Slashdot Mirror
Apple Clarifies 802.11g Controversy
Wireless Spider writes "A couple of days ago there was a controversy over the 802.11g data rates and supposed changes in IEEE specification. Apple has clarified this controversy, stating that nothing has changed in the spec. It seems the article from Computerworld was somewhat misleading. Quote from an Apple Vice President: "802.11g is still a 54Mbit/sec standard," Bell told MacCentral. "802.11b is 11Mbit/sec, but your actual throughput is somewhere between 4 and 5-1/2Mbit/sec. The number that's quoted is the data rate that's used between the radios (raw data rate, which includes the protocols etc.)" After reading this article featured on Macworld, 802.11g transfer rate controversy meaningless, says Apple, it seems clear that the people at Computerworld didn't do their homework for the article featured on May 22. Also, there seems to be a lot of politics between 802.11g and a supporters, and that every article posted on the Internet about this subject might not be true, or could be politically motivated."
Wow, if this isn't news for nerds I don't know what is.
I mean, good fucking lord.
I just, 5 minutes before this article popped up, showed a friend of mine the previous slashdot article saying that 802.11g's 54mbps is not-so. Damn contradictory news services! *shakes fist*
You're right, I wouldn't steal a car. But if it were possible, I sure as hell would download one!
2.5 million B.C.: OOG the Open Source Caveman develops the axe and releases it under the GPL. The axe quickly gains popularity as a means of crushing moderators' heads.
100,000 B.C.: Man domesticates the AIBO.
10,000 B.C.: Civilization begins when early farmers first learn to cultivate hot grits.
3000 B.C.: Sumerians develop a primitive cuneiform perl script.
2920 B.C.: A legendary flood sweeps Slashdot, filling up a Borland / Inprise story with hundreds of offtopic posts.
1750 B.C.: Hammurabi, a Mesopotamian king, codifies the first EULA.
490 B.C.: Greek city-states unite to defeat the Persians. ESR triumphantly proclaims that the Greeks "get it".
399 B.C.: Socrates is convicted of impiety. Despite the efforts of freesocrates.com, he is forced to kill himself by drinking hemlock.
336 B.C.: Fat-Time Charlie becomes King of Macedonia and conquers Persia.
4 B.C.: Following the Star (as in hot young actress) of Bethelem, wise men travel from far away to troll for baby Jesus.
A.D. 476: The Roman Empire BSODs.
A.D. 610: The Glorious MEEPT!! founds Islam after receiving a revelation from God. Following his disappearance from Slashdot in 632, a succession dispute results in the emergence of two troll factions: the Pythonni and the Perliites.
A.D. 800: Charlemagne conquers nearly all of Germany, only to be acquired by andover.net.
A.D. 874: Linus the Red discovers Iceland.
A.D. 1000: The epic of the Beowulf Cluster is written down. It is the first English epic poem.
A.D. 1095: Pope Bruce II calls for a crusade against the Turks when it is revealed they are violating the GPL. Later investigation reveals that Pope Bruce II had not yet contacted the Turks before calling for the crusade.
A.D. 1215: Bowing to pressure to open-source the British government, King John signs the Magna Carta, limiting the British monarchy's power. ESR triumphantly proclaims that the British monarchy "gets it".
A.D. 1348: The ILOVEYOU virus kills over half the population of Europe. (The other half was not using Outlook.)
A.D. 1420: Johann Gutenberg invents the printing press. He is immediately sued by monks claiming that the technology will promote the copying of hand-transcribed books, thus violating the church's intellectual property.
A.D. 1429: Natalie Portman of Arc gathers an army of Slashdot trolls to do battle with the moderators. She is eventually tried as a heretic and stoned (as in petrified).
A.D. 1478: The Catholic Church partners with doubleclick.net to launch the Spanish Inquisition.
A.D. 1492: Christopher Columbus arrives in what he believes to be "India", but which RMS informs him is actually "GNU/India".
A.D. 1508-12: Michaelengelo attempts to paint the Sistine Chapel ceiling with ASCII art, only to have his plan thwarted by the "Lameness Filter."
A.D. 1517: Martin Luther nails his 95 Theses to the church door and is promptly moderated down to (-1, Flamebait).
