Follow Slashdot stories on Twitter

 


Forgot your password?
Close
typodupeerror
×
Data Storage Hardware

Firms Get Away with Selling Untested DRAM 344

Posted by timothy from the cheap-n-cheerful dept.
peppytech75 writes "Melanie Hollands in IT Manager's Journal reports that 'In recent months, some Asian DRAM memory manufacturers have been getting away with selling untested ("UTT") DRAMs. Disturbingly, the practice seems to be getting traction at the lower portion of the module business. This is being done mostly by Taiwanese DRAM makers, who are undercutting the tier-1 guys by selling untested and unmarked parts.' What's the solution here? Or is there an actual solution to what amounts to pirate companies issuing counterfeit parts?" (IT Manager's Journal, like Slashdot, is part of OSTG.)
This discussion has been archived. No new comments can be posted.

Firms Get Away with Selling Untested DRAM More

Firms Get Away with Selling Untested DRAM

Comments Filter:

  • If you're stuck with one of these... (Score:5, Informative)

    by Anonymous Coward on Tuesday April 19, 2005 @08:48AM (#12280365)
    Here is the obligitory Memtest86 [memtest86.com] post. It's a great program, and chances are that you might already have a copy on your Linux install CD depending on the distro. There are even kernel patches [vanrein.org] that allow you to avoid the bad bits if they are isolated enough.
    • If someone is lucky she might obtain perfectly good memory at a cheap price, but does one make a living gambling in Los Vegas?

    • Re:If you're stuck with one of these... (Score:5, Informative)

      by Anonymous Coward on Tuesday April 19, 2005 @09:06AM (#12280501)
      Although Memtest86 is absolutely great for detecting memory errors, I perfer Memtest86+.

      It's a more updated version of Memtest86 (which was last updated in November!), from the x86-secret team. It'll do the same thing, just that it will identify all the new procs and chipsets better.

      http://www.memtest.org/ [memtest.org]

      PS: I find if the RAM has any errors, the Modulo-20 test will nail them. Methinks it's test number 11 in Memtest86+.

    • Re: (Score:3, Interesting)

      by account_deleted ( 4530225 )
      Comment removed based on user account deletion
      • Probably just lucky, Windows hitting different areas of memory to RH etc, as I've heard the reverse happen too (RAM failing on Windows but being fine on RH or other distros).
      • We used to have the same problem with OS/2 versus Windows back in the OS/2 days and all we could get was speculation out of the developers. I recently had a problem with a (linux) machine at work where it just wouldn't stay up for more than about 4 hours before it locked up. Knowing that this wasn't natural I tested a bunch of stuff and finally ended up downloading memtest86, which quickly diagnosted the bad RAM.

        Perhaps it's just a perception thing -- if Windows had been randomly crashing like that I woul

      • i had a dodgy motherboard (the RAM slots were dodgy). Linux installed but was a bit crashy, Windows wouldn't complete the install.

        Its because Windows and Linux tickle the RAM in different ways.

      • Re:If you're stuck with one of these... (Score:4, Funny)

        by Gr8Apes ( 679165 ) on Tuesday April 19, 2005 @09:51AM (#12280958)
        Because Windows has this interesting habit of loading crap you don't need into RAM, and swapping stuff you do need out into the pagefile. If the DLL that you don't need is loaded into the memory area that's bad, then nothing bad will ever happen, as that particular piece of code will never be attempted to be read.

      • Re:If you're stuck with one of these... (Score:5, Informative)

        by zakezuke ( 229119 ) on Tuesday April 19, 2005 @10:05AM (#12281064)
        I'm curious why Linux has issues with this... I had bad RAM for a while and didn't even know it running windows. It installed, and ran just fine for weeks. Installed Linux, and Redhat wouldn't even finish the install.. suse installed but then crashed at random times... etc.

        Was windows just getting lucky, or what?

        Are you sure it's a RAM issue. I found Redhat, and other distros hard to install when I had my old HP 2x burner. But when I upgraded to my DVD burner, the problems for the most part disappeared. It was as if the drives I was using didn't like the discs I burned, yet windows had no problem what so ever. I could install from my backup discs, never as much as an error making images, the evidence would suggest it made solid discs. To this day it remains a mystery to me, the fact that those discs still had the same problem, but if I copy those files to a HD from the very same discs, no problem.

