For some reason my reply did not get posted earlier today. Therefore, I will try to re-create it. But first I would like to thank everyone for the kind words today. I did respond to the poster through private mail and explained how I thought the newsletter may benefit readers. I also offered to send him an issue or two if he sent us an email address.
Since I did that privately, it occurred to me (no spam intended) that I should make the same offer publicly. I will be more than happy to send an abridged version of Tuesday's newsletter out to anyone who has never recieved a copy of the newsletter.
Also, on the subject of the website which sparked a great deal of interest and controversy this weekend, I have a funny anecdote. You know the story of the person who says he is not paranoid, he knows they are all out to get him. Well, as hard as I could try, I could not duplicate the problem with the website even though I believe everyone is telling me the truth, but maybe they are pulling my leg.
I use a DSL line to connect to the Internet and have my own file server that is hosted by my DSL provider. Everything is perfect. So, I went to my 56K modem and did a dial up to an ISP I use as a backup. Everything works and there is no problem with the website. I went through AOL at 28.8 and it still works. The whole world looked nuts and I was the only sane one it seemed. Then I went to my backup PC and that is where everything went bezerk. Problems on the 300MHz Win95 system.
So, what was the problem?
I send out the newsletter in PDF format that is password protected as added security. The Website must work because I have the password enabled on my system, which does not make sense to me. I went back to the word document itself, saved it in PDF format, reloaded it on the website and everything seems to work. Sometimes we are our own worst enemies. The security was to prevent the changing of the document such that it would be difficult for anyone to attach a virsue to it and send it along to someone else.
=======
Back to the questions asked in this post.
1. Your fee for RadarView was the steal of the century since it was at the very beginning and only for those charter individuals for the first year that acted as guinea pigs for the newsletter as we worked out form, function, and format. We all used to be able to go to the movies for under a buck and I remember gasoline at less than 30 cents a gallon when I went to college. Oh, the good ol days. But then again, QCOM was under $30 in the good ol' days last year.<GGG>
2. SGI dumping MIPS is a hold over from Beluzzo's CEO days. I am still hoping the new management sees the errors of their ways and stop the divestiture. SGI is about to spin off its Mediabase unit in an IPO that already had venture capitial funding so I am not sure why SGI needs the cash, especially since there are considerable tax consequences.
I have gone on record as saying that the best way to divest of MIPS, if that is still the plan, is to pass the shares onto the shareholders in return for SGI shares in a buyback and exchange.
SGI made a bad investment in Cray Research and are now trying to sell that unit off. This is but one of a few miscues by the company. Prior to his departure, Beluzzo had a string of "reorganizations and re-defining of the company." The road was to contract the company back to its basivc roots and to unload non core businesses. Whether it was NT or Linux software or processors or levels of system complexity that it would support, the entire company was thrown into chaos.
New management may decide it is in the best interest of the company to discontinue the divestiture and just maintain a smaller stake in MIPS. After all, with the new gaming systems hitting the market and the explosive growth seen in the IC sector, why get out before the real fun begins.
SGI is not strapped for cash and could very well leave their investment in MIPS alone and let it grow. IF the really needed the cash then they could sell MIPS a piece at a time. Right now, I cannot think of any better way to have my assets spread out in an investment. Besides, any cash they might need should be available after the IPO of the Mediabase unit.
Sometimes things just do not make sense. We are investors in MIPS and expect good things from it. We are also going to re-invest in SGI at the end of this month (Tax loss selling last year) since we view it as a turnaround candidate for 2000. Therefore, I cannot give you an explanation since we are bullish on both stocks and believe it is a win-win situation for both. If so, it makes no sense for SGI to divest itself of MIPS.
Price Target - not sure which you were referring to.
MIPS - $100 or better SGI - close to $18 in the time frame you are talking about, with MIPS being conservative. The dates are closer to the 12-18 months.
As far as the semi equipment cycle is concerned, I will be brief.
1. we have not even touched the peak of the cycle for the industry upturn. We are now starting to see the deliveries from the initial orders. We have the rest of this year for technology upgrades and some expansions.
2. We still have 1-2 years of a decent equipment cycle for the new fabs that are now being built and will come on line in 2002. A very aggressive construction schedule from announcement of a new fab to completion and start of production and equipment ramp, is close to 24 months.
3. Then we have all of 2003 for the as yet new fabs that will need to be built for the next technology and the 300mm fabs that are just now being announced.
I hate to oversimplify this in 3 bullets but we cannot be anywhere near a peak since we have yet to experience the long lead times and tight manufacturing capacity and deliveries from the equipment suppliers yet. We have not fully taken care of the initial fab retrofits and expansions like in Taiwanese foundries. we are probably just coming out of the infancy stage of this recovery. This recovery will be longer than the downturn.
If anything, this summer might be the peak of the retrofit and technology upgrade cycle with a few expansions thrown in. We might get a lull just before the new fab construction projects are completed and are ready for equipment and when the 300mm fabs are online. After that, we start the final round of expansion.
Consider this: If the recovery started in January of 1999, then any new construction does not come on line until Janaury 2001. I doubt it very much that too many companies were even thinking of upgrades, expansions, or new construction in the Q4 1998.
So, as I did more than two years ago when I called the recovery cycle in the industry for mid 1999 and did not waver like the anlaysts and media, I disagree with those that believe the peak is this summer. The earliest I can see the peak would be the summer of 2001.
Andrew
|