A.D. 1553: "Bloody" Mary ascends the throne of England and begins an infamous crusade against Protestants. ESR eats his words. A.D. 1588: The "IF I EVER MEET YOU, I WILL KICK YOUR ASS" guy meets the Spanish Armada.
A.D. 1603: Tokugawa Ieyasu unites the feuding pancake-eating ninjas of Japan.
A.D. 1611: Mattel adds Galileo Galilei to its CyberPatrol block list for proposing that the Earth revolves around the sun.
A.D. 1688: In the so-called "Glorious Revolution", King James II is bloodlessly forced out of power and flees to France. ESR again triumphantly proclaims that the British monarchy "gets it".
A.D. 1692: Anti-GIF hysteria in the New World comes to a head in the infamous "Salem GIF Trials", in which 20 alleged GIFs are burned at the stake. Later investigation reveals that mayn of the supposed GIFs were actually PNGs.
A.D. 1769: James Watt patents the one-click steam engine.
A.D. 1776: Trolls, angered by CmdrTaco's passage of the Moderation Ac
I voted on the 802.11g spec. We all knew the problems we would have with 802.11b integration (and which have been widely reported in various interoperability tests). We had to draw the line somewhere. And when you draw lines, someone will invariably take issue.
It is obvious that CW's reporter talked to someone who had an axe to grind. Maybe when we publish the spec in June (possibly July---yes, the IEEE also has a bureaucracy) that reporter will sit down and read it instead of reporting what someone else has said.
This assumes that the reporter can understand what he/she is reading (a BIG assumption these days with reporters).
It would have been good if they did this before they introduced the first (801.11b) wireless cards...
Now, the speed rating makes it seem as if 802.11a cards are several times faster than 802.11g cards.
Indeed, it does look as if someone is trying to create confusion.
Slashdot gets worse every day... Pipedot: News for nerds, without the corporate slant
It is perfectly reasonable to expect only 20 mbps throughput with a 802.11a or 11g network, for the same reason that 4-5 mbps is average using a 10baset hub or 802.11b. These are all shared mediums. Clients must use Collision Detection and avoidance. There is competition for the available bandwidth. All wireless must contend with clients that are connected at different rates. If a host is far enough from a 11a access point that it associates at 12 mbps, It's communications with the AP will take a longer timeslice from the available airspace. Clients associated at a higher rate will have their effective communication rate drastically effected.
Does it matter? Is it bad to market 11a and 11g at their 5x mbps? or 11b at 11mbps? Not really. (IMHO) Just like Hard drives are advertised at they size before putting a file system on them, it is up to the user to understand what the numbers really mean.
If you are the only client associated with an AP, your throughput will probably be much closer to the theoretical maximum, just as if there are only two things connected to a hub, their communications with each other will be better than if there were five.
Why is Apple responsible for defending 802.11g, and why is anyone attacking Apple for the shortcomings (if any) of 'g?
I have a Linksys 802.11g system, and if there is a problem with the design of the spec, that's the IEEE's fault, not Linksys, Apple or anyone else.
-twb
Most of the time the quoted speed is the RAW speed. a 100Mbps network card is doing 100Mbps in RAW speed and actual data level speed is much lower. So then, shouldn't they be always quoting the higher 54Mbit/sec as opposed to some 11Mbit/sec!!?!
Anyway, 802.11b is 11Mbps so I can't believe 802.11g would be the same. I am automatically decreeing that 802.11g is faster than 11Mbps...
Does sound like bad reporting. Shouldnt happen from technically saavy folks
So if I don't want any B clients on my networks all G max speed. But maybe my neighbor has an entire B network that overlaps from an RF point of view with my network? If I turn off the compatibility mode will I sink his B network?
Are they using the correct SI form of the prefix Mega, the now outdated binary form of the prefix Mega which has been replaced by the prefix Mebi?
I wish this stuff would catch on. It's useful.
You quote raw signal rate and actual throughput for b, but not for g, which is a bit misleading. For those who still haven't figured it out:
b: 11Mbps signalling rate, 4-5 Mbps effective throughput
g: 54Mbps signalling rate, ~22 Mbps effective throughput.
