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

[1155]

Look, this wasn't a troll remark, this was a serious remark. DDR is the cheaper of the two, and probably will be for some time. I don't see the point of spending all that cash on rambus when ddr will do it only slightly less better. Tom is a hardware nut. I still use pc100 ram, and I am satisfied. My comment was not a troll, and you guys know it.

Re:Experience tells us

[Anonymous Coward]

RDRAM is currently only about 20% more expensive than DDR and has been closing that gap pretty quickly recently. It's not really a price issue anymore. You may have had a point when RDRAM was 200% more expensive but that was a while ago.

Re:Experience tells us

[Anonymous Coward]

Of course this is only relevant if speed is important to you. Right now DDR keeps up pretty well with both Athlon and P4, so if you have a fixed amount to spend it makes sense to buy a faster processor and not the fastest memory.

But both CPUs will get much higher clock rates in the next few years, and it's not clear if DDR can keep up. If memory starts to seriously limit performance, then it will make sense to buy RDRAM. (Of course RDRAM + Athlon is not a possibility in the foreseeable future.)

Re:Experience tells us

[darketernal]

I'd have marked you Insightful, except that I don't have anymore mod points. Sorry.

DDR is CHEAP, and it's GOOD. But of late it's been getting rather expensive, so the gap is closing. I was still running on PC100 when Crucial was selling 256MB PC2100 sticks for, what? $35? I assumed that prices would stay that way for a while, but when I finally decided to upgrade, the price rose to $60-70.

First Windows 98 (bought the original version four days before Second Edition came out), the Radeon (bought it two weeks before the GeForce3 and the newer Radeons came out), and now this, I really don't have any luck...I either seize the opportunity too early or too late. Doh!

Re:Experience tells us

[Osty]

the Radeon (bought it two weeks before the GeForce3 and the newer Radeons came out)

You must have an awesome mastery of space/time, since there was a several-month time span between the release of the GeForce 3 and the Radeon 8500. If you truly were able to accomplish the same event exactly one time twice so that you did it two weeks before each event, you are truly a god among men.

Re:Experience tells us

[Trepalium]

The GeForce 3 and Radeon 8500 may not have been "released" on the same date, however they were announced within days of each other. They did this despite the fact they knew it wouldn't ship for months.

Re:Experience tells us

[sirsnork]

Actually Athlons won't get much faster. They are quickly approaching the end of their life cycle design-wise. Just as the Pentium 3 has done. That is why the Hammer series is in the pipeline. The problem for AMD that has been discussed before is. Will they get it finshed in time

Re:Experience tells us

[Osty]

Right, but the term "came out" generally implies "released", not "announced". Not that it matters. I was being facetious.

Re:Experience tells us

[jra101]

Actually the GeForce3 was announced around February and the Radeon 8500 was announced around July-August.

Re:Experience tells us

[trixterxp]

Well, from my experience, I would have to go with DDR, as it is more widely supported on newer motherboards such as the Asus A7V or higher. Still, RDRAM and DDR are very nice to have, especially when all your friends still have PC133 (hehe, dorks).

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.

Re:Experience tells us

[darketernal]

Right, let's clean this act up. My point was that the minute i bought the cutting edge, better stuff came out. Yes, I was wrong about the release within two weeks - i meant announcement.

Sorry for any confusion. meep.

Re:Experience tells us

[rosewood]

Intel killed the P3 @ 1.0 Ghz yet the tutaltin line has grown to 1.53 GHz

there is a lot more room left in that core then ppl will tell you

but AMD is also ancy to get that new architecture out

Rambus as a company

[Jesse Duke]

Given the bad performances of RDRAM due in large part to its insanely high latency, and Rambus' dubious business practices based mainly on trying to milk patents to leech on the entire memory industry's back, why on earth should anybody give then the opportunity to make a come-back ?

Re:Rambus as a company

[linzeal]

IS there anything else in memory arch besides rambus and DDR on the horizon?

Re:Rambus as a company

[Dudio]

The article you reference talks about RDRAM performance with the P3. The P4 presents an entirely different picture, as it fully utilizes the bandwidth of the dual-channel RDRAM architecture. See this article for details. A brief quote from the conclusion:
The memory benchmarks from above show that Pentium 4 really requires the 3,200 MB/s of data bandwidth supplied by the two Rambus channels. I doubt that it will perform as well with DDR-SDRAM, unless two channels will be used.

The business practices, of course, are a different story.

Re:Rambus as a company

[Glonk]

How is the parent post +4?

He links to an obsolete article from Q3 2000 about RDRAM on the Pentium III...

He talks about the "insanely high latency", and it's pretty obvious he's exaggerating slightly.

RDRAM's latency, particularly with the upcoming PC1066, is far better than people give it credit for. See this AcesHardware article.

PC1066 RDRAM latency for 128 bytes: 207 cycles
PC800 RDRAM latency for 128 bytes: 247 cycles
PC133 SDRAM latency for 128 bytes: 229 cycles

Slashdot moderators: Would it kill you to check the links before going points-crazy?

Re:Rambus as a company

[Anonymous Coward]

This is primarily a matter of the highly aggressive cache pre-load in the P4. With bandwidth to spare, it enhances performance, especially since it does a reasonable job of hiding latency in the underlying memory. If you don't have lots of bandwidth, it hurts rather badly, since it wastes bandwidth to hide latency that isn't actually present.

If Intel really cared, they could simply tweak the cache preloader and have a chip that performed similarly to the current P4 w/ RAMBUS.

Their reluctance to do this points to either impressive expected future RAMBUS performance (compared to DDR) or another chip in the later stages of the development pipeline (perhaps the nebulous yamhill of yore?).

Of course, this is all ass-wild speculation based on fairly slim facts, and the expectation of purely rational behavior from Intel.

Re:Rambus as a company

[coupland]

I agree but am compelled to ask: why was latency an issue w/ RDRAM before but now RDRAM is the greatest thing since sliced bread? I'm not upset that THG has posted a correction, I just didn't see a satisfactory explanation...

Re:Rambus as a company

[Compuser]

Presumably because bandwidth is becoming more
of a problem, esp. with how P4 is designed.

More factors than speed

[slithytove]

It seems to me that Rambus has offended so much of the industry that it even intel's continued (though lately lessening) support, or perhaps especially with intel's support it will fail to be implemented by the majority of m/b manufacturers.
Other avenues for gaining speed exist- like Nvidia's extra memory controller for the gpu in the xbox and higher end nForce chipset.

Re:More factors than speed

[slithytove]

To clarify on the second point:
The dual 64bit memory controllers of the Nvidia nForce northbridge part allow the cpu and gpu to access the same ddr ram at the same time and with the same speed. This doesn't help in non 3d apps, but then I haven't seen a lot of desktop apps besides maya and games that are dying for a huge performance upgrade anyway.
Larrence Leisig and I agree that there is a lack of killer apps to drive the demand to drive the r&d to keep moore's law going. On the other hand there are vast areas of application for embedded systems. Scientific and other cpu hungry apps will continue to be parralelized and run on ever cheaper distributed systems.

Re:More factors than speed

[Anonymous Coward]

Speaking about graphics, though, why can't these chipset makers refrain from beating the 3D graphics horse to death and work on something that's useful to users who aren't gamers, like infinitely scalable fonts, more tightly integrated video and audio, or embedded X11?

Re:More factors than speed

[Anonymous Coward]

A Lisp-machine graphics card! How cool would that be, eh?

common sense?

[Kargan]

With the exception of the shady business practices of Rambus, I don't fully understand why Intel dropped RDRAM in the first place. In every benchmark that I saw circa 7-8 months ago, the huge amount of memory bandwidth present gave Intel one of it's only advantages over the corresponding AMD CPUs.

It couldn't have just been the prices either, because Intel obviously knows they're not going to win that race.

Anyone?

Re:common sense?

[Greg Lindahl]

Intel had 2 reasons to bring out SDRAM based boxes:

1) Cost. Most consumers and business desktops don't care about speed, and RDRAM costs too much extra.

2) 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.

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).

Re:common sense?

[Greg Lindahl]

Actually, there have been multiple dual-bank SDRAM chipsets. While SDRAM and DDR do use more pins than RDRAM, that isn't necessarily a huge cost, especially since they're lower speed pins than RDRAM, which turned out (in practice) to be a pain in the neck.

Re:common sense?

