Posted
by
Hemos
from the finally-moving-forward dept.
jandrese writes "According to Cnet, Intel is finally getting around to supporting DDR SDRAM in their P4 chipsets. This is a good move on Intel's part, as they need to bring the cost of their P4 based systems down to compete with AMD, and moving away from Rambus is a good start."
This discussion has been archived.
No new comments can be posted.
Seriously when RAMBUS was 10x the price of SDRAM it seriously hurt, but now that RAMBUS is getting close to comparable, I don't see what the point is. In my neck of the woods PC-800 RDRAM goes for about 30% more than PC2100 DDR, which really isn't that much (and dual channel RDRAM is the fastest RAM platform out there). Given that the P4s one redeeming factor is that with RDRAM it has a serious memory advantage, I really don't see what Intel is thinking: Put a P4 with DDR DRAM and it'll get clobbered even more.
Interesting Point, but the price of DDR RAM is still to expensive to replace SDRAM right now. However, once the Northwood gets released in a few months, OEMs may move to DDR because of the sheer volume of DDR that is being used. DDR RAM, however is still more expensive; Crucial sells 256 Megs for 50 bucks, compared w/ 30 for some PC133...But I think thats artificial Christmas Price Inflation (TM).
In terms of high end Intel systems, DDR just isn't that way to go. A couple of months ago, when Intel got to 2GHz, they were beasting on similar Athlon systems. But now, AMD has gone on a tear, heavily ramping up their Palomino core. A 1.9 XP w/ DDR beats a 2GHz P4 w/ PC800 RDRAM in every category except for memory bandwith...If the Intel was using DDR as well, the Xp's margin of victory will be even greater. The P4 relies on fast, fast memory. Give a P4 slow memory, and it will freeze (P4 and SDRAM is a horrid combo). Since RAMBUS will soon be releasing pumped 133 MHz bus memory, I think this is the memory that will help Intel more than DDR. Intel is losing, and has ALWAYS lost, the price battle. I think that if Intel cuts memory performance to reduce price, they are losing their ONLY advantage over AMD systems.
"Crucial sells 256 Megs for 50 bucks, compared w/ 30 for some PC133...But I think thats artificial Christmas Price Inflation (TM). "
Please don't tell me you are seriously basing your opinions on the above! You are bitching about a $20 difference in RAM!!!!????!! At one time 2MB RAM cost $2000! 2MB to 4MB was thousands of dollars, yet you complain about $20? Yes, I know that everything is cheaper now, but you have zero perspective on value.
"but the price of DDR RAM is still to expensive to replace SDRAM" - ah, there was a point were RAM was too expensive to replace your floppy drvie. Again, this is dated but don't bitch because you can't buy that "extra juicy" gum instead of the "regular juicy" gum; we're talking 3-4 magnitudes of difference here.
Another OT topic-
"Intel is losing, and has ALWAYS lost, the price battle." I guess Apple should just pack it in?
It's nearly painful to watch these youngins complain about spending an extra $20 on memory!!
It's nearly painful to watch these youngins complain about spending an extra $20 on memory!!
You're missing the poster's point. His point was that the *current* prices of RAM, CPU etc. are in favour of AMD rather than Intel. For that, it doesn't matter what the prices of RAM was just under a decade ago (I remember those days too, and I know that I just payed 25CAND for 256 MB RAM... SD sure, but 256 nonetheless...).
Therefore, your arguement that it doesn't matter about the price difference because it is 20$ now instead of 2000$ doesn't make sense: the price is higher for a P4 system, even if it is only 20$, but that's an extra 256 MB (well, pro'lly 128 MB DDR) RAM... choose the system you want.
The guys at Intel aren't stupid. They invested millions into marketing themselves as a consumer brand before it was cool to do so, and it's been the best move they ever made. AMD, for all its trying, hasn't even registered on the radar of most consumers. Intel uses this to its advantage to charge a hefty price premium. After all, they're the Coca-cola while AMD is just the RC.
So Intel _is_ winning the price battle, since the winner of the price battle is the guy that gets to charge more money and still sell 80% of the processors, not the guy that sells them for half as much to push enough volume to break even. After all, Intel could sell their parts for much less than they do without actually losing money, but they don't have to.
And, as AMD's recent relabeling of their XP line has shown, clock speed is still king. Nobody has ever successfully dethroned it as the single number consumers care about above all others. Which is why Intel has won that battle as well.
P4's with DDR aren't in any way related to RAMBUS's performance as much as keeping low-margin systems affordable and still fast. That's why you'll see P4 Rambus and DDR boards out there, fighting it out for the price/performance sweet spot.
At this point, Intel is more worried about Sun than AMD, since Sun is the lone vendor not committed to Itanium/McKinley. They've also got the high-margin Xeon processors competing with Sun's mid-range offerings. This is where the interesting things are going to start happening, but you won't be hearing about it on Tom's Hardware.
All that is true. People will continue to buy Intel processors for the same reason everybody buys Coke, because it's there in front of them of the shelf and the name is stuck in their brains by advertising. It's the american way.
Now there are other factors. Until recently Windows didn't run too well on any of the available Athlon chipsets. From what i hear this has changed. On the other hand, the price/performance ratio assuming some sane OS like BSD or Linux was much in the favor of the Athlon.
My work has bought four Dell systems with the SDRAM package (my boss thought he got a great deal). They are all 1.7GHz systems, but the old 1.2GHz systems with the RAMBUS still at about the same speed. I'm basing this on Adobe Photoshop as we have the machines set up rather close to each other.
I told him that I could build some AMD machines that would be speed demons compared to the new Dells that he picked up, for about 500 less, but all he did was whine about the time it takes to support it if something goes wrong. He also has a problem paying 35 bucks for a cooler when there are $10 coolermasters (IMO these are all terrible) that will "do the same job."
(and dual channel RDRAM is the fastest RAM platform out there).
Speed isn't everything... RDRAM has a great deal more latency than DDR. In many cases, RDRAM performs significantly worse than even SDRAM.
Besides that, there's the evil factor, considering Rambus believed more in the policy of suing for royalties as a revenue model instead of producing and selling a decent product.
If SDRAM is challenging RDRAM with a cpu that can make use of much higher bandwidth RDRAM offers, I'd rather blame the software. It would mean that (pick any two or all three) data access is essentially random, in very small chunks or alignment is bad. Otherwise hardware prefetch (helps little on SDRAM) or bandwidth for could compansate for it. Few apps have to use such an access pattern, it is usually programmer's laziness. There is no excuse for bad alignment except in extremly tight memory, byte or word size data condition.
..you seriously think that for a majority of applications people have any clue on how to access memory? People write Visual Basic and Java..
Actually - fitting the data you use into L2 cache is much more important IMO.. I have seen factor of 3 improvements in some of my code.. Local alignment (UNder 4k blocks) matters less from my benchmarks..
I use Java to play Pogo games and "Shock the Monkey". I haven't seen Visual Basic in a while; It is not robust at all.
L2 cache is very important, and Intel's new chip will have increased cache. In the classic (slot)Athlons, the L2 ran at a third the speed of the chip, but nowadays, with these newfangled modern procs, L2 has gained in importance, running at full speed...But, then again, the Celerons and Durons have hacked, smaller, L2s...
Many (if not most) industry-specific Windows apps are written in VB. Need a data-entry database front end? VB. Need a telemarketing app? VB. Need to access scanned images of documents? VB. Need a really lame data conversion app? VB.
I don't claim to have huge experience - I've only been VB-aware for ~3 years - but every non-web niche Windows application I've ever seen has been a VB app. There's one possible exception, but the company that made it is out of business - to a large extent because of their development costs - though I'm not convinced they don't use VB.
People writing JVM's understand about memory alignment though. Anyway, IIRC the memory alignment hassles on RDRAM are about getting the structures to fit in 128 byte chunks because that's the minimum amount that can be requested from RDRAM.
Come on. The vast majority of apps don't need high bandwidth memory access. Most are busy spinning cycles waiting for user input. If you are finding you are getting poor memory performance, chances are you are running a lot of apps and the context switching is making your cache nearly useless. This is also where lower latency memory helps.
