Another memory standard backed by Intel that goes in gutter
http://www.theinquirer.net/default.aspx?article=34014
http://www.theinquirer.net/default.aspx?article=34220
nn_step, could you enlighten us about microbuffers?
Another memory standard backed by Intel that goes in gutter
http://www.theinquirer.net/default.aspx?article=34014
http://www.theinquirer.net/default.aspx?article=34220
nn_step, could you enlighten us about microbuffers?
Those are the most confused articles I have read on the inquirer in a long time, and that means something.
that they suck writting ?
Incoming new computer after 5 long years
YOU want to FIGHT CANCER OR AIDS join us at WCG and help to have a better FUTURE
wait are they saying they are going to use ANOTHER Rambus design >.<
if so they are probably talking about XDR...
Fast computers breed slow, lazy programmers
The price of reliability is the pursuit of the utmost simplicity. It is a price which the very rich find most hard to pay.
http://www.lighterra.com/papers/modernmicroprocessors/
Modern Ram, makes an old overclocker miss BH-5 and the fun it was
they are a goship machine nothing else they have to feed to the news world with gossip to survive
Incoming new computer after 5 long years
YOU want to FIGHT CANCER OR AIDS join us at WCG and help to have a better FUTURE
Intels not canning fb-dimms entirely, but rather playing with other promising memory standards. Eventually whatever proves the best for whatever market their aimed at will survive.
Let us be realists ok.
FB-Dimms have the potential for unlimited Memory changes and unlimited memory support. But Currently FB-Dimms are only using DDR2 memory and often are only found with a MAX of 2Gb of DDR2 on stick. However you can buy FB-Dimms with MORE memory but their costs are sky high $6,380.42+Shipping
On the other hand DDR2 is cheap $264.99+ shipping Have low latencies (yes DDR2 is now lower latency than DDR.. not really a shock)
DDR3 is going to come out without much news or problems. Supporting 8Gb Sticks of Memory with record low latencies, excellent speed, Consume less power than ANY Dynamic Ram out there( AKA DDR, DDR2, FB-Dimm, Rimm, SDR...) and DROP in price rapidly (be about a Year or two before it is cheaper than DDR2)
Fast computers breed slow, lazy programmers
The price of reliability is the pursuit of the utmost simplicity. It is a price which the very rich find most hard to pay.
http://www.lighterra.com/papers/modernmicroprocessors/
Modern Ram, makes an old overclocker miss BH-5 and the fun it was
Ya I just posted that link in the intel section . Told to come here and post it .
My brother in law who use to work for ATI and now Intel says that Intel may be buying Rambus . With a market cap of 1.7 billion it will be a cheap buy for intel. If this is true I bet AMD/ATI doesn't get its license renewed. That could really cut AMD and ATI off at the knees. I guess well have to wait and see if Intel pulls the rug out from AMD/ATI's feet.
i really doubt intel could pass anti monopoly laws buying rambus
Incoming new computer after 5 long years
YOU want to FIGHT CANCER OR AIDS join us at WCG and help to have a better FUTURE
Please explain that to me . I don't see were your coming from here at all.Originally Posted by leviathan18
Originally Posted by nn_step
So how does it stack up against this.
http://www.channelregister.co.uk/200.../rambus_xdr_2/
1) a Hell of alot cheaperOriginally Posted by Turtle 1
2) Better capacities
3) Similiar latencies for the First generation of DDR3 (aka 9 months later DDR3 will have better latencies)
4) lower bandwidth (but I doubt even a Octal core Conroe can use up all the bandwidth of a Dual channel DDR3 at full speed)
5) Alot lower energy usage (XDR is a massive current hog right now)
6) DDR3 Memory interface is ALOT easier to design and improve Performance wise.
7)or another way of saying it is "XDR is like Prescott and DDR3 is like Athlon64 in Socket 939"
Fast computers breed slow, lazy programmers
The price of reliability is the pursuit of the utmost simplicity. It is a price which the very rich find most hard to pay.
http://www.lighterra.com/papers/modernmicroprocessors/
Modern Ram, makes an old overclocker miss BH-5 and the fun it was
Rambus wasn't a bad design. It had loads od bandwidth but it had a downside, availability and it ran really hot. It's not Intel's fault that Rambus never took off.