[tkrotchko]

"I don't fully understand why Intel dropped RDRAM in the first place"

I think the reasons were pretty clear:

1) RDRAM was extremely expensive compared with SDRAM

2) The performance advantages were (and are) largely theoretical in desktop PC's

3) DDR RAM in practice showed itself to be faster than RAMBUS.

4) The Athlon chipsets supporting SDR/DDR combined with the cheaper costs of the AMD CPU and DDR RAM gave AMD based machines a huge cost advantage to vendors who chose to go the AMD route

[In fact, i would've bought a Dell last year except they only had P4's with RDRAM. That made a Dell computer not only slower, but more expensive than the Micron PC with the Athlon that I eventually bought. The price difference was significant, too]

Re:common sense?

[Ozx]

nForce

Re:common sense?

[afidel]

The problem with RDRAM is it moves the extra cost for extra bandwidth from the motherboard to the memory. Motherboards are cheap, even ones with 6-8 layers which is what you need for duel channel ddr. On the other hand RDRAM is expensive and because of the higher frequencies the yields are lowered significantly. So RDRAM trades off expensive testing, expensive parts in terms of yield for a little more work in the motherboard design and manufacture. RDRAM is trying to answer a question that doesn't exist. When they started designing rdram they thought we wouldn't be able to go past 4 layers in the pcb and that we wouldn't have materials that could run at great speeds without interference, the problem is we do. Dual channel DDR has already been shown to work, so we have near equivilant bandwidth with much lower latency. Sure the P4's prefetch helps hide latency, but not completely. An Athlon MP setup on an nforce style chipset would be huge on performance.

Re:common sense?

[screwtheNSA]

Since very few materials give near free-space resistance(Ohms) of 377 Ohms, we must separate the signal leads for the clock/s by at least 2-4 layers if the core speed is higher than 600 Mhz. Due to the shorter wavelengths of higher frequency CPUs, the crosstalk will negate the effort to increase FSB speeds and CPU clock rates while minimizing the latency built in with memory/CPU/FSB and controller speeds. Input and output signals will attenuate each other if they are not kept out of phase with each other constantly. In-phase signals cancel each other, but at the least, attenuate the integrity of those signals so that the data sent/received is garbled enough to make it useless for the CPU to use and transfer to the appropriate controller.

Rambus is a technology best left in the labs for the time being. Use what's best for your particular platform/s, use that if it functions well and causes no data loss or system lockups.

Board makers are too fast to accept anything that might grant a slight performance boost over a competitor's products. Choose the chipset/maker with extreme care and personal prejudice.

Bechmarks are tests that speak nothing of RAW power of the "system", but flaunt a CPU's core speed as the tell-all to total system throughput, which is NOT the real-world match to CAD, mathematica, or any pure processor hungry program that eats memory bandwidth and chews up CPU time performing several calculations and controlling functions in a split second. Heat is the "natural" by-product of clock speed/s, O/C'ing a CPU and then running a benchmark test to compare that against real world results is just a misleading application misused to amuse and confuse the actual processor's true capabilities.

The P-4 does NOT have any applications currently written that make full use of the high memory bandwidth it has, therefore, the current applications can't push the CPU to its MAXIMUM capabilities or speed/s.

The larger pipeline of the P-4 will slow the processor down if the data it handles is not written to make full use of that pipeline!

A quick perusal of the architecture will show you that the entire processor core is designed for programs and applications that CAN make full use of the pipeline, FSB speed, memory type and latency. Latency is both good and bad, if applied improperly, the CPU speed/FSB will become even more sluggish when the wrong memory specification is used with a chipset that over-allocates resources to the wrong areas; just as the right memory spec. used improperly will cause other failures with faster refresh rates, and memory dumps when the system is not "expecting" a memory dump. Limiting and eliminating board problems will become ever more problematic for main board designers with the ever increasing processor speeds as well as FSB speeds. A certain amount of "leveling" will be required for all component makers to play a little game of "catch up".

Stop designing CPUs with higher clock rates, and design a processor that USES the current speed EFFICIENTLY and PROPERLY, speed is NOT the do-all-end-all to the computer world. Show me even ONE application that suffers from NOT being run at 1 Ghz., nobody can I dare to bet!

Memory is now becoming the true bottleneck it once was when money was a factor, now it's compatibility between the "architectures" that has become the arch enemy. Rambus this, DDR that, SD there and so on...

ONE specification will rule Intel and another will rule AMD, you just have to sift through the B.S. spewed forth by those wanting your money.

Do a real test with your processor; UNDERCLOCK it, run it at the FSB speed if you can, or mod the BIOS to reflect that as close as you can. Does your application still run smoothly? Okay, slower, but stable, right? Now, return it to the normal speed, adjust the BIOS to reflect that as well, now run your application with different memory. Does the system still run stable? Time to swap to DDR next..same test, swap again; RD this time, and keep going. NOW, you have a base to judge what memory YOU can run stably, and without lockups.

Next, return the memory you have problems with to the store for a refund......YOU CHEAT!

I have a pair of 32 Mb. cards for my Compaq LTE5300, one is the correct speed, and one is slower, both cards work, but the slower memory caused the system to crash on cold boots, the "proper" speed card works no matter what. I have two LTE5300 notebooks, both do the very same thing when swapping these cards, so memory speed DOES make a difference, for me at least. I at least tested this in a real world application and based my findings from that. Your mileage may very.

Re:common sense?

[fred3666]

I don't know why the parent was left at zero. While not technically verbose, it is at least accurate. The person is quite correct that (at least some versions of) the nVidia nForce chipset supports dual channel DDR memory at a reletively inexpensive price. Say what you like about nVidia's chipset design but it does provide that inexpensive dual channel DDR memory is at least a proof of concept.

Re:common sense?

[Anonymous Coward]

You generally can't underclock modern microprocessors, because they make use of dynamic logic which his a limited charge life.

Re:common sense?

[choco]

A few years ago this "bandwidth/pin" thing would have been a killer difference. Adding pins was expensive, Pins and packaging were serious problems.

However that has changed considerably. uBGA and similar achieve huge pin densities at tiny cost per pin. It is harder to get your process right with BGA - but once you only have to do this at the design/setup stage, once it's right you get better quality, repeatability and yield with uBGA than you do with Fine Pitch packages. Multilayer PCBs are much less of an issue too - as are fine pitches in those PCBs.

>The only dual channel DDR system I know of is the upcoming Serverworks Grand Champion chipset for the P4 Xeon

Intel and others are working on several Dual DDR chipsets - "Granite Bay" is supposed to be released Q3 this year.

I hope not...

[andaru]

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

Sorry...

Re:I hope not...

[fredbsd]

Excellent. Can't believe this hasn't been scored funny.

I guess the kids who prowl around here would not remember Mr. Rambus anyway.

Very funny. Thanks for that.

Re:I hope not...

[j4ck50n]

I thought it was funny too!

Re:I hope not...

[yomegaman]

Yeah, but with Shaq on the shelf with a bum wheel they could use some help in the post...

Re:I hope not...

[Zinho]

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

Wow, if I had mod points I wouldn't be sure whether to mod you up as funny or down as flamebait...

I've got a friend that would consider that accusation fighting words - he's been carrying Kurt's bubble gum card in his wallet for 15+ years and reveres him as a demigod.

Of course, I never watched him play a game (that I know of), and couldn't careless personally (I really hope that my firend doesn't read /. or he'll probably hunt me down).

What is the world coming to...

Re:I hope not...

[maw]

What was funny was wneh Kevin McHale clotheslined him. Hehehe.

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

[Graymalkin]

While Rambus memory might not be the best design in the world your arguments for why it is garbage are retarded. The memory is interleved and runs hot. There's nothing wrong with interleved memory designs except the fact you have to buy two modules rather than one. If you use memory from different manufacturers in SDRAM based systems you can wind up with a system not booting too. It depends on how robust your memory controller is. They run hot but have twice the memory bandwidth of PC2100 memory. That sort of tradeoff is always inherent in a computer. You could say RDRAM is garbage because of the limited number of suppliers or patent issues or memory even the high latency. Instead you picked the fact it was interleaved and hot.

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

[Anonymous Coward]

>> It depends on how robust your memory controller is.

LOL! Talk about a fudged statement. Back to reality, there are many factors that could cause two sticks of RAM to not work alike. Quite more often than note, the marginality of one or both of the sticks of memory plays a larger part than anything else.

If you buy name brand SDRAM, it doesn't matter who it's from, as long as it meets the specs of the board it will work fine. RIMMs are hit and miss no matter how you mix and match them.