[I don't know anything about a, so I'll let someone else comment about that.]
10 PRINT CHR$(205.5+RND(1)); : GOTO 10
Reporters that don't do there homework and slashdot editors that don't check the facts before posting? What is this the New York Times?
Is it just me, or when something about a Vice President at Apple, did you expect to see Al Gore quoted?
______ This mind intentionally left blank.
this can also coincide with regulations from the FTC (correct me if i'm wrong), limiting 56K modems to actually having a maximum data transfer rate of 53Kbits..
so although Apple mentioned the article may have something to do with politics, i'm pretty sure there are regulations being set as well..
"The ones who dont do anything are always the ones who try to pull you down" -- Henry Rollins
After reading the article, I did a quick search for 802.11g throughput tests and 802.11a/b tests. I came up with two links:
Tom's Hardware 802.11g throughput tests
ExtremeTech's 802.11a and 802.11b throughput tests
There's going to be overhead with any protocol, but I would expect that wireless would have a higher overhead than wired protocols. There's certainly a lot of things you have to take into consideration for wireless throughput - obstructions, distance, error correction.
It's ok monkey boy .. I wouldn't put much stock in Computerworld as an authority on anything - Nick Petreley is just the Dave Barry of pop IT rags.
Just keep up the good work spreading mono around! It's quite contagious.
Paz, amor, esperanto es muerte
Computerworld reports that the IEEE has changed the 100BaseT spec to only run at 65Mb/s not 100Mb/s as initially specified, thus slowing down millions of computers world-wide. Additionally gigabit ethernet has also been affected by the IEEE bringing many critical business systems down to a crawl.
The only people who look bad as a result of this are silly chipset vendors and the 54g collaboration of idiots who put products on the market based loosely on the draft since now all their logos look stupid.
Wireless doesn't do collision detection... because you can't. You only do avoidance (more overhead)
The bottom line is, what number SHOULD we put on teh spec? Call it 11Mbps? It's only approximately that, and that doesn't really tell you anything about the spec. Calling it 54Mbps is totally, completely accurate, and those who misunderstand simply, well, do not understand.
He said *IF* there is a problem witht he spec, then it's the IEEE's fault, not someone elses. ANd he would be right.. if there was a problem in the first place.
Because in 100Mbps ethernet, the raw speed is NOT much slower.... the max theoretical speed a host can transmit on 100base with ethernet, ip, and tcp overhead is still over 90Mbps.. (I think it's near 97 Mbps, haven't calculated it for a few years). This number is even closer for 10Mbps.. (close to 9.9Mbps)
Nobody ever really kicked up a fuss about this because the speeds are so damn close... but in wireless, they are very different.
excuseme, desee decir esperanto es muerto ..
;P
amor, paz, esperanza, muelle
love, peace, hope, wharf?
It's 54,000,000 bits per second, which is a Megabit per second.. both under the old system AND the new one.
Yes, I realize this contradicts what you might think about a Kilobyte (now Kibi) being 1024 bits, and so on and so forth.. however data transmission speeds have ALWAYS been specified in metric units of bits per second.
A kilobit per second was always 1000 bits per second.
When someone says megabit, it always meant one million bits per second, not some strange power of two. That only comes about when you are dealing with memory.
With the internet, it got confusing because peopel started going from kilobits to kilobytes, or writing software to show upload rates without real knowledge of how thigns are technically specified, so it got muddy, and you have to guess what people mean.
However, in the case of 1.544Mbps T1, 10, 100, 1000, or 10000base ethernet, 11Mbps wireless, or 54Mbps wireless, we are talking about powers of 10
...or just a play on words with "Jobs" and "Wozniak"? :) Hell, if I had a name like that I would also be promoted to president of hardware product marketing.
You can't expect much more than 5 Mbps because around 5.5 Mbps is the theoretical bandwidth. Quoting 11 Mbps is highly misleading, as this is merely the signalling rate; there is some overhead in the physical-level protocol. If this sort of misleading labelling were used by ethernet, 100 Mbps ethernet would be advertised as 125 Mbps (the signalling rate is 125 Mbps, but it needs 5 bits for each 4 bits of real data transferred).
