Just wondering if there will be some RDRAM systems available when the 800mhz fsb technology comes out, or will there be only DDR systems available.
Also, is this coming out soon? I'm debating whether or not I should wait for it.
Thanks
Printable View
Just wondering if there will be some RDRAM systems available when the 800mhz fsb technology comes out, or will there be only DDR systems available.
Also, is this coming out soon? I'm debating whether or not I should wait for it.
Thanks
Intel will only release DDR chipsets for the prescott.Maybe other manufacturers (like SiS) will have RDRAM solutions.However RDRAM is a dying technology and i don't recommend it
This maybe of interest to you, I don't think I would wait for a RDRAM chipset though. Intel will release Canterwood and some of the Springdale chipsets will support DDR400 (and 800fsb). I would pass up Granite bay, and wait for one of these if your not in a hurry.Quote:
Originally posted by KANJI
Just wondering if there will be some RDRAM systems available when the 800mhz fsb technology comes out, or will there be only DDR systems available.
Also, is this coming out soon? I'm debating whether or not I should wait for it.
Thanks
On RDRAM:
http://www.xbitlabs.com/news/story.html?id=1044029944
Little cut and paste on the Canterwood and Springdale below, also from Xbits:
New Intel Chipsets: 800MHz Bus and DDR400 in Q2 2003 [9:43 pm] Rat
The launching of new CPUs with 800MHz bus scheduled for Q2 pushed Intel to introduce some changes to their chipset plans as well. By the time Pentium 4 with 800MHz bus is out, the company should be ready to offer new chipsets supporting them and offering adequate memory bus bandwidth. As you know, the bandwidth of the 800MHz Quad Pumped Bus equals 6.4GB/sec.
Almost together with the launching of the new Pentium 4 with 800MHz bus, Intel was planning to announce two new chipsets aka Springdale and Canterwood. However at first they were not intended for the faster 800MHz bus and were supposed to support only dual-channel DDR266/DDR333 SDRAM. Now Intel changed its initial plans in this respect accordingly.
Both chipsets, Springdale and Canterwood, will acquire 800MHz bus support. To be able to provide the desired memory bus bandwidth, both chipsets will also get the support of dual-channel DDR400 SDRAM, which bandwidth is most likely to equal the needed 6.4GB/sec.
As you remember, Intel used to vote against the DDR400 SDRAM standard, however now they had to change their attitude towards this memory type. But the DDR400 standard Intel is going to support will still be different from the currently existing “unannounced” DDR400 standard.
As I found out, DDR400 SDRAM will be supported by Intel chipsets only in case 800MHz system bus frequency is used. As soon as the system bus in your computer is set to 400/533MHz, the platforms built on the new Intel chipsets will automatically offer only two options: DDR266 or DDR333 SDRAM.
The company expects the “true” DDR400 SDRAM to appear in the market in Q1 2003. By January the major memory makers and suppliers should have their products ready for sampling. So far Intel has already sent out the preliminary DDR400 rev 0.9 spec to the memory guys, who claim to have the engineering samples meeting this specs already at hand.
Speaking about the chipsets coming out in Q2 next year to support the new Pentium 4 processors with 800MHz system bus, we would like to mention their following features:
Canterwood. The chipset is positioned as a high-performance solution, which should come to replace i850E and Intel E7205. It will support Pentium 4 with 800/533MHz bus and Hyper-Threading technology. Two DDR SDRAM channels will be capable of working with DDR400/DDR333/DDR266 with ECC and also Turbo Mode (a special work mode increasing the performance). Also the chipset will support AGP 8x interface and will go with an ICH5 South Bridge featuring SerialATA-150 support.
Springdale-PE. This mainstream solution will come to replace i845PE. This core logic will support Pentium 4 and Celeron CPUs with 800/533/400MHz bus and dual-channel DDR400/DDR333/DDR266 SDRAM. No ECC will be supported. Also the chipset will boast AGP 8x interface support and will go with an ICH5 South Bridge featuring SerialATA-150.