>> They run hot but have twice the memory bandwidth of PC2100 memory.
I can install a water cooling unit and overclock my 1GHz Athlon to 2GHz. You'll never see me selling one though. Every bit more of marginality makes the system that much more difficult to support. What's next? Mains leads to the motherboard because it's the only way to get enough juice?

Maybe for people who have to have absolute speed above everything else, but for anyone with common sense and an allowance (whether from Mommy or from the boss) SDRAM is a no-brainer.

>> That sort of tradeoff is always inherent in a computer.

Not in any computer I will ever build or support.

His argument is quite valid.

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.

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

[Chris Burke]

It depends on how robust your memory controller is.

While that is a true statement, it is not true that it only depends on the memory controller, which makes the thrust of the statement false. If the variances between your RIMMS is such that it goes outside of the tolerances of the channel (which are measured in tens of picoseconds), there ain't shit your memory controller can do.

They run hot but have twice the memory bandwidth of PC2100 memory.

Unless you are only using one significant digit, that's just not true. Dual-channel PC800 RDRAM would provide 3.2GB/s theoretical, while single-channel PC2100 is 2.1GB/s. 3.2 is not twice 2.1... It's just over 50% more.

But even that doesn't really excuse running so hot. Dual-channel DDR (which has 25% more bandwidth than RDRAM, not that it helps the platforms that have it) isn't incredibly hot.

Though oddly in the end I have to agree with you... Interleaved and hot are kinda silly reasons to avoid rambus. Especially interleaved, because that only -helps- . :)

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

[Pussy Is Money]

Garbage is perhaps not the best way to put it. Obviously there are tradeoffs. The argument I think the gentleman was making is that the tradeoffs that RAMBUS made are particularly bad. Especially considering the cost and cooling requirements of personal computers, interleaved and hot are two very big problems.

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

[Graymalkin]

I was assuming the dual channel PC1600 RDRAM which is the highest speed RDRAM I've read about. Of course this is a comparison between the slowest DDR RAM speed and fastest RDRAM speed but that is the qualifier for that statement. It was just to show that the heat generated is inversely proportional to the speed of the chip.

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

[Graymalkin]

What the fuck is it with people not reading everything people post. I was saying the argument of interleved memory and heat is a retarded basis for disliking RDRAM. I fucking said one of the problems with RDRAM is the high latency. Issues with a system's implimentation (ie. heat) as less important than a system's performance. The guy not liking RDRAM based on heat issues and the fact it is interleved is a bit ridculous when there are far worse problems with it, like latency as I originally said and you decided to repeat for your own benefit. RDRAM is a crappy technology and I wouldn't buy a system using it. I used to have a HX chipset in a Sony Vaio that used EDO RAM. The HX chipset was interleved so like you said I had to upgrade my SIMMs in pairs from the same vendor. This is a logistical hassle and prone to errors on my part and on the part of whoever is supplying me with the chips. I don't know why someone would design a system with more heat inherent in its operational design. I wouldn't personally as I am not a proponent of RDRAM or Rambus as a company. Because I said the guy was picking on two minor flaws in the design of RDRAM doesn't mean I have a Rambus t-shirt and posters on my walls.

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

[Anonymous Coward]

heat generated is inversely proportional to the speed of the chip

Inversely? Maybe you mean directly?

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

[choco]

>I was saying the argument of interleved memory and heat is a retarded basis for disliking RDRAM.

And as someone who build systems where reliability is a major issue - I am saying that heat is a damn good reason for disliking RDRAM.

If a component runs hot - that means it is using more power than a component which runs cool.

Fans create Noise

Fans are unreliable

Heat itself lowers reliability - a good "rule of thumb" is that every 10C rise in temperature halves the life of a component.

So Rambus = larger power supply, more cooling, more heat, more noise, less reliability.

Disclaimer - I am a bit biased - I mainly build systems which sit in racks in places like Telehouse, systems which are used for Telecomms therefore they must be seriously reliable. Heat and Fans matter more for me than they probably do for most desktop users. But the same basic principles still apply to everyone.

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

[Anonymous Coward]

"While Rambus memory might not be the best design in the world your arguments for why it is garbage are retarded."

I don't like RIM modules purely based on their name.

I guess you could say I'm ANAL

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

[Anonymous Coward]

But see, if you actually READ the article, it talks about the futures of DDR and Rambus. The article is stating that DDR is starting to approach its upper limit because of its parallel achetechture. Whereas Rambus is serial and can sustain higher clock rates.

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

[Graymalkin]

I think for many non-professional buyers heat is definitely not an issue in terms of overall lifespan of the system or in terms of noise. They are afterall buying a chip that gobbles down 70 watts needing a heat sink so large as to need mounting to the chassis of the systems. The P4 systems aren't exactly sipping electricity to begin with. The heat generated by the RAM is marginal compared to that of the processor. You definitely wouldn't stick a P4 system in the closet with shitty ventilation. The point I was trying to make which I think most people missed is that RDRAM has more flaws then heat which is an implimentation issue, not a design issue. There's more pressing design issues like high latency which I think take precedent over the heat problem which has more to do with the manufacturer of the RAM than with the actual design of the system.

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

[Chris Burke]

Well, that's pretty disingenious since the guy was talking about the incredible heat generated by a real RDRAM system, meaning one that ain't got PC1600 because you can't actually get any yet.

Re:first canadia post

[Anonymous Coward]

I bet you're one of them daft canadians.

Canadia, we're all aboot being sore losers.

I refuse to believe tomshardware.com anymore

[Talez]

After their stint with the "take the heat sinks off, watch the athlon melt" FUD, I refuse to believe anything that comes from tomshardware.com.

I just wonder how many poor saps were caught by that rather large untruth and decided to buy a processor that may not be suitable for their needs.

Re:I refuse to believe tomshardware.com anymore

[whoever_you_are]

I think the point of that article is not that the Athlon is going to burn up.

There was a rumor being spread by some kids with AMD processors that the Pentium 4 runs at half speed whenever you do more than just checking email with your computer. They had taken a new feature in the Intel processor and manipulated it into a fault. Tom's article was only trying to explain what the feature is and why it's good, since many people did not understand what it was about. He was not in any way saying that your heatsinks are going to fall off and your processor will burn up, if you use an Athlon processor. It was not an attack on AMD, it was just an explaination of a new feature found in certain other processors.

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.

Re:I refuse to believe tomshardware.com anymore

[Ozx]

Tom doesn't even write half of the content for Tom's Hardware anymore, anyway... People really should learn to read the author credits on each article...

Not that Tom and Tom's Hardware aren't a nVidia/AMD owned FUD machine with less brains that a small primate... Not passing a -j n to build kernels on a SMP AMD machine = stupid, showing your monitor refresh rate for LAME encoding = stupid, selecting items as "winners" that perform in the middle or toward the bottom of your own benchmarks = stupid, ...

Re:I refuse to believe tomshardware.com anymore

[Anonymous Coward]

I'm almost certain that you mean "than a small primate," so I'll only comment on this to say that Slashdot should consider an Edit option.

As for the substantive component of your post, the only concise way I can summarize my response, is to say that I agree with you entirely. Years of reading Tom's Hardware has left me with a rather disappointing picture of the slopiness of most enthusiast computer hardware sites.

Re:I refuse to believe tomshardware.com anymore

[Ozx]

I meant "that a small primate," you fat piece of crap...

Re:I refuse to believe tomshardware.com anymore

[Jemima Fei]

Though some might argue that the role of the preview option is to avoid such errors, I am inclined to agree with you. Either due to a lack of willingness to bother to preview, or simply because mistake initially escape your view, an "Edit" option would be most helpful. I'm almost entirely certain that many would argue that it would be abused (for instance, posting something insightful and receiving a bonus, and then swapping to an ASCII penis). Of course there are numerous safeguards that could be utilized to reduce the likelihood of this.

Re:I refuse to believe tomshardware.com anymore

[Sj0]

It didn't stop a bunch of fools from seeing that and saying "oops, my heatsync fell off! AMD 5UX0RZ!!! 1N73L is 1337!". Understand this: When we're talking about people too retarded to put on a heatsync correctly, the facts are just an unfortunate mishap.

Re:I refuse to believe tomshardware.com anymore

[Chris Burke]

Yes, I realize he doesn't write most of the articles. But it's called Tom's Hardware. By virtue of the name alone, the quality of the content, regardless of who it is written by, reflects on him.

