years.

  53. Re:Centrino is NOT the price/performance/powr lead by be-fan · · Score: 1

    Your perception is seriously off. First, there is no 2GHz Centrino. The fastest one is 1.7Ghz. Second, a Centrino is *extremely* fast clock-for-clock. Its even faster per-clock than a PIII. A 1.7GHz Centrino is about equal to a 2.6 GHz P4, or a hypothetical 1.9 GHz PIII.

    "tons" is not a precise metric of computing power. Relatively, a G3 has very little horsepower. Its got a pretty crappy FPU, and isn't that much faster clock-for-clock than a PIII anyway. A 900Mhz iBook is probably comparable to a 900MHz or 1GHz Centrino.

    --
    A deep unwavering belief is a sure sign you're missing something...
  54. The real problem is routing by Baldrson · · Score: 2, Informative
    Having set up a municipal mesh network it is pretty obvious the real problem to solve for WiFi isn't higher levels of integration so much as routing.

    Internet protocols were designed around wires and it shows when you go to wifi meshes. Meshes are critical due to the fact that meshes scale. If you are going to have a wifi node in every consumer device, as seems potentially viable, then you need to continually discover new routes and do so on nearly every packet. Route-flap is what you get, even with damping protocols, with current internet standards. You can end up waiting minutes for a route to stablize.

    Here's an algorithm for a mesh node that seems to work simply:

    Just keep a table of destination IP addresses in memory with a counter that decays exponentially with time.

    When the counter decays below some threshold, clip its IP address from the list. An IP address with no counter is considered to have a value of 0.

    Every time a packet acknowledgement comes through for a given destination IP address, add one to the counter for that IP address.

    Whenever a packet (not already awaiting acknowledgement) is 'heard' destined for an IP address, queue it for rebroadcast according to a priority established by the IP address's counter.

    Let packets that fall off the end of the queue due to low priority do so without further consideration.

    More complex algorithms are required for transmission power optimization, but even this simple algorithm shows how far off-base current internet protocols are for wifi.

    1. Re:The real problem is routing by sploxx · · Score: 2, Informative

      In IPV4, you can't allocate space for WiFi meshes anymore. ACK.

      But this is not true for IPV6. In IPV6 you can allocate big classless spaces and use these for your mesh networks. A 10 million city with one IP per PC, washing mashine, toaster... no problem.

      What you describe at the most important problem in todays IP world is not the basic protocol, i.e. IP. You are criticizing the routing protocols, they do not match well with fluctuating mesh networks. Indeed. They are designed for wired networks. But most of the internet traffic still runs over wires, and this will probably stay so even with competion from wifi. You can't really put all that bandwidth through wifi.

      There are solutions that work on top of IP, in a very clean way, and these solutions work well in the lab. In the field, there are not yet enough studies which support this research.
      The proposed new network architectures consist mainly of new routing protocols adapted to mesh networks. I see no problem with IP as the underlying protocol. It works, it's relatively reliable and it's already there. And the research groups (Are you professionally working on this topic? - I'm not. But this doesn't invalidate my arguments, I think...) are using IP on the lower level...

      For more information, try google for "manet". There is already much information out there.

      I can't really talk about you routing algorithm because I haven't seen it in simulation or reality. To state that a particular routing algorithm is good by just drawing it onto paper is a good start, but not a proof of concept. Maybe it works, maybe not. But try it out and build devices which connect to a mesh network (Mesh networking is one of the things I consider to play around with next in my freetime) and show the world that it works. But I'm sure that the underlying packets can IPV6 instead of your custom protocol without any functional penalty.

    2. Re:The real problem is routing by Baldrson · · Score: 2, Interesting
      But this is not true for IPV6. In IPV6 you can allocate big classless spaces and use these for your mesh networks. A 10 million city with one IP per PC, washing mashine, toaster... no problem.

      Well, yeah, and I did try to get Jon Anderson to include IPv6 in his first major release of LocustWorld's MeshAP for precisely that reason. To no avail however. He thinks the ability to set up private subnets, local to the mesh, is adequate to the problem at hand. He may be right if the problem at hand is simply getting out to a lot of MANs rapidly.