The few programs which might be blocking due to insufficient memory bandwidth (databases, games with large/many textures and the like) often do have the critical pieces tuned for cache performance.
Claiming that all apps should be tuned for optimal memory access is just silly. You do it where it's easy or where it increases performance significantly. Any more and its just a waste of time and money.
Before blaming things on lazy programmers remember that developer time isn't cheap. If I routinely get things done in one day and you take five days but get 5% better performance, who do you think is getting cut next time there are layoffs?
My comments were only applicable to programs that stress memory subsystem, otherwise you won't see much of a difference between sdram and rdram anyway. So I didn't (at least didn't intend to) claim that all (even most) apps should be tuned for optimal memory performance.
I don't agree with you comment about context switching emphasising latency. Context switching takes doesn't happen frequently, time slices are too long from a cpu perspective (linux defaults to 10ms, I guess that corresponds to ten million or so cycles.) Contexts are loaded in burst rates so latency would have little effect on overall performance. Offcourse it might be that context switching occurs more frequently because programs are releasing their time slices, but that would mean you don't need performance either since your processes are idle. A busy process still gets its ten million cycles, wasting a tiny percentage on context switching - high latency or not.
Until Intel get's the next gen of Northwood with 133Mhz FSB out the door (they are saying Q2/2002 last I checked) then the i845 with DDR-SDRAM will have a 100Mhz FSB (quad-pumped magic not withstanding). That means that any DDR you use will be running at PC1600 speed. You will have to wait for P-4s with a 133MHz FSB (Quad-pumped to 533Mhz) for the P-4 to be able to really use PC2100 DDR-SDRAM.
Benches of AThlons with PC1600 were underwhelming.
Besides that, there's the evil factor, considering Rambus believed more in the policy of suing for royalties as a revenue model instead of producing and selling a decent product.
The thing that nobody ever understood about Rambus is that they are not actually in the business of producing RAM. All along, they're business model was to come up with good ideas and license them out. There is nothing wrong with that.
The only thing that bothered me about Rambus is that they didn't disclose the patents they held while they participated in developing memory standards. That was crappy business.
And it is latency that usually counts, peak transfer rate is not sustainable anyway. A dual channel DDR-SDRAM platform would be faster than dual channel RDRAM platform, single channel ddrs are already competing with dual channel rdrams. With 166*2MHz DDRs on the horizon, I think it is a very sensible solution.
but now that RAMBUS is getting close to comparable, I don't see what the point is.
The point, in my opinion, is that Rambus has too high of a lawyer-to-engineer ratio for my tastes. I prefer a company that chooses innovation over litigation because I have a lot more faith in their product down the road.
"...Rambus has too high of a lawyer-to-engineer ratio for my tastes."
That sounds pretty reasonable...but:
"I prefer a company that chooses innovation over litigation because I have a lot more faith in their product down the road."
We are talking about memory here. What possible impact could future innovative products have on whether you buy an intel system that uses rambus or not, today?
I say buy the fastest thing you can at the price you are willing to pay. If rambus fits the bill, then go for it. Who cares what they put out down the line. It isn't likely to be compatible with your current system.
Assuming the litigious company (Rambus in this case) doesn't use the money to sue all their competitors out of business.
I'll support DDR SDRAM for that reason alone. The fact that it performs better in everything except MPEG encoding tests is just gravy.
Oh well, Rambus's stock value is in the toilet and there are rumors of the stockholders suing the principles for criminally bad management. I can't say I'd cry over it.
This isn't about price.
It's about Intel failing to grab more control of the technology that makes up a PC.
And that IS a GOOD thing.
I've got nothing against Intel.
But monopolies usually suck for everyone but the monopolist.
also consider that ddr is untested (well on th eopen market) with p4s... we don't know if it will work well at all. ddr wont give as high a bandwidth as dual channel rd-ram... will that small amount less be a bottleneck? who knows for now.. i say wait and see and then judge if the maybe $100 you save is worth it
of course dual channel ddr-sdram... gah... drool..
of course dual channel ddr-sdram... gah... drool..
Isn't the nforce dual channel DDR? I know I was seriously excited about it, but rather underwhelmed when the actual numbers came out. Then again nvidia has a funny way of doing that again and again: Releasing something with subpar numbers to get the moaners and whiners yabbling (in a way they'll soon regret), and to lower expectations, and then coming out with a final product that blows people away.
DDR-P4 combo not tested in open market? Intel's chipset is the third one! Both solutions from via and sis performed rather good, keeping up and sometimes even surpassing dual channel rdram solutions.
There are at least two family of boards operating on dual channel sdram principle, nforce boards and some server stuff company's (anand's older servers were powered from them, you can find the name in their older news archives.) One works with athlons, other with p3's but idea is similarly applicable to p4 chipsets.
ok my apologies for that... i guess i missed the reviews on those boards (anyone got links) i'm usually halfway decent about keeping up with new chipsets
yes i know the nforce is dual channel ddr, but it still needs work...
one thought i had was dual channel ddr with an smp athlon board since the nforce really doesn't seem to take much advantage of it
No dual channel Rambus, is what what give the
P4 its bandwidth advantage. But as Nvidia's
Nforce shows dual channel DDR is quite possible,
and gives even more bandwidth. Ram module
for Ram module, DDR has a high bandwidth and
is cheap.
Put your CPU to the same pricing that AMD is doing, do an equivalent $/mflop, you'll notice how much rambus memory isn't the biggest chunk of the pie unless you go to 1GB and above.
Oh, that would chunk in your profits... right:), better Rambust than you... Oh well... if at LEAST one of you two suffer, I'll live with that:)
The patent drove the cost of DDR RAM up so much relative to competing technologies that the tech died. Perhaps this will be a lesson for other companies that want to patent something in a world where there are alternatives.
I wanted to illustrate the similarities between this and Sony's patent related to Beta videocasette tapes, but it would have been sure to result in a flamewar.
The patent drove the cost of DDR RAM up so much relative to competing technologies that the tech died.
I'm hoping you meant to say RDRAM.
Anyway, the patents had nothing to do with the price differences between RDRAM and DDR SDRAM, it was all due to manufacturing costs. I remember a little over a year ago Kingston was bragging about their 30%(!) yield on PC-800 RDRAM chips. When 70+% of your product doesn't pass QA, that's definately going to drive your costs up! Additionally, manufacturers had a fair amount of retooling to do before they could make RDRAM, and high setup costs get passed on to the consumer. As I recall, RDRAM also has a bigger die size than DDR SDRAM (I could easily be wrong, it's been a while since I cared) which would also drive up costs.
In contrast, DDR SDRAM only required modifications to existing SDRAM tooling, and since the SDRAM manufacturing processes had been pretty much perfected already yield was high from the get-go.
Rambus' royalties on RDRAM were actually pretty low. I don't remember what they were, but I remember it being under 3%.
With the low margins on all memory types, even 3% is a significant hit.
That is true from the point of view of profit margins, but 3% is insignificant in relation to the price differential between RDRAM and DDR SDRAM, which was what I was trying to explain.
Yes, I realize that the price difference is not as extreme as it once was, I was mostly speaking from historical perspective.
I think it's been quite a while since anyone gave a serious fuck about RDRAM, but it was mentioned in Anandtech's article: 1T-SRAM has a 10-15% greater die size, and RDRAM has a 15-30% greater die size than SDRAM. I'm not sure exactly, but I know my numbers aren't more than 5% off.
The patent drove the cost of DDR RAM up so much relative to competing technologies that the tech died
Correction, the patent did not put up the price of the memory, it is impossible to make any IC without a fist full of patent licenses.
It was the insane greed of the RAMBUS management that has killed RDRAM, they thought they had a monopoly and demanded usurous royalties. It has taken a while to prove that they do not have a monopoly.
