Here's my notes (and interpretation) from the Rockwell conference call:
1. Rockwell and Lucent still have not agreed to interoperability specifications as of today.
2. Is there any product in the field that is upgradeable. No, and none shipping yet.
3. Will the first round of K56 be software upgradeable? NO.
4. Why are your ISP clients doubtful of the viability of K56 product? No comment.
5. Testing of PSTN done and results? Testing has been limited to internal Rockwell sites only. We believe we can get low 40k to low 50k throughput on PSTN.
6. Will the major ISPs be using ROK equipment? It's up to them to say.
7. Are there any OEMs on your list that are new customers? Most or all are existing Rockwell customers.
8. What is the design architecture? Our modems are DSP based, however, ours are ROM, low cost, type. Not upgradeable, but upgradeability costs.
9. Can you scale to beyond 64k? No. Our early tests don't even get beyond low 50k without more power being allowed into the network. Anything beyond 64k requires CO equipment. K56 is not upgradeable to DSL.
10. How many different chipsets are there? a) Non-programmable 56k. b) Upgradeable client. c) Upgradeable server. What is the time frame? Non-programmable in about a month. Upgradeable maybe a month later.
11. Why Microcom firmware? We're not using them, as of now. We sell them data pumps, they add their own firmware.
12. How about price & margins. (Paraphrased), we don't want to see the market bastardized.
13. Standards progress? Were working on a draft for a US standard by the end of the year, '98 for ITU. When do you expect USR interoperability? After an ITU standard is reached. Also, we expect the FCC waiver to increase power, but don't expect it this year.
14. Why do Cisco and Cascade say next year for deployment of K56? Dunno.
15. Why would a user buy one of your non-upgradeable modems? We are committed to maintaining backwards compatability on future K56 products. (In other words, it will never be standards compliant!)
16. Are you working with Intel regarding a software deployment of the data pump into the CPU? No comment, however, we're looking at it.
17. What retail prices to you expect on first deployment? Around $200 is our best guess.
18. What's your backlog on K56? Expect 1 million units per month by June. No comment on backlog. By June 2 or 3 or 4 million units produced by the end of June. (Nothing like specificity!)
19. What's the difference between ROM and Flash based architecture? $20-$50.
20. Why would I buy hard configured K56 unit, when it won't be standards capable, thus reducing the chance of making a 56k connection at a later date? Why wouldn't I buy a Flash based product and have an upgrade path? Folks, he never answered the question, except to talk about Lucent interoperability! He did say that lots of folks want cheap (a.k.a. inferior) products.
21. Sood Dood. Why did you not come out with upgradeable architecture first? We view both products as the same (Yeah, right - that was a non-answer.) ROM is just cheaper.
22. When do you expect the 1st ROK based ISPs to go on line? March is our expectation, so we can can support ROK client modems.
23. How do you plan to overcome USR's major name recognotion, and Rockwell's lack of it? We see that, and we're going to throw out weight into our customers marketing, starting in March. (No specifics given.)
24. What are v.34 inventories? We expect v.34 only products to keep shipping. (Yeah, right!) No further comment inre current order levels.
25. Tell us more about your testing to date. We don't want to disclose much here. Much testing in the labs, simulating real world. We're also relying on v.34 real world results and infering K56 results from that. We've done a few weeks of actual field test, but don't wish to discuss.
That was about 5 minutes of action packed into one hour. My opinion - they are behind time-wise and technology-wise, and they know it. They beat around the bush, and focused on FUD. Their rallying cry is "We're the open standard." Bullshit! |
