Is Rambus Destined to Return?

An anonymous reader pointed us to an article running over at Tom's that talks about the world of ram and criticizes the performance of DDR. The article goes into DDR333, DDR400, and Rambus, and explains the issues at higher clockspeeds.

Comments

Experience tells us

[1155]

Experience tells us that Rambus is faster.

Pocket books tell us that ddr is better.

Which will your wife let you decide on?

Re:Experience tells us

[Paladin128]

DDR is still faster than Rambus. PC1600 DDR SDRAM (100mhz DDR=200mhz effective), offers 1.6GB/sec of memory bandwidth, the same as PC800 RDRAM. Why does Rambus perform better on the P4?

The i850 chipset has a dual-channel RDRAM controller. It handles two channels of PC800 RDRAM, offering twice the memory bandwidth of PC1600, and about 25% greater than PC2100.

Conversely, the newer i845 DDR SDRAM memory controller offers single-channel support for PC2100.

If a dual-channel DDR motherboard was available for the P4, it would smoke the Rambus performance. Period. The article at Tom's stated that at the higher speeds DDR must have a relatively high CAS latency (2.5). This is still FAR lower than the latency in RDRAM. RDRAM is high-bandwidth, high latency. DDR is high-bandwidth, low latency.

I've been really disappointed with most articles at Tom's of late.

I hope not...

[andaru]

I always thought that Rambus was even more annoying than the rest of the Lakers.

Sorry...

As an small OEM computer maker, I hope not

[bravehamster]

Seriously, ignoring pure performance considerations, RDRAM is garbage. It has to be put in pairs, and if those pairs aren't made by the same manufacturer, I've seen motherboards refuse to boot. Heat is a serious issue, and I've burned one finger too many on those heat spreaders. I've also seen an analog cable coming from the cdrom get stuck between the RIMM's and melt to the heat spreader. And price is still an issue, although it's improved quite a bit recently.

Expensive + Has to run in pairs + Runs very hot == Useless to me.

Re:As an small OEM computer maker, I hope not

[Zathrus]

No, his points are valid.

Interleaved memory designs (interleaving on a slot basis rather than interleaving on the RAM stick itself) causes many issues. First off, you have to have more slots for equivalent upgradeability. And more slots requires you to have more layers on the motherboard due to increased number of traces (although, admittedly, RDRAM has vastly fewer traces than SDRAM even so). It also requires more real estate on the board, which isn't debateable. Second, you start running into timing issues more often with interleaving than standard memory clocking. Sure, as you say, it depends how robust your controller is. But, funny thing, RDRAM either has amazingly shitty controllers, or they're just vastly more prone to lockups when you have slightly differing speed memory.

As for heat - it's not a tradeoff issue. DDR didn't double the heat of standard SDRAM, and RDRAM isn't merely twice as hot as DDR. It's absurdly hot. And heat is a major computer issue already between CPUs, chipsets, and graphics cards throwing off oodles of heat as is. I don't know of a manufacturer that has a fan blowing specifically over the RAM, but RDRAM could certainly benefit from this. Heat kills systems (more specifically, thermal changes kill systems, but you'll get faster thermal changes with hotter components), so why design a system with RDRAM that is so much hotter than the alternatives? For how little (if any) of a performance gain?

Oh, and you claim RDRAM is twice the speed. Ok. Want to compare apples to apples? Put RDRAM in a non-interleaved system (yes, they're out there. They're even predominant) and the memory bandwidth is only slightly higher than DDR. Or compare it to an interleaved DDR system (again, they're out there). Boom. You have a DDR system with nearly as much bandwidth as RDRAM.

And, frankly, bandwidth ain't all it's cracked up to be. Funny how DDR systems routinely spank RDRAM systems in real world benchmarks (not pure memory bench's). Why? Because latency is king. Particularly if you're multitasking. You'll hit different areas of memory so much that bandwidth will make little difference compared to latency. And RDRAM has really, really miserable latency. And it gets higher as you add more sticks. So while it's great for some things (video editing/streaming, etc), it sucks for most applications.