I wanted to illustrate the similarities between this and Sony's patent related to Beta videocasette tapes
Sony never attempted to make Betamax a standard. They did not realise that the VCR would be used to show rented tapes. If the VCR had been used only for time shifting the Sony strategy was a rational one. Nobody cares that Tivo and Replay TV use incompatible file formats because the machines are not used for exchange of content. Once people demanded the ability to play pre-recorded tapes the Sony strategy failled.
And VIA sued back stating that Intel violated VIA's patents with the the Pentium 4 bus as well as some other reasons. You can follow the entire suit trail over at The Inq [theinquirer.net]
Well, two out of the three patents that Intel was suit VIA over have been thrown out of court. Intel only has one patent left to sue VIA over, and chances are, it's going to get tossed as well.
Intel isn't doing so hot these days. We'll see how well they fare once they go to the newer.13 micron procs with the larger L2 cache.
Of course, AMD's going to.13 in the near future as well... I love competition!
A "stealth introduction"? Doesn't Intel know about Slashdot? We geeks find out EVERYTHING, and as soon as one of us finds out and submits it to Slashdot, everyone knows. With media the way it is and Slashdot here, a quiet introduction is virtually impossible.
... as soon as one of us finds out and submits it to Slashdot, everyone knows.
Ahh, yes...
We are the SlashBorg. Your marketing secrets will be linked from an article in our website for the knowledge of all. Prepare to be Slashdotted. Resistance is futile.
And don't ever forget that it was a tentative of coup d'etat over the whole industry of memory and a direct attempt to kill the asian giants.
Whoever planned this, should know leave with honor through a ritual sepuku.
It's unfortunate that Rambus RIMMs are even more obsolete now that Intel's new chipsets are going to DDR. Not that I'll miss them, but its always unfortunate when early adopters get hosed with proprietary hardware. Anyone remember the Socket 4 P60/P66? Drop a ton of money on the new Pentium, and watch while everyone who waited is able to upgrade while you can't. Ditto to those with Asus P4Ts - not only are they hosed on processor options because intel changed pinouts for the new P4s, but now the RAM is obsolete too.
From the article: Intel is planning the stealth introduction of a chipset that will let computer makers connect the Pentium 4 to speedy DDR (double data rate) memory.
Speedy? Isn't DDR-SDRAM slower than RDRAM? Sounds a bit fluffy to me. What they really mean, but don't clearly spell out, is that DDR is faster than the normal SDRAM the 845 supports. But its still no RDRAM. Which I guess everyone here knew anyways.
Ahh well, I'm just grumpy b/c I convinced my mom to buy a P4T/Rambus-based P4 1.7Ghz, and now I have to ditch the Ram/Mobo/CPU to upgrade it. (I'd have given her an Athlon but the dustbunnies at her place are such that I'd be afraid of her burning the place down... remember that THG vid of the flaming Athlons?)
(*) DDR and Rambus transfer data at both the rising and falling edge of the clock cycle, thus doubling the effective bandwidth. Bacause of that they are often reffered to as 200MHz, 266MHz, and 800MHz respectively.
Anyway, the point is that DDR has greater bandwidth than Rambus. On top of that, Rambus has a pathetically high latency. Because of that Pentium 3 systems with PC133 SDRAM outperformed their Rambus counterparts most of the time.
As an aside, Intel decided to "fix" this flaw by making Pentium 4 waste four times as much memory bandwidth as Pentium 3 -- that makes P4 highly sensitive to memory bandwidth./. linked to a very interesting article discussing the P4 architecture a while ago.
Back to the point, Pentium 4 chipset uses two channels of Rambus memory that work in parallel. That gives it 2 * 1600 = 3200MB/s of bandwidth, which is greater than a single channel of PC2100 DDR (though it still has high latency). Problem is that you need to install memory in pairs (on RIMM for each channel), and each RDRAM channel can have only two memory slots. That means you are only one upgrade away from maxing out your memory. On the contrast, each DDR channel can have up to 4 memory slots and you can upgrade one slot at a time.
Also note that NVidia Nforce does the same thing for Athlons & DDR as the P4 chipset. Of course two channels of DDR have the bandwidth of 4256MB/s).
Ahh well, I'm just grumpy b/c I convinced my mom to buy a P4T/Rambus-based P4 1.7Ghz, and now I have to ditch the Ram/Mobo/CPU to upgrade it.
I'm so tempted to say "I told you so" -- which I would have if we ever spoke.
(I'd have given her an Athlon but the dustbunnies at her place are such that I'd be afraid of her burning the place down... remember that THG vid of the flaming Athlons?)
This is about the stupidest thing I ever heard. It's like claiming that Ford makes unsafe cars because their engines fry when you drain all the oil from them. Try this: remove the heatsink and fan from your P4 and see how long it takes for it to catch fire.
not to critisize your post, but i think you are forgtting the chipset itself... i think the memory handling of the chipset might make or break these new boards
Try this: remove the heatsink and fan from your P4 and see how long it takes for it to catch fire.
The Pentium 4 has a built-in thermal sensor and protection mechanism that will automatically throttle down the speed of the processor when it reaches a certain temperature. The Athlon's didn't have the same protection mechanism until the Palimino core came out (i.e.: the first Athlon MPs, the Athlon XPs and the newer MPs). The motherboards would also need to support the new sensor/mechanism rather than relying on rather not-always-so-accurate-under-the-processor diode.
The Pentium 4 has a built-in thermal sensor and protection mechanism that will automatically throttle down the speed of the processor when it reaches a certain temperature.
The thermal sensor isn't worth squat if the heatsink and fan are not present at all (err... I mean "fall off"), so Pentium 4 would have suffered the same fate. The sensor is useful if the fan stops or malfunctions somehow (but the heat sink is still there!) -- then the CPU temperature will gradually increase, and the thermal sensors on the CPU and board will do their magic. This can happen, BTW -- after years of work a fan can fail, just like any other mechanical device. However, a situation when the heat sink "falls off" has absolutely nothing to do with the real world. It's about as likely as the wheels of your car falling off.
No. Read the THG article. The same article states that the P4 *KEEPS RUNNING WITHOUT A HEATSINK OR FAN* while the P3 crashes and the Athlon fries itself.
I'm as big an Athlon fan as you can find, but if you are willing to let your HSF fall off, go with the P4
This is about the stupidest thing I ever heard. It's like claiming that Ford makes unsafe cars because their engines fry when you drain all the oil from them. Try this: remove the heatsink and fan from your P4 and see how long it takes for it to catch fire.
Ok, obviously the humour was lost on you guys. No CPU, AMD or Intel or otherwise, is going to burst into flames. She bought a P4 over an Althon because her husband, who gets all his tech info, a day late and a dollar short, from CNET, insisted on Intel Inside. "I'd rather pay the extra to get the real thing" I believe his words were. My first AMD was my 386DX/40.
As for the bandwidth issue, I guess I should have clarified by saying "isn't the DDR-SDRAM implementation intel proposes slower than their RDRAM implementation?". Which, AFAIK, it is.
Yes, it is true that DDRSDRAM is slower than RAMBUS, but its latency is much lower. For instance, CAS 3 SDRAM @ 150 mhz is actually much slower than CAS 2 SDRAM @ 133 mhz. So when you take DDR Ram at 266 mhz with low CAS 2 latency's its actuall quite powerful. Also to note it that the Intel RAMBUS platform uses dual channel RAMBUS. If Intel were to use dual channel DDR SDRAM (i.e. nForce) then you would have a lower latency and a larger memory bandwith to boot.
>Ahh well, I'm just grumpy b/c I convinced my mom to buy a P4T/Rambus-based P4 1.7Ghz, and now I have to ditch the Ram/Mobo/CPU to
>upgrade it.
Yeah, there are a lot of people in that same situation. All together, they are what drives the creation and marketing of dead-end tech. Uneducated consumers (along with the corporations who exploit them, or innovate through litigation) ultimately ruin the market of good products for everyone else.
>(I'd have given her an Athlon but the dustbunnies at her place are such that I'd be afraid of her burning the place down... remember
>that THG vid of the flaming Athlons?)