      As for my professional involvement, I am responsible for deployment and mantainence of a mesh network in a small metro area so yes to that extent I'm professionally involved. I'll probably do a computer simulation of the described algorithm before building any actual devices but yes testing of these ideas is important.

    3. Re:The real problem is routing by sploxx · · Score: 1

      Then... if you know what your problems are (i.e. no IPV6 in MeshAP) and you need it, why don't you fix it?
      I mean... it's certainly harder to get a new routing algorithm (i.e. a considerable piece of rather complex software) working than change the packet type from IPV4 to IPV6...

    4. Re:The real problem is routing by Baldrson · · Score: 1
      ? I don't follow.

      I said I thought IPv6 was a good idea, not that it was an alternate solution to the routing problem. The problems I'm running into now aren't lack of IP addresses, it is route flap.

  55. Re: integration by madmancarman · · Score: 1
    Integrated NICs may work fine when they work, but again - I've seen many a blown NIC card due to power surges/spikes. I'd rather swap a card and have a fully functional machine again than have a dead port permanently soldered onto the back of my computer....

    In a computer lab where I teach, we had 25 Toshiba Equiums with integrated NICs, and in the two years we had them in that lab, we probably sent back 4 or 5 motherboards with fried integrated NICs. Sometimes we could get them to limp along with an ISA or PCI NIC, but other times Windows would refuse to remember that we'd disabled the dead integrated NIC, so on restart it would try to reinstall the drivers.

    To be on-topic, I don't think this is likely to happen with intergrated wireless; I just question its inclusion at a time when people can't secure their own wireless access points.

    --
    First they ignore you, then they laugh at you, then they fight you, then you win. -- Gandhi
  56. Re: integration by GigsVT · · Score: 1

    Integrated video is nice for servers. It's nearly impossible to buy a cheap video card that doesn't have 3D anymore. My hardware buying coworker recently put a ATI 7000 series into a server, because it was the cheapest thing he could find!

    When you don't even need graphics at all, onboard video works great, and doesn't waste a PCI slot on something you don't really care about.

    Integrated NIC... Who cares? Again, it saves a PCI slot, and if you need to stick a new one in, they cost $10. Squirt hot glue into the old port if you are afraid of a braindead tech (or yourself) accidentally trying to use the dead one.

    Integrated sound, I'm more inclined to agree with you about. On board sound is usually kind of shoddy, but sufficient for most people.

    It seems your primary bitch is when it's difficult to disable on-board stuff. The only thing that I've ran into recently with that was "shared ram" on-board video that couldn't be totally disabled, so it still took up some ram even though it wasn't doing anything.

    That's not really a problem inherent with on-board stuff, that's just crappiness on the part of the motherboard assembler.

    --
    I've had enough abrasive sigs. Kittens are cute and fuzzy.
  57. Mobo features by t0ny · · Score: 2, Informative

    It will probably be easy for mobo makers to make use of this as an optional feature, much like SATA, USB 2.0, etc. Then you just need to have an antenna lead from the mobo, and enable it in the BIOS: it will work just like parallel/serial ports.

    --

    Manipulate the moderator system! Mod someone as "overrated" today.

  58. Re:Unintentional network compromises through bridg by drinkypoo · · Score: 2, Interesting
    The network name will probably be set to 'intel' and no encryption key set, just like linksys hardware comes up as 'linksys' with no WEP key. Windows configures new interfaces for DHCP by default. Hence if someone sat outside your office with a WAP (or a pc with a wifi nic acting like a WAP) providing dns service, with the network name set to 'intel' (in my example, anyway) then they could assign IPs to your machines and begin hacking.

    The solution is simple, however; Don't buy the systems with wifi in them, or be sure to configure them with a software image, unattended install, or similar feature to configure the card out of the box, or disable it.

