Slashdot Mirror
Bypassing Intel's Overclock Limit Reveals DDR2-667
BatonRogue writes "Slashdot posted a Tomshardware article talking about Intel's 10% overclock limit on their new chipsets not too long ago. The situation has just become even more interesting. AnandTech just posted a roundup of DDR2 memory that sheds some light as to why Intel may have implemented the lock. It seems that on the Abit board they tested, which supposedly bypasses the overclock limit, the first generation of DDR2-533 memory modules had no problems working at 667MHz. Could it be that Intel is keeping DDR2-667 support for yet another revision of their new chipsets even though the memory support is clearly here today?"
Will DDR2 memory let me hit twice as many arrows at the same BPM?
I seem to recall last week almost the exact same posting, however it was using AMD's name rather than Intel.
Anyone got any intel on that? (-;
Mike
Gamblers Forum
You know, I think we've all heard rumours of an intel conspiracy to make us buy the same product again and again for years now...
However, one must at least consider that they have a valid reason for this. Long-term stability maybe?
Oooohh... is that machine stability, or cashflow stability....
-s
Could it be that Intel is keeping DDR2-667 support for yet another revision of their new chipsets even though the memory support is clearly here today?
I'm suprised one must even ask this question, because the answer is absolutely Intel would do something like this. Keep in mind that the benefit of locking overclocking to only 10% is twofold for Intel. Not only do you disguise the now revealed fact that DDR2 is ready for faster speeds, but you stop people from buying cheaper chips and overclocking them. At least, that's what intel intended before the mobo manufactureres bypassed the locking. Think about it in this theoretical situation, why should I buy the $1000 3.4Ghz chip when I can buy the $500 3.2Ghz chip and overclock it 200Mhz?
To imagine that a company would manage the release of products. Its as if they had a plan. How could they keep new things from us that they have already finished? My goodness what is the world coming to? The next thing you will tell me is that they have products planned that go even faster.
Could it be that Intel is keeping DDR2-667 support for yet another revision of their new chipsets even though the memory support is clearly here today?
sure why not? intel could be postponing on brining out their biggest guns for later on when it would be better to shoot them off, they might also want to see what the competition will do and bring out before they bring out their best, but on the same note, who knows what else Intel might be holding back? They might have some other amazing thing that they are waiting to spring on the public at an appropriate time...
Then I suddenly remembered the warm-buzzy feeling I got the first/last time I tried to check my timing belt in my '80 Capri.
"I can't give you a brain, so I'll give you a diploma" - The Great Oz (blatently stolen sig)
it's the completed ones that are submarined that do.
and Abit's stock creeps up a bit...
mod parent troll down
hmm.....no.
I think it may just be that Intel doesn't want bad word of mouth. People see Fred's Intel based system runs awfully, little do they know he has it overclocked to the max and form opinion.
Intel is right. The world is not ready for Dance Dance Revolution at this breakneck speed.
I guess DDR2-666 was a bit too... evil. /lame_joke
Hmmm.
--Chag
Any DDR vs. DDR jokes have already been made.
Don't look a gift-horse in the mouth.
Could it be that Intel is keeping DDR2-667 support for yet another revision of their new chipsets even though the memory support is clearly here today?"
Wow, a company may be holding back on technology for a future money making opportunity? This has never happened before!
"I only know 2 things: The love for me, and the fear of me."
While I really dont like companies crippling the hardware I buy. Could you imagine a shovel that would stop you from moving sand too quickly ?
I have to say that if my systems are %10 slower but even %05 more reliable its a good trade off. I'm not saying that they are super unreliable now, given all that goes into making a computer behave and the number of suppliers involved. Its just that some of these machines are so fast that I there is no value to me personally in overclocking.
I think that businesses generally have enough to worry about with proper operation to not want to overclock their processors. If Intel is working with memory makers to insure stability they should be up front about it -- people would understand the overclocking then.
Overclocking adds another possible source of problems that most companies don't need.
Just because you can overclock your memory to that speed does not mean the manufacturer is greedy. It just means it can be done. Safely? Maybe. Possible? Certainly.
I hate sigs.
It is surprising. Intel surely would realize that whatever controls they put on there would eventually be bypassed. Sounds like a marketing decision rather than an engineering decision.
667 BPM is only a bit faster than the 600.0 BPM segment "MAX. (period)". If you're willing to count BPMs that occur only during freeze arrows, "The legend of MAX" already gets up to 666.0 BPM.
