-
Join the MyBroadband community
South Africa’s biggest forum. Discuss, discover, and connect with thousands of members.
-
Question of the Day: At what point do you sell your car?
WiMAX - the controversy continues
- Thread starter rpm
- Start date
vodacom3g
Vodacom Representative
Note that Intel says "Yes" in the article but then promptly mention only going into pilot now.
At the same time Intel calls HSDPA "mature and stable".
This is where we are today with WiMax, it's still early days and all their messing around with standards is not helping. Hopefully it will stabilise soon.
I would love to get 1Gbps connections. It would appear the writer of the article dont know the difference between connection speed (Gbps/Mbps/Kbps) and usage/cap (Gb/Mb/Kb). But I guess that is why he decided to do journalism...
vodacom3g
Vodacom Representative
This is how so much of the misinformation and confusion originates. Someone (seen as an authority) makes an incorrect statement in public, something like "WiMax" can give from 1Gb/s to 12Gb/s" (this article).
Those who understand writes it off a bad journalism but others read it and now it's become fact. After all "I read it somewhere"
We've got a few forumites that love this, taking a piece of information completely out of context and quoting it in a different context where it is completely invalid. Two that comes to mind: "HSDPA is cheaper to deploy than GPRS" and "Increase in radio throughput does not require an increase in backhaul".
I call these people 'armchair specialists or consultants'
. They read a web page or two and now suddenly they're 'experts'.To give a WiMax example, we've seen these parameters quoted for WiMax:
1) Up to 70Mb/s throughput
2) Uo to 70Km range
3) 802.16e works up to 120Km/h
Next thing we see is an article or statement by one of these 'armchair specialists' lumping them all together stating: "WiMax will give you 70Mb/s throughput at 70Km's while traveling at 120Km/h".
(BTW, I get 4Mb/s over 7Km (way out of distance spec) on WiMax - but with an antenna the size of a dining room table
)We saw it again earlier today in the WiFi article. A bunch of parameters, each valid on its own but mutually exclusive are bundled together and suddenly we get WiFi at 52Mb/s over 100Km using a watt or 2 on Omni antennas.
We need more rpm's (and Duncan's, Lesley's and Belinda's) in the industry; Journos who understaind their subject-matter!
TheRoDent
Cool Ideas Rep
WiMAX 802.16e mobility is unproven, largely undeployed and not really commercially available. That is the reality. Heck, most equipment manufactures are still going through the certification waves.
802.16d on the other hand is mature, fairly simple. However all the bandwidth claims are absolutely ludicrous. Distance claims too.
Yes, it's possible to do WiMAX up to 40kms.
Realistically, in that kind of deployment the performance of the base station will suck, so the reality is that radio planning works around 2-4km for cell deployments.
There is a lot of hype around WiMAX, and then there are the realities.
Telkom's responses to the EngineerIT questions posed in this article is actually pretty much on the money.
802.16d on the other hand is mature, fairly simple. However all the bandwidth claims are absolutely ludicrous. Distance claims too.
Yes, it's possible to do WiMAX up to 40kms.
Realistically, in that kind of deployment the performance of the base station will suck, so the reality is that radio planning works around 2-4km for cell deployments.
There is a lot of hype around WiMAX, and then there are the realities.
Telkom's responses to the EngineerIT questions posed in this article is actually pretty much on the money.
Last edited:
vodacom3g
Vodacom Representative
Exactly.
And the standards are not going to solidify if they don't stop the in-fighting.
TheRoDent
Cool Ideas Rep
I do portability on WiMAX 802.16d networks right now, without requiring all the insanity of ASN's and R5/R6 intefaces.
My prediction that the Chinese will commoditize 802.16d before 802.16e/m hits the markets is already turning out to be on the money. Companies like WaveSAT have OEM'd their technology to many Chinese companies who are now starting to churn out 802.16d gear like never before.
Wait for 802.16d in unlicensed bands (already available from some vendors), some Linux drivers and you end up with a cheap, fairly reliable air interface that is going to cause some serious disruption in the market. 16d already allows smaller, more tightly budgeted licensed operators to make good money in the African market.
802.16e/m is geared for Operator style deployments where project budgets run into the hundreds of millions of dollars. And it's certainly not ready for Operator prime-time yet.
My prediction that the Chinese will commoditize 802.16d before 802.16e/m hits the markets is already turning out to be on the money. Companies like WaveSAT have OEM'd their technology to many Chinese companies who are now starting to churn out 802.16d gear like never before.
Wait for 802.16d in unlicensed bands (already available from some vendors), some Linux drivers and you end up with a cheap, fairly reliable air interface that is going to cause some serious disruption in the market. 16d already allows smaller, more tightly budgeted licensed operators to make good money in the African market.
802.16e/m is geared for Operator style deployments where project budgets run into the hundreds of millions of dollars. And it's certainly not ready for Operator prime-time yet.
Unavailable?
