"So RDRAM costs an additional 57% for 256MB, or 30% for 512MB. Nice that it's no longer double the cost, but to me that is still a significant markup. Anyone know approximately how much of that is due to
(a) economies of scale,
(b) manufacturing cost after accounting for (a), or (c) patent licenses?"
Samsung makes most of the RDRAM sold--even Kingston RIMMs have Samsung chips. So you have a bit of a monolopy supply issue (Elpida and Infineon also make some, but Samsung accounts for better than 80% of production, if memory serves.
RDRAM has a bigger die penalty, but this has shrunk (no pun...) as Samsung shifted to.13-um process and 300mm wafers. Production and testing costs are about the same as higher-end DDR. DDR-II will be just as expensive as RDRAM (but will be made in far larger quantities, and have more competition).
Rambus' royalty on RDRAM is 1% of the selling cost of the chips, so has memory prices have plummeted, so have Rambus' revenues.
My original point is that when I decide to buy a computer, I pick a platform, not a memory. RDRAM is more expensive than quality DDR, but it amounts to less than the cost of shipping or a video card upgrade. For the last 2 years, I've been only building AMD systems. But the 2.53GHz P-4 looks pretty nice (for once). However, if you want to talk about a price difference, the premium you will pay for the higher-end P-4s makes the cost of memory wet change on the end of the bar.
Athlon is about done. Hammer is in the wings, along with DDR-2. The problem with Athlon is that AMD's implementation of the EV6-bus spec. limits FSB to 133MHz, so adding memory bandwidth above what PC2100 can deliver makes no difference. I guess AMD could implement a 166MHz FSB/Memory bus, but why invest any more in validating an aging platform? Put it into Hammer, which has on-die memory controller(s), and can consume all the memory bandwidth you want to feed it.
Don't own any RDRAM (using an Athlon+DDR mostly) but the "RDRAM costs much more" argument is bogus. Compare what Samsung originals PC800 cost compared to brand name--not generic or House Brand--true CAS Latency 2.0 PC2100. It's a wash.
Quality PC2100 is frequently marketed as PC2400. On www.pricewatch.com, the difference between PC800 and PC2100 is $5 for 128MB.
I don't pick my platforms for the DRAM. I went with an AthlonXP 2100+ (1733MHz). But if I was going to buy a Pentium-4, I would use the i850E with PC1066.
will we never learn? Always wait for the 3rd try from Redmond.
By then it will have just as many titles as the X-box, not burn-up in your bookcase, and have a controller not sized for Shaq. It may also not be quite so ugly.
This is even true with their lawsuits-- lose at the Federal district and appeals level, win at the Supreme Court (unless the friendly Bush Justice Dept. throws in it's hand first.)
http://www.aceshardware.com/read.jsp?id=45000219 RDRAM at 1066MHz has the same latency as DDR266 (aka, PC2100) and PC133. A little lower, actually, assuming that you have 2 RDRAM channels, as with the i850 chipset.
RDRAM is not a choice for servers because sheer capacity, chip-kill, and interleaving are critical--think "uptime" not playing Quake 3!
RDRAM is limited to 32 devices per channel. Assuming that 512Mbit (non-ecc) chips are used, you could theoretically put 2GB on a single RIMM. With two channels, that give you 4GB. Plenty for workstations and desktops, not even in the same league for servers.
that people who say "It's a well known fact" have absolutely no facts to back them up.
except for me;-p
Like everything else, it depends on the benchmark. Intel wins the ones they cheat best on, and AMD wins the ones that are crappy games (like "Serious Sam").
Here's a not-so-well-known fact: By the time AMD gets to a REAL 2GHz processor (Barton), Intel will be at 3.0GHz, and it ain't looking back.
Handing out technology is pretty much the mindset that has prevailed in the schools up to now, and it doesn't work. Teachers don't have the time or resources to effectively use the Macs/PCs they have, and most schools have no competent SysAdmin--they usually draft a teacher and they grudgingly do it for a year.
Talk to your local elem. school teachers, esp. ones with diverse classrooms, and get a feel for their challenges. Then tailor a technology approach that meets their needs; if you can find ways to improve the effectiveness of teaching, you will help more kids.
