SI
SI
discoversearch

We've detected that you're using an ad content blocking browser plug-in or feature. Ads provide a critical source of revenue to the continued operation of Silicon Investor.  We ask that you disable ad blocking while on Silicon Investor in the best interests of our community.  If you are not using an ad blocker but are still receiving this message, make sure your browser's tracking protection is set to the 'standard' level.
Technology Stocks : The New Qualcomm - a S&P500 company -- Ignore unavailable to you. Want to Upgrade?


To: quidditch who wrote (2140)10/10/1999 12:42:00 PM
From: Valueman  Read Replies (2) | Respond to of 13582
 
Ruffian:

12 months at 60,000 handsets per month does not allow for one million subs by 2001. In reality, there will be 50,000 at year end 99, and the manufacturing rate will be far below that 60,000 per month number. There is no way they can have enough handsets to service the number of subs they project. It is such a farce because everyone else knows this but management continues to throw out these numbers. Fortunately, I believe the Street will be happy with 200,000 subs by year end 2000.

As for GPRS/EDGE, neither use any CDMA technology. The 3G TDMA version is just that and will not use any CDMA technology. Cdma2000 and W-CDMA is where the world has to cough up the royalties and license fees.



To: quidditch who wrote (2140)10/10/1999 2:20:00 PM
From: Ruffian  Respond to of 13582
 
GPRS Lengthy Delays>

Lengthy Delays Plague GPRS
Overheating GPRS terminals cause vendors to drastically decrease
speeds.
Andrew Dornan, Data Communications

All over the Telecom 99 show, vendors are proclaiming that general packet radio service (GPRS)--a
big step towards third-generation (3G) wireless networks--is ready to roll. Ready to roll, it may be.
But ready to use, it is not.

The problem? Vendors have run into difficulties developing GPRS terminals. GPRS equipment to
build networks exists; vendors are right about that. But carriers won't be able to offer GPRS services
until the terminals arrive.

Right now, some vendors have prototype GPRS terminals that are being used in trials. In the eyes
of the GSM Association (Dublin), that means everything is on schedule, according to association
chairman Michael Stocks. "The terminals are here," Stocks told the Show Daily.

The vendors themselves tell a different story. "It will be at least a year before we are selling GPRS
phones," says Juha Junnonen, a mobile marketing manager at Nokia OY (Espoo, Finland). Alcatel
N.V. (Paris) says the same thing: "We won't be selling GPRS phones until Q3 2000," admits Ivan
Gouletquer, a GPRS demonstrator from Alcatel. "Neither will anyone else."

Nokia's GPRS demonstration at the show uses wireless local area network (LAN) cards, not GPRS
cards, for the actual radio interface. Ericsson LM (Stockholm, Sweden) hasn't even brought any
prototypes, preferring to show off its nonfunctioning "concept" 3G phones. Both of them--and
Motorola Inc. (Schaumburg, Ill.)--quote the second half of next year as their GPRS shipping target.

The delays are only half the problem. When GPRS terminals do finally arrive in the shops,
customers will find another disappointment: the phones and PC cards won't allow speeds anything
close to the 115 kbit/s so often touted by vendors.

Alcatel is one of the only vendors to have a working prototype GPRS phone at the show, and guess
its speed? A mere 14.4 kbit/s, no better than the bandwidth available using conventional
circuit-switched global system for mobile communication (GSM).

To make things even worse, the first GPRS terminals will only handle asymmetrical traffic. Nokia's
first GPRS card will only run at 14.4 kbit/s upstream--the same as GSM--and 43.2 kbit/s
downstream. Alcatel says the same thing about its GPRS phone, and Ericsson won't even put a
figure on speeds.

Panasonic Matsushita Co. (Yokohama, Japan) has announced a GPRS terminal that will handle 56
kbit/s in each direction, but right now, it's just an announcement. A mock-up of it is on display at
the Japan Pavilion. But potential customers face an even longer wait. "It should be available by
Christmas," says Hiroshi Tamano, associate director of sales and marketing at Panasonic. "That's
Christmas next year."