Springdale-P. This is a cut-down version of the previous chipset differing from Springdale-PE by the absence of support for CPUs with 800MHz system bus and DDR400 SDRAM.
Springdale-GE. This chipset will replace the today’s i845GE. It is none other but the already described above Springdale-PE but with the integrated graphics core.
Moreover, we managed to learn the official names of the upcoming Canterwood and Springdale chipsets. Canterwood will be launched as i875P, Springdale-PE will be called i865E, Springdale-GE will be announced as i865GE and the “lite” Springdale-P – as i865P.
Intel expects the chipsets supporting processors with 800MHz bus to make about 25% of all shipments in Q2, which will grow up to 50% in Q3. By the end of next year 65% of all Intel chipsets will support the new 800MHz processor bus.
Note that Springdale and Canterwood are also compatible with the future 90nm Prescott core due in the end of next year.
And then we have the SIS655:
http://www.anandtech.com/mb/showdoc.html?i=1781
-
Marv: Thanks for all that great information!
Yiotis: I should wait for the new stuff, but it doesn't look like I have a choice.
Basically, I need to have something in 1 month. The most I can wait is till the middle of March.
Do you guys think the chipsets with the 800MHz FSB and corresponding DDR memory will be out by then?
no, aprilQuote:
Originally posted by KANJI
Marv: Thanks for all that great information!
Yiotis: I should wait for the new stuff, but it doesn't look like I have a choice.
Basically, I need to have something in 1 month. The most I can wait is till the middle of March.
Do you guys think the chipsets with the 800MHz FSB and corresponding DDR memory will be out by then?
rdram will be back on top with the release of 1200. has 9.??GB/sec of bandwidth. from what i've seen on this site, i too thought it was dying, but it isn't. Rambus is still in full swing.
Dont you mean 1600? for an 800MHz FSB.Quote:
Originally posted by Philip
rdram will be back on top with the release of 1200. has 9.??GB/sec of bandwidth. from what i've seen on this site, i too thought it was dying, but it isn't. Rambus is still in full swing.
the next rdram is the 1200 version. now we have 1066, next is 1200.
PC1200 has 4.8GB/s of maximum bandwidth, IIRC. PC1200 is not available, that I am aware of, and no existing chipset supports PC1200. Rambus "full swing", as you call it, depends on chipset makers other than Intel at the moment. This of course could change.Quote:
Originally posted by Philip
rdram will be back on top with the release of 1200. has 9.??GB/sec of bandwidth. from what i've seen on this site, i too thought it was dying, but it isn't. Rambus is still in full swing.
Rambus is still fully pursuing and expanding their company and rimm technology. like i said, the next release of their ram is 1200. stated by their website. and yes, there is no support for their ram right now from intel, however, like you said, SIS will most likely produce boards for this ram. this stuff will be killer. however, it won't be available in mass till fall of this year.
rambus is fighting the inevetiable. even their parent company intel is moving to phase it out, and yes , it does keep fighting back, but that only ensures existence for a fnew more months. if rdram chipsets become scare, youll see manufactureres stop making rdram, driving up the price, so no matter what the speed, itll be very expensive, and drive most consumers away, and thatll be the end of rdram. i hope they manage to stay in there cause they do create some good stuff, but they dont seem to have a single vote of confidence in the iundustry from other big companies.
Dual channel PC1200 RDRAM has 4.8 GB/sec of bandwidth. Dual-channel PC3200 DDR (as used in the canterwood and springdale chipsets) has 6.4 GB/sec of bandwidth. The only way RDRAM could even match the FSB speed of the 800 MHz P4's (also 6.4 GB/sec) is to go quad-channel. SIS is develping a quad channel solution, but with 4 controllers, it's going to be much more expensive than DCDDDR and likely offer no benefit.Quote:
Originally posted by Philip
Rambus is still fully pursuing and expanding their company and rimm technology. like i said, the next release of their ram is 1200. stated by their website. and yes, there is no support for their ram right now from intel, however, like you said, SIS will most likely produce boards for this ram. this stuff will be killer. however, it won't be available in mass till fall of this year.
the next rimm is 1200mhz four channel ram providing 9.6gb/sec bandwidth. from www.rambus.com :D
edit: now imagine that dual channel.:D :D :D
wait, is that dual channel????