I think that the ideal device would be a PDA that is so ubiquitous and inexpensive that it is not worth stealing, and no great loss if damaged or misplaced. Now, design a classroom around that device-- the child carriers the PDA home or to school, but at either place it can be plugged into the desktop and become part of a more capable, flexible learning system, with a keyboard, mouse, or other input device depending on the child's need.
The main initial benefit of the EDA (let's call it) is to provide local storage of homework assignments, calandar, contact, basic reference information, and statistics on use. This ensures that kids can't forget their textbook, or homework assignment, or spelling list, or worksheet, because the teacher can synch every EDA in the class at the end of the day.
Unplugged, the EDA stores key imformation for homework, reading, and studies-- much like a handspring or palmpilot. Plugged into class net or a home PC, it is the front-end of a more powerful networked information device.
More ambitiously, use the EDA and the wired classroom to give teachers instantaneous feedback on student interaction, learning, participation. The Teacher's workstation would enable them to scan the entire class during a writing or reading assignment, enable or disable instant messaging or polling, and even measure the time use and interaction on a class assignment, realtime, or record statistics that can be analyzed later. This would also make standardized testing much more consistent across school.
Stop with the "Apples for the Students" already. It is having little positive impact on learning, burdens teachers that are already overloaded, and amounts to little more than a toy that teachers use to distract students while the provide individual attention on handle admin duties.
...mostly because my Grandpa kept fscking up WinMe trying to download pr0n, God bless him! I needed an OS that could handle real user accounts, and spending $200 for Win2K/XP was not an option (esp on their older PC with 64MB). So I installed RH7.2, gave them a little tutorial, and they just love it. Never turn it off, and they don't miss Windows one bit.
I also burned a CD for Grandpa of some choice stuff, all in mpg format;-)
Oh yeah, I remember! That crippled OS with only one desktop, no themeability, that's a magnet for every script kiddy and virus writer on the planet!
sometimes, you just need to run MSOffice or Freecell, and that's why God made VMWare (only a supreme being could have come up with that hack).
Linux just isn't going to make it on the desktop I guess--it gives you too many choices, it's free, it's stable, it's secure, and it's virus and disk-fragmentation free. Who would want that?
Tell ya what-- build an AOL client interface on top of embedded linux, and sell the appliance for $200 (FREE with a new subscription to AOL!) and you'd be surprised how many people would run Linux. But they wouldn't know it was Linux.
Until Intel get's the next gen of Northwood with 133Mhz FSB out the door (they are saying Q2/2002 last I checked) then the i845 with DDR-SDRAM will have a 100Mhz FSB (quad-pumped magic not withstanding). That means that any DDR you use will be running at PC1600 speed. You will have to wait for P-4s with a 133MHz FSB (Quad-pumped to 533Mhz) for the P-4 to be able to really use PC2100 DDR-SDRAM.
Benches of AThlons with PC1600 were underwhelming.
The "50 million" figure thrown around in the press is based on a broad extension of the original American's with Disabilities Act definition.
Traditionally the term "disabled " referred to a segment of the population, perhaps 4 or 5 percent, handicapped by blindness, deafness, problems with mobility or mental incapacity. Crafters of the Americans with Disabilities Act, or ADA, expanded that definition dramatically to where it now takes in 50 million people, including the mentally or emotionally unstable.
No one seems to know exactly how the population to be covered by the ADA was, or is, measured, but that enormous estimate often is cited. Most of that number are mental cases. The psychiatric industry's 300 or so various diagnoses were used in structuring the ADA , meaning that symptoms such as bad moods or anxiety may be taken as indicators of an illness requiring accommodation by the employer. The ADA does rule out direct protection in cases of active users of illegal drugs, pedophiles, voyeurs, compulsive gamblers, kleptomaniacs, pyromaniacs and several other particularly antisocial sorts found in psychiatric diagnostic manuals.
The ADA is a civil-rights law; it's protections span the spectrum of American life because, like racial-discrimination laws, it attempts to level the playing field absolutely -- from the public water fountain to bus transportation to restaurant service to job equality and more.