    --
    "You're right," Fisheye says. "I should have set it on 'whip' or 'chop.'"
  59. when it's all integrated, it'll be easy by Anonymous Coward · · Score: 0

    > why wouldn't any wifi enabled anything be able to do it?

    When a card, motherboard, and cpu are involved, the anti-privacy plan needs all card, motherboard, and cpu manufacturers to agree to make the hardware do this. When it's all on the CPU, only intel have to be convinced to do it.

    This is a good reason for buying AMD. Supporting a monopoly is not good for the consumer.

    (note that AMD may also do this. Market pressure might force them to do it, if intel do it, and M$ require it for M$ Windows, AMD would have to comply. But I still boycot intel, to prevent or lessen this monopoly problem in the first place.)

  60. I Agree by w42w42 · · Score: 1

    I agree with your sentiment. My question, and I'm not an experty by any means on WiFi, is this. If a new standard comes out, say 802.11musthave, do I suddenly have to buy a new motherboard (+ CPU/Memory/etc) to use that? It'd be awfully convenient for Intel now, wouldn't it.....

    I think that WiFi has gone through more upgrades lately that processor architectures, and perhaps Intel is looking for another upgrade gravy train.

  61. Re:Bluetooth? Apple! by Anonymous Coward · · Score: 0

    Apple's current PowerBook line comes with Bluetooth built-in out of the box, and it's iBook line (and every other product, except for the eMac) can have it added at the factory for $50.

    Just another reason to go Mac.

  62. Intel trying to create life? by Anonymous Coward · · Score: 0

    CPU's talking to each other over WiFi?

    MOBO's just to supply power and numerous empty sockets to add more CPU's as needed?

    Parallel processing?

  63. Good bye air gap security by Anonymous Coward · · Score: 0

    Hello NSA virus to activate built in wi-fi.

  64. Re: integration by Anonymous Coward · · Score: 0

    Wait a minute - you bought a new motherboard that was defective and you wanted to stick it into production anyway?

    The IDE controller is deeply integrated into modern chipsets. If it's funky, it's very likely something else isn't working also.

  65. Tell them you're writing windows drivers ... by konmaskisin · · Score: 1

    ...

    Why tell Intel the truth?

    Why not borrow documentation from a Windows software engineer who accidentally photocopies it at work or loses his/her backup copy while he/she is at your house for dinner.

    1. Re:Tell them you're writing windows drivers ... by gbjbaanb · · Score: 1

      because that Windows developer will have signed an NDA, and although losing it isn't a legal problem (unless it can be shown he was truly negligent), once you start using it, you will be charged with IP theft.

      I suppose you could stick it on Kazaa and have done with it though! :)

  66. Intel "Linux Support"? by konmaskisin · · Score: 1

    It's true that Intel Linux support sux and it's true that the government (which is full of idiots - think Ashcroft) thinks it shoul be "illegal" and that all large established vendors (with the possible exception of IBM) generally despise Linux (BTW they are going to be **seriously** burned - in the same way the U.S. automobile industry was by the Japanese - by Asian manufacturers of Linux devices and computers), but you are wrong on one point:

    Intel hardware and support sucks in general - not just for Linux. Intel Pro 2100 network card is a good card "except in Linux" or the 810 is a good video card except in X. No, it is expensive *crap*. Young teenage entrepreneurs in Taiwan are making better cards ... and in mainland china Intel ... well ...

    Microsoft will decline in market share and lose stock value (it's been out of whack for years 200/1 P/E nearly a trillion in capitalization and 95% market share can't last for ever folks) but they will be all right and will buy Disney or something and keep on trucking ...

    What I for one will really watching, and what I really hope happens soon, is the long slow tortuous death by stupidity and cupidity of Intel. It's going to be so surprising to the media so coverage will be silly but still it will be beautiful to watch them go down in flames.

  67. Re: intergated IDE, etc. by King_TJ · · Score: 1

    Well, I did think to try adding an IDE controller card - but the card I added didn't allow booting from CD, which I needed to get Windows 2000 or XP loaded on the box. (The existing install of 2000 refused to start up after the motherboard was swapped, since the incorrect HAL was on the drive.)