Besides, DDR jokes have been done.
But go forbid that the exact same engine with a different label is doing 20000 rpms no problem.
Moo!
Wasn't "Cartoon Heroes" already done, way back in DDR 8th Mix?
1) A major reason to not support the higher speed is that chip production yields are increased.
2) Why not speed-grade the parts and sell two models? Not cost effective at this time.
3) Futhermore to release one model how and then another model later maintains a more even revenue stream than two models now and then none later.
4) Most likely, spreading it out also increases total revenue due to the people who buy one and then upgrade to the next.
I guess DDR2-666 was a bit too... evil.
666 BPM? Isn't that "The legend of MAX"? I'd contend that the steps to "Maxx Unlimited" are a lot more evil than LoM's.
-- The legend.
Turn the question around. Instead of asking "why buy $1000 3.4 GHz chip instead of $500 3.2 GHz chip", ask this: how come Intel doesn't label that second chip "3.4 GHz" and sell it to you for $1000?
That would be $500 extra revenue for Intel. How come they don't do it?
Perhaps it's something to do with increased failure rates, warranty returns, and a negative hit on their reputation.
Intel grades their chips. They mark each chip with the speed that they feel comfortable selling with Intel's name and warranty.
If you want to overclock your chip, it's your chip; you bought it, you didn't license it with a stupid EULA! But the problem comes when another company (not Intel) buys a $500 chip from Intel, overclocks it, and then marks up the price and sells it in a system as if it were a $1000 chip. Intel gets nothing all the trouble, the cheater company gets the markup, and the end user gets the shaft.
In Intel's long tradition of pushing a new technology before it actually surpassed the previous generation (P2, P4, P4-Prescott) DDR2 is, at best, on par with systems already in place.
If you look at some benchmarks of DDR2 performance, you have to wonder why anyone would even consider buying it right now.
"Expect DDR2 memory at 533MHz to be comparable to DDR1 at 400MHz, but don't expect to see any "noticeable" memory bandwidth performance gains till DDR2 667 and above with low timings!"
my sig's at the bottom of the page.
Incidentally, my brother also says that Intel would be at a great advantage over its competitors if it could offer the 667 mhz, so it is clearly not hiding an existing feature to milk the market. The spikes are the reason.
Why, your describing that troll Ceren. I don't know how those *BSD nerds get a hard-on over her, if I did, even involuntary, I would do the right thing and kill myself.
Um, right. 1,000 terrahertz? ROFL, you must be smoking crack or something. Sure, its possible, if you want to pay a few grand for a motherboard.
marketing.
They don't want to keep pushing the envelope to quick. Otherwise, their future market is gone. Meaning, since technological advances are happening slower (in regards to chips) they have to milk the current tech as long as they can. Don't want to sell everything this year - there won't be anything next year.
That's a marketing decision.
Engineer: Our new chipset is capable of running DDR2-533 memory modules at 667MHz. We can jumpstart shipping...
Marketing: STOP! Right now there is no demand for such a thing. Let's wait till we are ready with a new promotion song from Christina Aguilera.
Engineer: But no one buys our chips because of the promo and the songs we have. Our customers buy because of the specifications. Right now we have...
Marketing: Who are you to tell ME what our customers want? Stick to your soldering gun and chips!
Engineer: But the Intel guys in their funny radioprotection overalls didn't accelerated sales....
Marketing: No buts! We start marketing this feature for christmas.
Engineer: But...
Marketing: SHUT UP, I'm the guy with the MBA! That's why I am here. I know what to do!
Engineer:...
Marketing:...
Grundgesetz * 23. Mai 1949 - 30. November 2007 - http://www.vorratsdatenspeicherung.de/
Wouldn't it be interesting if in 6 months time Intel puts the new speeds into yet another revision and we all refuse to pay for it or at the advertised price? merely because we BEAT Intel?, by six months?
How can you jump to that conclusion? There is a whole lot more to a new technology level being "here today" than a few chips being able to run at that level. Yield, reliability, availability of memory at reasonable price points and reliability, reliability of motherboard support, etc. all play a part. The famous Pentium floating point bug had rare effect except on scientific applications, but clearly that version of the Pentium wasn't even "here" when it shipped.
I currently have an ASUS based AMD64 system at home that I made the mistake of buying in the second month of availability. I can tell you from firsthand experience that it wasn't "here" when it shipped. Almost everything of any meaning has been replaced and the system still freezes solid twice a day. Only a hard reset brings it around. Pretty soon, I'll go another round of replace the processor... does it work... replace the motherboard... does it work... replace the memory... does it work. I'm betting this time I'll finally get there because someone has figured out a problem and fixed it in the latest releases of these 'stable' products.