        Another example, I thought I had a bad batch of ram. Tested bad, random reboots after being on for a while, crashing with CPU / memory intensive tasks. Drive me absolutely batty till I swapped out motherboard and the problems disappeared, and when I put in a lower speed chip in the same board, the problems also disappeared. I can only assume based on this evidence that the board in question didn't like running at 166mhz despite the fact that both are based on the same chipset, save the smaller north bridge heat sync.

      • Just this week I had exactly the opposite experience. I've had several Linux distros on this particular box, including the recent Ubuntu release. Installs were all flawless, system ran great, etc. When I tried to install XP on it, the installer kept crapping out. Come to find out, Memtest86 showed errors on a particular address every time I ran it no matter what RAM sticks were in the system. New mobo and CPU and now XP installs just fine.

    • Re:If you're stuck with one of these... (Score:5, Informative)

      by gravygraphics ( 548287 ) on Tuesday April 19, 2005 @11:41AM (#12282180)
      Memtest86 finds really bad ram, not good ram. Without having knowledge of how each chip is internally arranged, access to the chip's test modes and the ability to control the temperature, there is no way to finish testing a modern DRAM in our lifetime.

      Just take for example, the internal layout. If you had a 512M chip and you didn't know which cells were adjacent, you would have to write a single bit and read from every other word. We are talking x cells * y reads (*2 for writes). If you read 8 I/O's in parallel (remember I am talking about a chip, not a module) than we have 512M cells * (512/8)*2 = 7.2*10^16 OR 72 megagiga operations. Assuming you can keep about 200MHz worth of useful read/writes (remember most addresses aren't in the same page)than we are talking something like 11 years... for a single test that doesn't cover refresh, voltage/temperature margining.

      Oh one more thing. Tou are really not sure if when you write a 1, the device stores it as a high charge or a low charge. Without knowing this, you will have to redo that same pattern a BUNCH of times.

      Memtest86 is like a pilot walkaround on a plane. It can spot obvious things, but I sure hope I'm not the first one to fire up that jet engine.

  • unmarked and untested == pirated? (Score:5, Insightful)

    by qwertphobia ( 825473 ) on Tuesday April 19, 2005 @08:48AM (#12280369)
    I don't follow this analogy...

    • Re:unmarked and untested == pirated? (Score:5, Funny)

      by Anonymous Coward on Tuesday April 19, 2005 @09:03AM (#12280473)
      It would make sense if it was DD ARRRRRRR RAM.

    • Re:unmarked and untested == pirated? (Score:5, Informative)

      by yknott ( 463514 ) on Tuesday April 19, 2005 @09:06AM (#12280496) Homepage Journal
      If you RTFA, the author was saying that these unmarked and untested DRAM chips can later be marked as if they came from a Tier 1 manufacturer. These chips can then be sold for a premium, yet still less than the Tier 1 price. In that case unmarked and untested = pirated.

    • Re:unmarked and untested == pirated? (Score:5, Insightful)

      by akadruid ( 606405 ) <slashdot.thedruid@co@uk> on Tuesday April 19, 2005 @09:06AM (#12280506) Homepage
      Piracy = buzzword = whatever you want it to mean.

      My neighbour pirated my parking space. That guy pirated my seat on the train. All it means is 'they've got my toys, mummy'.

      In the UK, we have big posters at cinemas which declare 'Piracy funds Terrorism'. Which is beautiful, since its 100% true, and depends completely on people misunderstanding it.

      • Re:unmarked and untested == pirated? (Score:2, Interesting)

        by Anonymous Coward
        In the UK, we have big posters at cinemas which declare 'Piracy funds Terrorism'. Which is beautiful, since its 100% true, and depends completely on people misunderstanding it.