    My point is, PC's are designed to be a "box of slots". This concept is what initally made them dominate the marketplace. (If you didn't buy a PC or clone, it used to be, your other choices were systems that gave you a specific set of features which were largely non-upgradable. You wanted an upgrade? You sold the whole computer and got a newer model.) Now, we're going back to the "old model", where the PC has everything on-board, and most of the expansions slots are sacrificed in the name of "saving space".

    Micro ATX boards with everything integrated have their place - but it's not appropriate for the standard PC desktop, IMHO.

  68. Re: intergated IDE, etc. by gbjbaanb · · Score: 1

    Well, I did think to try adding an IDE controller card - but the card I added didn't allow booting from CD

    I think you've answered your own question about why mobo manufacturer's supply everything integrated on board. In most cases you can still disable the integrated feature (like sound graphics, lan, usb) and add your own card.

    So really apart from cost, there is no reason why its a problem. and the cost of mobos nowadays is really low...

  69. Re:Bluetooth? Apple! by General+Sherman · · Score: 1

    I know you're sort of half trolling, half being a mac promoter, and actually not lying, but why post as an AC? Is it really that bad to get a single -1 on your Karma (which you aren't even getting anyways).

    Grow some balls, man.

    --
    - Sherman
  70. Re:Unintentional network compromises through bridg by ForestGrump · · Score: 1

    No wait!
    If these machines have the default "intel" ssid.
    well, theyre blank machines anyway.
    Fine. make an image of the hdd. you won't get anything good anyway cuz the machine is new and has no sensitive data *yet*

    -Grump

    --
    Is it true that more people vote for the winner of American Idol, than vote for the president? -Ali G.
  71. Re: integration by nustar1 · · Score: 1

    I belive that 99.9 percent of the time integrated components work just fine. In the case of your law firm you could have put in a 3rd party ide card. And concerns of not being bootable, why not just use a bootdisk?

    Integration of components used to be a pain in the rear, but hardware has gotten better, and so have the drivers. I bought a Gigabyte mobo for about 130 earlier this year. It has integrated nic and audio including the other standard stuff. Should I have to buy a 100 mobo, then lay out thirty or fourty bucks more per part for extra cards?

    As for dead ports on the back of the machine, I imagine that you probably have open PCI slots on machines that are just dying to be used.
    Give them a shot.

    Integration is not a bad thing if it is implemented properly. But the devil is in the details on that one. 810 chipset video is nothing to write home about, but hey buyer beware!

    Honestly most people don't care about that stuff, they just want a good cheap computer, and if it came to saving 5 or 10 bucks on a port that they are never going to use, they would rather save the money.

  72. Centrino? by smart.id · · Score: 1

    Excuse me, but I thought this is what Centrino was. Can someone explain this to me?

    --
    blog & fiction: jd87
    1. Re:Centrino? by Hoser+McMoose · · Score: 1

      First and foremost, "Centrino" is a 100% pure marketing term, nothing at all new or technical behind it. All Centrino means is that you havea laptop that has an Intel Pentium-M processor with an Intel motherboard chipset and an Intel wireless chip. If you use a Pentium-M processor and Intel motherboard chipset but a non-Intel wireless chipset (eg one of the Broadcom chips that include 802.11g support which is lacking in the Intel chip), you can't call your laptop a "Centrino" laptop and more importantly, you don't get the millions of dollars in advertising money from Intel.

      Intel is just planning on integrating the motherboard chipset and the wireless chip onto a single piece of silicon. This has been the plan all along for their notebook chipsets, but now it looks like they're also planning on integrating the wireless chip into their motherboard chipsets for desktops as well.

  73. Re:Centrino is NOT the price/performance/powr lead by MarcQuadra · · Score: 1

    I think a G3 has more horsepower per-cycle than a PIII, I seem to compile things in less time on my G3/450 than on my PIII/600. Of course this is pretty subjective, but most other things seem faster on the G3 too.

