Gottfried, Sorry I didn't connect with you and others who might have attended the Magnetics Society meeting last night. If you were there, you might add something to my ramblings about what was said. I started writing this on the Innovex thread, so some of my comments are about head interconnects.
Anyway, I was at a meeting last night in Santa Clara for magnetics society people. The speaker was Mark Geenan from Disk Trend. Picture is still weak and not looking good for 98 recovery. He says the inventory issues are not going to change. Demand is slow. He said in the next 3 years we will see a shrinking component balance as new drives try to stay ahead of the falling ASP curve.
He said, and we all know this in our hearts, there is no need to keep adding capacity to the desk top. Asking engineers at leading PC makers what their customers really need, they all agree about 2 gig is enough. OEMs are adding capacity more because they can, and without raising cost, rather than due to user demand. It's a selling point that one model has a bigger drive, but capacity is well beyond need now.
They never mentions head interconnects at these meetings, but they said we would see less platters and heads over the next 3 years. The graph is gradual, not dropping off a cliff, but it's certainly not going up. On the bright side, the glut in head and media is so big that the incentive to cut components is not as high as it could be. He said most media makers are running at 15 to 20 percent of capacity. Billions over spent on media. Be glad you're not in Komag.
In the last 18 months, 10 million sq ft of manufacturing capacity was added.
Fujitsu projected to be a top 3 player in coming years. Samsung doing very well, and watch out for Hitachi
Material cost of HDD = 80%+/-. Cats reductions can only come from materials reduction. This will cause problems throughout the whole supply chain.
On the bright side, Geenan projects we are at bottom now, but expects a SLOW recovery. Again, there are no killer apps out there to drive demand. He does not expect any deice other than the PC to drive sales in the near term. Maybe 2 or 3 years out, home entertainment will be a big demand driver. Bandwidth will increase demand, but again, with high speed interactivity why store at home? He expects digital photography to increase demand this year.
IBM is making a mockery of portable drive market. Maybe Toshiba will not survive? In fact, he sugessts there has to be mergers and failures going forward to make the recovery possible.
He discounts the need for adding drives to make y2k transition. He does not see great demand from enterprise saving the day. He did raise the ratio of drives from 1.6 sold for every PC to 2, which means one in the box and one on the net, VPN, etc. I think he's missing a point here, but then he's mostly interested in large players and enterprise has got just as strong a pricing pressure issue as desk top, so any unit increases here are just threading water for HDD makers as ASP's fall.
I asked about drives for the low end, sub$600 and less. My question was how cheap could they go, assuming a floppy can be sold for about $15, what's the down and dirty price for a drive. He said $80, maybe as low as 50, but he really hoped not because it wouldn't be good for the industry.
All in all, it was not a very bullish session. As you can imagine, the statements from Mr Geenan brought some concerned responses from these engineers. Afterall, they spend their whole day trying to boost areal density, access speeds, and quality, but in the end they just can't do enough to make a difference anymore.
The upshot is HHD companies have to become better assembliers, increasing inventory turns. The next year will be a hard one. Geenan even suggested this whole scene was just delayed by the SEG Conner merge. In fact, the problems SEG has had digesting this merge has opened up opportunity to others which are no longer driving growth. Be greatful that Asia is having trouble because it could mean that the slash and burn tactics of Korea are not going to be so sevre. Still, expect big pushes from Fujitsu, who is not a Hutch or INVX customer.
Well, I'm rambling. It was an interesting evening, but not one to make you want to be a storage stock holder. Still, if he is correct, we are at bottom. I personally believe Geenan misses some points, but in general he's saying a lot of things that make sense.
****
I also spoke to a head engineer sitting next to me. He was with a Japanese disk drive maker. I asked him about what they were doing with connects. He said they all want TSA for GMR, but that HTI is not able to meet demand. His company must scale back plans for pico to nano sliders until this HTI supply issue can be worked out.
I asked him what he thought of HIF. He new about it, but didn't see it as a first choice. Probably a slowdown of pico head introduction will give INVX some added breathing space followed by a chance to win something with bonded HIF, but this is a battle to win back the high end. Thank goodness HTI does not have the manufacturing skills of INVX or they would have really made a big dent in INVX.
He said Quantum was a priority customer for Hutch and all others were getting what was left of capacity which is nothing at the moment. In the meantime, they can still buy TSA choices from NHK but these are also in short supply.
He suggested there are several good GMR programs coming out of Japanese head makers. APM is totally out of the picture for heads, but they have a strong HSA business. Sure looks like RDRT is having trouble.
He said he thought SEG was currently getting only 20% yields on their in house heads.
Regards,
Mark |
