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: sylvester80 who wrote (57200)	10/9/2000 5:25:25 PM
From: Jim McMannis	Read Replies (2) \| Respond to of 93625

Slyvester80, Are you long Rambus? If so, what's you exit strategy...6 weeks? 6 months?, 1 year?...when do you sell? JIm

To: sylvester80 who wrote (57200)	10/20/2000 6:34:37 AM
From: Bilow	Read Replies (1) \| Respond to of 93625

Hi sylvester80; Re smsperling prices and the price structure of RDRAM... You wrote: "Street prices show a totally different picture. Considering that smsperling is abti-Rambus, I no longer trust his numbers as they don't jive with street numbers (the same pricewatch numbers he's supposed to be using). So posting some anti-Rambus individual personal web page no longer means anything. Pricewatch real street numbers mean a lot more." #reply-14542226 Oct 9, 2000 I thought I'd take the time to answer this completely. Of course you have no excuse to call smsperling anti Rambus. I think the guy's more or less even handed. If he has an agenda, he sure doesn't show it very much, instead, he just shows the numbers. But how does he get those numbers, and why do they differ so much from yours? The numbers you are quoting are lowest prices for the cheapest version of the memory, while the smsperling figures are average prices for all varieties of the memory sold at that speed grade. I suspect that the smsperling figures include ECC memory, and that, in particular, makes the numbers higher than otherwise. To test this, we can try computing smsperling's figures... Smsperling is taking the average over 64, 128 and 256MB memory sizes. Let's take a stab at computing the average price of 128MB memories. When you go to pricewatch: pricewatch.com I got prices as follows: 128MB RIMM prices, pricewatch, 10/19/00 PC600: 165 170 170 170 195 195 218 234 259 259 266 294 294 294 315 339 349 422 426 469 PC700: 169 170 175 179 195 195 299 315 319 322 322 339 344 349 354 383 411 469 472 499 569 PC800: 186 196 206 208 208 219 225 259 265 275 283 284 289 298 309 309 316 328 328 329 330 332 334 335 355 362 362 363 364 369 385 385 394 404 407 411 425 427 445 451 453 469 517 519 520 562 629 743 795 There were also 4 RIMMs where I did not immediately see the speed grade. Their prices were: 279 296 469 799, and I just ignored them. I get the average prices in each category as follows: PC600 $260.45 or $2.03/MB (20 samples) PC700 $326.62 or $2.55/MB (21 samples) PC800 $370.76 or $2.90/MB (49 samples) My average (which is over a smaller sample than smsperling uses) his figures and your best price figures compare as follows: Bilow smsperling sylvester80 Average Average Cheapest ------- ---------- ----------- PC600 $2.03 $2.068 $1.29 PC700 $2.55 $2.379 $1.32 PC800 $2.90 $2.740 $1.45 I suspect that smsperling takes into account the extra memory included in ECC modules, which makes his numbers lower than mine, but he also includes 64MB and 256MB modules, which raise his figures as compared to mine. But you can see that his averages are reasonable, and that they are considerably higher than the lowest price figures. It is of some interest to see what the best prices/MB are for 64 and 256MB modules. It turns out that 128MB parts usually had the cheapest lowest prices per MB today: Best for 256MB are: (all figures pricewatch, 10/19/00) PC600 $355 => $1.39/MB PC700 $435 => $1.70/MB PC800 $365 => $1.43/MB Best for 128MB are: PC600 $165 => $1.29/MB PC700 $169 => $1.32/MB PC800 $186 => $1.45/MB Best for 64MB are: PC600 $ 95 => $1.48/MB PC700 $105 => $1.64/MB PC800 $109 => $1.79/MB The big question with RDRAM, is just why is there such a big difference between average and best prices? What is going on here? At first, I thought that maybe some of the prices were horribly obsolete, but then I took a look at the RDRAM prices provided by a single vendor. I chose the vendor that provided the cheapest 128MB RIMM, the $165 unit that gave your $1.29/MB memory price figure. That company, AC Micro, sells 10 varieties of RDRAM, in various sizes, speeds, and one variety has ECC. Here's there table of prices for today: RAMBUS 184Pin ECC 128M 800MHZ $424 $3.313 RAMBUS 184Pin 128M 800MHZ $347 $2.711 RAMBUS 184Pin 32M 700MHZ $ 59 $1.844 RAMBUS 184Pin 256M 700MHZ $435 $1.699 RAMBUS 184Pin 64M 700MHZ $105 $1.641 RAMBUS 184Pin 32M 600MHZ $ 49 $1.531 RAMBUS 184Pin 64M 600MHZ $ 95 $1.484 RAMBUS 184Pin 256M 600MHZ $355 $1.387 RAMBUS 184Pin 128M 700MHZ $169 $1.320 RAMBUS 184Pin 128M 600MHZ $165 $1.289 (register) store.yahoo.com From this single manufacturer, you can see how it is that average RDRAM prices are so much higher than cheapest. So what is the valid way to measure prices, cheapest or average? This is a question I cannot answer. I use the smsperling data because it goes back in time for a good long while, and is provided to us all with very little effort. If I compute the ratio of cheapest RDRAM to cheapest PC133 SDRAM, I will get ratios lower than the ratio I am getting, but the time series is what is of interest to me, not so much the actual value of the ratio. Maybe there is a source of spot pricing somewhere, one that goes backwards in time. If I find it, I will recompute my numbers. Here's the ratio figures for cheapest memory of various configurations, versus the smsperling numbers, PC800/PC133: Ratio cheapest 256MB: $355/$156 = 2.28 128% Ratio cheapest 128MB: $165/$ 62 = 2.66 166% Ratio cheapest 64MB: $ 95/$ 31 = 3.06 206% Ratio smsperling: $2.740/$0.873 = 3.14 214% I'd love to convert to cheapest prices, but I don't have a time series for them. -- Carl