What's the fundamental problem? Cellular terminals transmit using microwaves, which, like the
radiation used in a microwave oven, produce heat. The more time slots the terminal uses to
transmit, the hotter it gets. And because GPRS is always switched on, it continuously gives off
heat, unlike GSM phones, which only transmit very short messages to tell the base station where
they are. Wasting energy as heat also shortens battery life: Power consumption rises almost
proportionately to bandwidth.

Vendors are trying to find a way to focus the transmissions out of the equipment so that less is
used to cook the cards, but until then, the maximum bandwidth is about 20 kbit/s upstream and 40
kbit/s downstream, according to experts.

GSM Association chairman Stocks plays down the difficulties in tackling these problems. "It's a
challenge, not a crisis," he asserts. He maintains that GPRS terminals capable of handling
bidirectional 115 kbit/s will be available next year, but admits that he's not familiar with the technical
challenges faced by vendors.

The delays in rolling out GPRS terminals are likely to have some additional effects. In particular,
Wireless Application Protocol (WAP) will let service providers offer Internet access via mobile
phones, needs a packet-based technology such as GPRS to work effectively. WAP can be used
over today's circuit-switched mobile networks, but takes a long time to set up a call: Nokia, whose
first WAP phone has just shipped, is quoting a call setup time of up to 40 seconds. And
downloading Web sites at 14.4 kbit/s can be tedious.

All the same, GPRS promises to be well worth the wait. Higher bandwidths are only one of its
attractions. Other pluses include being "always on," eliminating the lengthy call set up times. In
addition, GPRS services promise to be low cost, because the data is carried over otherwise-unused
bandwidth.

While GPRS is on hold until terminals arrive, some users wanting higher-bandwidth mobile data
have another option: high-speed circuit-switched data (HSCSD), which enables users to set up
multiple parallel calls and then combine them together. The main problem with HSCSD, however, is
that users have to pay for multiple calls and suffer lengthy setup delays. Still, commercial HSCSD
terminals exist. One of the first, Nokia's HSCSD card, the CardPhone 2.0, offers 42.3 kbit/s
downstream and 28.8 kbit/s downstream, but calls still take up to 40 seconds to set up.

Peter Heywood contributed to this report



To: quidditch who wrote (2140)10/10/1999 7:25:00 PM
From: Clarksterh  Read Replies (2) | Respond to of 13582
 
Steve - Q gets royalties on all 3 flavors of 3G, whether 1XRT, 2XRT, W-CDMA, TDMA.

Only on CDMA kinds of 3g (it is possible they have some piddly IPR in 3g TDMA, but if so I am not aware of it, and it isn't likely to be material). But for CDMA versions of 3g, so far Qualcomm has always gotten the same royalties as for CDMAOne. Qualcomm has been very clear on this, and it is hardly a surprise given that their IPR is for mobile cell CDMA, regardless of data rate or modulation scheme.

2. This is regardless of the data-enhancing transmission of EDGE (for TDMA) or GPRS (GSM/W-CDMA). Engineer, if you could: what is the relationship between EDGE and GPRS and the CDMA RF interface, i.e., if CDMA= transmission of coded packets of data that are routed and selected by the handset or basestation ASIC according to the code identifier? Is this an overlay on an underlying GSM/GPRS or TDMA/EDGE interface or software protocol?

GPRS is just a protocol stack (handshaking et al) that allows a TDMA system to provide bursty bandwidth-on-demand and connects a device straight to the internet. But it still suffers from all the disadvantages of TDMA systems - you have to design the system knowing that some pesky user in the neighboring cell could transmit at the same time you do and corrupt your data. EDGE is primarily a change in the TDMA/GSM RF interface which allows more bits to be stuffed in per hertz and in this sense it is similar to HDR. Oddly (i.e. I'm sure there is a reason, but I don't know what it is) the first generation of GSM/NA-TDMA/CDMAOne did not take full adavantage of the Signal to Noise ratio. With a change in the modulation scheme they could have stuffed twice as many bits in a given hertz. HDR allows this for CDMA, and EDGE does the same for TDMA. Note that the oft touted figures for EDGE assume 3 times as many bits per hertz, but the reality is that they can't get that except in a lone cell (e.g. in cell within a subway station or a building) since the noise level will ordinarily be to high. But EDGE still suffers from the same draw backs of TDMA cell systems - limited frequency reuse.