Rambus is in full swing? Uh oh, who are they suing now? :DQuote:
Originally posted by Philip
rdram will be back on top with the release of 1200. has 9.??GB/sec of bandwidth. from what i've seen on this site, i too thought it was dying, but it isn't. Rambus is still in full swing.
If you need to upgrade in a month check out the 845E and PE chipsets with Corsair, Kingston or Crucial memory.
Actually, it's quad channel - four memory controllers instead of 2. Dual channel PC1200 RDRAM only has 4.8 GB/sec of bandwidth, not even as much as dual channel PC2700 DDR. I have a hard time believing that quad channel RDRAM will really be much of a competitor - the performance may be there, but the price will be high.Quote:
Originally posted by Philip
the next rimm is 1200mhz four channel ram providing 9.6gb/sec bandwidth. from www.rambus.com :D
edit: now imagine that dual channel.:D :D :D
wait, is that dual channel????
Price will not be high.. DDR is now more expensive then Rd at my Canadian vendor.. with 1066 edging out in performance.. price is not an issue anymore.. and with a quad controler.. the latency would be SWEET... Bandwith is irrelevant if it is more then the cpu needs.. therefor the speed is the real factor.. and DDR 400 usually clocked at 3XX to increase latency.. can't complete with 1200 mhz... ofcourse ddr 600 and stuff will come out but thats still 1/2
I heard that Rambus will have PC-1200 64bit memory out which would give it 9.6gbs in a single channel.
The price increase I was referring to is the price of the controller ASIC, not the ram. I do expect RDRAM prices to climb as demand falls off - even PC1066 is no longer the fastest stuff on the market, and intel will be putting out no more RDRAM chipsets, so demand will fall off considerably. But even if the RAM stayed the same price, putting 4 memory controllers on a chip will increase the price of the chip.Quote:
Originally posted by MAVrecK
Price will not be high.. DDR is now more expensive then Rd at my Canadian vendor.. with 1066 edging out in performance.. price is not an issue anymore.. and with a quad controler.. the latency would be SWEET... Bandwith is irrelevant if it is more then the cpu needs.. therefor the speed is the real factor.. and DDR 400 usually clocked at 3XX to increase latency.. can't complete with 1200 mhz... ofcourse ddr 600 and stuff will come out but thats still 1/2
Also, while more controllers will decrease the latency from the CPU to the memory controllers, it will not affect the latency of the controllers to the RAM at all - and that's where RDRAM has latency issues. DDR 400 is almost always low latency, and even most PC3500 is CAS 2.5 or less now (mushkin, kingston hyper-x, corsair).
Bandwidth is (clock speed) x (bus width). Bandwidth is the total amount of data the memory can send to the controller in a certain amount of time. Latency is how long it takes from a request for data until it is sent on the memory bus. Clock speed is irrelevant except in how it affects bandwidth and latency. RDRAM is high clockspeed, low bus width. DDR SDRAM is high bus width, low clock speed. The result? They perform about the same. 1200 MHz RDRAM at 16 bits bus width (standard 1-channel) would be far slower (less than 1/2 after latency) than 600 MHz DDR SDRAM at 32-bits (standard 1-channel).
But of course, DDR will be used in dual-channel, and RDRAM in quad channel, so it all evens out, roughly.
64 bit single channel RDRAM will never happen - certainly not at 1.2 GHz. The reason RDRAM can run at such high clock speeds is that they can keep the trace count down and terminate and isolate each one. A 64-bit bus at 1.2 GHz would be nearly impossible with current PCB techniques. I suspect that what you heard was the quad-channel PC1200 info, which is 4 16-bit busses and has 9.6 GB/sec of bandwidth.Quote:
By Voodoo 5 5500 I heard that Rambus will have PC-1200 64bit memory out which would give it 9.6gbs in a single channel.
Of course, I could be wrong!