So yes, I think it's bad to market 11b at 11 Mbps or 11g at 54 Mbps, as these are not their throughputs; they should be marketed at 5.5 Mbps and 22 Mbps, respectively.
10 PRINT CHR$(205.5+RND(1)); : GOTO 10
...that every article posted on the Internet about this subject might not be true, or could be politically motivated.
I'm not sure the age of the submitter, but if this comes as a surprise to anyone you really should be ashamed. Just because it's in print, on TV, or online does NOT make it true.
-- El Sacarino tiene gusto de la chocha
I find the discussions about bandwidth (real and potential) less interesting than whether new AP's will have good backward compatability with a and b clients, have better range, and other usability issues.
It would be nice to stream high-quality video over wireless links, but that's what wired segments are for. Other factors are more important for the 802.11x's (most applications; most people). Like for instance, I'd like to see a breakdown of how many web surfers a 'g' access point could handle in a mixed-card environment.
Right on!
Walk between your 802.11b Access-Point and your laptop with an 802.11b PC Card while talking on your 2.4GHz cordless phone and chances are the data connection will be interrupted, but your phone call will be ok. I have seen it happen time and time again.
With so many devices in the same spectrum, there will only be more problems like this. It seems to me that moving to a different spectrum (such as 802.11a provides) makes the most sense, even if you need three times the Access Points to provide the same service.
"They that can give up essential liberty to obtain a little temporary safety deserve neither liberty nor safety."
Our company has been purely 802.11b within the office complex for over 2 years now. Every client machine uses 802.11b, be it desktop or notebook. Our servers, of course, are connected to fast ethernet.
... it's not -that- much better on a busy LAN, and overall, the lower range makes it useless for us.
We've recently moved the network to 802.11g draft standard using linksys access points and linksys network adapters... which use the Broadcom chipset. I'm personally using a 17" PowerBook with an Airport Extreme (also Broadcom) card.
On the whole. we get between 15-22Mbps.
We did some pretty minor testing with 802.11a, and although I agree that it's throughput is -better- then 802.11g
802.11a will likely die out as 802.11b/g gain more and more popularity... the "non-standardness" of 11a, coupled with it's range issues, make it a hassle for most network admins.
bash-3.00$ uname -a
SunOS panda 5.10 Generic sun4u sparc SUNW,Ultra-2
I have a question.
Since 802.11g and b are backward compatible.
It would seem the controversy stems from the fact that, if you already invested in 802.11b equipment, mixing 802.11g in with your environment is going to cause the 802.11g access point to step down or send RTS/CTS signals after each packet as a courtesy to 802.11b equipment trying to communicate in the same area.
So, here is something I propose then:
Say you decide to deploy 802.11g equipment in your wharehouse. You have not invested in anything WiFi and you have a nice radio free environment.
So you deply your 802.11g network in your wharehouse and everything is ducky.
Now, along comes Joe Shmoe. Joe Shmoe decides he is going to open a Steppen Brew right next door to your wharehouse.
He has this brilliant plan about offering Customers free internet access while they sip there latte's.
So he deploys a 802.11b access point on his roof next to your wharehouse operating with 802.11g equipment.
All of a sudden, you start getting complaints about crappy through put on your Wharehouse wireless LAN.
You can't seem to figure it out, but your 802.11g network is now half the network it was when your deployed it.
So you look for anyone using 2.4Gigahertz bluetooth devices, remote phones, cordless radio headsets...etc.
Nothing?
In short, the question is: will 802.11g equipment step down in the presence of any 802.11b device, or does it only step down if that device is actually transmitting on your network?
Couldn't find anything in the specs that would rule out this completely NASTY scenario.
Anyone care to comment?
-Hack
Got Geometrodynamics? Awe, too hard to figure out? Too bad.
Contrast this with 802.11a which is in an entirely different Spectrum range, which IT OWNS, specifically for wireless networking.
That's not true; the 5GHz U-NII band is unlicensed as well. There are 5GHz cordless phones already, and I suspect in the future we'll see more non-networking 5GHz equipment.
It seems to me that what the IEEE decided to do was to label the spec with the actual throughput speed as opposed to the raw one. That makes sense and I don't know why it wasn't done with b. But apparantly some people took this to mean the raw speed had been reduced from 54 to 20 which would have meant a sizeable reduction in actual speed.