Faster than a speeding bullet, Greater than the speed of ...uuh ZzzSSSHOOOM!!! ... DAMN YOU INTEL!!!
Intel reliability standards call for infant mortality less than 500 units per million and a service life of 7-10 years depending on product. The goals of overclockers are very different.
Ever heard of a schmoo? It a contour graph of what frequencies a part will work at at a given voltage and frequency (and that's only at one set temperature). Ofthen is is very blob like and can have holes of instability or islands of stability. You need to pick product offerings that not only yield well, but are far enough on any contour line from the schmoo that you can test the part at a few values of voltage and frequency with confidence. Tester time is a bottleneck, and Intel test more than anyone else.
Yes, downbinning occurs for marketing reasons, but generally for economy products. Believe me, Intel and every company fights for yields at the top bin for every flagship product. Even when downbinning occurs, it is done to match actual yield to orders, not out of any desire to hold back.
So you can beleive the last three paragraphs or you can believe that Intel is being forced to slow the rate of product advangement by the Carlye Group. It's up to you.
...is going to like this as well. Remember, they've usually had less of a requirement for memory bus. With them going dual-core, while retaining socket compatibility, they only have one connection to feed two processors. At DDR 400 today, they're already looking at 2xDDR333. While I'm sure the equivalent from Intel would like to have 2xDDR667...
:)
Overall, looks like there's going to be a lot of competition going on in the CPU biz for a while to come. And obviously a bunch of memory brands. In fact, I think there's pretty good competition all around in hardware, and things are starting to look up in the OS world as well. I predict good times for the consumers in the time ahead
Kjella
Live today, because you never know what tomorrow brings
a few weeks ago there was a deal about dell dimension 8400 with p4 3G HT, 1G DDR2, ati x800xt video card for less than $1100 after 15% coupon and mail in rebate.
I wonder if anybody who bought that system ever got any problem for stability?
Why does the subject of overclocking have to come up over and over again. I personally wish this topic could be banned from /. posts. This is my view .. you buy a system (unoverclocked) and 90% of the time you are using it, it's processor level is under 10% usage. Granted if you are a heavy video game player then get a damn console system. They have smaller processors but are optimized for game play and run more stable than any other computer system. Or look into other processor alternatives .. for gaming I have a dual Xeon 2.8 GHz system .. now those 2.8GHz processors cost you under 300 a piece. I use the system for work related things as well, but I would bet that I would run smoother in high resolutions than some lame overclocked 500+ dollar single processor system. Not to mention that I am stable. I have yet to see someone overclock any type of processor and see it run perfectly stable. Overclocking is for script kiddies and metrosexuals!
[off topic]*lol* troll?![/off topic]
Moocowsia: But go forbid that the exact same engine with a different label is doing 20000 rpms no problem.
see, that's the caveat... the "no problem" part
mod me redundant, but obviously, intel does product testing and quality control. sure it may work with "no problem" on your board, but not every board. surely there's a threshold they try to hit...
for a minute there, i lost myself...
that the chip manufacturers have designed their chips to work at certain frequencies, and through the manufacturing process/refinement of the design, most parts hit it, some are too slow, and some are much faster than expected. Once in a while, most of the parts for a given design can run much faster (Celeron 300A was used as an example in another post.)
Now, what if those memory makers design and test those parts to stringent standards, which define and limit the functionality. If you move outside of those "standards", you could move into un-guaranteed territory and possibly do better or worse.
In these cases, raising the voltage to 2.0V or slowing the latency to 6 tCK will put it outside of the JEDEC-defined standard, yet can help get DDR2 parts to reach 800MHz clock or faster.
Let me repeat that: outside of the JEDEC-defined standard.
The standard is there for interoperability between suppliers/VARs/etc. Commodity markets require this.
Sometimes commodity chips can do much more, slightly outside the standard, depending on a lot of variables. The variables are usually what the standard tries to minimize or control to increase the stability of the entire system.
AC, since I do work for a major memory maker
The problem with the benchmark:
They were running Windows XP.
Any pre-emptive OS will give you terribly innaccurate results for low-level hardware testing.