Nope. I never saw ANY such video. In fact, I SERIOUSLY doubt you did, either. The smoke erupting from the chip in the aforementioned video was not from the CPU itself, but the EXCESSIVE amount of crappy thermal paste left on it when they removed the heat sink. The "fire" that some people say they saw was a reflection of the non-contact thermal probe (which emits an ornage-reddish light).
Even so, how many houses have reportedly burned down from the heat sink falling off? In fact, how many heat sinks have reportedly "fell" off while the machine was running and not being moved around? Hmmmm?
I'd have given her an Athlon but the dustbunnies at her place are such that I'd be afraid of her burning the place down... remember that THG vid of the flaming Athlons?)
Dust bunnies don't kill Athlons. Neither does the fan failing, providing you have a motherboard that shuts the system down once the temperature goes too high. In order to kill their Athlon, THG had to run quake and pull both the heatsink and fan off. It only worked because there was nowhere for the heat to go when the processor was at 100%, and locked there, and the temperature rose faster than the sensor was able to detect.
Kinda unlikely to happen in the real world, unless you like playing Quake while your mates pull bits of your PC.
I recently had a P4 1.4Ghz with 256megs of RDRAM. I fried that processor. How, you ask? By breaking the fan clips and seeing if it would boot without a fan on. (It doesn't, dont run a proc without a fan or other cooling solution, unless u want to waste money).
Anyway, i had to buy a new Motherboard, Proc, but not ram. However, i was sufficiently impressed with the new Athlon XP line. (especially the low cost of the XP 1600+).
Basically, i ended up updating my system to an AMD Athlon XP 1600+, Abit KG7-Raid MoBo, and with 256megs of Crucial PC2100 DDR Sdram at 266mhz. This machine SCREAMS compared to my P4 with DDR, and the processor is the exact same clock frequency. RDRAM is faster in benchmarks, but for me, the price vs. performance for the AMD was unquestionable around christmas time and now i'd think twice about getting an AMD vs. Intel if Intel supports DDR.
p.s. I know RDRAM is faster, but my DDR just seems more stable and more responsive, just me, who knows.
Have a good one,
AJ
Ditto to that, I've got a p4 1.5 with 512 MB PC800 rdrram in 2 chips, and a 1.3 amd with 512 MB ddr2100 in 1 chip. The AMD is hands down faster in everything I've had occasion to run. It boots about 10-15 seconds faster. I'll buy the fact that RDR will excell under certain applications, I've just never encountered one.
I was changing the fans (the one i had on was a 7200RPM fan and it was just to loud for me to sleep). As i was taking it off, i broke one of the socket clips that it connects to by accident. So i wanted to see if i ruined the chip and just turned it on with the fan resting on top of it and it shut off after only 4 seconds. So then that ruined the chip. Granted it was a stupid thing to do on my part, but im much happier with my new system then old! And, i learned something, so oh well thats life.
AJ
It's not that they were 'getting around to it'. That's just plain sour grapes and company bashing.
It's that they had a timed agreement with Rambus to be exclusive. They thought they were doing the right thing to boost performance and it didn't work out in all areas of the business model. At least they stick to their agreements.
There is no truth to 'Rambus exclusivity' whatsoever. Intel has been shipping the SDRAM version of the i845 chipset for several months, and it is the fastest ramping chipset ever in the industry. The only reason got the gap between the SDRAM and DDR versions is the time to validate it.
This is very wrong.
The Rambus-Intel liscensing deal has RDRAM as the exclusive high speed memory subsystem for Intel until Jan 1 2002. The penalties for breaching this were pretty stiff, so stiff in fact that when combined with the loss Intel would take from the drop in Rambus stock (they own ~15%) it was thought that it would literally cost Intel over 1 Billion to go DDR before then. What kind of behind the scenes deals went on to allow this chipset to ship before Jan 1 I am not privy to.
The contract between Intel and Rambus was that Intel could not manufacture nor introduce any products that use newer and competitive memory technologies outside of Rambus RDRAM and SDRAM. DDR SDRAM was a newer memory technology and one that was aim right at RDRAM. That's why Intel sat on their hands on a product that would work with DDR SDRAM... that was until the i845 was introduced.
Intel and Rambus has since re-negotiated their contracts with each other and the "exclusive" clause has been tweaked, but not completely removed. Check out The Inq or The Register for more information about the previous and the recent contracts between the two.
As we all know, the P4 is designed to take advantage of high memory bandwidth-- which is why the 845 chipset (or whatever) that used SDRAM was such a flop.
When the P4 was first introduced, Intel claimed that it was designed specifically for RDRAM. If this is true-- and I suppose it doesn't have to be-- then is it possible that the new DDR stuff will actually perform below RDRAM systems? Is the only advantage going to be price?
I'm not really a tech guy, so this is an honest question. I'm not a Rambus fan-- I've got a PIII with the 820 chipset, and I'm not particularly fond of it-- but could it be that the company that everybody hates is actually the better way to go in this case?
Of course, everybody around here is going to be be gushing over DDR over Rambus-- if they choose Intel over AMD, which doesn't seem likely-- but it seems that Intel is either stepping backwards or conflicting with their past words.
The long and short of it is that RAMBUS - the company - created a type of memory which was legally provable to be different from the known standard, and then they tried to claim that they had a patent on SDRAM as well. The chip itself is a perfect example of "Market-engineering"... it was marketed as being much faster and more powerful than SDRAM or DDR, but in real-life conditions it was almost laughably slow and even unstable - overheating being one of its worst problems. As a result, it is frankly the honest-to-god truth that RAMBUS just wanted to usurp the memory market with an inferior product. Don't take my word for it though - go read the articles about RAMBUS at Tom's Hardware [tomshardware.com] and see it for yourself. The numbers there don't lie, at least in my experience. Intel stuck with RAMBUS because they stood to gain a strong ally in the hardware market, but RAMBUS buggered it all up by being greedy...
The P4 was designed for truly immense memory bandwidth and very large on chip cache. The current generation of P4 is suffering from the cache being significantly below it's design point, which is somewhere around the 2Mb mark. Obviously with a cache this big the latency of the ram itself isn't such a problem - hence Intel signing up to RDRAM.
Anyway, the actual question was:
but could it be that the company that everybody hates is actually the better way to go in this case?
Not really. There's not that much difference in bandwidth between DDR and RDRAM anyway. And Rambus need to go broke to remind the industry in general that we won't tolerate that kind of behaviour. Unfortunately they have some huge contracts, the PS2 being probably the biggest, so it seems unlikely they are going to go chapter 11 in the near future.
The other reason why the Pentium 4 isn't performing as well as it should is that it was built for fast streams of data into it's SSE2 execution units. The problem with the L2 cache isn't really that it is well undersized, it's that the x86 execution units suck altogether. The other problem is the 20-stage pipeline (which was required to ramp the speed up well beyond 1.1Ghz which they were stuck at with the 0.18-micron Pentium III) and they are still using the relatively poor x87 FPU units. The Athlon's FPU units blow away the x87 FPU units both the Pentium III and the Pentium 4 (the PIII and the P4 use almost the same FPU units).
Ace's Hardware [aceshardware.com] and Real World Tech [realworldtech.com] have great discussions and articles on the downfalls of the Pentium 4 processor.
by Anonymous Coward writes:
on Wednesday December 12, 2001 @07:14PM (#2696437)
People seem to forget that RDRAM has gone down DRAMATICALLY in price. The days of $900 128mb RIMM's are long gone. You can order 256mb RIMM's for like $76 on pricewatch.com now
Yes DDR is still cheaper per mb, but RDRAM isn't that much more expensive. Especially when you consider how fast & stable it is in the i850 chipset by Intel.
Is it still true that the more RDRAM you add to a system, the system's latency will increase because of its architecture? If this is so, I would stay away from RDRAM even if it's on par with DDR-SDRAM in terms of price.
We've all seen the THG memory bandwidth benchmarks that show RIMMS are still way ahead of DDR DIMMS in terms of bandwidth. The CPU, well that's another story, but nonetheless Intel is not taking back what they've said, they're simply offering another choice to consumers, which is a good thing. I don't think we should critisize them for offering up a system with "slightly" lower performance. How rare is it, when a company has been forced (a la AMD and VIA) to offer more choices for the increasingly price/performance consious consumers? Pour a cold one for the little guy.
