C-Cube Message Board - Msg: 2904461

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

CUBE 36.62

-0.1%

Nov 14 9:30 AM EST

Public Reply Prvt Reply Mark as Last Read File

Previous 10 Next 10 Previous Next

To: BillyG who wrote (26278)	12/8/1997 6:26:00 AM
From: Carnac	Read Replies (1) of 50808

>It sounds like RamBus will do the trick. Is SDRAM >fast enough? What about DDR-SDRAM? Embedded memory >would be a clean solution. Today's Concurrent RDRAM or 100 MHz SDRAM would be sufficient. In HDTV (MPEG-2 High Level) there are 2+ million random blocks being read from frame buffers. The transdecoder simply reduces the size of these blocks by about 4:1, but you still have as many blocks. This means a page miss every 500 ns on average. Of course block fetching is only about half the DRAM traffic, so you need to sustain a page miss every 250 ns or so. This 2 million blocks is similar to 2 million triangle texture operations you see in a 3-D engine. This is why some 3-D chip vendors say they can do HDTV decode in 1999 (not just MPEG-2 Main Level). Sadly for Cube, as per Intel's settop PC they are being squeezed by consumer-placating software-hardware hybrid decoders on the PC (and will continue to be so even into HDTV future), low cost encoder chips for consumer electronics, mass DSP array on the high end of encoding (to a lesser extent), and dirt-cheap decoders on the settop box front (although I think the settop box will VERY soon go the direction of hybrid software-hardware too... c.f. Chameleon chip from SGS-Thomson and other media processors being developed at Sony (Zanker division again)). Also for the low cost PC, there should soon be x86 chips with significant media acceleration such as Cyrix's MXi that should not have trouble handling any kind of DVD bitstream (although Mediamanics is bad at execution). For Cube's sake, the Intel 740 better not have any video acceleration on-board (you can almost rely on Intel to do the stupid thing... like their MMX-1 instruction set: a truly brain-dead solution). Alex needs to get damn agressive.

Report TOU Violation

Share This Post

Public Reply Prvt Reply Mark as Last Read File

Previous 10 Next 10 Previous Next