If he doesn't want to look like a dipshit, he shouldn't put dipshits on the payroll.

Re:I refuse to believe tomshardware.com anymore

[Anonymous Coward]

Well I guess that you were here before her...

(529611 558551)

Re:I refuse to believe tomshardware.com anymore

[fred3666]

I think the author went out of his way to say that your Athlon is going to burn up if your heat sink falls off. Actually they made a video.
And while I do remember quite a few of those plastic 486 fans falling off, I have yet to see an Athlon fan ever come off on its own during normal (ab)use.

I lost faith in THG when they started sticking $300 GeForce3 cards into systems with onboard graphics (for testing purposes). Talk about not comprehending the value segment.

Re:I refuse to believe tomshardware.com anymore

[Anonymous Coward]

And we are supposed to roast taco for katz's fucked articles? use your brain instead of thinking like a sheep. The general public will say that tom wrote it and it will reflect poorly on him. But you and I both have brains and WE DO NOT HAVE TO BE SHEEP. Tom did not write it. Tom hates Rambus. Tom spent the better part of half a year roasting Intel over Rambus. You could say Tom made a name for himself by exploiting Intel's fuckups. I'm a Smart Mother Fucker! 166 tested IQ. I have not ever seen Tom use flawed logic. And he only does hardware part time. The rest of the time he is a Trauma Surgeon. SO, since I have read all your posts AND you seem like a smart mother fucker too, QUIT BEING A SHEEP!

Re:I refuse to believe tomshardware.com anymore

[Ozx]

I'll sheep you, you little poop...

Covering Research Costs

[Alien54]

DDR333 manufacturers are starting to get the feeling that the technology is essentially moving too fast to allow companies to reap the benefits of their investment in research and development.

This can be a problem. You should be able to make back the money so you cover your costs. Unfortunately, you may have to have deep pockets to stay in the game for a long time.

toms

[Anonymous Coward]

Why does slashdot post a story every time Tom has a new article? Agreed some of the reviews are really well done but it seems like 2/3 of tom's articles get posted on slashdot. There are lots of other hardware sites out there besides Tom's that have great reviews and information. Just a thought..

Re:toms

[Dave_bsr]

" Why does slashdot post a story every time Tom has a new article? Agreed some of the reviews are really well done but it seems like 2/3 of tom's articles get posted on slashdot."

Probably because we all argue and yell at tom and each other. discussion is good...then again, any article comparing any two products is heavily debated, from distro's to text editors. I guess tom must be special.

not that profound of an article

[2ms]

It's basically just regurgitating a fact we've known ever since we first saw the design of the PIV: The PIV was designed under the assumption that Intel would be able to force the public to spend way too much on CPUS and memory simply b/c there would be no alternative just as there hadn't been for the previous 20 years. In other words, the PIV was designed for RDRAM.

It's not news that a CPU which was designed to use one memory doesn't perform as well when using a different kind of memory. The PIV needs memory bandwidth desperately. How was Intel supposed to know that all those Alpha engineers would go to AMD and give the public a decent alternative? How were they to know that the public would have the option of getting better performance for less than half the money? The PIV is the last processor of an era AMD just put an end to. It's foolhardy to derive the future of a memory technology from its performance in conjunction with a misdesigned processor. If you could test an Athlon with DDR versus RDRAM, of course, the DDR would perform better. Please let's not post "news" from Tom's anymore - Tom's has just really gone to crap.

We'll give 'em the niche market.

[Yo Grark]

Much like SGI boxes for the average desktop, Rambus plays it's part to a niche market. (Though I'll be damned if I'm going to stay in a job that makes this ram standard build)

When price isn't an issue, and politics not a motivater, it's amazing what ends up in the niche.

-Slashot is like a sewer, what you get out of it, pretty much depends on what you put into it.. - Updated Tom Lehr.

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.

OFFTOPIC?!?!

[Luminair]

This gets a 2 and Offtopic?

*DOWN WITH SLASHDOT MODS*

Go back to RUSSIA, YOU MODS!!!!!!!!!!!!

That's the most informative post here. A lot better than the other garbarge you mod up.

Mod me down, FOR AMERICA!!!!!!!

Re:toms [OT]

[VA Software]

If you see an interesting article on another hardware site then submit it. The submit story link is on the left hand side of all /. pages.

How about this FUD:

[Breakfast Pants]

Have a look at the SUSE linux kernel compilations with dual athlons here. As everyone with dual processors knows, compiling kernels is one of the main advantages of having dual CPUs. In this test however the athlon dual boards got worse scores than their single processor counterparts. If they had just run make with the "-j 2" option they would likely have gotten around 80% faster than than the single processor boards.

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:doesn't it depend...

[Wesley Felter]

If I wanted a high-performance Pentium 4 system I'd wait a little while for the Grand Champion chipset that supports two DDR channels.

Re:doesn't it depend...

[ryusen]

now that is intresting... but based on the performance of the nForce (where the memeory bandwidth is twice the fsb, but performance wasn't increased that much) how much better do you think the grand champion is going to be unless they pump up the fsb again...
the new p4 is going to be 133x4=600
dual ddr, assuming it's 333, would be 166x2x2=666
assuming each channel is the same width (i can't remember right now) would that extra 66mhz.x64bit make that much of a difference? might be intresting for overclocking though... according to Ace's the NForce has really good overcloackability. i wonder if it's cause of the eaxtra memory bandwidth headroom...

Re:doesn't it depend...

[Wesley Felter]

The Pentium 4 FSB is 3.2GB/s; even the fastest DDR is only 2.7GB/s, so the Grand Champion uses two 1.6GB/s DDR channels to get balanced performance.

I think the Grand Champion HE has four channels, so you'd get 6.4GB/s; that's probably only useful for workloads with a lot of DMA traffic (e.g. disk and network I/O).

Ouch, troll, and he's wrong too!

[Luminair]

Good try, but you're TOTALLY WRONG.

Been reading too much Tom's Hardware Misinformation Guide?

Hello?

[Anonymous Coward]

Umm, does somebody else want to comment on this? Hello?

Problems with RDRAM

[Wyatt Earp]

I have a few problems with RDRAM.

The first one is the price. It's simply more expensive than DDR Ram as far as I can see.

Secondly, it appears to me that we are getting to the spliting hairs, angels dancing on a needle level here with RAM. Unless it dramaticly increases my boot time, time to do things in a word processor, makes mySQL fly like a greased dolphin, gives me kickass FPS in UT, or makes my G4 fill my bowl with Fruit Loops in the morning, I'm not going to spend one cent more for RAM than I have to.

If RDRAM is 101 dollars for 256 and DDR is 100 dollars for 256, I'm going to go with the DDR and the hardware that supports it.

Re:Problems with RDRAM

[Chris Burke]

Well, not to make Tom's or RAMbus's argument for them... But just look at the benchmarks of the P4 2.6GHz/RMBS vs 3.0GHz/DDR vs 2.6GHz/DDR. It's clear that at those clock speeds RDRAM works much better with the P4. Much better. So if you are doing -anything- where such an increase is meaningfull -- which includes 3d games like UT -- then yes, you should go with the RDRAM.

But if what you get with DDR is "good enough", then of course you should go with that, because it is cheaper.

Re:Problems with RDRAM

[afidel]

Yes, but what about price/performance. I can get 768MB of PC2100 ram for what 512MB of RDRAM costs. Which will I see better real world performance with, 50% more ram or 33% faster ram? Guess it depends on your applications. I know that if I'm running a large animation in 3DS-Max that the extra ram (and not swapping) will get me done hours or days faster than the faster ram!

Re:Problems with RDRAM

[Chris Burke]

If price/performance was that important, you'd be using an AMD system, and RDRAM wouldn't come up. :)

Re:Problems with RDRAM

[afidel]

actually I do =) 2XAthlon 1.2Ghz plus 1.5GB ram=smoking and it didn't cost all that much, think it was like $550 for MB+CPU's+Ram, compare that to the cost of a 2.2Ghz P4 that costs that much!

DDR Rules!

[Anonymous Coward]

What do you mean? DDR is fantastic! I just can't coordinate my moves rightly. Every time I get an RIGHT-LEFT-RIGHT-LEFT I fall down on the ground.

Re:DDR Rules!

[$uperjay]

Funniest... post... EVER!

This Topic Sucks...

