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 : 3DFX -- Ignore unavailable to you. Want to Upgrade?


To: Joe C. who wrote (10363)1/29/1999 1:23:00 AM
From: Patrick Grinsell  Read Replies (1) | Respond to of 16960
 
Calling all...netfriends?

Here's the mission, should you accept it: I'm looking at putting together some estimates for 2Q, but I'd like you guys to give some input here to expand my vision. Here's some of my starter data to help us get going.

1) Board level costs for Voodoo3 should be virtually identical to those of Banshee. Memory is slightly more costly (of the 200mhz variety) but is of similar size (16 mb). This may be offset by the fact that V3 uses .25u and so the chip itself should be cheaper. I'm looking for total cost difference to be minimal.

2) At $129 and $179 retail what kind of margins will those numbers yield? The Banshee comparison may be helpful. Is a 40% retail markup right?

3) What will be the product mix? Given the success of V2 12MB or V2 8MB we I'm guessing the 3000 will outsell the 2000 by a 2 to 1 margin. Anybody have hard sales data on the V2 12 MB vs. the 8 MB to give us an idea of the market fragmentation?

4) I guess we'll be capped at 1.5M units for now. How many units do we think will be sold in Q299? Obviously 1.5M is the highest. What's the lowest? We sold about 750K units (roughly est.) during our crappy Q3 and 2M during a great Q4. I'm thinking 1M units to be the minimum with a large potential upside with volume sales of Voodoo3. I think Voodoo3 has much more potential than Banshee and they just sold 1.5M. Note that Diamond and Creative were each 10% of the business last quarter. I'm no longer worried about 3dfx using these guys for sellthough. 3dfx seems to be making the sale, not them.

5) By 2Q, STB should be 3dfx's only customer. I wouldn't say that's true the other way around. How much revenue do you guys think we can expect from STB's other sales? Specialized tech. looks to be about 4.6M quarterly according to STB 10K.

6) nVidias public offering has made 8.6M from an original 3.0M for a 5.6M gain. Nice! That's about triple 3dfx's Q4 net income.

7) Tax rates anyone?

8) Interest income?

I already have my estimates but I think using the borg collective might be much more insightful...not to mention more accurate.

Pat



To: Joe C. who wrote (10363)1/29/1999 6:07:00 AM
From: Michael G. Potter  Read Replies (2) | Respond to of 16960
 
Joe C. (& Pat),

3Dfx has publically stated that they're looking for a European partner for the V3 launch. They'll have some problems as it'll probably only be for the V3, but it will also be a good source of profits for the cardmaker so they probably will be able to strike a deal. With only one chosen, the gray market problem may not be too severe.

They also plan on selling many of the V3-2000's to motherboard makers in Taiwan. That means that Juarez's capacity will not be needed for those sales.

Finally, there are a fair number of contract board houses (Solectron, Jabil Circuit, and EFTC come to mind right away) that could be used if additional capacity is needed. The cost may not be all that more than making it themselves.

The real benefit to being able to make the 500K boards is the volume leverage you get.

Michael



To: Joe C. who wrote (10363)1/29/1999 10:52:00 AM
From: Daniel P. Dwyer  Read Replies (2) | Respond to of 16960
 
Doesn't 500,000 graphic cards per month equal 1.5 million per quarter? Do not understand your post. :)
Dan Dwyer