Redhat launches a new version, and I can still get into their main ftp servers after the news is posted on/.
Clearly something paranormal at work here-- not surprising given the Roswell beta and Enigma release, plus all of the shite going down on the planet beyond the remote valley of geekdom.
I've had no problems compiling and using custom kernels under the 7.2 beta, aka "Roswell." You just need to patch the stable 2.4.12 tree for ext3 support, or use the -ac kernels, which already incorporate ext3 support. I've found 2.4.12-ac3 to be a very stable kernel with good memory and VM behavior. RedHat kernels are closer to -ac kernels right now, it would appear, than to Linus' main kernel tree.
For once, the retailers had product at launch--some of them even jumped the gun. So you can buy your boxed set RIGHT NOW all over this fair land (USA).
Which I told/. three days ago, and the submission was, of course, rejected. There are several unofficial mirrors that have been serving Enigma (7.2) isos for days, but you're on your own.
please moderate this whining drooler down. What a pile of eurofag drivel. reiserfs? It's in there. Xfs? Yup. KDE 2.2.1? Uh-huh. Please, catch a freaking clue before you spout off and look like the idiot you are.
shut your pie holes and go enlist in the US military. Soon to be manditory for citizenship anyway. You will learn to use a weapon, which could come in handy after we have defeated the Islamic masses and want to become a democracy again.
The ends this test of the emergency trolling system. Had this been an actual troll, it would have contained references to Nazism, profanity, and or the word "WH00T!"
or the latest version of the comparable VIA Apollo Pro chipset. Search pricewatch for the terms "FCPGA2" or "tualatin". Such board include:
Abit ST6/ST6-RA, Abit VH6-T Asus TUSL2, Soyo SY-TISU
Some board makers have been using the i815-B but don't mention it in the board specs ; marketing hasn't caught up with production yet.
BTW, the newest Celeron 1.2GHz uses the Tualatin core; no hardware prefetch and it uses a 100MHz FSB, but it can overclock to 1.5GHz with at most a minor voltage jump and standard cooling.
If you want to consider a real nightmare scenario for AMD, perhaps IBM, MicronPC, Tiny (UK) and other major OEMs who have dumped AMD in the past five months have taken a look at the upcoming Intel "Northwood" P-4 (.13-micron process die-shrink, 512kb L2 cache, support for 533MHz FSB) and compared it to the long-delayed Palomino and realized that the P-4 is about to beat AMD's brains out.
the price of the CPU is irrelevant next to the cost of supporting any system that uses a VIA chipset. Gateway's AMD systems are not competitive anyway--they never used DDR, and MicronPC kills them on price.
AMD is hurtin' fer certain, and you can tell just by looking at their prices-- they charge $100 for a 1.4GHz Athlon because they can't sell if for more in this market next to Intel's GHz onslaught. Hell, a Celeron 1.1GHz costs the same! If you don't think that GHz sells, AMD just confirmed it.
Wait until they release earnings (losses) for this quarter. They may have held onto market share (doubtful) but at a terrible price. On the other hand, Intel is keeping their ASPs high, but they must be paying $ out the ass for marketing--not just their own, but their "partners'" as well. I can't tell you how many times I've gotten a mailing from Gateway that was all Intel, and wondered who paid for the mailing. Ditto on those Sunday inserts from BestBuy and CompUSA that are all Intel, even though they have AMD systems on the shelves. Then there are incentives paid to sales staffs to push a specific product or line.
Sooner or later, this has to show up on Intel's bottom line, but they are masters at dressing the numbers, and offsetting operating expenses in their core area with stock sales or changes in depreciation.
Slashdot Mirror
...timed for the launch of the t-bred (AthlonXP on 13-um process).
Intel 2.66GHz and 2.6GHz P-4s are already on pricewatch (alternatively for 400Mhz and 533Mhz FSB). Only $633! Such a deal!
"So RDRAM costs an additional 57% for 256MB, or 30% for 512MB. Nice that it's no longer double the cost, but to me that is still a significant markup. Anyone know approximately how much of that is due to
.13-um process and 300mm wafers. Production and testing costs are about the same as higher-end DDR. DDR-II will be just as expensive as RDRAM (but will be made in far larger quantities, and have more competition).