        Having made a complaint to ASA about the sensationalist ad they show before the feature, I found out the above statement is based entirely on one case where someone alledged to be associated with the IRA was caught selling pirated cassette tapes at a car boot sale. Nothing to do with movies, and hardly a major source of funding for

      • Re:unmarked and untested == pirated? (Score:5, Interesting)

        by Mysticalfruit ( 533341 ) on Tuesday April 19, 2005 @09:52AM (#12280972) Homepage Journal
        You should make up some stickers that say "Ignorance funds Tyranny" and modify their signs a bit!

    • Re:unmarked and untested == pirated? (Score:5, Insightful)

      by FidelCatsro ( 861135 ) <fidelcatsro&gmail,com> on Tuesday April 19, 2005 @09:10AM (#12280536) Journal
      I was about to say the exact same thing , This is capitalism in action .

      If these chips sell to people under no false pretense about what they are and there is a market for them then what exactly are they doing wrong .
      If they are mislabeling them then yes that is very much illegal , but mostly they make no claim to this , If you want to risk buying these chips then fair enough .
      They are mostly not pirates though and labeling them that because they are undercutting other firms sounds like a dubious marketing ploy.
      The major risk as I see it is a batch of modules gets into a major user (think IBM, H-P, and/or Dell) and fails (probably in Asia). The user goes publicly ballistic over the combination of faulty material and the supplier's inability to control the quality of its material. The press runs with it and the unlucky DRAM supplier's stock gets hammered. Some time afterward, it emerges that all the DRAM suppliers have this risk and then they all go down.
      If this hapens then its the fault of the companys such as IBM ,HP or dell for not testing these products before shipping , i very much doubt that IBM would fail to run a memory test before shipping a server though.

      If as she says they are being sold as tested moduals then this is illegal and the practice can be stoped fairly easily and is no threat to the Semi conductor bussiness .

    • You don't understand, ok.

      Pirated = doesn't fit in my current bussiness model.

      You want your music in MP3?? PIRATE!!
      You like your TIVO?? PIRATE!!!
      You want to sell lower quality products at a lower price?? PIRATE!!

      You see, it's the commerce equivalent to "terrorist". One size fits all.

  • We've been burned (Score:3, Insightful)

    by eyegor ( 148503 ) on Tuesday April 19, 2005 @08:50AM (#12280373)
    We've been burned several times recently buying lower tier RAM. Out of a lot of 20 pieces, nearly a third was DOA or died horrible with a month of installation (and yes, I know how to install RAM).
    • I know how to install RAM

      What's there to know when the task is simply plugging a DIMM in a slot? Dude, it is 2005, it's been really a while since this task required skill. :-)

    • Learning the hard way (Score:3, Insightful)

      by Tassach ( 137772 )
      Todays lesson is that cheap parts are inexpensive because they are crap quality.

      If you want it to work right, buy parts from a tier-1 vendor from a reputable reseller. Buying brand-x crap might be cheaper today, but it's more expensive in the long run as you'll have to replace it sooner, and waste more of your time tracking down wierd errors caused by flakey hardware.

      I hope the lesson wasn't too expensive for you. Next time, shell out a few extra bucks and get Crucial or Kingston RAM.

  • For me, great. (Score:5, Interesting)

    by Vo0k ( 760020 ) on Tuesday April 19, 2005 @08:51AM (#12280379) Journal
    I definitely prefer to go to shop, get the die, plug it in, run a test program for a few hours and have it replaced if I find any errors, than to pay some 80% extra for a sticker saying that some malaysian kid did it for me.

    • Re:For me, great. (Score:5, Insightful)

      by FireFury03 ( 653718 ) <slashdot@NoSPAm.nexusuk.org> on Tuesday April 19, 2005 @08:56AM (#12280422) Homepage
      I definitely prefer to go to shop, get the die, plug it in, run a test program for a few hours and have it replaced if I find any errors, than to pay some 80% extra for a sticker saying that some malaysian kid did it for me.

      Yes, because waiting for my mail order RAM to turn up, finding it's buggered and then having to spend a month trying to convince the supplier to get their finger out doing their slow-as-treacle RMA procedure is such a good use of my time... (Not to mention the very real chanced that the replacement RAM will be just as screwed)

      • Re:For me, great. (Score:2, Informative)

        by DataPath ( 1111 )
        Especially when they're just reshipping returned RAM. That's when you find a new RAM supplier.