[Kubik - The Original]

Who gives a crap if RAMBUS is on the way back? All /. users know that even if the number of channels double, Bill Gates will just jack up the memory usage of Windows. Then everyone will just start bitching about it. Then Jon Katz will just post an article about something that has nothing to do with "News for Nerd". The cycle never ends... ... ...

Re:This Topic Sucks...

[Anonymous Coward]

Why is this marked flamebait.

This *should* be marked as funny.

Y'know why?

Because its funny.

I think the editors around here look after Katz the way a mother bear looks after a cub.

My time with rambus has been awesome

[t0qer]

About a year or so ago, intel was doing demo's of Q3A at fry's to show off the new P4. Economy was good, I had like 4k from PTO that I aquired between jobs so I said what the hell and dropped $1500 on a 850dgb board, processor, case and 128 of ram.

I gotta say, this stuff is hot, my friends have all gone off and bought gforce3's, amd's with DDR. I thought these new cards/systems would have score winmarks well above my own (around 3800 with a gforce2gts) but I was surprised to see they only score 1000 or so more than me.

Out of curiosity, we put one of those GF3's in my system. Without fail I would score about 400 to come in around 6300 3dmarks above my buddies amd1.6. My P4 is just 1.4. Yet even with a lower clockrate the memory bandwidth made a huge difference.

I'm not trying to cause a ruckus here, anyone with deep enough pockets (or access to enough systems) can just as easily do the same testing I did. Bottom line whether or not the moderators like it is rambus systems do provide the absolute best possible performance in 3D gaming. It certainly was expensive when it came out but now with the falling prices of all ram, it's within reach of anyone that want's that extra "oomph" in thier system.

Does anyone know of any AMD boards that use rambus? I'm sorta curious what kind of scores those get in comparison to the intel one's. Anyways thats my comment.

Re:My time with rambus has been awesome

[Anonymous Coward]

Wow! Talk about contradicting practically every hardware site out there!

Everywhere says that except in Quake III, an AMD DDR system rated 2000+ (1.66GHz) will match or beat a P4 system at 2GHz with RAMBUS...

So your mates have seriously poor systems - did they buy then from QVC?

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).

Re:My time with rambus has been awesome

[Zathrus]

Congrats on using a bad benchmark.

Go read the hardware sites. Quake3 does ABSURDLY well on a P4 for some reason. Nobody can explain why. Jon Carmack doesn't even understand why last I read. But because of this, it's not a wonderful benchmark for AMD vs P4 comparisons. If it's all you play, then that's fine - benchmark it off that. Otherwise go look at other gaming benchmarks, like Serious Sam and Anandtech's new Unreal2 benchmark (which is of debatable value, admittedly).

Of course, then you might realize that people who have a clue are right, and that RDRAM costs 2-4x as much as DDR for no performance gain. Or for a performance loss in some cases.

And no, AMD doesn't use RDRAM. Nobody's even bothered to even design an AMD motherboard that uses Rambus. Partly because it makes no sense - AMD is still mostly used by people who are cost conscious, and RDRAM isn't desired in that catagory. Partly because it would be relatively difficult to design such a beast, due to lack of support from AMD. And partly because there's no performance advantage in the real world.

Oh... and even in Q3... consider how much more you spent for RDRAM, a P4, and the premium on the motherboard as compared to a comparable Athlon system. Then figure that out as a percentage of system cost. Then figure out how much performance percentage you gained. I bet the first is greater than the latter.

Rambus? Naw

[Nall]

I'm generally not too fond of Rambus. Why you ask? Look at what's it's done with the N64. Framerate problems, jaggies...etc. I'm just fine with my SDRAM, thankyouverymuch.

Re:Rambus? Naw

[Anonymous Coward]

yea, I always knew it was that damn rdram giving the N64 "jaggies".

Maybe that's why the PS2 isn't fuly anti-aliased either.

Grow up.

What I got out of it was..

[Anonymous Coward]

That Intel needs Smallbus, err, Rambus memory to keep on par with AMD chips.

This is the 'return of Rambus!'?

Please. SDRAM is the standard. DDR is entrenching into that market. Rambus? It's like the Mac - some people wonder, 'What's that?' while the techs laugh at people who have it.

Rambus is a horrible. The technology? No, the company. Not by speed, but for business should we continue ignoring it. They are a horrible company, and despite their products, should not be dealt with as a result.

The technology.. Isn't really any better or worse than SDRAM/DDR save for price. I've seen boxes refuse to boot when two different brand yet same speed/size SDRAM chips were inserted into a computer. I've seen bits 'o SDRAM cause page faults, kernel panics, etc.

I've seen Rambus do the same. *shrug*

Re:What I got out of it was..

[cheezehead]

Please. SDRAM is the standard. DDR is entrenching into that market. Rambus? It's like the Mac - some people wonder, 'What's that?' while the techs laugh at people who have it.

Maybe. But, often it's the non-techs that make technical decisions, at least in a business environment. I'm sure companies like Dell are selling plenty of systems with RAMBUS in it.
Having a technologically superior product does not mean you'll succeed commercially. Having an inferior product does not mean you'll fail. Unfortunately.

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%

[Chris Burke]

Bad as in you have to be aware of dielectric losses in the PCB material.

Which is pretty damn bad! ;)

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

Oh, shit. Well, a word in x86 land is 16 bits, so I'm in the clear, right? I mean, all the macros for 32-bit integers are all "DWORD", right? :)

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.

Re:Tom, as usual, not 100%

[cheezehead]

That's all nice and well, but I am a bit bothered by the fact that there is controversy about the info on THG. I used to trust the articles, now I'll have to apply my own judgement (oh dear...:-).

Re:Tom, as usual, not 100%

[Anonymous Coward]

I'm not in a position to evaluate Rambus; my peers and superiors, however, have the assignment for a, uhm, big project. I've poked around the real world, and only found Rambus as they compete in the DRR world, and been less than impressed with Intel's implementation. I'm interested in seeing the real-world consequences of using a Rambus communication protocol. Is there any freely available data that I can send along to the evaluators to say that Rambus is worse than the alternatives? Posting anonymously for obvious reasons.

QDR. . . .

[Com2Kid]

QDR is coming out soon, (though they are calling it something else as I recall, no idea why, QDR is such a nice logical name, even the laymen can understand it) and it (seems to be?) but a mere advance of DDR technology.

Not to mention how far up Nvidia has managed to scale DDR RAM. Heh. I would like to see RAMBUS get that. :) (it never would, latency is too high, part of the base rambus technology)

RAMBUS would settle down good in a video toaster type of applicance, but that is about it. Video editing seems to be one of its few strong suits.

Besides, I would like to see a Motherboard that is halfway cheap and can support 3-4GigaBytes of RAMBUS RAM. :)

Re:QDR. . . .

[TeknoHog]

QDR is such a nice logical name, even the laymen can understand it)

At least it doesn't have the associations with East Germany.

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:Cost control

[Anonymous Coward]

How do you figure that? It's cheaper to make a system with 128MB of really expensive memory plus a few GB of PC2100 than it is to make a system with 512MB of PC2100? How? This defies all logic.

Re:Cost control

[cheezehead]

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"?

Intel's McKinley chip reportedly has 3MB of Level 3 on-chip cache. Not exactly what you are proposing, but the same basic principle.

Slightly off-topic: there's an interesting column in Embedded Systems magazine where the author expresses concern about cosmic rays flipping bits in this cache. Apparently Intel acknowledges that this may be a problem (they have studied it). Intel claims a 'normal' user should experience this problem once every 1000 years. However, as the author points out, what if every airplane is equipped with a McKinley chip? Apparently there are 7000 planes in the air at any given moment (on average), so would that mean 7 plane crashes a year due to this problem?

To get on topic again: your idea is interesting, but maybe we should try to avoid running monstrous applications that need ridiculous amounts of RAM.

Having said that, I did run into a memory problem when I had to edit a 270MB text file the other day. For some reason, the 512 MB of memory in my machine wasn't enough. Emacs wouldn't load the darn file ("buffer size exceeded"), and Wordpad hung. I tried Notepad (I know, I'm nuts...), and it actually worked! The machine started thrashing like crazy, it took several minutes to scroll, but eventually I managed to do the minor edits. Yes, Linux probably would have done it without a problem, but that was just not an option, so save me your flames. And I hate vi, so don't bother...

Re:Cost control

[Anonymous Coward]

Try Texpad, you'll like it.

textpad.com

Re:Rambus is a bunch of goons...