If you want real results, don't run tests from an operating system--run them from the bootloader (with interupts disabled). Myself, I would probably modify something like Memtest86, and have it time the tests. This would reduce problems due to chipset differences and chance. Any test that can't be exactly repeated with the same results is marketing fodder.
The reviewers otherwise appear to have done a good job. The two test rigs used very similar hardware. They even thought to disable unneccessary peripherials in the bios. They probably even ran each of their tests after a fresh Windows install.
This doesn't mean that their results are wrong, in fact their results are about what I would expect (after having poked around the internet looking for technical information on DDR1 and DDR2). Their presentation is a bit misleading: They test the DDR1 memory with tighter timings than the DDR2. Since these timings reflect how much the memory can get done in a clock, is it suprising that DDR2 requiring 4 clocks for an operation running at 667MHz is equal to DDR1 requiring 2.5 clocks for an operation running at 500 MHz? (For the math impaired this means that with their setup the DDR2 can do 166 million ops per second, and the DDR1 can do 200 million ops per second). I suspect that the reason they tested this way is that the DDR2 chips don't quite have the timing of their DDR1 counterparts yet. Even so, if you claim to be comparing two different technologies, wouldn't you at least run one test with them set up equally?
are you talking about bus speed? apple's (ibm's, really) g5's bus runs at half the speed of the processor, so the bus is 1.25GHz on the dual 2.5GHz powermac. what's the problem here?
There was 66, 166, 266, 366, 466, 566, but get to 666, OH NO! Now it's 667! and then they reverted back with 866.. Of course, there's goobers that try to justify the change as if there was some technical reason than admit why it was really relabled. Boo to bowing to religious superstitions. Quick, change the BSD mascot!
Sounds like a fun time I had with my car alarm. It never quite sounded *loud* enough. Then, when I got a new car and moved the alarm over, the guy installing it clipped the "muting" wire on it. Apparently the previous installer (Visions) either didn't know about the muting wire or saw fit not to off the option of unmuting it.
Intel would never cripple the functionality of their products to spurn sales of other products. Nope. NO way.
AMD Rules.
I think it was pretty clearly intended as a joke. He forgot the "beleagered" flag that tips us off in all the Apple stories.
"The manufacturers are simply not leaving as much margin as they once did."
Neither are consumers.
god all those replies and no-one's got it right yet. perhaps it's too obvious: 666.666 (recurring, that must be really evil) rounds to 667. the nearest integer, easier for marketing purposes i guess. i.e. it's slightly slower than the rated speed (oh no! we'll have people complaining they're being ripped off now..)
This is my Sig, this is my Gun. One is for Slashdot and one is for Fun.
For example: An Agilent 16900 logic analyzer comes standard with a 400MHz/1MB trace buffer. If you want a 550MHz/32MB trace buffer, you pay them some money and they provide you with a software key to "unlock" the feature.
Do you think they lost money when they sold you the base unit? No, it's pure profit to them and they actually save lots of money by making just one type of analyzer and not dealing with any field installable hardware upgrades.
Somehow it seems unethical to me - since I've already paid for the faster and deeper memory and have to pay again just to be able to use it.
--
This space for rent.
A friend of mine worked at the sales call center at Mindspring, Back In The Day(tm), when you had to put your account number as the PPP login. He had a customer cancel service because he had no way to change her account number to get rid of the 666 in the middle.
My new
Just because something is stable when it's overclocked doesnt mean that it's stable. If it was a server, even a problem that caused a lockup every 6 months or year wouldnt be acceptable. That freak incident could happen next year, or tommarow.
I'm not against overclocking or an Intel fanboy, I run an overclocked barton at home. But I dont think Intel would hold back, especially with the athlon64 out there and with the negative prescott press, Intel might be a bit more cautious than usual.
I'm an Intel girl, in a Bill Gates world.
Bound in barbed wire, thrown in the mire!
You can bash my skull, I will be your gull!
[Chorus - Bill] Come on, Intel, visit my hell.
Wasn't old Tom right tho?, that DC current is best for short distances? Why doesn't the modern PC have an integrated power backup? If it ran on straight DC that would eliminate the spikes and we could go on to a safer tomorrow. AC power supplies are evil. The college-educated who are making simple processes overly complex and failure prone aren't far behind. Edison laughs at us from the grave. America used to be a place to excel but now we found out how to profit out the yingyang from neverending Sales... which holds people back. We have become a nation of control freaks. One look at our political system that can only come up with two white guys running for President should tell us a thousand words. Maybe the reason we're getting the dregs on cpu and memory is that NASA is scooping up the best for sending to Mars eh? I underclock my stuff so it will still be running in 2030. You can't beat a 3.4 downclocked to 667 hehehehe. Nitrogen power Forever.
