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 : C-Cube -- Ignore unavailable to you. Want to Upgrade?

To: DiViT who wrote (35281)	8/19/1998 12:45:00 PM
From: E. Mark	Respond to of 50808

The beef might be that we're still waiting patiently at 18. Not a huge problem for a long term investor, but we've seen some self proclaimed long termed investors fly the coop at the slightest scent of trouble. Perhaps that bird is not so rare after all though. Instead, he wears the feathers of a chicken while singing the song of a crow.

To: DiViT who wrote (35281)	8/19/1998 1:12:00 PM
From: Stoctrash	Read Replies (2) \| Respond to of 50808

<<Management has vision. Has executed that vision, continues to execute the vision. So where's the beef?>> In my studies and travels I've come across one thing that every analyst wants to see....and that's Revenue Growth. Bottom line is nice, Asset growth is nice...but everyone HAS to see Rev. Growth or they don't touch. See how few big names follow CUBE.... this lack of top line is why, IMO. So...the BEEF is in the Revenue numbers for Divi-Dog, without it she's gunna still be in the same ballpark.

To: DiViT who wrote (35281)	8/19/1998 9:29:00 PM
From: Ron Mayer	Read Replies (1) \| Respond to of 50808

VLIW vs SMP vs SIMD David wrote: ' [quoting someone] "The future is all about parallelism, and there are easier ways to get parallelism than to build a complex [VLIW] machine ... In short, it's better to have two simple machines than one complex one. I think it's absolutely clear we will move to SMP on a chip." ... Furthermore, Cube tossed out the VLIW based encoder last year in favor of a Sparc based one.' But is it SMP? Or 'just' fast. If not, iCompression may be one of the more interesting MPEG companies doing SMP-on-a-chip: icompression.com : "Two on-chip digital signal processors. They execute Java(TM) Virtual Machine instructions." This is interesting because the well-defined multithreading specification seems to have put a lot of consideration into SMP-systems (did you ever read the parts of the JVM spec that refer to read,load,use,assign,and store operations and how they interact with shared "main memory" and thread-specific "working memory"). Of course it's arguable of whether or not Java has enough disadvantages in real-time-systems to make up for that benefit. . I believe both SMP and VLIW techniques have very interesting potential, and which is better may be application dependant: Note that Texas Instruments quite effectively uses both approaches: ti.com "TMS320C6x DSP ...(VLIW) architecture to achieve high performance through increased instruction-level parallelism" ti.com "TMS320C80, is a single-chip, parallel processor that can be used for applications such as real-time audio/video processing..." . However for most applications I'm still not convinced either SMP-on-a-chip or VLIW will be more cost-effective than RISC or DSP with SIMD extentions (like Intel's MMX or Sun's VIS) and special purpose hardcoded logic where extreme perfomance is needed. Anyone have any differing opinions? Anyone know which companies favor which kind of architure (VLIW, SMP-on-a-chip, SIMD, etc)? These are my guesses: Phillips: vliw eet.com C-Cube: vliw first, now risc(sparc)+hardcoded-logic? Message 5537340 iCompression: smp(2-java-cores)? icompression.com Chromatics: vliw first, now changed to smp Message 5537150 IBM: ? TI: vliw AND smp products targeted for digital video eetimes.com ti.com Intel RISC + DSP-like, dual-issue media processor eet.com Equator vliw Message 4618302 Innovacom :-) systolic arrays? Message 797091 Any corrections/additions?