(a) economies of scale,
(b) manufacturing cost after accounting for (a), or
(c) patent licenses?"
Samsung makes most of the RDRAM sold--even Kingston RIMMs have Samsung chips. So you have a bit of a monolopy supply issue (Elpida and Infineon also make some, but Samsung accounts for better than 80% of production, if memory serves.
RDRAM has a bigger die penalty, but this has shrunk (no pun...) as Samsung shifted to
Rambus' royalty on RDRAM is 1% of the selling cost of the chips, so has memory prices have plummeted, so have Rambus' revenues.
My original point is that when I decide to buy a computer, I pick a platform, not a memory. RDRAM is more expensive than quality DDR, but it amounts to less than the cost of shipping or a video card upgrade. For the last 2 years, I've been only building AMD systems. But the 2.53GHz P-4 looks pretty nice (for once). However, if you want to talk about a price difference, the premium you will pay for the higher-end P-4s makes the cost of memory wet change on the end of the bar.
Athlon is about done. Hammer is in the wings, along with DDR-2. The problem with Athlon is that AMD's implementation of the EV6-bus spec. limits FSB to 133MHz, so adding memory bandwidth above what PC2100 can deliver makes no difference. I guess AMD could implement a 166MHz FSB/Memory bus, but why invest any more in validating an aging platform? Put it into Hammer, which has on-die memory controller(s), and can consume all the memory bandwidth you want to feed it.
Don't own any RDRAM (using an Athlon+DDR mostly) but the "RDRAM costs much more" argument is bogus. Compare what Samsung originals PC800 cost compared to brand name--not generic or House Brand--true CAS Latency 2.0 PC2100. It's a wash.
Quality PC2100 is frequently marketed as PC2400. On www.pricewatch.com, the difference between PC800 and PC2100 is $5 for 128MB.
I don't pick my platforms for the DRAM. I went with an AthlonXP 2100+ (1733MHz). But if I was going to buy a Pentium-4, I would use the i850E with PC1066.
Nice chat with Diane Rehm yesterday on NPR.
Don't worry, this promo is free!
http://www.wamu.org/ram/2002/r2020422.ram
How can you not like a book about Golden Retrievers and Border Collies? Why kind of twisted, hate-filled bags of dirt haunt ./?
I mean, it's not like Jon's little advert blotted out crucial news like another 2.4.19 Rev xxx Linux kernel patch.
will we never learn? Always wait for the 3rd try from Redmond.
By then it will have just as many titles as the X-box, not burn-up in your bookcase, and have a controller not sized for Shaq. It may also not be quite so ugly.
This is even true with their lawsuits-- lose at the Federal district and appeals level, win at the Supreme Court (unless the friendly Bush Justice Dept. throws in it's hand first.)
this is not hard.
ps, substitue your running kernel version for "2.4.18" in the above.
http://www.aceshardware.com/read.jsp?id=45000219
RDRAM at 1066MHz has the same latency as DDR266 (aka, PC2100) and PC133. A little lower, actually, assuming that you have 2 RDRAM channels, as with the i850 chipset.
RDRAM is not a choice for servers because sheer capacity, chip-kill, and interleaving are critical--think "uptime" not playing Quake 3!
RDRAM is limited to 32 devices per channel. Assuming that 512Mbit (non-ecc) chips are used, you could theoretically put 2GB on a single RIMM. With two channels, that give you 4GB. Plenty for workstations and desktops, not even in the same league for servers.
that people who say "It's a well known fact" have absolutely no facts to back them up.
;-p
except for me
Like everything else, it depends on the benchmark. Intel wins the ones they cheat best on, and AMD wins the ones that are crappy games (like "Serious Sam").
Here's a not-so-well-known fact: By the time AMD gets to a REAL 2GHz processor (Barton), Intel will be at 3.0GHz, and it ain't looking back.
Intel didn't beat AMD. It just outran it.
Handing out technology is pretty much the mindset that has prevailed in the schools up to now, and it doesn't work. Teachers don't have the time or resources to effectively use the Macs/PCs they have, and most schools have no competent SysAdmin--they usually draft a teacher and they grudgingly do it for a year.