[Anonymous Coward]

It sure would be nice if the slashcode could know if a poster is has an IQ of under 35 and send a powerful shock through his keyboard, incapacitating him for a couple days. Might cut down on the moron factor of most threads.

Rambus

[themurray]

It runs REALLY hot. The machine one of our techs has needed an extra fan to cool the memory!!

But really Rambus is not the solution, but another technology will finally arrive. Damn it, I want those quantuam computers with 3D optical storage!

How much does Tom get kick-backs for supporting Rambu$? They are one of poster boys for Patent reform for both consumers and the patent holders.

DDR doesn't slow P4, it;s pipeline does

[Gavitron_zero]

The prob with the P4 is it's crazy 20+ stage pipeline. It's design is almost counter-productive... I mean, it scales to really high Mhz pretty easy, but all of those pipeline flushes kill it in anything that isn't a 3-d game. And since most 3-d games use the graphics card almost exclusively now...I kinda wonder why you would want such a crazy processor.

The one thing it's good at, it doesn't get to do...

Re:ALARM! TOGLODYTE MOLD-DAH (Malda)

[Anonymous Coward]

its rosie pal and her 5 sisters to you, BIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIOOOOTCH.

yo homie yo rap yo NIGZ lalala kn33gr0.

MEEPT!!

[Anonymous Coward]

I'm interested in how the folowing comment will be moderated down:

Fuck you all, even though I love you all.

MEEPT!!

Re:ALARM! TOGLODYTE MOLD-DAH (Malda)

[Anonymous Coward]

I feel bad for your wife.

After feeling up his fiance, I felt pretty good. Of course I was drunk at the time. Man, that bitch has fucked up teeth, and a stinky pussy to boot.

PS - I just heard the news on slashdot - Kathleen Fent's box was rooted. Not that it comes to a surprise to anyone, she goes down faster than a jew that just spotted some pocket change in the street.

What's worse is that the "Will you marry me" story was a scam, they were actually engaged Feb 12th.

Don't disgrace a P4 with anything less than RDRAM

[Jchrome]

I didn't like Rambus either. Hated them and the bs lawsuits they busted out with. However, I also knew just how much faster RDRAM was, and reading this just confirms what I've known and been enjoying for months: a P4 paired with RDRAM not only rocks, but it's simply the only way to go if you're getting one. A 20 stage pipe on a proc demands the higher thoroughput dual channel RDRAM offers. If saving a buck is the prime concern, build an athlon XP box!

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:Is Intel's DDR implementation bad, or not?

[nelsonal]

There are at least two, the SiS 645 is fairly new, and not troubled by legal/licensing issues. The VIA P4X266 which I believe there are only a few (lower end) mobo companies making, because of the brewhaha Intel and Via are having over P4 bus licensing issues. Cheapo boards with each chipset are starting at about 70 before shipping on pricewatch.

Re:Is Intel's DDR implementation bad, or not?

[Gogo Dodo]

There's also the ALLADiN-P4 (their caps lock key, not mine), though I haven't seen any motherboards with that chipset.

Metal Will Never Die

[Anonymous Coward]

Fuck Yeah!

Re:I don't have time to read the article

[Anonymous Coward]

The register is reporting that CmdrTaco's fiance Kathleen Fent will be participating in an anti-Beowulf experiment. Any man with $5 is invited to help determine how much of a load she can take before being sexually satisfied. The movie will be available from Think Geek.

Re:Rambus is a bunch of goons...

[Anonymous Coward]

Perhaps, but a system like that would stop your post, so perhaps its premature to endorse this system from your viewpoint.

Re:ALARM! TOGLODYTE MOLD-DAH (Malda)

[Anonymous Coward]

I meant Rosie Palm and her FIVE Sisters of Delight.

Have you ever had your G.F. spit your hot load of cum back in your mouth to be sick? Or do you sniff your earwax?

Re:I don't have time to read the article

[Anonymous Coward]

I heard she has sucked on 37 dicks in a row!

Re:ALARM! TOGLODYTE MOLD-DAH (Malda)

[Anonymous Coward]

joseph? is that you joseph?

Show some backbone

[coupland]

Thomas Pabst (who we all respect) posted scathing reviews not only of Rambus the company but also of Rambus the technology. If he is recanting he should do it in person, not through a couple of stoolies. By withdrawing such a controversial statement Tom's site is calling into question both the technical and political validity of his write-ups.

Don't get me wrong, THG rocks and I respect Tom's advice. He knows 10x more than me about hardware. But he should explain why this review is so opposed to the ones he wrote himself...

Re:Show some backbone

[Anonymous Coward]

This article was not written by Tom, read the byline, it was Frank Völkel.

Re:ECU APPSTATE

[Anonymous Coward]

what does NT mean? i am hearing it often and stoopidity bothers me. please explain

Seriously.

[Chris Burke]

Now, unless you're doing video encoding, I don't know why you'd want a P4 in the first place...

But to even entertain the thought in the first place, you can't be on an ultra-tight budget! So why try to save what will ultimately be a small % off the total price at the expense of performance?

It just doesn't make sense.

Though I still want to see a dual-channel DDR chipset for P4. :)

Re:Seriously.

[junkster191]

Well- the other reason I might want a Pentium 4 is I can't stand how loud every AMD box I've ever owned has been. I realize they have more significant power/heat issues than Intel, but how hard is it to design an effective, quiet fan and a good muffling case? I bought a Athlon XP 1600+ but still use my old Pentium MMX 200 90% of the time because the noise drives me crazy!

Re:Seriously.

[Paladin128]

There is a solution. Use an Alpha PAL8045 heat sink. It mounts directly to your motherboard, and allows you to choose your own 80mm fan. Unless you need the serious cooling to overclock, I'd suggest a Papst 80mm ultra-quiet fan. It does the job well, and you can use them to replace other case fans as well.

If you need help finding one, reply to this message.

I, myself, prefer a loud system, or at least a loud hard drive :)

--Aaron

Re:Test costs remain high, apps support remains ni

[kilrogg]

For those who don't know, IC testers are big ass machines used to production test chips (i.e. screen out the good chips/dice from the bad ones).

Intel uses testers from schlumberger (their reps are quick to point that out). Typically, a tester cost anywhere from 1-10 million dollars + plus they require a lot of maintenance, calibration, etc. Basically, the faster and more pins you need, the more it cost.

I've worked with some schlumbeger 'KX' testers, they're a big pain in the ass, unreliable, and are badly designed: just shutting the thing off can break it!! (especially if you use the emergency off button).

There is another choice, however, you can use 'bist'(built-in self test) and have the chip basically test itself :-). This allows companies to get away with using cheaper, more reliable testers.

Quantity vs Quality

[daviddennis]

First, it looks to me like RDRAM is still about double the cost of SDRAM, according to Tom's Hardware's own price guide.

They have $93 for 512mb SDRAM and $175-250 for 512mb RDRAM.

My question is this: Let's say I have a choice between 512mb of SDRAM and 256mb of RDRAM. Would the SDRAM not almost always be faster because RAM, however slow, trumps swap space every time?

In other words, isn't the amount of memory I have more important than how fast it is?

Many moons ago, I had a SGI O2 workstation. Tremendous memory bandwidth, but memory that cost 10x more than anything else. As a result, it could be embarassed by lesser machines, since I couldn't afford to load it up with RAM.

I see Intel repeating the same mistake when it decided to focus on RDRAM.

Apple is putting L3 cache in their G4s so that the use of expensive RAM is confined to a relatively small and affordable amount. I can upgrade my PC133-equipped G4/450 dual processor to the latest 1ghz dual processor, put my 1.5gb RAM in it, and fly. That seems like a good compromise to me, maybe better than going to DDR, which I would have to buy new.

Thoughts?

D

Re:Quantity vs Quality

[Jemima Fei]

First, it looks to me like RDRAM is still about double the cost of SDRAM, according to Tom's Hardware's own price guide.
They have $93 for 512mb SDRAM and $175-250 for 512mb RDRAM.


Tom's price guide is not usually the place to find the lowest prices on hardware. But it's also not entirely fair to compare RDRAM prices with normal SDRAM prices, because of the performance difference and shrinking platform availability for non-DDR SDRAM.
If you look at pricewatch.com, though we can find some prices like these:
Samsung 512MB RIMM for $156 + $9 shipping from some provider called 11cb. I simply picked this as the cheapest Samsung-labeled provider, since Samsung provides some of the best RDRAM. Keep in mind for interleaved operation that you'll actually be using two RIMMs, so you might instead want to compare 2x256MB or simply look at 2x512MB for your other RAM platforms.