PC Lab Qmicra V2 Case SFFi7 950 4.4GHz 200 x 22 1.36 volts
Cooled by Swiftech GTZ - CPX-Pro - MCR420+MCR320+MCR220 | Completely Silent loads at 62c
GTX 470 EVGA SuperClocked Plain stock
12 Gigs OCZ Reaper DDR3 1600MHz) 8-8-8-24
ASUS Rampage Gene II |Four OCZ Vertex 2 in RAID-0(60Gig x 4) | WD 2000Gig Storage
Theater ::: Panasonic G20 50" Plasma | Onkyo SC5508 Processor | Emotiva XPA-5 and XPA-2 | CSi A6 Center| 2 x Polk RTi A9 Front Towers| 2 x Klipsch RW-12d
Lian-LI HTPC | Panasonic Blu Ray 655k| APC AV J10BLK Conditioner |
if ddr3 is gonna be seen at ddr3-1600 (800mhz), i'd tend to agree, that's 25.6gb/s theoretical we're talking about.. and if ddr2 is any indication, speeds upward of 1000-1200mhz wouldn't surprise me, so 38.4gb/s when pushed a bit. a nice 4.8gb/s per core... assuming intel ever does something about that damn FSB bottleneckOriginally Posted by nn_step
Got a fan over those memory sticks? No? Well get to it before you kill them
I don't know it seems to me that with the intel token ring that will be used on nehalem and the modular design of nehalem with the Intel high performance GPU also on die. and 1node for every 2cores its seems to me this is exactly what intel needs. I mean Intel will get a memory system that has the XIO memory controller interface cell/ XMC memory controller/ XCG clock generator that allows for 4 differant settings/ Micro threading / send 4 data request in the same amount of time it now takes to send one.
This looks to be a perfect fit to Intels token ring design. IBM and sony have both licensed the memory controller . But that about it as far as whats applicable here.
Also do you have a link to where the XDR memory is said to run hot and use a lot of power.
Your absolutly correct . The trueth come out . The memory makers conspired against rambus. So who's to say they didn't make them run hot also and create the low availabilities.Originally Posted by Nanometer
OK, so does anybody have any information about what exactly "Microbuffers" are supposed to do?
And yes, I googled.
If this info is coming out of anywhere near RAMBUS' organization I'm not going to give much about it.
The only people talking about MicroBuffers right now is Rambus and Rambus.
Fast computers breed slow, lazy programmers
The price of reliability is the pursuit of the utmost simplicity. It is a price which the very rich find most hard to pay.
http://www.lighterra.com/papers/modernmicroprocessors/
Modern Ram, makes an old overclocker miss BH-5 and the fun it was
DDR2 is going to be around for a long time. So don't plan on DDR3 replacing it so soon even when support starts to fall in place. DDR2 has a long life span ahead, and don't even worry about more speed. 1300MHz DDR2 is on the DDR2 roadmap, and we will finally be able to run higher dividers. Can't believe the day will come when we rate RAM in gigahertz!
PC Lab Qmicra V2 Case SFFi7 950 4.4GHz 200 x 22 1.36 volts
Cooled by Swiftech GTZ - CPX-Pro - MCR420+MCR320+MCR220 | Completely Silent loads at 62c
GTX 470 EVGA SuperClocked Plain stock
12 Gigs OCZ Reaper DDR3 1600MHz) 8-8-8-24
ASUS Rampage Gene II |Four OCZ Vertex 2 in RAID-0(60Gig x 4) | WD 2000Gig Storage
Theater ::: Panasonic G20 50" Plasma | Onkyo SC5508 Processor | Emotiva XPA-5 and XPA-2 | CSi A6 Center| 2 x Polk RTi A9 Front Towers| 2 x Klipsch RW-12d
Lian-LI HTPC | Panasonic Blu Ray 655k| APC AV J10BLK Conditioner |
Me thinks that DDR2 will not go beyond 1066 MHz! Hell I think that Intel will push hard for fast adoptation of DDR3-1333!
There are faster DDR2 modules allreadyOriginally Posted by Nedjo
If only two globally really small companies are offering one pair of DDR2 modules faster then 1066 MHz in quantities that are represented with two, or hopefully three digit numbers, then it’s really nonexistent
I find this amusing and interesting . note the 5 year deal.
http://arstechnica.com/news.ars/post/20060103-5885.html
Now this is why I find the above article amusing
http://www.thestandard.com/article/0,1902,28998,00.html
If Intel does buy Rambus. in 4 years amd will not get its license renewed.
Also notice how the AMD fan-boy in the first link says that Intel may follow AMD as they have in so many other area's . Thats absolutely hilarious. As Intel and Rambus have had a full cross licensing agreement in place since 2001. AMD fan-boys crack me up.
Last edited by Turtle 1; 09-09-2006 at 12:08 PM.
a lot of things can happen between now and 2011Originally Posted by Turtle 1
what makes you think that amd won't get it's license renewed ? or even if it needs to have a the license renewed ?
the license is for the memory controllers that exist now , and until 2011 , nothing for the memory it's self
what makes you think that intel is going to buy rambus ?
intel is already following in amd steps , x86 32/64 cpu , memory controller on cpu , intel making it's version of hypertransportAlso notice how the AMD fan-boy in the first link says that Intel may follow AMD as they have in so many other area's . Thats absolutely hilarious. As Intel and Rambus have had a full cross licensing agreement in place since 2001. AMD fan-boys crack me up.
do i have to go on ?
Bookmarks