RIMMS are way ahead of DDR DIMS bandwith wise, but as THG also shows us [tomshardware.com], performance wise, single channel DDR is as fast (less that 1% slower) than RDRAM based systems. This of course is on chipsets by companies that typically have trouble keeping up with intel. Not to mention, another channel can be added (easily ?), ask NVIDIA. Latency kills the RAMBUS advantage. Intel isn't so much offering lesser performance for less money. They are offering equal performance for less money. Hopefully, this option will cut down Rambus prices.
The funny thing about the DDR memory is that when you run it its about faster and better and that is what the thing of it is. I always said that when it comes to DDR memory that I would buy a chunk of chocolate and eat it before the DDR memory would address a bit. And you know what? I was right !!
Back in October at an expo the Intel guys did say this chipset was indeed hitting in Jan/Feb. But they're also looking to bring a 533MHz FSB using 166MHz DDR by June.
Hasn't anyone being paying attention to VIA? What about the P4X266 chipset, even Tyan has a board with it: http://tyan.com/products/html/trinity510.html
Intel motherboards and chipsets are fine however you don't have to wait for Intel to come out with a DDR chipset for your P4. VIA has one already.
The VIA chipset you would really want to look at is the P4X266A; like the KT266A it performs much better than its predecessor. Despite this, some early benchmarks of the 845D look like it outperforms all other DDR solutions. The real advantage to this chipset will be major motherboard manufacturers picking up DDR P4 platforms, ie. ASUS, Abit, Gigabyte, etc. They wouldn't touch the P4X266A because of lawsuit threats.
yea.. and isn't VIA in the process of getting sued by Intel for using a chipset that don't have the license for?
VIA believes because they own a large portion of a company that has the rights from Intel, that they, too, can benefit from these rights. This really isn't the case though according to Intel's lawyers.
I really don't wanna use a chipset that will be illegal in a few months, and thus, lose all support. do you?
lost the url's for these, but prolly from anandtech.com
1) Since Intel is sueing VIA, VIA *must* be in the wrong. Just like Rambus suing Infineon (I forgot how to spell the companies name, sorry). Come on, you have no idea of the cross-licensing issues.
2) I really don't wanna use a chipset that will be illegal in a few months, and thus, lose all support. do you?
Using an illegal chipset? What the hell are they going to go, make you return the board? Loose support? Exactly how many times have you reflashed your BIOS? Dude, you really need to get with the program.
No thanks. As the owner of several VIA chipset boards I can personally vouch for their unstability. If an Intel solution(i845D) offers performance on par with the VIA solution, I'll take the Intel solution ANY DAY. It's worth not having all kinds of hardware conflicts(SB Live anyone?) and crashes.
The i845 actually does and has supported DDR SDRAM since it was designed. Only, Intel has only allowed motherboard manufacturers to produce SDRAM-based boards thus far, allegedly for "validation" reasons, although clearly pressure from Rambus has had something to do with it.
Then there are the DDR P4 chipsets from both VIA and SiS. Don't forget about that.
The sad part is that us AMD enthusiasts used to be able to get DDR memory really cheap because the demand was low. Now that all the P4 people are looking to buy DDR, the price is going way high. Wouldn't it be ironic if it went as high as RDRAM? That would be a real bum move on Intel's part.
For those who missed it. In 1992 Rambus joined an industry consortium (Joint Electron Device Engineering Council JEDEC) made up of companies seeking to develop a *royalty-free* standard for the next generation of memory chips. The resulting standards (SDRAM, and DDR RAM) have been widely adopted during the past few years.
Then, about a year ago, Rambus let the lawyers loose. They claimed that, despite its participation in JEDEC, it owned patents that were being infringed upon by any company making SDRAM or DDR RAM chips without a license. Moreover, Rambus claimed it was entitled to damages in the form of retroactive royalty payments.
Intel's chipsets for P4 have been using DDR for a while now. In fact, most OEMs have been developing P4 systems with DDR for months now. I can't tell you the last time we sold a RAMBUS probe. Further more, all of intel's future processor chipsets will use DDR as well.
I work for a test equipment supplier which will go unnamed.
Why is Intel doing this now instead of releasing
a DDR SDRAM chipset all the way since the very release of the Pentium 4 -months- (years?) ago?
Did they originally make only the RAMBUS chipset available because they still favored RAMBUS over other memory types or did they simply rush P4 to the market before DDR SDRAM chipset was ready?
There were several articles a year ago, even before the original P4 release, where the Intel representatives confessed that they were not all that happy about RAMBUS rollout and that promoting and pushing RAMBUS memory down consumers' throats did them more harm than good. And then later they announced P4 which for a while could be used -only- with RAMBUS. Were they not honest about their feelings toward RAMBUS?
No more Rambus! (Score:1)
Ah, DDR-SDRAM! (Score:2, Funny)
*slurp* *CRUNCH* AARRGHH!! *bleed profusely*
*reads the rest of the story*
Ah, shit.
Isn't it too late to worry about this? (Score:5, Insightful)
Seriously when RAMBUS was 10x the price of SDRAM it seriously hurt, but now that RAMBUS is getting close to comparable, I don't see what the point is. In my neck of the woods PC-800 RDRAM goes for about 30% more than PC2100 DDR, which really isn't that much (and dual channel RDRAM is the fastest RAM platform out there). Given that the P4s one redeeming factor is that with RDRAM it has a serious memory advantage, I really don't see what Intel is thinking: Put a P4 with DDR DRAM and it'll get clobbered even more.
Re:Isn't it too late to worry about this? (Score:2)
I don't think this is to replace their RDRAM chipsets, but instead their "bargain" PC133 SDRAM chipsets (which I imagine are seriously choked).
Re:Isn't it too late to worry about this? (Score:3, Interesting)
In terms of high end Intel systems, DDR just isn't that way to go. A couple of months ago, when Intel got to 2GHz, they were beasting on similar Athlon systems. But now, AMD has gone on a tear, heavily ramping up their Palomino core. A 1.9 XP w/ DDR beats a 2GHz P4 w/ PC800 RDRAM in every category except for memory bandwith...If the Intel was using DDR as well, the Xp's margin of victory will be even greater. The P4 relies on fast, fast memory. Give a P4 slow memory, and it will freeze (P4 and SDRAM is a horrid combo). Since RAMBUS will soon be releasing pumped 133 MHz bus memory, I think this is the memory that will help Intel more than DDR. Intel is losing, and has ALWAYS lost, the price battle. I think that if Intel cuts memory performance to reduce price, they are losing their ONLY advantage over AMD systems.
Re:Isn't it too late to worry about this? (Score:1, Interesting)
Please don't tell me you are seriously basing your opinions on the above! You are bitching about a $20 difference in RAM!!!!????!! At one time 2MB RAM cost $2000! 2MB to 4MB was thousands of dollars, yet you complain about $20? Yes, I know that everything is cheaper now, but you have zero perspective on value.
"but the price of DDR RAM is still to expensive to replace SDRAM" - ah, there was a point were RAM was too expensive to replace your floppy drvie. Again, this is dated but don't bitch because you can't buy that "extra juicy" gum instead of the "regular juicy" gum; we're talking 3-4 magnitudes of difference here.
Another OT topic-
"Intel is losing, and has ALWAYS lost, the price battle." I guess Apple should just pack it in?
It's nearly painful to watch these youngins complain about spending an extra $20 on memory!!
Re:Isn't it too late to worry about this? (Score:1)
You're missing the poster's point. His point was that the *current* prices of RAM, CPU etc. are in favour of AMD rather than Intel. For that, it doesn't matter what the prices of RAM was just under a decade ago (I remember those days too, and I know that I just payed 25CAND for 256 MB RAM... SD sure, but 256 nonetheless...).