. 3G, as such, is meant to refer to enhanced data and enhanced speed of that data transmission. Thus, W-CDMA or TDMA 3G, if they truly exist, are different radio media of transmission to accomplish this, and if CDMA interface is the medium, then Q's IPR is relevant.

It amazes me how often I have to repeat this. Qualcomm's IPR is required for any mobile cell CDMA system. Period. No 'ifs', 'ands' or 'buts'. At some point someone may find a better system for mobile cell comm than CDMA, and at that point Qualcomm has a problem. But that is hardly immanent. In addition, it is possible that someone will find something that really enhances mobile cell CDMA, and is so important that Qualcomm has to trade IPR even-steven. Again this is not a huge threat, although probably bigger than the first. (MUD and smart antennas for CDMA fit into this category, although at this point they are still nowhere near as important as CDMA itself.)

When ERICY capitulated, Gregg wrote, and it was generally believed, that point #1 above would result. It was never clear to me why ERICY's contractual deal with Q on the patent litigation and the infra would necessarily bind other GSM vendors/carriers, such as NOK and in Euroland the GSM carriers or TDMA.

It doesn't. The patents do.

Plus, the lingering question whether ITU principles on non-discriminatory, non-monopolitic terms for licensing would, at the end of the day, pose a problem.

I read non-discriminatory to mean not differentiating among users of your IPR. You are probably better able to judge this than I but lose more sleep over Maurice's rants than this issue (i.e. not much).

5. Why have we heard nothing from ETSI recently. Did it's leverage on resisting use of CDMA-based interface in Europe cave when ERICY caved?

It's a non-issue. Undoubtedly some people will move to EDGE and GPRS, but given their recent delays I think it reasonable to assume that many European providers will move straight to W-CDMA when it becomes available. Qualcomm will collect good royalties on W-CDMA, and will probably sell a healthy number of chips into that market (albeit probably not with the same market share as for CDMA-2000).

Clark



To: quidditch who wrote (2140)10/10/1999 10:37:00 PM
From: Bux  Read Replies (2) | Respond to of 13582
 
Clearly GPRS is in trouble. And the funny thing is it's getting toasted not by HDR CDMA but by itself! I find it telling that even the developers say it won't be ready for another year. I wouldn't be surprised if that date is pushed further out or the project is scrapped altogether. I have always suspected that GPRS and EDGE are mostly smoke and mirrors to help maintain GSM infrastructure sales. They are clearly not as advanced as CDMA HDR.

Qualcomm gets no royalties from GPRS or EDGE even if they do get commercially deployed so this is a win for Qualcomm and reports of the poor showing of GPRS at the tele-conference might even explain some of the recent strength in QCOM shares as the picture becomes more clear to the investment community.

It was never clear to me why ERICY's contractual deal with Q on the patent litigation and the infra would necessarily bind other GSM vendors/carriers, such as NOK and in Euroland the GSM carriers or TDMA. Plus, the lingering question whether ITU principles on non-discriminatory, non-monopolitic terms for licensing would, at the end of the day, pose a problem.

The deal between E and Q doesn't bind the rest but the other significant players had already licensed with Q and E was the only significant player debating that they had significant IPR in Q's implementation of CDMA. Qualcomm has always been willing to license their technology on fair terms that are not monopolistic. That doesn't preclude QCOM from getting the gorilla's share of the royalties, it just means they can't be discriminatory and license willy-nilly terms, different for each license without justification. Of course some of this could be considered subjective but I don't see the ITU becoming a road-block here.

A lot rides on the marketability of GPRS and EDGE since operators are making choices today based upon future upgrade options. Since neither GPRS or EDGE is available yet, operators have been relying on the word of those who are developing these technologies. Tero has assured me these are viable technologies that will be widely deployed but I have serious doubts and the fact that there is not much to demonstrate now can't put the carriers minds at ease. Time will tell.

Bux