        (true story - I worked at a computer store. Not that computer stores aren't guilty of reselling returned defective computer parts as new.)
    • Do you also replace the bad module when you find it?

    • Whatever. (Score:2, Insightful)

      by cnelzie ( 451984 )
      First off, the difference isn't 80%.

      Secondly, how much is your time really worth?

      For me, (and I live within 5 miles of multiple PC stores), buying RAM, taking it home, installing it, finding it to be bad (After running a 45 minute or longer Memtest86) and then returning it to the store would more then cost me more then my average hourly rate at the office.

      I would rather pay the few extra dollars, get home and have an extremely low chance of installing bad RAM into my PC, then have the possibil

      • Re:Whatever. (Score:3, Insightful)

        by gowen ( 141411 )
        (After running a 45 minute or longer Memtest86
        yeah, but you don't have to sit and watch Memtest you know? Find something else to fill that time : Mow the lawn, tidy the house, hang some wallpaper, have sex, masturbate frantically ... do whatever you'd normally do on a lazy Sunday afternoon, and you'll find you won't have wasted anything like as much RIAA-style "virtual money".

        • Re:Whatever. (Score:2, Funny)

          by Anonymous Coward
          Can't masturbate frantically. The computer is busy running MemTest, no access to porn. Sex is out of question too, no IRC, same reason.
      • I'd run the test anyway. (I remember buying a "wonderful" Kingston die for a server, lifetime warranty and all, horribly expensive, and finding it faulty) And it's not like I spend 45 minutes in front of the PC, I just start the program and go watch TV or something. I have to install the die, no matter if it's brand or not. So, for now no extra costs.

        The costs start only when the die appears to be faulty. At worst 1 in 20 case. Then I need to return it to the store, and repeat the procedure. So divide you
        • The kingston stick probably died because you mishandled it. I am willing to bet money you weren't wearing an antistatic wrist strap when you opened that package.
    • I'm worried it will lead to higher prices though. Memory manufacturers might end up charging a premium for tested ram as untested ram becomes common. This may be beneficial for large outfits like Dell who might be able to save some money with an in house testing facility, but this might end up costing more for people building their own systems.

      Think of it this way. If I want to build a system, I can get $100 of tested ram or $90 for untested ram. Now the $10 difference might sound good, but if it ends

      • Re:For me, great. (Score:2, Insightful)

        by Anonymous Coward
        If people are savvy enough to build their own systems, they are certainly savvy enough to test the memory themselves, so I see no problem here. Of course the "split" may increase the tested RAM prices a bit (smaller market share), but I don't think it would be a significant rise. Certainly not bigger than introduction of "lifetime warranty" dies besides the standard warranty ones.

        If you save $10, remember the chance of getting the faulty RAM is quite low. You -may- spend more on shipping (if you didn't buy

    • Re:For me, great. (Score:3, Insightful)

      by Lonewolf666 ( 259450 )
      Then you still might end up with some marginal RAM that happens to work fine during your test, but fails under slightly less favorable conditions. Like the next hot summer (the performance of digital circuitry degrades with high temperatures).

      When the german C't magazine did a RAM test a few years ago, they worked with a company that specializes in such tests. The used test environment can reportedly (IIRC) simulate borderline conditions and test the module under these. It does also cost a lot more than a

      • Re:For me, great. (Score:3, Informative)

        by Sique ( 173459 )
        But Germany is another case after all, as is the whole EU: Here every seller of technical equipment has to give a 2 year warranty to every piece of hardware he sells. So if a RAM trader buys those untested RAMs and sells them within EU borders, he has to replace every single RAM that fails within 2 years for free (except he can prove the customer is at fault).
        In fact according to the law he has to give the money for the original RAM back, but he can instead try to 'better afterwards' a.k.a. repair or replac
        • Actually, after six month the burden of proof shifts to the buyer.
          But this was not my point. I was trying to point out that doing a reliable memory test may take more than just plugging the stick into your mainboard and running a test program. So it may be a good idea to buy from a company that does it's own testing, hopefully to professional standards.