Talk to your local elem. school teachers, esp. ones with diverse classrooms, and get a feel for their challenges. Then tailor a technology approach that meets their needs; if you can find ways to improve the effectiveness of teaching, you will help more kids.
I think that the ideal device would be a PDA that is so ubiquitous and inexpensive that it is not worth stealing, and no great loss if damaged or misplaced. Now, design a classroom around that device-- the child carriers the PDA home or to school, but at either place it can be plugged into the desktop and become part of a more capable, flexible learning system, with a keyboard, mouse, or other input device depending on the child's need.
The main initial benefit of the EDA (let's call it) is to provide local storage of homework assignments, calandar, contact, basic reference information, and statistics on use. This ensures that kids can't forget their textbook, or homework assignment, or spelling list, or worksheet, because the teacher can synch every EDA in the class at the end of the day.
Unplugged, the EDA stores key imformation for homework, reading, and studies-- much like a handspring or palmpilot. Plugged into class net or a home PC, it is the front-end of a more powerful networked information device.
More ambitiously, use the EDA and the wired classroom to give teachers instantaneous feedback on student interaction, learning, participation. The Teacher's workstation would enable them to scan the entire class during a writing or reading assignment, enable or disable instant messaging or polling, and even measure the time use and interaction on a class assignment, realtime, or record statistics that can be analyzed later. This would also make standardized testing much more consistent across school.
Stop with the "Apples for the Students" already. It is having little positive impact on learning, burdens teachers that are already overloaded, and amounts to little more than a toy that teachers use to distract students while the provide individual attention on handle admin duties.
...mostly because my Grandpa kept fscking up WinMe trying to download pr0n, God bless him! I needed an OS that could handle real user accounts, and spending $200 for Win2K/XP was not an option (esp on their older PC with 64MB). So I installed RH7.2, gave them a little tutorial, and they just love it. Never turn it off, and they don't miss Windows one bit.
;-)
I also burned a CD for Grandpa of some choice stuff, all in mpg format
wow, look at that. You might want to talk about something you know, like selecting drool recepticles.
Oh yeah, I remember! That crippled OS with only one desktop, no themeability, that's a magnet for every script kiddy and virus writer on the planet!
sometimes, you just need to run MSOffice or Freecell, and that's why God made VMWare (only a supreme being could have come up with that hack).
Linux just isn't going to make it on the desktop I guess--it gives you too many choices, it's free, it's stable, it's secure, and it's virus and disk-fragmentation free. Who would want that?
Tell ya what-- build an AOL client interface on top of embedded linux, and sell the appliance for $200 (FREE with a new subscription to AOL!) and you'd be surprised how many people would run Linux. But they wouldn't know it was Linux.
Until Intel get's the next gen of Northwood with 133Mhz FSB out the door (they are saying Q2/2002 last I checked) then the i845 with DDR-SDRAM will have a 100Mhz FSB (quad-pumped magic not withstanding). That means that any DDR you use will be running at PC1600 speed. You will have to wait for P-4s with a 133MHz FSB (Quad-pumped to 533Mhz) for the P-4 to be able to really use PC2100 DDR-SDRAM.
Benches of AThlons with PC1600 were underwhelming.
The "50 million" figure thrown around in the press is based on a broad extension of the original American's with Disabilities Act definition.
Traditionally the term "disabled " referred to a segment of the population, perhaps 4 or 5 percent, handicapped by blindness, deafness, problems with mobility or mental incapacity. Crafters of the Americans with Disabilities Act, or ADA, expanded that definition dramatically to where it now takes in 50 million people, including the mentally or emotionally unstable.
No one seems to know exactly how the population to be covered by the ADA was, or is, measured, but that enormous estimate often is cited. Most of that number are mental cases. The psychiatric industry's 300 or so various diagnoses were used in structuring the ADA , meaning that symptoms such as bad moods or anxiety may be taken as indicators of an illness requiring accommodation by the employer. The ADA does rule out direct protection in cases of active users of illegal drugs, pedophiles, voyeurs, compulsive gamblers, kleptomaniacs, pyromaniacs and several other particularly antisocial sorts found in psychiatric diagnostic manuals.