-You may license this sig for only $6.99.
"Microsoft can goad you into building on a framework that is wholly controlled by them"
actually the correct term is bribe; and its not me getting the money its the VCs
Now it seems Mono will simply morph into a commercial concern with backing from various industry heavyweights... that is unless the VCs consider it ripe for picking - in which case I will suddenly find myself workking for Intel... IBM... HP... you pick one, the VCs will.
Mono in GNOME, not likely at all I'm afraid... now if I had only kept my dumb mouth shut about reading the ECMA specs...DOH' CURSES!
Before adopting WHATWG, read the moonlight.NET EULA [http://www.microsoft.com/interop/msnovellcollab/moonlight.mspx]
Actually, since the (linear) range of a is about one third the range of g, you need 3^2 or nine times as many access points to cover the same area.
Myself, I've just avoided buying 2.4GHz cordless phones since I use 802.11b at home. 900Mhz is just fine quality wise.
--
The internet is the greatest source of biased information in the history of mankind.
Oh, sure. *I* posted this when the original article came up, and nobody cared. But then some fly-by-night company nobody's ever heard of named 'apple' steals my comment, and suddenly it's news :)
...the land of illusions. The speed of your CPU turns out to be a myth and your 801.11g-card is subject to controversy.
You know it makes sense, a little reminder from jointm1k.
To prove the system works, Jobs revealed that Apple Senior Director of Hardware Product Marketing, Greg Joswiak, is, in fact, the result of an iClone experiment combining the genes of Jobs and Apple co-founder Steve Wozniak. "Half Steve and half me!" Jobs said. "He's great for Hardware Product Marketing, and we grew him in just three weeks!"
Except that there are no licensees of the 2.4 and 5 ghz spectrum ranges... they are unlicensed spectrums, and the 5 ghz spectrum is not "just for 802.11a or wireless networking technologies"
Yes, right now the 5 ghz range has less using it, but rest assured that will change, because everyone "knows" more ghz means better!
Except when it doesnt. 5 ghz is a higher frequency, thus offers lower penetration through walls and other line of sight obstacles. Which is the major reason we all use wireless technologies- so we dont have to drill holes through our walls and run cables through them.
802.11a is a mixed blessing. It doesnt collide with 802.11[b|g], but it does collide with the studs, pipes, conduit, drywall, and insulation in your walls a lot easier.
You have to remember, Apple doesn't really offer a huge product line like some vendors. They have a core set of laptops, desktops, one type of server product, and several accessories and gadgets (mainly the iPod).
The Apple "Airport Extreme" was the first commercial 802.11g device to market - and Apple did their best to put a "spin" on it that it was somehow their own invention. ("That's right folks... good old Steve J. is bringing you the next insanely great thing. Faster wireless than anyone else offers!") Can't really blame them.... They were the only one willing to stick their neck out and start selling the product at the time. Everyone else waited until Apple had it on the shelves before rushing to release their own.
If people start publically attacking the 802.11g spec now and making it look bad, Apple stands to lose the most from it. They've already built all of their systems with it either integrated inside, or upgradable by expansion board.
You mean that there are only the two of you working at Xerox?
No the term is Goad, as in:
.Net framework being standardized by the EMCA, when there is no chance in hell Microsoft is going to open up or standardize the parts required to work with Windowz."
.Net's framework.
:-)
"Make a big deal about syntax and other irrelevant aspects to the
At least, any that mean jack. (The secret API calls they build into everything to screw competitors, historically, now and damn well in the future.)
As for your VC scam, you would have to be OUT OF YOUR MIND, to develop a competing standard against any of
I don't mean just the ideas copied from Sun's Java, such as virtual machines etc, either.
The company (Micro$oft) has 43 Billion in cash reserves.
Your VC friends are about as smart as you are.
Finally, I am not talking about Mono in ANYWHERE, whether it be GNOME, Windows or KDE.
I am talking about the fact there are coders in the MONO project that could and should be working on GNOME's bugs, faults.
Instead they are working on a Microsoft deterrant to adopt Linux. We already have that, because KDE and GNOME are barely profesional grade desktops as IS.