    The G3 has much better integer performance than a PIII of similar clocking, and FPU scores are neck-and-neck. I don't know about you, but I tend to make use of the integer units about 100 times more often than the FPU units.

    I think our 'feel' for instructions-per-clock has been whacked by the P4, which does amazingly little per-cycle. The PIII was a hot slow CPU (for it's clock), and the P4 is even worse.

    --
    "Sometimes, I think Trent just needs a cup of hot chocolate and a blankie." -Tori Amos on Nine Inch Nails
  74. Why is this on Slashdot? by jwr · · Score: 2, Interesting

    Why do I read about this on Slashdot, instead of an article about Intel declining to provide drivers or specifications for their wireless part of Centrino?

  75. Re:Centrino is NOT the price/performance/powr lead by Dahan · · Score: 1

    The CPU in Centrino machines is a Pentium M, which is basically a souped-up Pentium III (includes SSE2 instructions, for example). It is not in the same family as the Pentium 4. I got a Centrino laptop for my dad (Dell Inspiron 500m), and while I was setting it up for him, netbooted NetBSD and ran some benchmarks and compiles to get a feel for the performance. I was impressed by the speed and battery life; it's a bit faster than the PIII per clock, but has much lower power consumption. I had expected it to be in the P4 family, with the associated low instructions-per-cycle, but was pleasantly surprised. I wouldn't mind having a desktop machine with a Pentium M processor... I'm not a gamer and don't need a superfast CPU; I'd be happy with a slower CPU that didn't need to run its fan all the time.

  76. Turned off by default in the BIOS? by screwballicus · · Score: 1

    To respond to those who are making the slippery slope contention, I respond, so what. The slippery slope argument seems to be...

    "what if in the near future, are new computers boot with an active wifi ap built in!"

    That's obviously not dealing with the issue at hand, which is a wifi chipset, but without the necessary hardware built in, but it seems to be popular, so to address that...

    There seems to be at least one fairly simple solution to this. So incredibly simple that I feel silly proposing it, but,

    Why is it oh so insecure for our new computers, hypothetically, to boot with integrated wifi...by default disabled in the BIOS? What danger is there? If Grandma before booting her computer plays with the jumpers on her motherboard or goes into her BIOS to change the setting, she'll be open to haxors?

    Yeah, wifi is in most cases inherently less secure than plain ol' RJ45, but I tend to think disabled but integrated wifi is a pretty slim threat. Your computer's open to hacking if someone gets into your office and changes a setting in your BIOS. If malicious hackers are fiddling around in your BIOS, you've got bigger problems.

  77. How badly does intel want chip ID? by leehwtsohg · · Score: 1

    It seems chip ID didn't work te first time. People were too paranoid. Others were saing that you have an ID anyway, because you have one on the ethernet card.

    Hey! So why don't we just put the card on the chip? No one can argue with that!

    But it could also be that they just want a monopoly over wireless... but why? why not put the sound on the chip? Or the graphics chip? Is it really that much more energy efficient to have the wireless card on chip, but it doesn't help much for graphics? Anyone?

  78. IT SHOULD BE MODDED DOWN BUT AS REDUNDANT by Anonymous Coward · · Score: 0

    Slashbots are already experts at giving head.

  79. I call BS on this one by Anonymous Coward · · Score: 0

    I let them ream my tight asshole in return.

    Kathleen.


    This has to be BS. Tight asshole? Fent's asshole is bigger than the goatse guy's

  80. Linux is succesful because.... by LnxAddct · · Score: 1

    Linux is successful because it is modular. It is what helps it make it so great(besides the awesome community behind it). If Intel starts integrating everything then they are going the way that Microsoft has gone. Everything must be upgraded at once and when one thing breaks, everything must be replaced. Companies should learn from linux and make their products modular. It was great when you could just pull out a burnt card and get a 15 dollar replacment for it. The way wireless is going, I can't see a single standard being around for long. We'll be at 802.11z at 1000gbs before you know it. (That is somewhat of an exageration) My point is that, if i bought a 802.11g compatible motherboard, in 2 or 3 years g won't be the standard anymore. Joe Linuxhacker may get around this with his '1337' skillz, but Joe Schmo and Joe Corporate will most likely feel forced to upgrade. This move is partly intel making its customers life easier, but partly ensuring a future income for intel.My company could probably use its current computers for 10 more years. They used their last ones for at least 6. This integration type stuff I feel is trying to stop that and accelerate upgrades. Its much harder to fix a broken integrated board. Just my 2 cents.

