Slashdot Mirror
Second-Gen DDR SDRAM On The Horizon
cplcap writes "This story in The Register picks up on Samsung's new DDR-II Chips, pushing DDR's speed up to 533 Mb/s and a 4.2GB/s memory bus. Prototype 512MB DIMMs are being produced, and IBM has developed a chipset to take advantage of the speed. There's a little more meat in Samsung's official press release."
Is it a whole new form factor so everyone had to redesign the motherboards and to force incompatability with older systems??
This is important because industrial and corperate-mission-critical is older equipment. and an upgrade path for ram is still important.
Do not look at laser with remaining good eye.
It seems not so surprising that IBM would flock to new and "exciting" technologies and improvements like this one. One can only wish that such forward thinking would shake-down through the rest of their organization.
The geeks in the back rooms, and the engineers in the streets certainly see the benefits from this new technology. Perhaps someday the slimmer, sleeker, more lithe and less middle-management IBM will show itself and we'll see this sort of stuff rolling downhill more quickly.
In the past, IBM has been near the forefront of going after new improvements. They sometimes just can't market their way out of a wet paper bag. Often sad, but true.
Awk! Pieces of eight. Pieces of eight. Pieces of seven... ERROR: General Protection Fault. [Paroty Error.]
It is nice to know what's coming, but Q3 2003 is a long ways away, and the modules being ready doesn't necessarily mean there will be a chipset ready that supports them. I'm still waiting for P4 motherboards to support 333Mhz DDR. Tom's has a review of the VIA P4X333 here, but we haven't seen any motherboards with this chipset yet. The VIA KT333 chipset currently has around 16 Athlon motherboards shipping with 333Mhz DDR support.
-- Adam
Hahahhahahha!
When will *processors* keep up with memory!
Oh, you have me laughing.
Memory technology is so far behind processor technology it ain't funny. Hence L1 and L2 caches, L3 caches are becoming necessary now as well.
something tells me that the "computer age" is getting a bit unbalanced... we're nowhere NEAR reaching the hardware limits that we already have (home useage speaking at least) and we're already surpassing whats more than enough? sounds a bit unneccesary, at least for the moment...
Since the slashdot article about IBM 210Ghz transistors I've not seen any further mention of these super-transistors or the possibility of building chipsets from the technology... And yes, I mean 210Ghz not 210Mhz.
-- A change is as good as a reboot.
I understand that there are physical limitations well beyond my comprehension that are factors in all of this, but it seems that any time one of these improvements comes out, be it RAM or CPUs or any number of other upgrades, a new chipset has to be developed to support it?
I'll tell you why: because marketers understand that some people (many of which frequent this site) will pay plenty extra to have the latest technology, no matter what.
I, for one, am sick of it. For once I'd like to be able to upgrade my CPU or RAM without having to buy a new motherboard and re-install my entire OS.
Sorry for the rant, but I think the fact that every incremental hardware update requires a new chipset is noteworthy.
"Ask not what your country can do for you." --John F. Kennedy
Hell, it makes you feel old when 512Mb is being bandied around as the standard memory size. I remember getting all excited about a 512k(!!) trapdoor expansion for my Amiga, for which I paid 100 quid for.
;)
Oh yeah, and this whole website was fields back then... far as the eye could see.
http://www.davetansley.com - you proba
You could be sure to say that the Memory development line isn't exactly following Moores law. But that's okay tough, since that wasn't Moores intension.
A feeling of having made the same mistake before: Deja Foobar
The importance of faster, cheaper RAM is obvious, however I do find it hard to get excited about a technology that will at best be in new systems in about a year, probably later. RAM speeds have skyrocketed, but unfortunately bus speeds have'nt, unless you're an overclocking guru this RAM just won't mean much to overall performance gains - just yet. But all the same, every step forward is a great step.
What with IBM making the G-4s for Mac, and making this chipset, d'you think they could finally bring Macs out of the PC100 world, and get a memory pipeline worth having?
if ($it != $onething) {$it = $another;}
First, they only make G3s for Apple currently. Also, Macs use PC133, and the newest server line uses DDR. They do need to work on the processor side though, supposedly the CPU itself can't interface with DDR, so a lot of the bandwidth goes unused.
The two big reasons for the generational change are
Yes, this makes for backward-compatibility problems.
Yes, the Committee (JC-42.3) put a huge amount of work into making DDR-II as backward-compatible as possible
Yes, we're starting work on DDR-III. You'll have to wait until 2006 or so.
Target speeds for DDR-II were set at 600 MT/s for fully-loaded systems and 800 MT/s for embedded stuff like graphics.
The signal-integrity issues for DDR-II are ugly, but we met the margin specs with lots of conservativism thrown in, so once we get hands-on time with systems you'll probably see the numbers exceeded just as the original DDR targets were.
Flame away. You can get more info at JEDEC or Advanced Memory International.
so did DDR333 put any _real_ speed increases into today's system?
Gordon.
The actual clock speed is 200mhz and 266 mhz ddr for 400 (PC 3200, @ 3.2 gb/s), and 533 (PC4300 @ 4.3 gb/s), respectively. TheInquirer points this out, along with some more specs. Also, way back on May 2, JEDEC approved the standard.
Well, I've managed to break both win98 and win2000 by switching motherboards. (Crashing all the time, finding lots of new devices, messing up IRQs, etc.)
I think most other OSes handle that stuff better...
Probably 533Mb/s per data pin at 266MHz.