Besides the rather assanine API constructs that contributes to USER FRUSTRATION, they could be working on cleaning up GTK, and GNOME applications.
If Email had the feature, I would reach through this message and shake your freakin neck for wasting and draining the Open Source communities direction and focus on MONO crap'ola which is nothing but a roose...
But since you are very far away, your very safe...
But if your ever in Madison Wisconsin, I will shake you so hard your brain will fall out!
Then we can go and get a pizza and I can convince you it is all a bad dream, through hypnosis.
Then send you happily on your marry way to make GNOME a great desktop APP, along with a great API to build a nice user interface we can all understand.
-Hack
PS: See, sun setting, and millions of users dumping MS Office and walking into the sunset happily with thier copies of GNOME Office apps...
Got Geometrodynamics? Awe, too hard to figure out? Too bad.
Ha!
:-)
My Karma power bitch slaps your TROLL!
You whiner.
Why, even the GOD Shiva herself wouldn't dare touch me my Karma it is so high!
Phhhhhfffftttt! NA na naaaaa NA!!!
-hack
Got Geometrodynamics? Awe, too hard to figure out? Too bad.
Yeah, you're correct there. Apple's wireless is a relative no-brainer to set up (at least compared to many wireless PC config. utilities).
Still, that's improving on the PC side as well. I recently set up some Belkin 802.11g wireless stuff for a client, and it allowed actual passwords too. (Even showed what they converted to in hex, in a seperate "info" window below as you keyed it in.) It also featured auto-detect.
The Belkin hub had an integrated web-based interface, so using the included Windows setup software was optional (and I saw no point to using it at all, since the web interface was functionally equivalent).
One thing I really did like on Apple's software is the way it automatically checks for firmware updates, and lets you update an Airport base with a single click if newer firmware is available. (By contrast, I needed a firmware update for the Belkin setup mentioned above, and it never informed me an update was available. I just happened to check their web site out of habit, and found it there. If I hadn't done it, I wouldn't have gotten one of the 802.11b client cards to connect properly!)
Let me qualify by first saying I've worked designing network equipment at a lower level, and have analyzed this stuff in detail. Of course, that doesn't make me right all the time.. I just mean, I have actually researched this stuff somewhat seriuosly, and looked at it with scopes, compared products, etcetera.
.01 microseconds.And when a card receives a frame, it expects to clock the data in at.. you guessed it, 100 megabits per second, ,or a bit every .01 microseconds.
First, the switches DO matter, because despite what you might have been told, switches are NOT all capable of switching at wire speed. If you don't think the switch has an effect, get a better switch and try again, you'll see. I've done it, it's true. It was true of 10Mbps switches, and is even more true of 100Mbps switches. Switches store & forward packets, and make decisions, and that takes time. In good switches, it takes very little time. Switches also have a limited amount of backplane bandwidth...if you have a 12 port switch, and you think it can actually switch six pairs of computers at wire speed at full duplex, well, unless it has a backplane bandwidth of 1.2 gigabits per second, it's not gonna come close.
Thirdly, the reason I staretd at the TCP layer was very specific: Because people are comparing TCP related transfer speeds, like with FTP.
I didn't START at the TCP layer, I FINISHED there.
If you re-read my post, you'll see that the difference in overhead between looking at what you can expect all the way down at TCP and what you can expect from the raw hardware is almost negligible.
The reason it matters is because those of us who actually do real network engineering have an understand of what all the terms mean, and how they interrelate, and we don't just think "100Mbps means I can transfer a 100 megabit file in one second".. and the same has *always* been true of the wireless protocols. We actually read up on how things work, and make our own decisions.
His card was labelled as 100Mbps because WHEN YOU PUT A FRAME ON THE NETWORK, you do it at exactly 100Mbps. The spacing between the bits is precisely
The networking is called "100 megabit" because, on a busy network, if you look at the "ether" part of ethernet, you will find there are exactly 100 million bits a second in use, if it's at 100% capacity. Nobody said all thsoe bits were actual userland data, and, in fact, any decent network admin realizes they are not.
Yes, they could explain that actual speeds are different.. or anyone could, you know, LOOK IT UP.