  • Why do people buy cheap ram? (Score:5, Insightful)

    by flibble-san ( 700028 ) on Tuesday April 19, 2005 @08:51AM (#12280380)
    These days I don't reccomend anyone to buy cheap no-name unbranded RAM. Of all the PC hardware problems I've had over the years, about 80% are down to bad faulty generic RAM. I know only use Crucial or Kingston. They check the RAM, I know the RAM I buy is going to be working. RAM is one of the most important parts of any computer system. Is it really worth saving the £3-£5 by getting cheap unbranded RAM? As the saying goes, you get what you paid for.
    • So true but as with everything in todays world, people look at price and not value.

    • Re:Why do people buy cheap ram? (Score:4, Informative)

      by chiph ( 523845 ) on Tuesday April 19, 2005 @09:05AM (#12280495)
      Same here. I used to buy whatever was cheapest, but after the time that a series of flakey bugs was solved by switching to good quality DRAM, I'll never go back. I probably spent two days troubleshooting it, which at my hourly rate, is many times the amount I "saved" by buying cheap memory.

      Blatant promotion: I've never had a bad stick from Crucial [crucial.com]

      Chip H.
      • Only had three pieces of RAM from Crucial (1x512MB PC2700 DDR333, 1x1GB PC4000 DDR500, 1x512MB PC2100 DDR266), but all of them have been error free as far as I can tell too. Another thing I like about Crucial is they deliver quick and for free (for orders >£25 anyway) - usually next day delivery or at most day after, with tracking (at least here in the UK anyway), and I'd definatly recommend them.
      • They sent us ram wich didnt work too well, it was for revision 1 of our motherboard and we had revision 2. I called them up and told them that memtest86 said their ram was bad. They didn't treat me like a moron: reseat the ram, we dont support linux, etc. The guy asked me the model number for my motherboard. He said that there were two different revisions and that the ram sent to me was for the other revision. They overnighted the correct ram and paid for return shipping. I understand things like this happe
    • Why do people buy cheap ram?


      Because when we shop on the net we're trained to seek out the lowest price, at the expense of quality. And it's not exactly like the places selling dodgy RAM will label it as "Dodgy!".

    • Re:Why do people buy cheap ram? (Score:5, Informative)

      by vadim_t ( 324782 ) on Tuesday April 19, 2005 @09:17AM (#12280593) Homepage
      I just buy ECC RAM.

      Sure it's more expensive, but it's great. If the computer does something strange I know that I can check /proc/ram or /proc/mc/0, see the statistics and instantly find if the memory is seeing errors or not. Here I do see a corrected error or two sometimes, although very infrequently. But it's indeed very nice to know it's been corrected.

      However, even if it's ECC I still wouldn't like at all knowing that it's not been tested. ECC has limits to the corrections it can make, after all.
    • Cheap ram is great. Run it through memtest. If it passes, it's just as good as the expensive stuff. If it doesn't, then return it, because it's defective. The savings in my experience are more than $10, and all of my systems with "cheap" ram work great.

      In 2001? a friend of mine got a great deal at fry's on 1gb DIMMs. This is when memory was PRICY. A weekend of testing found the good ones, the rest went back.

      I've had very good luck with cheaper memory. The only time I've been a memory snob is when I bought
      • I agree, with the caveat that memory fails down the road at an unspecified time more frequently than any other solid-state component, and I only buy RAM with a lifetime warranty. Also, most of the cheapest RAM is made up of modules that are less than general purpose, and only work with some of the more accomodating chipsets.
      • The invisible hand will "correct" for this. If places like Frys, BestBuy, etc. buy this untested RAM and get a lot of returns, that costs them money and you can bet that next time around they won't buy from these manufacturers. When that happens enough times, these guys will either start testing RAM or go out of business.

        It's a pain for the consumer (to return bad RAM... I've had to do this often enough that I stopped buying RAM from Frys) but the problem will eventually be solved by "evolution" -- compani
    • Corsair also sells good stuff, and they have pretty good support (just in case you do get a bad stick) for a product that's generally treated like a commodity rather than a retail good.

    • Is it really worth saving the £3-£5 by getting cheap unbranded RAM? As the saying goes, you get what you paid for.

