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 : Wind River going up, up, up! -- Ignore unavailable to you. Want to Upgrade?

To: William Sheppard who wrote (3560)	8/27/1998 12:53:00 AM
From: Snowshoe	Read Replies (2) \| Respond to of 10309

WSJ article on internet appliances... Companies Gear Up for a Boom In Appliances That Use the Internet August 27, 1998 interactive.wsj.com

To: William Sheppard who wrote (3560)	8/27/1998 8:27:00 AM
From: bob	Respond to of 10309

Bill, They normalize to 20 Mhz. The benchmarks compare "computational efficiency" and not speed. That means, they compare something like execution time per megahertz, or effectively how many clock cycles it takes to get the job done. Not very meaningful except in an academic sense. What counts are not clock speed related items, but work completed per second at the desired system price point. If one were to factor in cost, one of the other chips would probably beat the PSC1000 because some were 8 bit chips that are real cheap. If you only need X amount of work for the task, then a processor that can do 3X at 2 times the price would have a better price performance, but would still be too expensive and 2/3 of the processing power would be going to waste. BTW, the PSC1000 sells for about $10 in volume. Bob

To: William Sheppard who wrote (3560)	8/27/1998 3:39:00 PM
From: Urlman	Respond to of 10309

Re: SwiftX for PSC1000/ShBoom more headers Aÿ << Previous message Aÿ >> N e x t message Aÿ Author: Aÿ Elizabeth D. Rather Aÿ authorAÿprofile emailAÿreply viewAÿthread postAÿnewAÿAúAÿpostAÿreply subscribe Email: erather@forth.com Date: 1998/08/26 Forums: comp.lang.forth, comp.arch.embedded view for bookmarking AÿAúAÿ text only Aÿ -------------------------------------------------------------------------------- Keith Wootten wrote: > > In article <35DB870B.DAF457DF@forth.com>, Elizabeth D. Rather > <erather@forth.com> writes > >The following is a brief commercial announcement: > > > >As of today, we have added the Patriot Scientific PSC1000 (ShBoom) chip > >to the list of products supported by our SwiftX product line. This is > >an extremely fast chip; benchmarks comparing it to other SwiftX targets > >are available at www.forth.com. > > Thanks for this, your benchmarks are very interesting. Could you > explain what is meant by clock speed normalisation - does this mean that > the chips were run at their maximum rated speed? What speed grade of > ShBoom did you use? Normalization in this case means we multiplied all the raw speeds by a ratio to get the speed that "would have been measured if they all ran at the nominal rate" (20 MHz in this case). For example, the 68332 was 16 MHz, and the ShBoom 80 MHz. One might say, "but what if I cranked the 332's clock up?" The answer is, if the 332 and ShBoom ran at the same clock rate, ShBoom would still be 4 times faster. Granted, it's not the only way to compare chips, but as a measure of processor efficiency, we find it interesting. > I know that the Patriot ShBoom was originally a Chuck Moore design - > given your association with CM, do you have any insights into its > development? He has been designing stack-based chip architectures since about 1980. We have worked with the Novix, Harris RTX, and now this one. They have a lot of things in common, but some differences. They're all very interesting, and very fast! Cheers, Elizabeth -- =============================================== Elizabeth D. Rather (US & Canada) 800-55-FORTH FORTH Inc. +1 310-372-8493 111 N. Sepulveda Blvd. Fax: +1 310-318-7130 Manhattan Beach, CA 90266 forth.com "Forth-based products and Services for real-time applications since 1973." =============================================== --------------------------------------------------------------------------------