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 : QUANTUM -- Ignore unavailable to you. Want to Upgrade?

To: Stitch who wrote (7326)	2/15/1998 9:29:00 PM
From: Chuck Bleakney	Read Replies (2) \| Respond to of 9124

Usually new technology is introduced when it makes sense to do so. Commonly what pushes one into new tech is that the old tech has been pushed hard already... and the new tech approach has lots of leeway for future growth but also has a learning curve and its associated risks. For instance, I know we considered PRML several generations prior to actually implementing in anything. Usually it is a better choice to cut your teeth on a product with very limited exposure. Sometimes this is done by knocking off a variant of a real product that incorporates the new technology and then either using it in house or possibly giving them to a very close client to evaluate for/with you. You learn the hard way not to try to introduce too many new things in any one generation... It'll bite you for sure.

To: Stitch who wrote (7326)	2/16/1998 2:31:00 AM
From: Gottfried	Respond to of 9124

Stitch, re design goal specificity: typically we used to get a capacity objective and then performed simulation to study trade-offs between tpi, bpi and data rate with error rate as the output. Error rate goals (especially hard error rate) are usually non-negotiable. Soft error rate is flexible, since more powerful correction codes can improve it to the required hard error rate (at a cost). The detection method is dictated by the signal resolution that results from the bpi. Resolution lower than about 60 % requires PRML, i.e. something other than peak detection. GM