      Pardon the US prices
      Crucial CT6464Z40B 512meg pc3200 $60.00 shipped
      Lowest bid 512meg PC3200 $30.00 shipped
      Lowest bid 1024meg pc3200 $65 shipped

      What do you get with the brand name? Lifetime warranty, assurance of compatibility, known reliability. Good resale value, esp with odd chip types no longer made.

      What do you get with the lowest bid? Half the price, might carry a lifetime

  • So what else is new? (Score:5, Funny)

    by Mad Hamster ( 870092 ) on Tuesday April 19, 2005 @08:52AM (#12280386)
    I've heard of a company in the northwestern US which has gotten away with selling untested operating systems for years.

    - Did I make first post?

  • You get what you pay for. (Score:4, Insightful)

    by Anonymous Coward on Tuesday April 19, 2005 @08:52AM (#12280390)
    Quality really seems to be a thing of the past. Cheaper != Better.
    • This seems to especially be the case with floppy disks. Anyone know anyone who sells floppies where you'll actually get the data off more than a third of them? It all seems to be cheap junk, some of the floppies I buy won't even format out of the box.

  • Lotsa cheap ram! (Score:4, Informative)

    by imroy ( 755 ) <imroykun@gmail.com> on Tuesday April 19, 2005 @08:55AM (#12280417) Homepage Journal

    Solution:

    1. Compile a Linux kernel with the BadRAM [vanrein.org] patch.
    2. Run Memtest86+ [memtest.org] to get a list of bad areas.
    3. Profit!... erm, I mean a Linux system with lots of cheap ram!

  • Really? (Score:3, Interesting)

    by TardisX ( 15222 ) on Tuesday April 19, 2005 @08:56AM (#12280421)
    Duh. I once received a SIMM where one of the chips was mis-placed on the PCB - the last two legs were actually hanging off the end into space.

    Whenever there is competition there will be cost-cutting. The heavier the competition, the heavier the cost-cutting.
  • ...as it seems products are rushed to market without significant testing. Take the Treo 650. They "tested" the device, but later found out (after release) that people who used it in the real world couldn't use the new file system because it didn't store things the same way.

  • When enough sellers (Score:3, Insightful)

    by Dorsai65 ( 804760 ) <dkmerriman.gmail@com> on Tuesday April 19, 2005 @08:57AM (#12280430) Homepage Journal
    lose sales (and their reputations) because of this, the problem will die out.
    • When enough sellers lose sales (and their reputations) because of this, the problem will die out

      Actually that's a fallacy. There's a long established principle in economics that whenever the cost of discernment reaches a critical level the cheap crappy look-alike beats the high quality product. This becomes a run-away situation as the economies of scale kick in as well, making the price differnetial larger and the market flooded with more lousy product increasing the consumers cost of discenrment.

      Thi

  • Freemarket (Score:5, Insightful)

    by XorNand ( 517466 ) on Tuesday April 19, 2005 @08:58AM (#12280436)
    What's the solution here?
    Huh? The freemarket seems like an ideal solution to me. Why do we need an external solution that entails fining/regulating them? If a company makes inferior product, odds are that they'll lose a large percentage of their customers. They'll be forced to either change their practices or go out of business.

    • Re:Freemarket (Score:2, Interesting)

      by zerkon ( 838861 )
      because granny and grandpa don't know the difference between BrandX ram ($45) and Crucial ($55)

      which do you think they'll pick
      • You inference is kinda insulting to your everyday computer user. You don't need to be a computer guru to understand basic economics. Whether it's brake pads, kitchen knives or RAM, there's always reason why some brands are more expensive than others. The onus is on the buyer to educate themselves before making a purchase. There's only so much hand holding a company ought to do.

    • Re:Freemarket (Score:2, Interesting)

      by Anonymous Coward
      It's no solution if you're the one who bought the bad RAM in the first place! This kind of application of the so-called free market always entails people get conned out of their money in the first place...

      Free market does *not* mean you can pass-off your shit as gold until you are caught.

