To: Chip Anderson who wrote (7704 ) 9/26/1998 4:44:00 PM From: Michael G. Potter Read Replies (1) | Respond to of 16960
Paul Wilson apologized for taking a while to get back to me (a few days) because they still do not have an investor relations person on board. However, since I'm a Finance guy I was very happy to talk to someone who spoke my language. I greatly enjoy your posts because they deal with the market and the competition and I need to factor that in as well. If you want to read an almost perfect description of what happened with the Voodoo2 chipset sales, take a look at the second chapter of _The Fifth Discipline_ by Peter Senge. The beer game described in it is a great learning tool. It even has retail, wholesale (the board makers, maybe Ingram's), and the OEM (3dfx). I've seen it happen time and time again in the computer industry. It happens with disk drives (I've lived and invested through two boom/bust cycles there), microprocessors (Intel has just fought through an overstock of chipset in their channel), RAM, and PC end units. It is easy to rant and rave about how 3dfx should've seen it coming and planned differently, but they're also trapped in the PC retail system. Their customers WANTED the chipsets - they had very high demand and glowing reviews all over the web and gaming magazines. They wanted to capture as much of the sales as possible as soon as possible and there was little incentive to lower board prices as they were selling well at the release price. I think that the combination of steady OEM sales (remember that Dell, Compaq, and Gateway's sales are not flat all year, they are somewhat seasonal) with a high margin retail part is the best strategy for 3dfx. I also think that being fabless and not having to carry RAM and board inventory is crucial at their current size. It allows them to spend their cash on R&D and marketing instead of capital equipment and infrastructure. The only problem that I see with that model is who viable is the retail market going to be for upgrades in the future? A voodoo or voodoo2 board added to an existing 2d card makes sense. Banshee makes sense as an upgrade if you're a casual gamer. However, how much more power will you really need? (I know, they were saying that about the Apple II with 64K years ago <grin>.) I haven't joined any of the DD online Quake II games because I do not own Quake II. I do play games, but I'm much more of a Talonsoft customer (or Dark Omen for 3D games). My S3-based 2D card just doesn't cut it, but my voodoo2 with it is enough. I'm going to get a Banshee because they're very cheap, but my next new video card will be when I upgrade my computer in a year or so. I'll buy the best that I can afford and wouldn't expect to upgrade my video card from the one that comes with the system then as well. What I hope will happen is that 3dfx's high-end "retail" card also ends up being the high-end OEM systems. As they become the dominant video card supplier, they leverage Glide and their other technologies to maximize their profits and to stay ahead of their competition. 3dfx does become a "platform" and we all become very rich on their success. The chance of making very high returns on my investment summarizes why I'm still invested in 3dfx. I'm going to try to take a little time tomorrow to try and list out what I think the main threats are in the near future. Michael ps - sorry for the long post, I'm taking a break from helping my wife out in the garden right now and the longer I type the more rest I get.