Silicon Investor (SI) -- The First Internet Community

STOCKTALK

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 : Qualcomm Incorporated (QCOM) -- Ignore unavailable to you. Want to Upgrade?

To: engineer who wrote (43368)	10/5/1999 12:40:00 PM
From: slacker711	Read Replies (1) \| Respond to of 152472

Samsung chipset.... edtn.com Samsung pits new CDMA chipset against market leader Qualcomm By Mark LaPedus Electronic Buyers' News (10/04/99, 3:35 p.m. EDT) Hoping to lessen its dependence on Qualcomm Inc., Samsung Electronics Co. Ltd. has made good on its earlier promise to develop wireless handsets based on its own chipset line. Recently, Samsung announced plans to build handsets that would use an internally-developed chipset-the MSMScom4000-which is designed for CDMA applications. The chipset, also known as a baseband processor, supports voice-processing and other functions in cell phones. Samsung and other OEMs that make CDMA handsets must currently procure these critical components from third-party suppliers, namely Qualcomm, which has a market leading 89% share of the segment, according to analysts' estimates. DSP Communications Inc. also sells chipsets for CDMA handsets. In addition to building a captive supply, Samsung is developing its own devices for competitive reasons. Qualcomm at present also sells handsets based on its own chips, although the San Diego-based company recently said it is considering selling off its cell-phone business. Still, Samsung is taking no chances. The company, which buys millions of dollars in CDMA chipsets from Qualcomm every year, today said it has successfully completed development of the rival baseband controller. Samsung will use the chipset in its IMT-2000 handset line, a product that will be launched by the middle of next year. Samsung is also developing other cell-phone chips in-house, including analog-based components and RF ICs.

To: engineer who wrote (43368)	10/5/1999 12:50:00 PM
From: Joe NYC	Read Replies (1) \| Respond to of 152472

engineer, If AT&T sells CDPD as a modem for laptops for downloads or to be used by GUI browsers, I can see your scenario of running out of capacity quickly. But if used by mini-browsers in the handsets, it sends and receives only tiny bits of data, probably no more than one packet at the time, which is sent in a fraction of a second. It's virtually free, since it just fills gaps. I would guess it is a lot more efficient than circuit switched data offered by Sprint. I don't know the mechanics, but my guess is that it takes up constant bandwidth whether or not you are sending data. Do you know if they have any kind of optimization or compression for this? Joe

To: engineer who wrote (43368)	10/5/1999 12:53:00 PM
From: FWS	Read Replies (1) \| Respond to of 152472

Over on the COMS thread there is a plethora of web applications being offered on the PALM VII and more to come. Doesn't the palm use CDPD?? How do you see CDPD competing with CDMA? Is there a data limit to CDPD that leaves it at a competitive disadvantage to CDMA? Thanks and sorry if i don't have all the terms down just right. I'm just a layman at best... chris

To: engineer who wrote (43368)	10/5/1999 4:08:00 PM
From: w molloy	Read Replies (1) \| Respond to of 152472

CDPD and GPRS CDPD is made as an overlay system to the existing analog network. what is does is watch the over the air radio signals and find call channels that are not in use. When it detects this, it then attempts to use this unsed channel for data. If another call channel comes along, then CDPD must detect this and drop off the air, even in mid packet. As the demand for bandwidth grows and the demand for data grows, these two will collide even faster than that described by BUX. CDPD works now only because it has a small following. If it is pushed to a large following and an large user base, then it will log jam not only itslef, but it would severely limit the capacity of the TDMA voice netowrk it is overlayed on. It must drop transmission during the most critical phase of a TDMA call, the data carrier timing burst. The demand for bandwidth might be infinite, but the bandwidth available isn't. In CDPD systems, the voice service is paramount. Lets say there is one spare channel - channel x. All other channels are servicing Voive calls. We begin transmitting packets on channel x. Another voice call comes along. Our CDPD service is bounced and the packet is dropped. At a random time period later, we try to resend the packet. We search for a channel. If one is available OK, if not, we wait for another random time period. The net effect is a reduction in the effective data rate that the application using the CDPD service sees. The nominal rate for CDPD is 19.2 kbps. This can be reduced to as low as 8kbps ( which I saw in a lab test with 12 CDPD units jostelling for service). There is no additional stress on the underlying network. All the stress is focussed on the CDPD service itself. I take your point regarding the packet being dropped during the carrier timing burst. Don't you think AT&T system engineers would have taken this into account before offering the CDPD service? (well - maybe not...) GPRS is the GSM equivalent to CDPD. From a software viewpoint it looks very similar indeed. There is no point in migrating from CDPD on IS-136 to GPRS on IS-136. w.