The ADA is a civil-rights law; it's protections span the spectrum of American life because, like racial-discrimination laws, it attempts to level the playing field absolutely -- from the public water fountain to bus transportation to restaurant service to job equality and more.
Redhat launches a new version, and I can still get into their main ftp servers after the news is posted on /.
Clearly something paranormal at work here-- not surprising given the Roswell beta and Enigma release, plus all of the shite going down on the planet beyond the remote valley of geekdom.
I've had no problems compiling and using custom kernels under the 7.2 beta, aka "Roswell." You just need to patch the stable 2.4.12 tree for ext3 support, or use the -ac kernels, which already incorporate ext3 support. I've found 2.4.12-ac3 to be a very stable kernel with good memory and VM behavior. RedHat kernels are closer to -ac kernels right now, it would appear, than to Linus' main kernel tree.
For once, the retailers had product at launch--some of them even jumped the gun. So you can buy your boxed set RIGHT NOW all over this fair land (USA).
/. three days ago, and the submission was, of course, rejected. There are several unofficial mirrors that have been serving Enigma (7.2) isos for days, but you're on your own.
Which I told
please moderate this whining drooler down. What a pile of eurofag drivel. reiserfs? It's in there. Xfs? Yup. KDE 2.2.1? Uh-huh. Please, catch a freaking clue before you spout off and look like the idiot you are.
RH 7.2 is released, and the FTP servers are not /.-ed yet:
ftp://ftp.redhat.com/pub/redhat/linux/7.2
A packet tsunami from /. readers with broadband should give the RIAA a whole new understading of the concept of "smother."
shut your pie holes and go enlist in the US military. Soon to be manditory for citizenship anyway. You will learn to use a weapon, which could come in handy after we have defeated the Islamic masses and want to become a democracy again.
The ends this test of the emergency trolling system. Had this been an actual troll, it would have contained references to Nazism, profanity, and or the word "WH00T!"
or the latest version of the comparable VIA Apollo Pro chipset. Search pricewatch for the terms "FCPGA2" or "tualatin". Such board include:
Abit ST6/ST6-RA, Abit VH6-T Asus TUSL2, Soyo SY-TISU
Some board makers have been using the i815-B but don't mention it in the board specs ; marketing hasn't caught up with production yet.
BTW, the newest Celeron 1.2GHz uses the Tualatin core; no hardware prefetch and it uses a 100MHz FSB, but it can overclock to 1.5GHz with at most a minor voltage jump and standard cooling.
If you want to consider a real nightmare scenario for AMD, perhaps IBM, MicronPC, Tiny (UK) and other major OEMs who have dumped AMD in the past five months have taken a look at the upcoming Intel "Northwood" P-4 (.13-micron process die-shrink, 512kb L2 cache, support for 533MHz FSB) and compared it to the long-delayed Palomino and realized that the P-4 is about to beat AMD's brains out.
Or not.
the price of the CPU is irrelevant next to the cost of supporting any system that uses a VIA chipset. Gateway's AMD systems are not competitive anyway--they never used DDR, and MicronPC kills them on price.
AMD is hurtin' fer certain, and you can tell just by looking at their prices-- they charge $100 for a 1.4GHz Athlon because they can't sell if for more in this market next to Intel's GHz onslaught. Hell, a Celeron 1.1GHz costs the same! If you don't think that GHz sells, AMD just confirmed it.
Wait until they release earnings (losses) for this quarter. They may have held onto market share (doubtful) but at a terrible price. On the other hand, Intel is keeping their ASPs high, but they must be paying $ out the ass for marketing--not just their own, but their "partners'" as well. I can't tell you how many times I've gotten a mailing from Gateway that was all Intel, and wondered who paid for the mailing. Ditto on those Sunday inserts from BestBuy and CompUSA that are all Intel, even though they have AMD systems on the shelves. Then there are incentives paid to sales staffs to push a specific product or line.
Sooner or later, this has to show up on Intel's bottom line, but they are masters at dressing the numbers, and offsetting operating expenses in their core area with stock sales or changes in depreciation.