Test costs remain high, apps support remains nil

[Anonymous Coward]

The fact is that most IC testers today support the lower-speed parallel connections. High-speed serial connections like Rambus and SERDES require very expensive mixed-signal testers with expensive and complicated load boards (the PCB between the tester itself and the chip). These high-speed serial I/Os on the memory ICs themselves are also generally much larger than on a DRAM, probably by a factor of 5. So, you don't get die savings, you don't get lower test costs, and most of all you don't have any processors whose front-side buses exploit this. Plus, you have very expensive target products in terms of motherboards to support the Rambus ram requiring tight trace routing and signal isolation, and their very limiting 28ohm max impedance (at least with the PC800 RDRAM), almost completely opposite in difficulty to DDR. So where's the advantage?

If you also figure that the memory controllers for Rambus are configured for dual-channel operation, it becomes much clearer that the advantage is not in the memory architecture itself but in the controllers. Suppose a server board manufacturer decides to support quad-channel PC2700 1GBx4. That's 10.8 GB/s of potential memory bandwith on sequential accesses! There's hope with chipsets like the Nvidia nForce420 dual-channel DDR, but the Athlon FSB is the limiting factor there. And let's not get into the infamous first-access latency issues which I hope they're finally addressing.

Rambus is also notorious for poor tech support. I worked for a major silicon vendor using their core, and they never responded to our requests for minimum PLL-to-Rambus core distances. It was abjectly ridiculous, but not surprising considering that regular SDR/DDR memory interfaces outnumbered Rambus designs 100:1. Have things changed? Considering what their legal bills have been lately and an erosion of their tech support, I doubt they can afford to improve it much.

doesn't it depend...

[ryusen]

on what systems you are working with? if you want a performance p4 system then obviously you use rambus... and if you want an amd system you use ddr(since there is no rambus/athlon chipset)
and until there is a rambus/athlon chipset i don't really think we can gague the real world implication of it...
either way i have better things to do with a few $100 than put it into a more expensive chipset/cpu/memory rig. if you have the extra money and the rambus system gives you what you want, then more power to you. overall, right now, you can't say either system is "the best" in ever possible catagory

Re:common sense?

[VAXman]

Speed. RDRAM looked fast because it was implemented with multiple banks. You can do the same thing with SDRAM, if you like. And that would give an apples to apples comparison.

The whole advantage of RDRAM is high bandwidth/pin, and the fastest RDRAM has more than double bandwith/pin than the fastest DDR. RDRAM is very cheap to make dual channel because it has fewer pins. It is very expensive to make a dual channel DDR system because it requires that many more signals. The only dual channel DDR system I know of is the upcoming Serverworks Grand Champion chipset for the P4 Xeon which is very high-end (and no doubt expensive).

Tom, as usual, not 100%

[Chris Burke]

It wasn't a bad article... I mean, the facts -do- show that the p4 runs better with RDRAM, and he addresses the consequences of that quite well, and quite neutrally. For that I commend him.

But he does misrepresent some issues. For example, signal integrity issues. I can say with complete assurance that Rambus is loaded with signal integrity issues. These issues get -very bad- as the clock frequency goes up. Also Rambus is -not-, strictly speaking, a serial bus. First, it is 16 bits wide, while pure serial would be 1. Second, the depiction of a DIMM as being a unterminated stub with significant SI issues is correct, but this doesn't go away with rambus, and this definition of "serial" fails as well. While the signals do pass through a RIMM continuously, eliminating the RIMM itself as the source of major SI problems, you still have each and every RDRAM device itself acting as an unterminated stub, each of which causes reflections of its own. Especially for devices with tolerances as low as RDRAM, this can be difficult to manage. While in the balance I'd have to concede that at a given clock frequency RDRAM has the SI advantage, remember that RDRAM needs 4x the clock frequency of DDR to match bandwidth.

Or you could have 2 channels of rambus, and only need 2x the frequency. Well, 2 channel DDR is becoming a reality. Not only does nForce support it, Sledgehammer will as well. Neither of these are Intel platforms, but I would guess that going dual-channel would be a natural step for VIA and others competing with Intel chipsets. It would especially make sense for p4, as it would more than make up the memory bandwidth disparity that currently exists.