That gives us 533*64/8 = 4.2GB/s for a 64bit bus.
That sounded like a Troll.
Exactly what DDR bottleneck are you talking about? Specifically, where Windows has a workaround and Linux doesn't.
Memory bandwidth has always been a bottleneck on systems, and it probably always will. However, this is a hardware issue and not an OS one, as far as I understand it.
Finally, an admin "in the know" that chose a Mac server was probably on drugs. PPC hardware is very nice, but the extra cost -- not to mention the skills needed to deal with the lack of server software... a questionable choice for any serious environment.
Learning HOW to think is more important than learning WHAT to think.
i wish i had your network card...
The article says that IBM made a chipset to take advantage the new memory speed, but what CPU does the chipset support? Athlon? P4? G4/G5? Anyone? Bueller? Bueller? Frye?
"I'm The Bounty Bear. I will find him anywhere. I'm searching."
You really should stop hoping that better hardware will fix the boogered Linux VM [software]. Then again, if you knew any better, you would be using FreeBSD if the bee in your bonnet is VM implementation.
Gah! I fucking hate marketting! Why can't they just call it QDR, because that's what it is. Quad Data Rate fucking DIMMs. God damn it, motherfucking, sun of a bitch!
Have you not seen the new intel servers IBM is releasing like the x360 and x440?
Need more PCI? Add a drawer of 12 and plug in a cable. Need more processors? Buy another four way and plug them together, you have an eight way.
Hot swap a failed memory dimm lately? You can in a x440.
There are a lot of cool tech coming from IBM in the xSeries servers. There are only so many marketing guys out there
But it sure is easy to bash IBM, so people do. They are changing. You think the layoffs of the last year or two are getting rid of the good people and not the middle management?
As a rock-in-roll Physicist once said, No matter where you go, there you are.
Depends on what you do. Some applications are particularly memory intensive, like MPEG-4 encoding. The faster the memory, the better. Other apps are limited by the CPU, peripheral bus or graphics card.
If you can't beat them, embrace and extend them.
I suspect that there were really two factors at work in the short DDR-I lifetime.
First, Intel muddied the waters with the big exclusive Rambus push. While there was DDR work going on prior to the Rambus push, there was some very real contention in carrying both programs through development. This doesn't even mention quite a bit of "wait and back the winner," at many levels of the industry.I suspect that the success of the Athlon competing with PIII had almost as much to do with DDR success as Rambus prices.
Second, there were very real signal integrity issues that had been skirted for quite some time, and really came to the fore with DDR. That took some time, but more thought has been applied forward to DDR-II, so it shouldn't be as painful.
The living have better things to do than to continue hating the dead.
Since it keeps all that backwards compatibility.
On a serious note, keeping backwards compatibility both stifles development and raises cost. I don't complain that new computers are cheaper and faster than they ever have been.
Moderation: Put your hand inside the puppet head!
Actually, yes. The problem is the Athlon's FSB is 133MHz DDR aka 266MHz. If you overclock the FSB (Chipset proc rather than RAMchipset) to 166MHz DDR you do actually get a significant speed increase.
Key words there are 'daily work'. If that is using things like word processors, spreadsheets, web browsers and the like, then that is correct. If your daily work is encoding video or audio, making heavy use of Photoshop, Lightwave or Maya, then it makes a difference.
If you can't beat them, embrace and extend them.
3D Volume Holographics Optical Storage NanoTechnology wants to replace all memory devices with one storage device to handle all
system memory requirements replacing DDR, SRAM,DRAM, MRAM, FRAM, CD, DVD, TAPE, HARD DRIVES, OVONIC, etc.
www.colossalstorage.net
The Xserve has a DDR bus.
The G4 will remain at 133 or 166 MHz because all effort is going into the G5's pipe.
Lies about crimes
The story on the IBM site linked from the slashdot story claimed 100Ghz chips within two years (although we're talking comms chips) "IBM expects the new transistor will result in communications chips at speeds of 100 GigaHertz (GHz) within two years -- five times faster and four years sooner than recently-announced competitive approaches." So naturally I was curious what that might mean in terms of regular chips. Oliver
-- A change is as good as a reboot.
- ...what CPU does the chipset support...
None as far as I could tell. From the Samsung press release it looks as if the chipset IBM developed isn't a "chipset" as you're using the term. You're thinking Northbridge (or MCH in Intelspeak), i.e. a "chipset" that controls your CPU, memory, AGP, IDE, PCI, (and in combo with a Southbridge, or ICH in Intelspeak) USB, Parallel, Serial, ADB, ISA, IrDA, Audio, SmartCard, or whatever.The IBM-developed DDR-II chipset in question here is simply a memory controller, i.e. a small (but important) part of the Northbridge in what you're calling a chipset.
There comes a time in every man's life when he must say, "No mother! I do not want any more Jell-O!"
I had a high school English teacher who once said, "Brilliant people make complex things simple."
Thank you for the clarification. Brief, to the point; the perfect response.
"I'm The Bounty Bear. I will find him anywhere. I'm searching."
The data lines are still clocked at DDR, but there are twice as many of them.
DDR-II is basically equivalent to on-chip interleaving - the data path is clocked the same but twice as wide. Well, board makers are now doing motherboard-side DDR interleaving themselves - see recent boards based on the ServerWorks GC-LE or Intel E7500 chipsets. Some manufacturers might just decide not to bother with DDR-II.
test