Silicon Investor (SI) -- The First Internet Community

STOCKTALK

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 : Intel Corporation (INTC) -- Ignore unavailable to you. Want to Upgrade?

To: Mary Cluney who wrote (142580)	9/1/2001 12:58:34 AM
From: BelowTheCrowd	Read Replies (3) \| Respond to of 186894

>computers that are more than three years old, have and will have a difficult time keeping up swith software development I don't think I disagree with you on much, except your insistence that there is somehow a 3 year lifetime on a PC in "normal use." There have been times when even a 2 year old PC became quickly obsolete, there have been times where software development took longer to catch up. Right now, we appear to be in one of the latter times. As I said, we are achieving acceptable levels of performance from most of our desktops even on some three year old machines running the latest software, up through XP. > What is acceptable for a a worker to wait 25 - 35 seconds to get a response from his Excel spread sheet today , is not going to be acceptable in the next two years < Even with that three year old PC, you're not waiting 25-35 seconds for a spreadsheet to caluculate. Not even the more complex models used by option traders and other people who run fairly difficult models (And for those types of people I'd generally recommend an upgrade in any case). Yes, there are economists and others who try to model entire economic systems who probably run into long waits, and I probably woudn't recommend a "run of the mill" PC for them. That's assuming, or course, that the spreadsheet is all contained in memory. If you're running a query or calculation of any kind that requires disk access, then you may very well have long waits. But upgrading to a faster processor doesn't help one bit. Unfortuantely, disk-drive access times have not followed Moore's law. If you're doing anything disk intensive, a faster processor buys you nothing. That's one of the problems I think Intel has to deal with, and it's the reason I always ignore all the processor benchmarks everybody here wants to talk about endlessly. The processor doesn't operate in a vacuum, it's at the core of a whole bunch of other components, many of which severely restrict practical-world performance. >giving employees the capabilities only of only what he currently needs is equivalent to a football quarterback throwing the ball to where the receiver is at the time he releases the ball, rather than to where he needs to get to.< Where he "needs to go" is determined by the available software. Right now, giving him a processor five times faster doesn't get him anywhere, as he's using the same old software. When the newer software is available, I'm the first in line to say we need to upgrade the hardware to accommodate it, but not before. And for several reasons: * I can't make any reasonable recommendation about what to upgrade TO. I have no idea what the future software may require of the hardware. Even if I specify a top-end machine, I may find myself in 12 months going back and saying, "oops, guess we have to upgrade again." * I can't justify the expense of buying and installing the new hardware without ANY idea of what benefits they will bring. They won't do much for current software, and I can't even begin to quantify what unknown future software will bring. * The people who own the purse-strings have come to realize that hardware gets cheaper over time. Even if I know that I'll need 1.5Ghz machines all around in 18 months, they won't want to upgrade now. They'll wait a year for those 1.5Ghz machines to get a lot cheaper. These are rough times for business. Anything without a documentable benefit and ROI just isn't getting done. And right now there's not much of an ROI to hardware upgrades in most cases. It'll change, but I don't expect it to change too fast. Note, I'm not saying that I would run out and look for ancient machines when we need to buy additional stuff. When we need to buy new, I generally recommend something in the middle of the range. Certainly I can justify an extra $50-100 when I already need to buy something. Very different from trying to justify outright replacement of every box in the house, "just in case." mg