How much faster is this new generation of DDR-2 compared to the old DDR? I didn't see any graphs at AT showing the speed bump.
How does this technology work? I'm disappointed that this isn't talked about in the article.
So you overclocked a CPU by 25%. Because it's not the only part in your computer, your system runs 10% faster. So what?
I understand overclocking 8088 from 4.77 to 8MHz to speed up compiles. Or projects like hymn that remove artificialy imposed limitations which really hinder users. But what is achived here that can not be better done with a dual processor, perhaps a nice G5?
Not to be pedantic but he actually lives across the street.
Hmmm, why is it when somebody wants to do something on here they start with "Not to..."
Well, I guess it's a good thing to lose customers like that. I mean, if they can't get over some random numbers in their PPP login, how often do they call to complain about that 'Peer' person and how he always resets their connection?
flamebait, but here goes.
i can personally remember 50 or so mildly overclocked systems i've built that were rock-solid: they'd prime95 or 3dmark all week with no problems.
Because I think the exact number is something like 666.66... Which as you ought to know, rounds to 667. But 333.333 rounds to 333. See? :)
"Could it be that Intel is keeping DDR2-667 support for yet another revision of their new chipsets even though the memory support is clearly here today?"
Of course it could be. Why release DDR2-667 support today when you're still making reams of cash on DDR-533? The only way to make technology profitable, since margins are so low on hardware, is to squeeze every penny you can out of one technology before releasing another.
Besides, I doubt that they had "no problems" with DDR2-667. The memory likely was not designed to run at that speed, and I am sure that eventually it would fail, and by fail I mean there being an error, not necessarily the complete failure of the memory device. Signal integrity is a major problem at these speeds and unless the memory busses are terminated specifically for the speed at which they are intended to operate, it is unlikely that reliable operation can be expected.
Terminations at the higher speed would most certainly be different than at the lower speed because of the reduced reactance of the input capacitance at the bus loads. It is likely that a smaller series terminator would be necessary to prevent overdamping, which might cause the memory to miss a transition.
I would personally feel much more comfortable knowing that my memory was terminated specifically for 667. Now, if you have a high-speed scope, surface mount soldering skills, and lots of time, you could change the resistor packs on your motherboard and DIMM and check signal integrity yourself. It would certainly be a fun project worthy of a slashdot article.
Also, only the very first 486SXs were DXs with the Math Co-processor disabled. It was much cheaper to make a new die w/o the unit, put it in a plastic (not ceramic) cover. I believe they also went with BGA type interfaces instead of the pins in the ZIF socket.
I've never done the overclocking thing, but regardless, I found your posting very informative and interesting. I now know a lot more about the issues of both overclocking and chip design in general. Thanks!!
Ever since I first used a multiprocessor box (under RH 6.x), I have wondered why it isn't more popular - it was quite obvious to me that a dual or quad processor box with slower, cheaper processors provided much better interactive performance than a single fast (=expensive) processor. It was great having X11 display processing done on one CPU while compiles etc. ran on another. However I don't play 1st person shooters, so I suppose that the game engines don't support spreading their processing across processors yet.
It's easier to be a result of the past, but more fun to be a cause of the future! http://www.spacefinancegroup.com/
Some of the material comes from bin split, but not most of it.
Your strategy works to an extent, but also would result in a huge amount of overtesting. Test time is at a premium, so why would they test a 2.6 GHz chip at 3.2, 3.0, 2.8, 2.6 (i.e. 4 times the amount of test) when they only have to test their high-volume parts once at 2.4 on their cheap testers to get the results they want?
In other words, they test just as many as they need to test at 3.2 to satisfy the market requirements for 3.2. The rest of them they don't even bother testing that fast. They just instantly test at the lower speeds and sell em that way.
This is still a feature of IBM mainframes and higher-end servers (and some other companies too). It's called Capacity on Demand. You buy e.g. a 64-way machine but pay for a 32-way one. Then one day when your business is really busy, and you need the extra capacity, you pay them, they send someone out to flip a switch and you get to use that capacity. When you're done with it, they send the engineer back to reset the switch and you're back with a 32-way machine. It's not a secret. It's a marketing feature and is touted as such. PHBs love it.
Stick Men
Much better value for money, better engineering and 64 bits as well. IMHO of course.