From SDRAM and DDR SDRAM I'll just pick Crucial/Micron, while they won't be picked as the high-performance provider (people would be more apt to pick Mushkin or Corsair for performance) you'll see much less flakiness than with a non-labeled generic provider.

(Shipping not mentioned with Crucial, check their site)

Micron 512MB PC133 CL2.5 $139 + $10 shipping from "Alpha International Business Inc."

Crucial 512MB PC133 CL3 $139 Crucial 512MB PC2100 CL2.5 $152

Now for the faster DDR I'll pick the lowest reputable name-brand item, since Micron/Crucial don't offer all speeds of DDR, currently.

Corsair 512MB PC2400 CL2 $187 + $9.74 shipping from Googlegear.com

Mushkin 512MB PC2700 CL2.5 $211 + $9 shipping from Mushkin


Now, I don't intend for you to read too much into this, but provided you stay with "non-crap" providers of memory, the closer you come to the performance levels of RDRAM, the less you see a price difference in favor of SDRAM.


My question is this: Let's say I have a choice between 512mb of SDRAM and 256mb of RDRAM. Would the SDRAM not almost always be faster because RAM, however slow, trumps swap space every time?

In other words, isn't the amount of memory I have more important than how fast it is?


If you aren't being limited by the amount of system memory, then no. Provided that for your application at hand you don't need more memory than you currently have available, swap access differences really aren't an issue. Does it matter that you have 1GB of memory if you only use a small portion of it for something other than disk cache, when compared to 512MB of memory with much more bandwidth?
If you don't need or can't use the bandwidth, then of course it's not overly useful. Or if you need to access more data than you can realistically ever store in memory, then there will be a point where memory bandwidth is made mood by increased disk access. It's a matter of application and necessity of your processor.

The Pentium 4 sees very realistic gains from using RDRAM versus DDR memory, because of how it was designed.
At one point Intel was being embarrassed by the absurd cost of RDRAM, but times have changed. It's continued to go down in price, and DDR and normal SDRAM have recently increased in price.

Re:Quantity vs Quality

[Detritus]

In other words, isn't the amount of memory I have more important than how fast it is?

No. If your software's working set is smaller than the amount of physical memory, you are better off with the faster memory. You can create software workloads that make either configuration look better.

Re:Quantity vs Quality

[Anonymous Coward]

No one's talking about SDRAM here, it's all DDR RAM now. Get with the times. The price difference between RDRAM and DDR RAM is much smaller than with SDRAM.

Why does everyone hate rambus?

[TheKingOfCowards]

I've noticed that alot of slashdot posters hate rambus with a passion. Why is this? I have not really been paying attention to the history of rambus vs. DDR. Other than hearing stuff with "rambus sucks" or somthing like that in them. Could someone post an article or explain the history of rambus and why people hate it?

Re:Why does everyone hate rambus?

[Jemima Fei]

Historically RDRAM has been plagued by cost, which has deterred its adoption, but this isn't why average reader of Slashdot dislikes it. They might claim that they think it's technically inferior (PC600 and PC800 have more latency than SDRAM, but PC1066 and PC1200 RDRAM will likely be out within the next year), I think a large majority of hatred of RDRAM comes from Intel and Rambus's business practicates.
Intel and Rambus were hoping to strangle the market into adopting RDRAM in order to hurt Intel's competitors, and when this failed (RDRAM's prices lead people to adopt PC133 and then DDR), they attempted to obtain royalties or sue developers of alternative memory technologies for patent infringement of one form or another.

Re:Why does everyone hate rambus?

[Anonymous Coward]

Also, Rambus sat silent during memory technology meetings where technical things were discussed. They were required to disclose any patents they had. But they had a few in the pipeline that had not been official yet, and they were things other companies were going to use. So after the conference they made damn sure the patents went through and sued the other companies for infringing on patents that hadn't been patented yet.

Re:Why does everyone hate rambus?

[bani]

rambus is disliked because they comitted fraud.

Fortunately, in May 2001 a Virginia jury convicted rambus of fraud.

Unfortunately, they fine the jury imposed on Rambus ($3.5m) was reduced to a mere fraction of the original penalty ($350,000) by state laws capping the limit of punitive damages.

A mere slap on the wrist for a company which acted so unethically.

Highly Doubtful

[CMiYC]

I work for a test and measurement company and we sell logic analyzer tools for both DDR and RAMBUS. I service at least 1 site of each of the major computer manufactuers and I can tell you none of them are even considering RAMBUS. In fact, I can't remember the last time someone asked me about it. The only thing I have consulted customers on is the future of DDR. If anyone was interested in RAMBUS, I'd at least be hearing murmurs. Keep in mind I am only looking at computer manufactuers, not the 3rd party Asian motherboard manufactuers. Who knows what they are doing.

Tom contradicts himself?

[DrMrLordX]

Instead of focusing upon some benchmarks with overclocked CPUs squaring off against one another, why not pay attention to another article from Tom's that shows some benchmarks of the P4 2.0 ghz operating at factory spec on a number of different P4 boards with different memory configs(P4 + i850 + RDRAM vs P4 + Sis645/ViaP4X266/ViaP4X266a/i845 + DDR). As you can see from that article, at least on the pre-Northwood P4s, DDR did pretty well on the non-Intel chipsets, particularly on the Sis645 when PC2700 was used.

Re:Tom contradicts himself?

[phoenix123]

That's right folks. This article (RIMM vs. SD-RAM) at Tom's Hardware shows even more the self-contradiction within that site. That time Rambus was smacked down comletetly. It even has figures and (some kind of) proof for its conclusions. Something I am missing at the newest pro-RIMM article
Have you noticed? There is no nothing to back-up their claims! No cool diagram showing some magic number difference on Tom's Hardware???.

Seems like Rambus found a good use for all those precious marketing-$s.
Don't trust Tom's anymore. It's labelling ads as editorial columns!

If I have to decide...

[Decimal]

I'll take the Glad Bag!

Re:you're not funny you fag troll

[Anonymous Coward]

Ah,but who is the bigger pansy: the original idiot with no life or you, the guy critizing the no lifer?

I say you.

sheep, ignorance, patents...

[rebelcool]

they've relied alot on patents with a few shady practices. Your average slashdot reader doesnt know anything about ram design though, so they tend to just follow the loud critcizers..

I know of no company doing analog BIST right now

[Anonymous Coward]

I'm sure there is already memBIST built into the memory for automatic test and repair, but that's just at a logical level. Nearly every RAM out there today has it, and especially embedded SRAMs that I'm used to dealing with.

The real issue here then becomes analog BIST. To test analog, you frequently have external known reference voltages and impedances. How are you going to do this if you have problems? Usually the SERDES cells have a loop-back mode where TX and RX are tied together, but that says nothing about their interoperability with components on the outside - only with what's inside. On that basis, you can exclude any bonding/package problems (but assembly yields are typically high anyway), and external drive capabilities under certain and varied conditions. And, again, we want to make sure we have working problems on the outside so we don't have the same problems that arose with the i815 chipset so that customers are told about a kludge for a solution. I'll bet diamonds to dollars Intel didn't get enough apps-side support from Rambus to be able to solve that problem properly, turning that chipset into garbage.

Tom's Credibility

[Anonymous Coward]

After Tom's Hardware printed that article where all they glossed over the PS2 and Gamecube as if they were useless (and all the information about those 2 consoles seems to have come straight from Microsoft) and then devoted 12 pages to details about the Xbox, I'll never really believe what they write about products again. I'm not even really sure if I believe all their AMD benchmarks, anymore. They were clearly bought by MS for that, so they probably were bought by AMD for those other articles.

Re:Tom's Credibility

[Anonymous Coward]

Yah, God forbid they actually like a Microsoft product.

The highest performing chips ever use rambus

[ppetrakis]

If you doubt these words take a look at Compaq's
Alpha EV7 with it's RAMBUS controller ON CHIP.
Just because PC architecture is limited by chipsets with limited memory bus bandwidth does not mean there are no other uses for such a memory architecture.

Peter

Re:I know of no company doing analog BIST right no

[kilrogg]

For a simple enough high speed link, you can do I/O bist using external loopback. Or alternatively, use a second know good device on the loadboard to test it. Of course it should be used only for detecting manufacturing defects, you need to do a lot of testing up front to convince yourself that the part will actually meet specs across process variations.

