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.
Politics : Formerly About Applied Materials -- Ignore unavailable to you. Want to Upgrade?


To: Sun Tzu who wrote (30707)5/28/1999 11:47:00 AM
From: Jack Kunkle  Respond to of 70976
 
<P.S. When a commodity goes into a bear market, its price falls below its production cost until supply drops below demand for a few months. Ponder on this a bit. >

The collapse of Asia illustrates your point exactly. It is your statement <for a few months> that concerns me.

<I need the same amount of DRAM, but now I pay less>

I hope that we can get voice recognition into the mainstream so that you can use your Palm Pilot with more memory.

Regards,

Jack Kunkle :')



To: Sun Tzu who wrote (30707)5/28/1999 10:07:00 PM
From: semi_infinite   Read Replies (1) | Respond to of 70976
 
SunTzu,
<<So what does all this have to do with DRAM and SemiCaps? Well, in case of DRAM I have not been able to
justify using more than 256MB in my machine since 1994. Granted, it cost me a lot more to put all that
memory in my Sun workstation in '94 than it does today, but that is exactly the problem for DRAM makers;
I need the same amount of DRAM, but now I pay less. As for the semicaps, you can't really justify new
purchases to your customers if they are not making any money. So far the commodity chip makers are in a
self destructive upgrade cycle. Eventually (soon?) there will be only 3~5 players in each category.>>

OTOH, I have not touched, within the past year, any Unix box with less than 2G DRAM. Even then, the models are limited by memory and CPU. Even on machines with 16G and 8 R10010 CPU's, the models are only tolerable and definitely not real time. Simulations and visualizations are sequential and slow. Yes I am pushing the envelope but clearly, the models are still very simplistic compared to the real physical world. There are still much work to be done on software and hardware. Computing technology is not close to being satisfactory unless Excel spreadsheets and Word are all that's needed for further human advances.

When is it going to stop? Not in our life times. Not until we can analyze and model everything that's matters to our existence and fast enough for it to matter. And yes, I expect that I or my children will have affordable home 3D theater and gigabyte networks, and we will find a use for it.