Therefore, your arguement that it doesn't matter about the price difference because it is 20$ now instead of 2000$ doesn't make sense: the price is higher for a P4 system, even if it is only 20$, but that's an extra 256 MB (well, pro'lly 128 MB DDR) RAM... choose the system you want.
MIKE
Re:Isn't it too late to worry about this? (Score:1)
Re:Isn't it too late to worry about this? (Score:5, Insightful)
The guys at Intel aren't stupid. They invested millions into marketing themselves as a consumer brand before it was cool to do so, and it's been the best move they ever made. AMD, for all its trying, hasn't even registered on the radar of most consumers. Intel uses this to its advantage to charge a hefty price premium. After all, they're the Coca-cola while AMD is just the RC.
So Intel _is_ winning the price battle, since the winner of the price battle is the guy that gets to charge more money and still sell 80% of the processors, not the guy that sells them for half as much to push enough volume to break even. After all, Intel could sell their parts for much less than they do without actually losing money, but they don't have to.
And, as AMD's recent relabeling of their XP line has shown, clock speed is still king. Nobody has ever successfully dethroned it as the single number consumers care about above all others. Which is why Intel has won that battle as well.
P4's with DDR aren't in any way related to RAMBUS's performance as much as keeping low-margin systems affordable and still fast. That's why you'll see P4 Rambus and DDR boards out there, fighting it out for the price/performance sweet spot.
At this point, Intel is more worried about Sun than AMD, since Sun is the lone vendor not committed to Itanium/McKinley. They've also got the high-margin Xeon processors competing with Sun's mid-range offerings. This is where the interesting things are going to start happening, but you won't be hearing about it on Tom's Hardware.
RC Cola still tastes better =:-) (Score:1)
Now there are other factors. Until recently Windows didn't run too well on any of the available Athlon chipsets. From what i hear this has changed. On the other hand, the price/performance ratio assuming some sane OS like BSD or Linux was much in the favor of the Athlon.
Re:Isn't it too late to worry about this? (Score:1)
I told him that I could build some AMD machines that would be speed demons compared to the new Dells that he picked up, for about 500 less, but all he did was whine about the time it takes to support it if something goes wrong. He also has a problem paying 35 bucks for a cooler when there are $10 coolermasters (IMO these are all terrible) that will "do the same job."
Re:Isn't it too late to worry about this? (Score:5, Insightful)
Speed isn't everything... RDRAM has a great deal more latency than DDR. In many cases, RDRAM performs significantly worse than even SDRAM.
Besides that, there's the evil factor, considering Rambus believed more in the policy of suing for royalties as a revenue model instead of producing and selling a decent product.
Re:Isn't it too late to worry about this? (Score:1)
hm.. (Score:2, Interesting)
Actually - fitting the data you use into L2 cache is much more important IMO.. I have seen factor of 3 improvements in some of my code.. Local alignment (UNder 4k blocks) matters less from my benchmarks..
Re:hm.. (Score:1)
I use Java to play Pogo games and "Shock the Monkey". I haven't seen Visual Basic in a while; It is not robust at all.
L2 cache is very important, and Intel's new chip will have increased cache. In the classic (slot)Athlons, the L2 ran at a third the speed of the chip, but nowadays, with these newfangled modern procs, L2 has gained in importance, running at full speed...But, then again, the Celerons and Durons have hacked, smaller, L2s...
Re:hm.. (Score:2)
I don't claim to have huge experience - I've only been VB-aware for ~3 years - but every non-web niche Windows application I've ever seen has been a VB app. There's one possible exception, but the company that made it is out of business - to a large extent because of their development costs - though I'm not convinced they don't use VB.
Re:hm.. (Score:2)
Or something like that.
Dave
Re:Isn't it too late to worry about this? (Score:2, Insightful)
The few programs which might be blocking due to insufficient memory bandwidth (databases, games with large/many textures and the like) often do have the critical pieces tuned for cache performance.
Claiming that all apps should be tuned for optimal memory access is just silly. You do it where it's easy or where it increases performance significantly. Any more and its just a waste of time and money.
Before blaming things on lazy programmers remember that developer time isn't cheap. If I routinely get things done in one day and you take five days but get 5% better performance, who do you think is getting cut next time there are layoffs?
Re:Isn't it too late to worry about this? (Score:3, Informative)
I don't agree with you comment about context switching emphasising latency. Context switching takes doesn't happen frequently, time slices are too long from a cpu perspective (linux defaults to 10ms, I guess that corresponds to ten million or so cycles.) Contexts are loaded in burst rates so latency would have little effect on overall performance. Offcourse it might be that context switching occurs more frequently because programs are releasing their time slices, but that would mean you don't need performance either since your processes are idle. A busy process still gets its ten million cycles, wasting a tiny percentage on context switching - high latency or not.
Its almost 4am here, I'd better sleep now.
How fast do you think a P4 with PC1600 will be? (Score:1)
Benches of AThlons with PC1600 were underwhelming.
Re:Isn't it too late to worry about this? (Score:2)
The thing that nobody ever understood about Rambus is that they are not actually in the business of producing RAM. All along, they're business model was to come up with good ideas and license them out. There is nothing wrong with that.
The only thing that bothered me about Rambus is that they didn't disclose the patents they held while they participated in developing memory standards. That was crappy business.
DDR has better latency (Score:3, Interesting)
Re:Isn't it too late to worry about this? (Score:2)
The point, in my opinion, is that Rambus has too high of a lawyer-to-engineer ratio for my tastes. I prefer a company that chooses innovation over litigation because I have a lot more faith in their product down the road.
Re:Isn't it too late to worry about this? (Score:1)
That sounds pretty reasonable...but:
"I prefer a company that chooses innovation over litigation because I have a lot more faith in their product down the road."
We are talking about memory here. What possible impact could future innovative products have on whether you buy an intel system that uses rambus or not, today?
I say buy the fastest thing you can at the price you are willing to pay. If rambus fits the bill, then go for it. Who cares what they put out down the line. It isn't likely to be compatible with your current system.
Re:Isn't it too late to worry about this? (Score:2)
I'll support DDR SDRAM for that reason alone. The fact that it performs better in everything except MPEG encoding tests is just gravy.
Oh well, Rambus's stock value is in the toilet and there are rumors of the stockholders suing the principles for criminally bad management. I can't say I'd cry over it.
Re:Isn't it too late to worry about this? (Score:1, Interesting)
It's about Intel failing to grab more control of the technology that makes up a PC.
And that IS a GOOD thing.
I've got nothing against Intel.
But monopolies usually suck for everyone but the monopolist.
Re:Isn't it too late to worry about this? (Score:1)
of course dual channel ddr-sdram... gah... drool..
Re:Isn't it too late to worry about this? (Score:1)
of course dual channel ddr-sdram... gah... drool..
Isn't the nforce dual channel DDR? I know I was seriously excited about it, but rather underwhelmed when the actual numbers came out. Then again nvidia has a funny way of doing that again and again: Releasing something with subpar numbers to get the moaners and whiners yabbling (in a way they'll soon regret), and to lower expectations, and then coming out with a final product that blows people away.
Check your facts before you post (Score:2)
There are at least two family of boards operating on dual channel sdram principle, nforce boards and some server stuff company's (anand's older servers were powered from them, you can find the name in their older news archives.) One works with athlons, other with p3's but idea is similarly applicable to p4 chipsets.
Re:Isn't it too late to worry about this? (Score:1)
yes i know the nforce is dual channel ddr, but it still needs work...
one thought i had was dual channel ddr with an smp athlon board since the nforce really doesn't seem to take much advantage of it
Re:Unfortunate (Score:2)
No dual channel Rambus, is what what give the
P4 its bandwidth advantage. But as Nvidia's
Nforce shows dual channel DDR is quite possible,
and gives even more bandwidth. Ram module
for Ram module, DDR has a high bandwidth and
is cheap.
Hey intel, here's a good idea to LOWER the cost (Score:2, Insightful)
Oh, that would chunk in your profits... right
Propritary... (Score:1)
Archtecture: Not yet. Still need to reveal how this (P4) works.
Names: No. Pentium(tm)
RAM: Yay! A victory!
Intel still has a way to go, but is definitly a good start.
No comments about spelling/grammer, please.
Patents kill your tech off! (Score:2, Interesting)
I wanted to illustrate the similarities between this and Sony's patent related to Beta videocasette tapes, but it would have been sure to result in a flamewar.
Re:Patents kill your tech off! (Score:4, Interesting)
I'm hoping you meant to say RDRAM.
Anyway, the patents had nothing to do with the price differences between RDRAM and DDR SDRAM, it was all due to manufacturing costs. I remember a little over a year ago Kingston was bragging about their 30%(!) yield on PC-800 RDRAM chips. When 70+% of your product doesn't pass QA, that's definately going to drive your costs up! Additionally, manufacturers had a fair amount of retooling to do before they could make RDRAM, and high setup costs get passed on to the consumer. As I recall, RDRAM also has a bigger die size than DDR SDRAM (I could easily be wrong, it's been a while since I cared) which would also drive up costs.
In contrast, DDR SDRAM only required modifications to existing SDRAM tooling, and since the SDRAM manufacturing processes had been pretty much perfected already yield was high from the get-go.
Rambus' royalties on RDRAM were actually pretty low. I don't remember what they were, but I remember it being under 3%.
Re:Patents kill your tech off! (Score:1)
With the low margins on all memory types, even 3% is a significant hit.
Re:Patents kill your tech off! (Score:2)
That is true from the point of view of profit margins, but 3% is insignificant in relation to the price differential between RDRAM and DDR SDRAM, which was what I was trying to explain.
Yes, I realize that the price difference is not as extreme as it once was, I was mostly speaking from historical perspective.
Re:Patents kill your tech off! (Score:2)
Re:Patents kill your tech off! (Score:2)
Correction, the patent did not put up the price of the memory, it is impossible to make any IC without a fist full of patent licenses.
It was the insane greed of the RAMBUS management that has killed RDRAM, they thought they had a monopoly and demanded usurous royalties. It has taken a while to prove that they do not have a monopoly.
I wanted to illustrate the similarities between this and Sony's patent related to Beta videocasette tapes
Sony never attempted to make Betamax a standard. They did not realise that the VCR would be used to show rented tapes. If the VCR had been used only for time shifting the Sony strategy was a rational one. Nobody cares that Tivo and Replay TV use incompatible file formats because the machines are not used for exchange of content. Once people demanded the ability to play pre-recorded tapes the Sony strategy failled.
affordable gaming... (Score:1)
Intel's not stupid ... (Score:1)
Re:Intel's not stupid ... (Score:1)
VIA (Score:1)
Re:VIA (Score:1)
Re:VIA (Score:2)
Intel isn't doing so hot these days. We'll see how well they fare once they go to the newer
Of course, AMD's going to
Stealth Introduction? (Score:3, Redundant)
-------------------
Re:Stealth Introduction? (Score:1)
Ahh, yes...
We are the SlashBorg. Your marketing secrets will be linked from an article in our website for the knowledge of all. Prepare to be Slashdotted. Resistance is futile.
remember! (Score:1)
Whoever planned this, should know leave with honor through a ritual sepuku.
Rambus - now even more obsolete! (Score:3, Interesting)
From the article: Intel is planning the stealth introduction of a chipset that will let computer makers connect the Pentium 4 to speedy DDR (double data rate) memory.
Speedy? Isn't DDR-SDRAM slower than RDRAM? Sounds a bit fluffy to me. What they really mean, but don't clearly spell out, is that DDR is faster than the normal SDRAM the 845 supports. But its still no RDRAM. Which I guess everyone here knew anyways.
Ahh well, I'm just grumpy b/c I convinced my mom to buy a P4T/Rambus-based P4 1.7Ghz, and now I have to ditch the Ram/Mobo/CPU to upgrade it. (I'd have given her an Athlon but the dustbunnies at her place are such that I'd be afraid of her burning the place down... remember that THG vid of the flaming Athlons?)
Re:Rambus - now even more obsolete! (Score:4, Troll)
False. Comparing just bandwidth:
100MHz SDRAM -> 800MB/s
133MHz SDRAM -> 1064MB/s
100MHz DDR -> 1600MB/s (*)
133MHz DDR -> 2128MB/s (*)
400MHz RDRAM -> 1600MB/s (*)
(*) DDR and Rambus transfer data at both the rising and falling edge of the clock cycle, thus doubling the effective bandwidth. Bacause of that they are often reffered to as 200MHz, 266MHz, and 800MHz respectively.
Anyway, the point is that DDR has greater bandwidth than Rambus. On top of that, Rambus has a pathetically high latency. Because of that Pentium 3 systems with PC133 SDRAM outperformed their Rambus counterparts most of the time. /. linked to a very interesting article discussing the P4 architecture a while ago.
As an aside, Intel decided to "fix" this flaw by making Pentium 4 waste four times as much memory bandwidth as Pentium 3 -- that makes P4 highly sensitive to memory bandwidth.
Back to the point, Pentium 4 chipset uses two channels of Rambus memory that work in parallel. That gives it 2 * 1600 = 3200MB/s of bandwidth, which is greater than a single channel of PC2100 DDR (though it still has high latency). Problem is that you need to install memory in pairs (on RIMM for each channel), and each RDRAM channel can have only two memory slots. That means you are only one upgrade away from maxing out your memory. On the contrast, each DDR channel can have up to 4 memory slots and you can upgrade one slot at a time.
Also note that NVidia Nforce does the same thing for Athlons & DDR as the P4 chipset. Of course two channels of DDR have the bandwidth of 4256MB/s).
Ahh well, I'm just grumpy b/c I convinced my mom to buy a P4T/Rambus-based P4 1.7Ghz, and now I have to ditch the Ram/Mobo/CPU to upgrade it.
I'm so tempted to say "I told you so" -- which I would have if we ever spoke.
(I'd have given her an Athlon but the dustbunnies at her place are such that I'd be afraid of her burning the place down... remember that THG vid of the flaming Athlons?)
This is about the stupidest thing I ever heard. It's like claiming that Ford makes unsafe cars because their engines fry when you drain all the oil from them. Try this: remove the heatsink and fan from your P4 and see how long it takes for it to catch fire.
Re:Rambus - now even more obsolete! (Score:1)
Re:Rambus - now even more obsolete! (Score:1)
Re:Rambus - now even more obsolete! (Score:2)
The thermal sensor isn't worth squat if the heatsink and fan are not present at all (err... I mean "fall off"), so Pentium 4 would have suffered the same fate. The sensor is useful if the fan stops or malfunctions somehow (but the heat sink is still there!) -- then the CPU temperature will gradually increase, and the thermal sensors on the CPU and board will do their magic. This can happen, BTW -- after years of work a fan can fail, just like any other mechanical device. However, a situation when the heat sink "falls off" has absolutely nothing to do with the real world. It's about as likely as the wheels of your car falling off.
Re:Rambus - now even more obsolete! (Score:1, Informative)
I'm as big an Athlon fan as you can find, but if you are willing to let your HSF fall off, go with the P4
Re:Rambus - now even more obsolete! (Score:2)
Ok, obviously the humour was lost on you guys. No CPU, AMD or Intel or otherwise, is going to burst into flames. She bought a P4 over an Althon because her husband, who gets all his tech info, a day late and a dollar short, from CNET, insisted on Intel Inside. "I'd rather pay the extra to get the real thing" I believe his words were. My first AMD was my 386DX/40.
As for the bandwidth issue, I guess I should have clarified by saying "isn't the DDR-SDRAM implementation intel proposes slower than their RDRAM implementation?". Which, AFAIK, it is.
Re:Rambus - now even more obsolete! (Score:2, Informative)
Re:Rambus - now even more obsolete! (Score:1)
Re:Flaming Athlons? (Score:1)
>upgrade it.
Yeah, there are a lot of people in that same situation. All together, they are what drives the creation and marketing of dead-end tech. Uneducated consumers (along with the corporations who exploit them, or innovate through litigation) ultimately ruin the market of good products for everyone else.