  • Nothing New. (Score:5, Insightful)

    by necrodeep ( 96704 ) * on Tuesday April 19, 2005 @08:58AM (#12280441)
    The old saying stands true here... You get what you pay for. If you are going to only pay pennies on the dollar for Memory. Well you should expect a high number of failures.

    If your system memory is mission critical, you probably are going to buy top-shelf rather than bargain-basement, aren't you?
  • Some PC manufacturers get away with selling PCs without MS Windows preinstalled.

    If the product is clearly marked as untested, it's my responsiblity to test it (and replace if it's faulty). I pay less, I take more work on myself. It's my choice. Or I risk using a faulty device, but that's my choice again.

    Of course this cuts into market share of the "quality brand manufacturers", and they aren't happy about that, but that's a perfectly honest competition.

  • Stupid business practice... or not? (Score:3, Insightful)

    by n54 ( 807502 ) on Tuesday April 19, 2005 @09:03AM (#12280475) Homepage Journal
    Unless these companies are specifically targeting anybody willing to do the testing themselves & go through any replacement hassle I don't see what they're aiming at... that nobody will notice?

    How many end users would be remotely interested in doing this stuff? And they say they're going to increase this practise and that others might do the same?

    I would imagine factory testing isn't just to check the chips themselves but also to check up on the manufacturing process itself, how low quality are they aiming at? If they're hell bent on producing worthless trinkets they might as well make glass beads.
  • UTT stands for Ultra-Turbo-Technology. It's an advanced memory system created by everyone's favorite company, Rambus.

  • Analogy for the world (Score:5, Insightful)

    by jason.hall ( 640247 ) on Tuesday April 19, 2005 @09:04AM (#12280486)
    I believe we're slowly becoming a nation (world?) completely driven by prices at the expense of quality. I continuous hear things like "Why did he buy a Lexus for $50,000 when a new Hyundai is $15,000?" "This CD-R is $.10 each, this one is $.09. Why would anyone buy the $0.10 one?" People don't always get there's more to a product's specs than the price.
    • That is very true , its also works the other way , i know alot of people who belive Just because something costs more that it will instantly mean its better , and don't bother to look for bargins(as in good/"aceptable for the task" quality for a good price .
      Its the same old problem of people being ill informed and stuborn.

    • Yep (Score:2, Interesting)

      by jrushton ( 806560 )
      But then I pay more for organic food, I build PCs with good quality parts and pay more for an elegant case rather than one that looks like a chav car.... and would prefer a government that plans for the long term.

      Ooooh look at me I'm in the minority.

    • Re:Analogy for the world (Score:5, Insightful)

      by drinkypoo ( 153816 ) <drink@hyperlogos.org> on Tuesday April 19, 2005 @09:45AM (#12280885) Homepage Journal
      The hyundai might be more reliable, actually. Lexus/Toyota has done some really stupid things. A car that costs $45,000 should not have ball joints that must be replaced with the entire upper suspension arm... where did all that money go, anyway? We're mostly slaves to advertising. Granted many people are free from that kind of bullshit, but most humans are susceptible.
    • That's why I don't buy cooking gear at any of the *mart/Target stores anymore. My experience is that they sell cheap crap that's no fun to work with and falls apart within a few months. The little kitchen store where I do most of my shopping now seems to be doing a thriving business and sells good quality stuff. So what if I spent $75 on a chef's knife? If I take care of it, it'll last longer than I will. Ditto on the pots and pans I've picked up there. The blender I picked up there would put a jet engine t

  • Simple Solution (Score:5, Insightful)

    by mrRay720 ( 874710 ) on Tuesday April 19, 2005 @09:05AM (#12280491)
    Really, it's easy. If they're selling untested, unmarked parts and this is a problem, just don't buy untested, unmarked parts! Let the market sort itself out. If the market decides that the cost saving here isn't worth it, the demand for slightly cheaper untested parts will surely dry up, and the manufacturers will catch on and stop trying to sell them.

    If there are enough people out there though who DO want the cost saving brought on by buying untested crap - let them! Nobody says you have to buy cheap crap if it's on the shelf. You get what you pay for. You want good quality - pay good money. You want bad quality - pay peanuts.

    Basic Economics, really. And it's not as if the likes of Crucial, Corsair, Kingston etc. are doing it.