That largely depends on what you mean by "commercially available".
If you mean that in most locations that there is no WiMax commercial service available, whether 16d or 16e, then you are correct.
If you mean that carrier equipment that supports 16e is not commercially available, you are wrong. There are several commercial operations that offer full mobility for high speed data right now, the best currently in South Korea. (That network doesn't support voice right now but this is a regulatory, rather than a technological, restriction).
I could also, for instance, say that "LTE is unproven, largely undeployed, and not really commercially available. That is the reality." This statement would be true, but not particularly useful.
I could also say that "GPRS is proven, widely deployed, and commercially available" but this would not be particularly useful either. These statements reveal little in terms of a practical comparison between LTE, WiMax and GPRS.
HSDPA here falls somewhere between GPRS and WiMax/LTE. It is not as widely deployed as GPRS, but not as fast as WiMax or LTE. The difference is that while LTE is not available at all right now, WiMax is.
This is separate from the question of whether or not WiMax will be commercially successful. But the fact that it has not yet been so says very little about whether it will be.
True, but also not terribly important. Wave 1 equipment has been widely available for some time. Wave 2 equipment is coming online now, ahead of certifications. Some vendors wait for standards and certifications and some don't.
As pointed out elsewhere in the thread, largely because of poorly written marketing text that leads to worse-written articles, where the theoretical maximums are combined.
WiMax is no different from GSM or CDMA in this respect. One must balance coverage, number of clients, and bandwidth per client.
The real difference is that compared to the other technologies available for carrier deployment at this moment, WiMax offers more bandwidth per base station and more bandwidth per client than just about anything else. Yes, you can trade away that bandwidth for coverage (and still end up getting more bandwidth per client than competing technologies) but there is a point of diminishing returns (about 8km) and to some extent it's not a good use of the technology.
Just because WiMax can't live up to the erroneous claims made for it doesn't make it unviable, unproven, or unavailable.
That largely depends on what you mean by "commercially available".
If you mean that in most locations that there is no WiMax commercial service available, whether 16d or 16e, then you are correct.
If you mean that carrier equipment that supports 16e is not commercially available, you are wrong. There are several commercial operations that offer full mobility for high speed data right now, the best currently in South Korea. (That network doesn't support voice right now but this is a regulatory, rather than a technological, restriction).
I could also, for instance, say that "LTE is unproven, largely undeployed, and not really commercially available. That is the reality." This statement would be true, but not particularly useful.
I could also say that "GPRS is proven, widely deployed, and commercially available" but this would not be particularly useful either. These statements reveal little in terms of a practical comparison between LTE, WiMax and GPRS.
HSDPA here falls somewhere between GPRS and WiMax/LTE. It is not as widely deployed as GPRS, but not as fast as WiMax or LTE. The difference is that while LTE is not available at all right now, WiMax is.
This is separate from the question of whether or not WiMax will be commercially successful. But the fact that it has not yet been so says very little about whether it will be.
True, but also not terribly important. Wave 1 equipment has been widely available for some time. Wave 2 equipment is coming online now, ahead of certifications. Some vendors wait for standards and certifications and some don't.
As pointed out elsewhere in the thread, largely because of poorly written marketing text that leads to worse-written articles, where the theoretical maximums are combined.
WiMax is no different from GSM or CDMA in this respect. One must balance coverage, number of clients, and bandwidth per client.
The real difference is that compared to the other technologies available for carrier deployment at this moment, WiMax offers more bandwidth per base station and more bandwidth per client than just about anything else. Yes, you can trade away that bandwidth for coverage (and still end up getting more bandwidth per client than competing technologies) but there is a point of diminishing returns (about 8km) and to some extent it's not a good use of the technology.
Just because WiMax can't live up to the erroneous claims made for it doesn't make it unviable, unproven, or unavailable.
Roman4604
Executive Member
I would also add that 'pupa' Intel has yet to make their 1st move into the end-user equipment space, but its coming near the end of the year with the 5150 & 5350 WiFi/WiMAX modules.
So far they've bowed to laptop manufacturer pressure and not made them manditory for Centrino2 certification, but I think in time they will close the price gap between the WiFi only & WiFi/WiMAX modules (currently ~$12-20) and may start applying their considerable influence on the hw community.
They have to, they're in the business of selling chips and they wont ship in volume if there are no networks to use them on. So to stimulate network deployment they're going to need to force the hw industry to ship the CPEs beforehand.
Will be interesting.
So far they've bowed to laptop manufacturer pressure and not made them manditory for Centrino2 certification, but I think in time they will close the price gap between the WiFi only & WiFi/WiMAX modules (currently ~$12-20) and may start applying their considerable influence on the hw community.
They have to, they're in the business of selling chips and they wont ship in volume if there are no networks to use them on. So to stimulate network deployment they're going to need to force the hw industry to ship the CPEs beforehand.
Will be interesting.
Last edited: