DDR3 RAM Explained 200
Das Capitolin sends us to Benchmark Reviews for an in-depth feature on DDR3 memory that begins: "These are uncertain financial times we live in today, and the rise and fall of our economy has had [a] direct [effect] on consumer spending. It has already been one full year now that DDR3 has been patiently waiting for the enthusiast community to give it proper consideration, yet [its] success is still undermined by misconceptions and high price. Benchmark Reviews has been testing DDR3 more actively than anyone. ... Sadly, it might take an article like this to open the eyes of my fellow hardware enthusiast[s] and overclocker[s], because it seems like DDR3 is the technology nobody wants [badly] enough to learn about. Pity, because overclocking is what it's all about."
Just a tad over the top? (Score:5, Insightful)
Re:Just a tad over the top? (Score:5, Insightful)
Right now, all too many people I know are finding themselves out of jobs, with no good-paying alternatives in sight. I just attended my girlfriend's college graduation ceremony yesterday afternoon, and the guy sitting behind us was a 40-something year old who decided to go back to school last semester, because he couldn't make it anymore in the construction business. He said he worked in construction for 18 years, and until 3 years ago, it was a good career for him. But in the last few years, things have gotten so bad, many people are resorting to selling off the trucks and equipment they used in their trade, just to keep the bills paid and to stay afloat. They're seeing their work dwindle to the point where they can do it as a side job, but can't guarantee they're always busy. Therefore, he finally decided to go back to school and start a new career path.
My g/f is in a similar dilemma. Here is she. fresh out of school with a degree in psychology, and really can't do a thing with it except continue on to earn a Masters' in psych. After that, she could open her own practice (MORE $'s on top of huge student loan debt!), or possibly partner with someone else - with results varying depending on what part of the country you decide to live in. She's thinking about going for a double major, with the 2nd. one in business
Anyway
Re: (Score:3, Insightful)
Back to RAM though, I don't see how this is any different than with DDR and DDR2. At first, the new technology was barely faster (sometimes not at all) then the old one, was not very widely supported, and of course cost more. I don't see why this should be any different now with DDR2 and DDR3. A slowdown in the US economy isn't going to bring technological progress to a halt.
Re: (Score:3, Insightful)
The only reason to buy DDR2 over DDR is because it is cheaper (and compatible with more stuff you want).
At the moment is DDR3 cheaper? No. The last I checked it's about 4X more expensive or more.
Who cares if RAR is 10% faster (just making up figures
Re: (Score:3, Interesting)
Actually I don't think DDR2 was (is?) significantly faster than DDR for normal PCs.
The only reason to buy DDR2 over DDR is because it is cheaper (and compatible with more stuff you want).
That would be a horrible reason to make new RAM, change the connection just enough without any improvement to the hardware. Not that I'd put it past some hardware makers.
Your statement about speed might have been true when DDR2 had just launched (wasn't paying attention back then), since the low-end DDR2 memory modules have the same transfer rate as high-end DDR modules, but DDR2 has topped off at a transfer rate of just over 3x DDR's limit. Most people I know have gotten the more reasonably priced
Re: (Score:2)
I think there are only a few benchmarks/apps where the mem transfer rate becomes a big bottleneck. So far most of the real world benchmarks out there seem to indicate just like the DDR vs DDR2 days, DDR3 is not worth it.
This is because once the data set is big enough - disk becomes the bottleneck. Whereas below a few MB, a lot fits in the CPU cache - processing loops in games etc, so you don't get such a huge hit in performance for having a slower channel. In fact when there's a cache miss, l
Re: (Score:2)
Re: (Score:2)
That seems to be an important area now that every piece of hardwares seems to have a visual GUI.
Re: (Score:2)
Re: (Score:2)
The US economy isn't in a recession (last GDP was +.6), but web journalism certainly is in the boom times of economic fear mongering.
Re: (Score:2)
It's not a bad article and covers all the relevant points of DDR3. Well worth reading by people not familiar with DDR3. It's not an unreserved sales-pitch for DD3 as it does detail the disadvantages of the technology as well. The only thing the article seems to gloss over in terms of negatives is the ratio of performance to cost. Right now, I'm seeing DDR3 chips starting at around double their DDR2 equivalents and then rising substantially for the very fastest. For most people, I would say that this is not
Re: (Score:2)
Re:Just a tad over the top? (Score:5, Interesting)
Re: (Score:2)
same old crapola, with new gfx effects is about it.
Except, the new games are speed code in Visual Studio and
are massively resource intensive and require a Quad core to
play it like Supreme Commander with anything more than about
5 armies on the screen it starts to crawl.
After years and years, nothing really ground breaking.
Same old same old, so I am not gonna upgrade my PC when
I don't need the power.
Same deal for 90% of businesses, they don't need new
Re: (Score:2)
You sound like some of the guys here that say the same thing, until it comes on Fox News, in which case it's gospel.
Re:Just a tad over the top? No ECC = NO buy (Score:5, Interesting)
Anyhow, I'm on a 975X chipset with 4GB of DDR2 800 MHZ unbuffered ECC memory machine now. Not a single unforced error since I bought this machine and assembled it December, 2006. Nothing. This unit is primarily a gaming rig, the 3DMark 2006 score is 11500 with an 8800GTX, all with ECC memory.
The most irritating thing for me is, looking at the great new CPUs available, is the utter lack of any unbuffered ECC memory in the DDR3 range. This to me is unacceptable. I refuse to compromise so I will wait. Intel has a motherboard featuring DDR2 800 fully buffered memory for the 'high end workstation' , D5400XS, this is $600+. Supermicro offers something similar.
The X38 does DDR2 ECC, and for whatever reason the X48 took that away. I don't get why Intel wants to deny us DDR3 unbuffered ECC? Its really a selling point, a very good thing. If you overclock, its nice to have because it can tell you the limits minus the guesswork, not that I would bother with OC personally.
Fundamentally, without ECC, do you even know if the memory works at all? My experience leads me to believe that without ECC present, the answer is no at all.
Re:Just a tad over the top? No ECC = NO buy (Score:4, Interesting)
Re: (Score:2, Interesting)
Re: (Score:2, Insightful)
I have yet to have a customer offer a contract for a full ECC system, but I expect that type of customer to be ordering in the dozens or hundreds. For one machine or 100 I would still give 48 hours of memtest. For ECC I would still give 48 hours of memtest.
If you can recommend a better test, I am open to suggestions.
Re: (Score:2)
Memtest86 (and Memtest86+) do not do a good job of finding RAM that is borderline working. They are only able to find RAM that is definitely not working. By "borderline", I'm talking about systems that fail under moderate to heavy load in random fashion (usually bluescreens or random ap
Re: (Score:2)
How about a DDR2 versus DDR3 chart? (Score:5, Insightful)
Re:How about a DDR2 versus DDR3 chart? (Score:5, Informative)
Re: (Score:2)
Re: (Score:2)
I'm still waiting for an AM2+ chipset that will support DDR3, as the Phenoms (I think) have a memory controller that supports it. That should give the AMD chips a boost when compared to the current crop of Intel chips as the on-chip memory controller should allow for better usage of the RAM, but again, I'll wait until a benchmark confirms it.
From what I've read lately, Intel's Nehalem architecture, which features an on-chip memory controller and QuickPath interconnect (HyperTransport competitor), will be available around the same time AMD DDR3 platforms are available (maybe sooner). Therefore, instead of getting a boost from DDR3, AMD may get trumped by Nehalem.
Intel motherboards, at least to me, seem to be 2-3x as costly as AMD varieties, and don't always offer the same BIOS flexibility.
"2-3x as costly" seems a bit high to me, but I
Using His Jet Analogy (Score:5, Insightful)
"Jet engines are inherrently capable of much greater speeds than propeller engines!"
"OK, so show me one that goes much faster than the prop driven spitfire?"
"Well, uh, there's the Gloster Meteor!"
"It does about 500mph*, right? That's not bad compared to the Spitfire XIX's 460mph. But there are tons of Spitfires out there, available cheaply vs. paying several times the cost for the Meteor for about a 10% speed improvement." *note: F-3 variant, not the "overclocked" tweaked versions that set speed records.
"Well, but the point isn't that it's better today. It's a better technology! It'll be better in the future! Props will never go supersonic. Jets can potentially go several times supersonic."
"That is cool. Doesn't really help me today though, does it? I'm still paying several times the cost for a small improvement, today."
"Yeah, but if you don't buy jets now, how will their prices ever come down? How will we ever reach the heady perfection they're capable of?"
"Again, not helping me today, is it?"
"But! But! It's really cool!"
Yes, the technology shows promise. But its future promise with only small increases today doesn't justify its current high cost.
If more people bought it, the cost would come down over time and more investment would mean unlocking more of that promise. Which is great in the future but gives you very little today in exchange for that high cost.
The argument he seems to be making is that everyone should adopt it right now, not because it actually gets them much for their money but because their investment will enable him to buy even faster stuff for a lower price later.
Not really compelling.
Re: (Score:2)
+1 insightful
Re: (Score:2, Funny)
Re: (Score:2)
Over the years I've learned to let the other guy pay the R & D cost. I've watched the prices drop on several items of high tech. Instead of overclocking, I just wait.
I waited until 4 banger calculators were under $100 before I bought one that ate 4 AA batteries in 6 hours. (Th
Re: (Score:2)
not cost effective for the performance gain (Score:5, Informative)
Re:not cost effective for the performance gain (Score:5, Informative)
Re:not cost effective for the performance gain (Score:5, Informative)
Re: (Score:2)
DDR3 is still 5-10 times the cost of DDR2, and the performance gain is not big (maybe 10% at best) on overall system performance.
And to take it even further, that's just the cost of memory. A quick NewEgg search for DDR3 motherboard came up with a whopping 10 boards. The cheapest are 150 for open box items. The typical price is over 300.
If you compare (using Newegg) otherwise identical motherboards where the only difference is DDR2 or DDR3 (e.g. ASUS P5K DELUXE [newegg.com] vs ASUS P5K3 DELUXE [newegg.com]), the price premium for the DDR3 motherboard is typically $50. Those "typical" $300 DDR3 motherboards are high-end enthusiast boards and their DDR2 equivalents are pretty darned expensive, too (typically $250).
I'm not saying it's better to buy DDR3 now, but I don't think it's as bad as you and the GP make it out to be. One advantage of going with the newer mem
Re:not cost effective for the performance gain (Score:5, Interesting)
If you want real performance improvement get a 64 bit OS and 8GB DDR2 instead of 2GB DDR3. It will probably cost less and you'll notice the performance improvement as fewer accesses to HD (given you're OS knows how to pre-fetch intelligently).
Re: (Score:2)
$ cat
MemTotal: 8310468 kB
MemFree: 5979288 kB
Buffers: 38932 kB
Cached: 1234492 kB
-snip-
SwapTotal: 0 kB
SwapFree: 0 kB
-snip-
Normally the Cached value is sat around the 6GB mark. PAE is a dirty, filthy hack, but it works rather well. The main limitation is the 4GB per-process virtual memory space - but under 32bit Linux can handle up to 64GB of RAM.
Re: (Score:2)
These hard drives are what's slowing us down.
Which is exactly why we need more RAM: RAM is a cache for your hard drive. The more RAM you have, the less your hard drive gets accessed so the less its slowness matters. If your RAM is large enough to hold your entire working set (data, programs, shared libraries), then your computer hits the disk the first time it needs each file and then only touches it for writes afterwards.
Just as much, I assure you that memory speed is a very serious concern. Read up on the memory wall [wikipedia.org].
Re: (Score:2)
Re: (Score:2)
Re: (Score:2)
Memory Bandwidth... (Score:5, Insightful)
Re: (Score:2, Insightful)
LANs are already fast -- when did you last saturate a GigE switch ? Right now, for most applications, the bottleneck is in the disk.
Re: (Score:2, Interesting)
I'm typing this, and by the time I release the key I'm pressing for the current letter the computer has already received it, sent it to the right app, which put it in the right text-box, updated the undo buffer, re-rendered the window, checked if it completes a word, and spell-checked the new word if it does (and readied a list of likely corrections if it isn't correct).
It's hard to get enthused about x% lower memory latency when it still takes me 100m
Re: (Score:2)
Build A Cluster... (Score:2)
Overclocking is stricly amateur level (Score:2)
Re: (Score:2, Interesting)
Every machine I have had since my 120 Pentium Doom special o/c'd to 133 has been overclocked with no problems. Once you have everything stable there is almost always at least 10% with room to spare.
Re: (Score:2)
Yeah, but that's Opterons. You could probably smear them with jelly instead of thermal paste and they'd still work.
But for most users, people would prefer the "certainty" of running under a manufacturers specifications to the "risk" of overclocking their CPU.
Re: (Score:2)
Performance wise, a mid-level PC does fine for me (my latest one is an AMD dual core with a Nvidia GT8600 GPU). But the reliability better be above average. That means
-no overclocking
-reliability enhancing extras where they are not too expensive (ECC RAM is one of those)
-choosing vendors that have a history of good quality rather than flashy features
-not necessarily the very latest technology, as older stuff is often more mature
Of course, this will not solve sof
Don't [you] just [love] these... (Score:5, Funny)
Re: (Score:2)
Teenage enthusiasm (Score:5, Insightful)
It's always nice to see a tech writer full of teenage enthusiasm, but this article goes a little over the top.
It's supposed to be an article about a performance enhancement, and there's barely any performance values at all (except for the theoretical peak throughputs on page 3, and we know how much that means).
I think what the guy really wanted was to write about planes, not about computer hardware.
No advantage to DDR3 (?) (Score:5, Interesting)
First of all, as the technology was brand new at the time, the motherboard, although capable of 1333mhz ram, it only detected it as 1066 (we double checked they sold us the right stuff), so we manually set the RAM in the BIOS to run at 1333.
After all the setup, on otherwise almost identically configured machines, we found absolutely 0 performance gain on the DDR3 machine over the other 3 DDR2-800 machines. Although one might argue that our applications we were using to test were not so memory intensive, the fact is it was a computationally intensive task that regularly accessed about 200-300mb of data from ram. I would think that even if everything would be pre-fetched into the 8MB CPU cache before it was used, we should at least see some small difference.
In the end, it seems that we spent an extra $800 for no noticeable performance gain.
Re:No advantage to DDR3 (?) (Score:5, Interesting)
- We were using XP Pro x64 edition
- I believe there was 4GB RAM, possibly 8GB
- I tested exactly the same program that we wrote and compile with the same input data and timed it between the DDR2 and DDR3 machines.
- The processing took exactly the same number of seconds on both platforms. (Approx 20sec). Testing on older Xeon 3.6 gave me approx 50sec, and Pentium D-3.2 at 30 seconds.
Re: (Score:2)
It all depends on how the memory is accessed. If you keep accessing elements at (seemingly) random locations in the data set, then it's likely that you'll h
What about the 4GB limit in Vista 32? (Score:4, Insightful)
Re: (Score:2, Interesting)
Re: (Score:3, Informative)
Actually you can address your additional memory using PAE though I'm not necessarily recommending you do this. I use a 64-bit Linux install and seem to be doing okay. I have had a few FLASH problems, but only an odd non-start on YouTube pages, fixed by hitting reload. It honestly doesn't bother me and has never happened during the playing of a FLASH movie, only on starting one. I expect this to be fixed soon enough and I'm much happier running the 64-bit version. Personally, I'd install the 64-bit version. [wikipedia.org]
Oops - missed a tag. (Score:2)
Well, it was bound to happen sooner or later.
Re: (Score:2)
SDRAM not SRAM (Score:2)
I've learned my lesson before (Score:2)
Bending the facts, paid advertising? (Score:4, Insightful)
And the fact is that the double prefetch buffer is the sole reason for the double bandwith and the double latency. As a matter of fact the speed of the individual memory chips on the ram module are the same as on ddr2 (see that table in the article, just divide the ddr3 clock by 2 to get the corresponding ddr2 speed) - but instead of reading 1 bit from 4 chips at once into the prefetch buffer (that is four bit prefetch buffer), they are reading from eight at once (so that's the 8 bit prefetch buffer), so double the amount of data can be read in the same time (hence the double bandwidth). However because the memory chips are the same speed as in ddr2, the time needed to program the individual chips stays the same - so because the clock is double the speed, it takes twice as many clocks to tell the individual chips which bit we want to fetch. And that bullshit about lower latencies is also not quite right: ddr3-1600 cl6 is the same latency as ddr2-800 cl3 - and such modules have been sold for years.
Of course ddr3 is better, because it has higher bandwidth, and absolute latency is not worse than ddr2's. Also there are in deed technological advances (e.g. the lower voltage). But this article is still not exactly honest.
Wait for Nehalem (Score:2)
Incompatablity (Score:2)
I couldn't find DDR3 memory confirmed to work with the motherboard I wanted!
There are so many stories in the reviews sections of NewEgg and other sites with people complaining that boards like the Asus Maximus fail to work with many and various brands of DDR3 memory.
The only brands I could confirm to work, were "out of
Huh? (Score:2)
While I understand it might be your particular hobby, it's not the case for most others.
I used to upgrade my computers every 3 years, which would roughly give me a 3x improvement in raw processing power (not to mention video, etc.).
IBM AT, 486-33DX, P-90, (AMD)P300, (AMD)P1.2GHz, P4 2.7 GHz.
That last one is something like 4-5 years old now (I forget) and it's only reaching the end of it's value as a cutting-edge game machine. I haven't found a lot of persuasive ev
Re: (Score:2, Informative)
Re: (Score:2)
Re: (Score:2)
Re:stupid (Score:4, Informative)
I wouldn't call it a success, either. I'd wager that figure is 90%+ copies that came with new PCs. The large majority of which probably end up in a corporate setting where it was re-imaged with XP Pro (happens all the time where I work and for our clients).
Re: (Score:2)
I might just take that bet.
The OEM system install has been the norm in the home and SOHO markets for thirty years. The bare bones PC is for the enthusiast or the IT pro.
I am not convinced that consumer sales of Vista can be so easily ignored.
Operating System Market Share [hitslink.com] {April 2008]
These web-based stats
Re: (Score:2)
Vista is a bundle success (they've succeeded in bundling it with a lot of systems), not a sales success.
It would be interesting to know how many Vista-licenses that's been chosen when there's been XP as an alternative.
Today I hear a lot of "I bought a new machine and it came with Vista. Can you help me get rid of it?".
Fortunately we have a license that allows our personnel to use our site XP-license
Lack of choice (Score:2)
Did the customers actually want vista, or was it forced upon them by their choice of computer?
A better metric would be to see how many copies of vista sell with XP sitting next to it, both available equally.
Otherwise, it's like saying that 99.999% of people prefer to breathe oxygen - like there's a choice.
Re: (Score:2)
And the Internal Revenue Service has received "donations" from 200+ million Americans and thousands of businesses. What it is selling must be amazing.
Just because the copies went out the door, does not mean everyone asked to buy them.
Re: (Score:2, Funny)
Burning down the factory and collecting the insurance?
Re: (Score:2)
Re: (Score:2)
Services.
NBC was the dominant radio network in the thirties, so big that it had two AM and one shortwave feed. You might not be able to afford a concert ticket, but big band jazz and the NBC Symphony Orchestra broadcasts were free.
The movie studios had a rough time of it until they hit on the formulas that would draw audiences to a cheap - air-conditioned - respite from the stresses of home and work.
Re: (Score:2)
Why pay for the highest clocked processor when you can get the exact same core(and performance) with the cheaper one?
Re: (Score:3, Insightful)
Re: (Score:3, Informative)
Yes, mark it at a lower speed, and sell it as such.
Why do you think they "lock" things?
Re: (Score:2)
Re: (Score:2)
Re: (Score:2)
Oh, might also mention for you other doubters, Im running a nearly silent zalman cpu cooler. The only thing you hear in the case is the pair of seagate hard disks.
Ive been gaming, capturing audio(I record my own music), encoding video and audio, watching movies, downloading things.. If that's not stable enough for you, you have
Re: (Score:2)
Re: (Score:2)
out of Q6600 clocked to 3.4 ghz, and it will go to 3.6 ghz,
but it is branded for just 2.4 ghz.
The new 45 nm cores supposedly go a little faster even.
You also get to up your bus speed, and other things as well
if you cards can handle it.
It takes some tweaking if your flying blind, but reading
some hardware sites you can just back it off a notch from
their best and it works fine.
Re: (Score:2)
Re: (Score:3, Insightful)
* BIOS
* BMC
* WLAN Card
* Disk Controllers
etc. pp.
Re: (Score:2)
Have a look here: http://www.coreboot.org/Supported_Motherboards [coreboot.org]
Re: (Score:2)
I have a passively-cooled PSU and video card, and two 120mm fans running nearly silently at minimum speed, and still I overclocked my Core 2 Duo nearly 50% just for the heck of it.
It's not all about fans roaring like jet engines or ridiculously intricate wate
We Drink Ritalin (Score:5, Funny)
Re: (Score:3, Interesting)
> If PC gaming is dying, HTPC gaming can revive it.
Considering the HTPC itself doesn't seem to be gaining much traction these past couple years, and consoles have been encroaching (albeit very slowly) on the HTPC space, I'm interested to hear what your view on the topic is.
Re: (Score:2)
Re: (Score:2)
Why would Dance Dance RAM you? (Score:2)
Re: (Score:2)
Re: (Score:2)
Re:Not Needed (Score:4, Informative)
And the actual reason memory manufacturers have such a hard time selling memory performance is an extension of this; most purchases of memory have an implicit tradeoff: amount versus speed.
In most cases the real memory pain will come when you hit swap. That means you will almost always notice having less memory more than you'll notice having slower memory. So basically the only ones who end up having 'fast memory' on their system purchase checklist is those for whom money isn't an issue at all (ie, there is nothing else you could buy for that money that would improve your system (or life) more than faster memory). Or those who have very small applications they need to execute blazingly fast. Such as benchmarks.
Until those problems get fixed, faster RAM won't make a bit of difference to the end-user.
More accurately, faster RAM will make less of a difference than more RAM.
Re: (Score:3, Informative)
I think I'm being trolled.
Re: (Score:2)
A 1 hour
The power consumption is NOT linear with the voltage but square. So if you lower U by 17% then P gets lower by 21%.
Mentioning Voltage and not mentioning the Resistance make the comparison bad. If the memory is the same resistance in both, then the math holds up. If the resistance (affecting current drawn)isn't the same, then that is like comparing 2 cars. One runs 4 hours and the other runs 6 hours. Which gets better gas milage? What size is the tanks? What speed and
Re: (Score:2)
For the money, maybe you need to consider a cluster instead. I think 16 to 32 cores m