:D
Benchmarks? I would think it would be faster actually, due to the efficient architecture of RDRAM. Granite Bay has more BW and lower latency than RDRAM and still gets beat by it.Quote:
1200 MHz RDRAM at 16 bits bus width (standard 1-channel) would be far slower (less than 1/2 after latency) than 600 MHz DDR SDRAM at 32-bits (standard 1-channel).
Latency has little to due with performance in RAM benchamrks. Just compare RIMM4200s with PC1066. The difference is negligeable even with the differences in latency.
Where are you getting your information? Due you know how complicated and expensive it will be to verify the memory controllers for the DC DDR chipsets? It's insane.Quote:
putting 4 memory controllers on a chip will increase the price of the chip.
I would show you the 64-bit RIMM whitepaper, but the Rambus website is down right now. In any case, 64-bit RIMMs are supposed to be demonstrated this year and will be used on the Sis 659 if it is released.Quote:
64 bit single channel RDRAM will never happen - certainly not at 1.2 GHz. The reason RDRAM can run at such high clock speeds is that they can keep the trace count down and terminate and isolate each one. A 64-bit bus at 1.2 GHz would be nearly impossible with current PCB techniques.
Since neither 16 bit 1200 MHz RDRAM or 600 MHz 32 bit DDR will likely ever see the light of day, benchmarks would be difficult. The reason I say 16 bit 1200 MHz RDRAM would be slower is because it has (16 bits) x (1200 MHz) = 2.4 GB/sec of bandwidth, whereas 32 bit, 600 MHz DDR has (32 bits) x (300 MHz x 2) = 4.8 GB/sec of bandwidth. Like I said, thats single channel. DDR would be used in a dual channel config, RD in a quad channel config.Quote:
Originally posted by elimc
Benchmarks? I would think it would be faster actually, due to the efficient architecture of RDRAM. Granite Bay has more BW and lower latency than RDRAM and still gets beat by it.
Latency has little to due with performance in RAM benchamrks. Just compare RIMM4200s with PC1066. The difference is negligeable even with the differences in latency.
I agree, the DC-DDR verification is insane. That's why I think quad channel RDRAM, with the same number of signal traces, but at 6 times the clock speed will be even harder!Quote:
Where are you getting your information? Due you know how complicated and expensive it will be to verify the memory controllers for the DC DDR chipsets? It's insane.
This is a quad channel RIMM - 4 channels of 16 bits each. Like I said it was.Quote:
I would show you the 64-bit RIMM whitepaper, but the Rambus website is down right now. In any case, 64-bit RIMMs are supposed to be demonstrated this year and will be used on the Sis 659 if it is released.
A quad channel RIMM would provide 9.6GBs. That is plenty fast to destroy DDR SDRAM which would have to be dual channel in order to compete.Quote:
The reason I say 16 bit 1200 MHz RDRAM would be slower is because it has (16 bits) x (1200 MHz) = 2.4 GB/sec of bandwidth, whereas 32 bit, 600 MHz DDR has (32 bits) x (300 MHz x 2) = 4.8 GB/sec of bandwidth. Like I said, thats single channel. DDR would be used in a dual channel config, RD in a quad channel config.
RDRAM verification is tricky, but not insane. The timing specs are very tight and well laid out. DDR SDRAM verification, however, is very tough. Actually, I think that DDR SDRAM has uses four times the number of signal traces that RDRAM uses, and the signaling is much poorer. This has lead to DDR SDRAM boards that are constantly late. Also, you may be forgetting that RDRAM is double pumped, so a 1200MHz RDRAM module would only be twoce as fast as a 600MHz DDR SDRAM module.Quote:
I agree, the DC-DDR verification is insane. That's why I think quad channel RDRAM, with the same number of signal traces, but at 6 times the clock speed will be even harder!
Right. Four channels on one RIMM will give you 9.6GBs per stick.Quote:
This is a quad channel RIMM - 4 channels of 16 bits each. Like I said it was.
[(64 bits) x (300MHz x 2)] / 8 = 4.8GBsQuote:
(32 bits) x (300 MHz x 2) = 4.8