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 : Rambus (RMBS) - Eagle or Penguin -- Ignore unavailable to you. Want to Upgrade?

To: Dan3 who wrote (32778)	10/25/1999 7:56:00 AM
From: Bilow	Read Replies (1) \| Respond to of 93625

Hi Dan3; Well, I finally went and looked at the SharkyExtreme benchmarks, and here is what I saw. First of all, I should note that I am not a high-level guy like you. I design things from the gates up to the chip level. I don't spec systems or anything like that. A man's got to know his limitations, and one of my limitations is that I don't know much about high modern PC architecture. That means I have to simplify things before I can understand them. The question is how much does RDRAM help? Unfortunately, since RDRAM and SDRAM take different chip sets, this isn't easy to figure out. But we can look at the combination of RDRAM+i820@133MHz and SDRAM+440BX@100MHz, each with a 600MHz Katmai processor: In addition to comparing RDRAM to SDRAM, I am also comparing the coppermine to the previous silicon. First column is 100MHz SDRAM, Katmai, 440BX. Second coumn is 133MHz RDRAM, Katmai, i820. Third column is 100MHz SDRAM, Coppermin, 440BX. Katmai Copper ------------ mine 440BX i820 440BX /100 B/133 E/100 Speed Improvements -------------------- ------------ BM SDRAM RDRAM Copper RDRAM Copper ---- ----- ----- ------ ----- ------ 10.1 67.4 71.9 69.1 6.7% 2.5% 10.2 9570 9450 9440 -1.3 -1.4 11.1 60.1 64.8 63.2 7.8 5.2 11.2 84.0 88.9 94.2 5.8 12.1 12.1 43.9 44.4 56.0 1.1 26.1 12.2 3055 3050 3240 -0.2 6.1 13.1 224 244 268 8.9 19.6 13.2 78.9 79.4 86.2 0.6 9.3 Delay -------------------- BM SDRAM RDRAM Copper ---- ----- ----- ------ 14.1 110.9 97.5 93.3 13.7 18.9 14.2 318.6 279.1 279.1 14.1 14.1 RDRAM Copper ----- ------ Average Improvements: 5.7% 11.2% You noted that some of the 5.7% RDRAM improvement will disappear when the bus on the BX design is changed from 100 to 133 MHz, and the SDRAM is switched to PC133 (with CAS delay of 2, presumably). I believe that this is true, but some of the benchmarks listed above are CPU only, and will not change much. As an AMD investor, my favorite quote was: "In our own testing we achieved a top speed of 840MHz with our Pentium III 733 Coppermine CPU, although stability was barely practical for the average user to endure and it was only possible to hit on a specially unlocked engineering sample CPU provided by a friendly OEM system builder." The Athlon still looks like a winner to me. -- Carl

To: Dan3 who wrote (32778)	10/25/1999 9:08:00 AM
From: Bilow	Respond to of 93625

Hi Dan3; Tom's hardware has a direct comparison between RDRAM and VC SDRAM, both on 133MHz FSB: www6.tomshardware.com Direct comparison of preliminary Intel i820 and VIA Apollo Pro 133+ platform How much 'worse' is VIA's Apollo Pro running with PC133 virtual channel SDRAM than Intel's great i820 with 400 MHz direct RDRAM really? We ran Coppermine on both: <<<graphic>>> You can see that the differences aren't big. In office applications i820 has a slight edge of up to 3% over the Apollo Pro 133+, but in 3D-games the VIA chipset scores even better. Thus there' shouldn't be too much to worry about using an Apollo Pro 133+-platform with Coppermine. www6.tomshardware.com We should note that VC SDRAM doesn't have the die penalty of RDRAM, and probably runs very close to the speed of PC133. What I would like to know is whether that VC SDRAM is 2 or 3 cycle delayed. My guess is that it would be 3 cycle delay, and the newer stuff will be faster... For the non-technical types out there, the conclusion is that RDRAM isn't worth the trouble. -- Carl