Speaking of nForce, another thing I take issue with is the suggestion that the nforce's DIMM-slot population problems are indicative that DDR is crippled by SI issues. I think more likely is that this was the first chipset designed by a company whose experience lies solely with graphics cards, on which the ram is directly soddered to the PCB. Lack of experience in the harsher SI conditions of a computer motherboard are to blame.

Speaking of DIMM population, it's hard for me to see only having 2 DIMM's on some boards as a particularly black mark for DDR... That leaves you with 2GB per channel, the same as RAMBUS.

So, he was right about some things, insightful on others, but the picture is -not- so clear-cut in the image of rambus Inc.

Re:Tom, as usual, not 100%

[sigwinch]

For example, signal integrity issues. I can say with complete assurance that Rambus is loaded with signal integrity issues. These issues get -very bad- as the clock frequency goes up.

Bad as in you have to be aware of dielectric losses in the PCB material. I remember seeing reflectometer graph of a Rambus system where the plateaus were noticeably sloping from dielectric loss.

Also Rambus is -not-, strictly speaking, a serial bus.

Serial != 1 bit. Serial == takes more than one clock cycle to transfer a word.

Re:Tom, as usual, not 100%

[Perdo]

The Column was done by Frank Völkel. Based apon his lack of technical documentation, I'm going to guess it's just his opinion. I doubt very seriously that Tom Pabst himself agrees with the article. Tom tends to be much more objective in his articles than Frank is.

Cost control

[Waffle Iron]

From the article:

The days of cheap memory are over.

They say this because of the huge expense needed to provide 512MB or more of ultra fast memory. But what if they added yet another level of "cache"?

Put in 128MB or more of super-fast RAM (faster than today's RDRAM or DDRAM, maybe using an exotic bus) backed by gigs of cheap, easy-to-make memory (PC266 DDRAM or slower). The cheap ram is still orders of magnitude faster than a disk drive. Manage them with hardware that does page swapping similar to virtual memory.

You could get good system performance and lower overall cost.

Re:My time with rambus has been awesome

[Chris Burke]

Were the AMD systems Duron-based? :)

What about games that -don't- love the P4, like, say, -any other game- (even those based on the Q3 engine)? :)

But no one needs to do the same testing you did. They can just look at all the tech sites. Hey, you already visited Tom's Hardware to read this article, check out who -he- thinks has the "best possible performance in 3D".

At one time, your "best possible performance in 3D gameing" applied... That time was the year 2000, and Q3A was the most demanding benchmark anyone could cook up It is now 2002, and the world has moved on from Q3A, and P4 lost that crown. But nice try.

You should have said "media encoding", because then you'd have been right even today.

As to AMD using Rambus... It'd suck. P4 does better with RDRAM than DDR because it's a highly pipelined, high-clocked machine that craves bandwidth. The K7 is a very wide machine, and for it the worst thing that can happen is having to stall waiting for data. The latency of RDRAM would kill the K7. You'll note that the dual-channel nforce (higher theoretical bandwidth than the i850, and 2x the KT266A) doesn't outperform VIA's chipset. An likely reason would be that the KT266A has lower latency, and that more than makes up for the extra bandwidth (which K7 doesn't need).

Is Intel's DDR implementation bad, or not?

[landley]

The guy starts out the article saying that Intel's DDR implementation was crippled for political reasons. He also states that Athlons benefit from DDR more than P4.

Then the political aspect is ignored and he talks almost exclusively about technical issues about why Rambus might theoretically be better, and uses existing intel chipsets as evidence.

Hello? Answer the question, please? Has Intel ever come out with a non-crippled DDR chipset for the P4? How do Intel's DDR P4 chipsets compare to non-intel DDR P4 chipsets? (ARE there any non-intel P4 chipsets?)

How much of the problem is political, and how much of it is a real technical issue?

Re:I refuse to believe tomshardware.com anymore

[Chris Burke]

Um... How was that FUD?

It's true. If the heatsink falls off your Athlon it is toast. (note that just in the last week or so a board was released that supported the XP's thermal diode... but for all other boards/chips, you still get toast)

Tom isn't the genius a lot of people think he is (or that he'd want you to think), but that was not FUD.