>(I'd have given her an Athlon but the dustbunnies at her place are such that I'd be afraid of her burning the place down... remember
>that THG vid of the flaming Athlons?)
Nope. I never saw ANY such video. In fact, I SERIOUSLY doubt you did, either. The smoke erupting from the chip in the aforementioned video was not from the CPU itself, but the EXCESSIVE amount of crappy thermal paste left on it when they removed the heat sink. The "fire" that some people say they saw was a reflection of the non-contact thermal probe (which emits an ornage-reddish light).
Even so, how many houses have reportedly burned down from the heat sink falling off? In fact, how many heat sinks have reportedly "fell" off while the machine was running and not being moved around? Hmmmm?
Re:Rambus - now even more obsolete! (Score:2)
Dust bunnies don't kill Athlons. Neither does the fan failing, providing you have a motherboard that shuts the system down once the temperature goes too high. In order to kill their Athlon, THG had to run quake and pull both the heatsink and fan off. It only worked because there was nowhere for the heat to go when the processor was at 100%, and locked there, and the temperature rose faster than the sensor was able to detect.
Kinda unlikely to happen in the real world, unless you like playing Quake while your mates pull bits of your PC.
Dave
I Broke it I Bought it! (Score:1, Informative)
Anyway, i had to buy a new Motherboard, Proc, but not ram. However, i was sufficiently impressed with the new Athlon XP line. (especially the low cost of the XP 1600+).
Basically, i ended up updating my system to an AMD Athlon XP 1600+, Abit KG7-Raid MoBo, and with 256megs of Crucial PC2100 DDR Sdram at 266mhz. This machine SCREAMS compared to my P4 with DDR, and the processor is the exact same clock frequency. RDRAM is faster in benchmarks, but for me, the price vs. performance for the AMD was unquestionable around christmas time and now i'd think twice about getting an AMD vs. Intel if Intel supports DDR.
p.s. I know RDRAM is faster, but my DDR just seems more stable and more responsive, just me, who knows.
Have a good one,
AJ
hands down AMD out performs (Score:2)
Re:You are a dumbass (Score:1)
AJ
Getting around a lawsuit more like. (Score:1)
It's that they had a timed agreement with Rambus to be exclusive. They thought they were doing the right thing to boost performance and it didn't work out in all areas of the business model. At least they stick to their agreements.
Re:Getting around a lawsuit more like. (Score:2)
Re:Getting around a lawsuit more like. (Score:1)
The Rambus-Intel liscensing deal has RDRAM as the exclusive high speed memory subsystem for Intel until Jan 1 2002. The penalties for breaching this were pretty stiff, so stiff in fact that when combined with the loss Intel would take from the drop in Rambus stock (they own ~15%) it was thought that it would literally cost Intel over 1 Billion to go DDR before then. What kind of behind the scenes deals went on to allow this chipset to ship before Jan 1 I am not privy to.
Re:Getting around a lawsuit more like. (Score:2)
Re:Getting around a lawsuit more like. (Score:1)
Intel and Rambus has since re-negotiated their contracts with each other and the "exclusive" clause has been tweaked, but not completely removed. Check out The Inq or The Register for more information about the previous and the recent contracts between the two.
Initial Designs (Score:3, Informative)
When the P4 was first introduced, Intel claimed that it was designed specifically for RDRAM. If this is true-- and I suppose it doesn't have to be-- then is it possible that the new DDR stuff will actually perform below RDRAM systems? Is the only advantage going to be price?
I'm not really a tech guy, so this is an honest question. I'm not a Rambus fan-- I've got a PIII with the 820 chipset, and I'm not particularly fond of it-- but could it be that the company that everybody hates is actually the better way to go in this case?
Of course, everybody around here is going to be be gushing over DDR over Rambus-- if they choose Intel over AMD, which doesn't seem likely-- but it seems that Intel is either stepping backwards or conflicting with their past words.
Re:Initial Designs (Score:1)
Re:Initial Designs (Score:3, Interesting)
Anyway, the actual question was:
but could it be that the company that everybody hates is actually the better way to go in this case?
Not really. There's not that much difference in bandwidth between DDR and RDRAM anyway. And Rambus need to go broke to remind the industry in general that we won't tolerate that kind of behaviour. Unfortunately they have some huge contracts, the PS2 being probably the biggest, so it seems unlikely they are going to go chapter 11 in the near future.
Wankers.
Dave
Re:Initial Designs (Score:1)
Ace's Hardware [aceshardware.com] and Real World Tech [realworldtech.com] have great discussions and articles on the downfalls of the Pentium 4 processor.
VIA (Score:1)
RDRAM not to bad.... (Score:3, Informative)
Yes DDR is still cheaper per mb, but RDRAM isn't that much more expensive. Especially when you consider how fast & stable it is in the i850 chipset by Intel.
Re:RDRAM not to bad.... (Score:1)
C'mon folks (Score:3, Interesting)
Re:C'mon folks (Score:1)
The thing about this is... (Score:1, Offtopic)
... and more on the way (Score:1)
Oh, and the P4 mobile also in Q1 2002.
Gimme gimme gimme!
Old News (Score:4, Informative)
Intel motherboards and chipsets are fine however you don't have to wait for Intel to come out with a DDR chipset for your P4. VIA has one already.
Re:Old News (Score:1)
The performance is almost that of an RDRAM solution, but much cheaper.
Re:Old News (Score:1)
Re:Old News (Score:1)
VIA believes because they own a large portion of a company that has the rights from Intel, that they, too, can benefit from these rights. This really isn't the case though according to Intel's lawyers.
I really don't wanna use a chipset that will be illegal in a few months, and thus, lose all support. do you?
lost the url's for these, but prolly from anandtech.com
Re:Old News (Score:1)
2) I really don't wanna use a chipset that will be illegal in a few months, and thus, lose all support. do you?
Using an illegal chipset? What the hell are they going to go, make you return the board? Loose support? Exactly how many times have you reflashed your BIOS? Dude, you really need to get with the program.
Re:Old News (Score:1)
What's wrong with the SiS? (Score:4, Interesting)
I thought that this chipset looked good enough.
not really new (Score:2, Informative)
Then there are the DDR P4 chipsets from both VIA and SiS. Don't forget about that.
DDR getting more expensive because of this (Score:2, Insightful)
It wasn't so much the price of RAMBUS... (Score:4, Informative)
For those who missed it. In 1992 Rambus joined an industry consortium (Joint Electron Device Engineering Council JEDEC) made up of companies seeking to develop a *royalty-free* standard for the next generation of memory chips. The resulting standards (SDRAM, and DDR RAM) have been widely adopted during the past few years.
Then, about a year ago, Rambus let the lawyers loose. They claimed that, despite its participation in JEDEC, it owned patents that were being infringed upon by any company making SDRAM or DDR RAM chips without a license. Moreover, Rambus claimed it was entitled to damages in the form of retroactive royalty payments.
And then the lawsuits began....
-Derek
Well, this is really old actually. (Score:3, Interesting)
I work for a test equipment supplier which will go unnamed.
Square Peg and a Round Hole (Score:1)
The P4 is built to utilize memory bandwidth, not lower latency. The AMD's chips are the opposite.
While a P4 PC would cost less with DDR-SDRAM, it would cop a slight performance hit.
DDR and P4 (Score:1)
Question: (Score:1)
Why is Intel doing this now instead of releasing a DDR SDRAM chipset all the way since the very release of the Pentium 4 -months- (years?) ago?
Did they originally make only the RAMBUS chipset available because they still favored RAMBUS over other memory types or did they simply rush P4 to the market before DDR SDRAM chipset was ready?
There were several articles a year ago, even before the original P4 release, where the Intel representatives confessed that they were not all that happy about RAMBUS rollout and that promoting and pushing RAMBUS memory down consumers' throats did them more harm than good. And then later they announced P4 which for a while could be used -only- with RAMBUS. Were they not honest about their feelings toward RAMBUS?
DDR is obsolete (Score:1)