  81. Integrated Hard Drives?... by LnxAddct · · Score: 1

    How long until hard drives are integrated? All this integration is starting to get a little ridiculous. Microsoft is doing away with Independant Software Developers (i.e. they are coming out with their own AV, and they now have integrated zip utils and cd burning utils), and Intel is doing away with Integrated Hardware Developers. When will the madness stop?

  82. Re:Centrino is NOT the price/performance/powr lead by be-fan · · Score: 1

    Mine's not a matter of feel. Benchmarks show that a G3 is only a bit faster than a PIII clock-for-clock. This was true back when Boot compared a 300MHz G3 to a 300MHz PII (same core architecture as the PIII) and its true today. The Pentium-M is significantly faster than the PIII clock for clock. Either way, a 900 MHz will lose very badly to a 1.7GHz P-M.

    --
    A deep unwavering belief is a sure sign you're missing something...
  83. Re:Hospitals by ericspinder · · Score: 1
    My infant son has had a couple of problems, so I have spent a fair amount of time in the hospital over the last year and a half. Both hospitals had signs out front saying "Turn off Cell Phones". In both places I was told where to use my phone from staff directly and they gave me "interference" as the reason for not using it anywhere near the equipment (I asked, nicely). It might fall into the catagory of an abundance of caution, perhaps they haven't yet "certified" all of their equipment.

    Clearly your workplace is using the Nextel cellular system for your voice (and maybe data) communications, and I am sure that your workplace has taken great care with the deployment. However, I do wonder how it all will hold up when everything is broadcasting radio signals, cellphones, PDAs, desktops, heart monitors, baby monitors (everything with a processor) etc.; not to mention tracking tags for beds, lunch trays, wrist bands, and everything else that get wheeled about the place. I am sure that questions like that won't escape the over cautious. Don't get me wrong, I love wireless, but adding features to the motherboard before some are willing to adapt (or even live with), might not be the best idea. (Grain of salt) I am sure that Intel will have a way of turning it off in BIOS for people who don't/ can't use it.

    BTW, my son is is doing ok and do have a great deal of respect for the fine Nurses and Doctors who have cared for him.

    --
    The grass is only greener, if you don't take care of your own lawn.
  84. Re:Hospitals by Anonymous Coward · · Score: 0

    BTW, my son is is doing ok and do have a great deal of respect for the fine Nurses and Doctors who have cared for him.

    That is very good to hear. Seems alot of people who are priveledged enough to have it take for granted their health care... glad to see someone who has the extreme respect - and success - for their nurses and doctors as you do.

  85. High Altitude Seg Faults - The Plusser Effect by Anonymous Coward · · Score: 0

    [bullshit=on] As somebody in the know, [credibility=off] I do worry that these new WiFi enabled equipment could be the next mobile phone when it come to interference of avionic systems; especially as many modern microprocessors are prone to [hang=on] soft faults at altitude due to the effects of the upper atmosphere. [[ insert rim shot here ]]

    Earth to Plusser... WARNING ... You have entered the Bogosity Zone... and your mom says it's time to come in now.

  86. Approved portable electronic devices? by Watchman_ds · · Score: 1

    I was on a flight last week where they specifically mentioned that Centrino laptops must be kept off during the flight.

    So if there is integrated wi-fi on the motherboard, will that be the end of laptops on airplanes? Will I have to be content playing the gameboy version of Splinter Cell? :(

    --
    Sigs are for lusers. Hey! wait a second...
  87. Please accept my humble appologies. by Anonymous Coward · · Score: 0

    I'll adjust my Karma down appropriatly.