Run a skew lot and cross your fingers...

[Anonymous Coward]

Oh, and by the way, I like that nice shiny red button behind the Scumburger testers. :D

RAM in Laptops

[XBL]

To be honest, I really don't plan on buying another desktop system ever again. If needed, I'll build one out of spare/cheap parts. Why? Because I use my laptop anymore 99% of the time.

There will NEVER be a laptop with RIMMS in them because they are too damn hot. Unless the design of them drastically changes in some unknown way, this "NEVER" is a fact.

I think there are some DDR laptop solutions in the pipe now. Yet, there is the problem of the slower system bus speeds on laptops, so it will not matter much until that's fixed too.

Hard to find rambus now

[Anonymous Coward]

I couldn't get any for a gateway e4400 I have at work (not what I would have spec'd, but no matter) from Crucial or Kingston. I finally bid on Ebay.

The REAL speed of DDR...

[SkOink]

If you guys think DDR isn't fast enough, just TRY to pass Paranoia on maniac mode... It's got more steps than a ladder to the moon. :D

Woah, you meant RAM?

Rambus Comeback @ Tom's Hardware

[Anonymous Coward]

Sounds like Tom @ Tom's Hardware is finally selling out to Intel. Can't beat'em - join'em, right Tom???

Rambus is crap. P4's were designed for Rambus memory - that is why they run better with it. All this means is that BOTH the P4 and Rambus are crap designs.

And no, we didn't make a mistake in judgement in the beginning about the Rambus/P4 designs. The market speaks for itself. P4's are only being sold to technically un-inclined users who don't know any better. They are not being used in critical or demanding markets and never will be.

Re:Quantity vs Quality -- RAMBUS now affordable

[cheezehead]

In other words, isn't the amount of memory I have more important than how fast it is?

I think this is probably true.

However, as you point out, the price difference between RAMBUS and SDRAM is now very small. According to Sharky Extreme the difference between 512 MB of SDRAM and RDRAM is about $80, and DDR RAM (PC2100) is actually more expensive than RDRAM! So if you plan to put 2GB in your machine, SDRAM is appreciably cheaper, but if you plan to do that, you probably plan on some serious hardware as well, so you'll probably spend $3000+ (probably coulnd't get a motherboard that would take 2GB SDRAM anyway...).

My point is that both SDRAM, DDR RAM and RDRAM have come down in price dramatically in the past year (although memory prices seem to be on the rise again). The price difference is very small when compared to the total price of the machine, so why bother? I have nothing against DDR RAM, but it'll have to win on technical merit nowadays.

As an example, I had to specify and buy a PC for my job some weeks ago. Now, this PC will be running a very specialized application, and nothing else. No CD burning, MPEG/MP3 encoding, no image processing, and no games. I like a cool machine as much as the next guy, but I simply could not justify putting more than 512 MB in this machine. Same for the hard drive, 40GB should be more than enough. A decent monitor was a requirement however. So why save $80 on memory when we're spending $700+ on the monitor alone?

Summarizing: if 512 MB is enough for you, why bother? If it isn't enough, you'll likely spend a lot of dough anyway, so again, why bother?

Uh, 256MB RDRAM is CHEAPER than DDR right now

[Heretic2]

You arguement is invalid. DDR isn't so cheap anymore. The only thing DDR has going for it--besides the Athlon--is super high density. 1GB modules.

RAM is not an issue in the long run

[Anonymous Coward]

The funny thing is, everyone is guessing "which will be the better RAM in the future?", the truth is DDR is the best bang for the buck right NOW. In the long run, RAM won't even be on the motherboard! It'll be on die with the chip. Remember the days when Cache was on it's own seperate chip on the motherboard? There are already plans right now to put ALL of the ram on the processor die. We already have processor designs w/ 8 meg caches... how long before we hit 256 megs and don't need ram? or 512 megs? The primary problem with RAM is that it is designed to be a cheap,intermediary storage for data between the processor (and cache) and the hard drive. As the difference in speed between the processor and the ram increases (which it will exponentially b/c ram cannot increase in speed at the same rate the processors can), cache sizes will continue to grow. At some point, the cache will become large enough to act as if it were RAM, and RAM as we know it today will likely disappear or become imbedded into hard drives as a larger buffer. Then, we'll have to figure out a way to focus on increasing hard drive speed by trying new devices... like using RAM drives w/ internal power supplies to prevent our data from being wiped ;-) I believe at the current rate of increasing processor speeds and cache sizes, we'll see RAM vanish within the next 10 years... Maybe by then we can work on faster hard drives ;-)

The falling off heat sink thing is FUD!

[evilpaul13]

The falling off heat sink thing is FUD. In the several year period working repairing PCs on a daily basis, I came across ONE heatsink having fallen off. It was on an old K6-2.

I've never seen a newer Socket 370 or Socket A HSF fall off. They are simply on much too securely. This included boxes that were moved by idiots to different locations so gently that the HDDs were destroyed. The HSF was still attached though.

What percent of CPUs do you think have their heatsinks fall off? Less than 1% probably.

Making people worry that their computers will melt down over a less than 1% chance of something happening, is spreading Fear, Uncertainty, and Doubt.

Tom is wrong!

[PhoenxHwk]

Ok, I'm going to try to keep this short even though I could talk about RAM for days on end.

There are quite a few inaccuracies in Tom's article here. From calling Rambus serial to some of his crazy statements about DDR not handling impedance balancing as well as Rambus, I'd call him insane.

However, its a statement like 800 Mb/s/pin is better than 333 Mb/s/pin that really gets me fired up. Yeah, those are the speeds, but he's losing so much of the big picture that its scary. The more important thing to account for is the geometry of the RAM configuration which includes things like the number of data pins in parallel. For instance, a DDR bus has 64 while Rambus uses 16. Lets see here, multiplying 64*333 and 16*800 (pins*(Mb/s/pin)=Mb/s total) shows that DDR333 is now twice the performance of Rambus. Yikes.

And then there's his comments about CAS latency. Tom says that CAS 2 will be the factor that makes DDR333 outperform Rambus because CAS 2.5 is too slow. Lets just sketch out what CAS means: it is the time between when a read command is issued to the RAM and when the first piece of data comes out, measured in clock cycles. So basically, he's saying that .5 cycles per random access (to DDR) will make the difference. It won't. First, we're talking .5 cycles out of ~10 (for a purely random access), or a 5% increase. Second, bursting from the RAM almost entirely disguises this so that CAS makes up less than 1% by the time its all said and done. This happens because burst accesses (which are almost all that takes place in a PC, due to the cache structure) pipeline and end up hiding the CAS time over a long period.

Gads, I've been doing too much work with RAM lately.

Re:Sure we can...

[evilpaul13]

Sure we can, you've never seen the i820 PIII chipset with Rambus? It wasn't as fast a the BX with PC100. The PC800 also cost 4x more.

The P4 is a dog. A P4 platform requires: a more expensive CPU, more expensive RDRAM, and the expensive i850 chipset motherboards. And then, it still isn't much faster in apps where it isn't beat by the Athlon.

But, I would recommend it highly to people who use the Lightwave 3d modeling app, and Windows Media Encoder!

Rambus *is* a specialty memory

[Anonymous Coward]

I see Rambus from two perspectives. One is from the perspective of a home computer enthusiast. From this vantage, it is expensive and hot. There is also the fact that a system based around DDR memory performs plenty well for practically everything I do (including games). However, this isn't the extent of the market.

I also work as a design engineer for a defense contractor building computers for radars. These systems are extremely memory and I/O intensive. When they aren't running computations, they are doing corner turns (transposition in matrix speak) This requires tremendous amounts of memory bandwidth, and we use Rambus memory in some of our designs.

Of course, our customer is more interested in performance than price (within reason), and we have the option of using liquid cooled and conduction cooled boards, so heat, while still an important issue, isn't as big a deal.

As for stability, we run our systems through environmental tests that would totally destroy any home computer, and the rambus memory not only holds up - it keeps running. Of course, so does SDRAM.

The long and the short of it is, Rambus just isn't a great choice right now for home computing. They are not, however, without merit - and will likely have a place in the industry for some time.

editing large files

[TheLink]

If I don't need to insert data I use hiew (hacker's view).

I've used it to fix other people's software many times. One case had tcl source code stored in an Oracle database with no official way of altering it. So hiew it was...