    Why is this even an issue? I think it's commonly accepted wisdom EVERYWHERE that going for the lowest bidder will give you cheap rubbish. Computer components are no different.
  • What's the big deal? A lot of products on the market are sold as untested. Out of a batch of 1000 products, only a small number might actually be tested. I know it's different with electronics, but I wouldn't have any problem buying something that I am fully aware is untested if it's cheaper.

    -Jesse

  • Not necesarily a problem (Score:4, Insightful)

    by Anonymous Coward on Tuesday April 19, 2005 @09:06AM (#12280504)
    This isn't necessarily a problem. If the yields on the DRAM chips are high enough, then it can make sense to NOT test the individual chips and instead wait to do the testing at the module (DIMM) level. If the chip yields are perhaps 95%, then the chance that a DIMM will be good is 0.95^8, or 66%. That may be high enough to make it worth while to avoid the cost of the chip testing.

    Now, if the chips are not tested AND the DIMMs are not tested, well that's another story...

  • tests shmests (Score:2, Interesting)

    by Anonymous Coward
    seems like untested ram would be the perfect compliment to the worthless hd that comes in dells.

  • We.. (Score:3, Interesting)

    by Turn-X Alphonse ( 789240 ) on Tuesday April 19, 2005 @09:17AM (#12280589) Journal
    We live in an era where somepeople consider £30 ($60) CD drives "disposable". The least of our worries is testing something. Because hey to these people whats the difference between 6 months and 12 years? After all it's just "throw away".

  • Big Deal... (Score:5, Interesting)

    by jeremy_faller ( 877248 ) on Tuesday April 19, 2005 @09:30AM (#12280701)
    The article states that the manufactorers are finding it cheaper to have customers test the RAM, and return it. This has been the case for a number of products for a number of years. Anyone remember Motorola pagers (the POCSAG and FLEX ones)? They sold millions without testing them. It was cheaper for MOT to sell them without testing them, and just accept a number of returns.
  • If the memory is truly untested and it is being sold as being untested: fine. What I would be afraid of with these parts is that they have been tested and that they have failed some tests. The failures may not have been total, just enough to not cause them to not be sold as prime parts. Even though I would't mind saving money by buying factory "seconds" underwear, somehow I don't think it worthwhile to save money buying factory "seconds" memory.

  • The solution is to test it yourself (Score:5, Informative)

    by gotan ( 60103 ) on Tuesday April 19, 2005 @09:44AM (#12280878) Homepage
    Whenever we buy new RAM, mostly as part of new PCs, we run Memtest86 [memtest86.com]. It's easy to do, it takes a while so do it overnight. There's so much that can go wrong with RAM, even with "good" RAM: it might not work together with the board, the SPD-timings might be off, whatever. Every once in a while we find some RAM that doesn't work for us and return it to the shop. We never had any problems at all to get it exchanged.

    For hardware-sellers it's probably more expensive if they have to factor in a certain return-rate (and the overhead for that) so they will look to it that the RAM they buy is ok. That way market forces will work for the benefits of all of us: untested RAM will, in the end, be more expensive than tested RAM. It's much easier and cheaper to do RAMtesting factoryside than having it returned by millions of customers.

    Of course that doesn't work if you buy your PC in a supermarket, but even for cheap PCs it's better to configure them yourself than buying crap. That way you can specify exactly where to save money and if anything breaks you get it fixed much quicker.
  • Not really. Untested does not mean conterfeit. It just means less quality control. If your facility makes 1000 units an hour, and it takes 30 minutes to test each unit, this means you will need 500 test fixtures and additional labor. If when tested, only 2 units out of 1000 fail, you have to ask yourself... is it really worth it to test? Sure, those 2 bad units are going to piss off a couple customers, but 99.8% customer satisfaction is completely acceptable in any market. Given the time required to t
  • I wonder how long until some enterprising RAM vendor partners with Microsoft and markets "Beta" RAM. Stable enough for your enterprise's production environment, yet rushed to market so fast that they couldn't waste time with such petty concerns as "testing". Wheee!

  • Scary Stuff! (Score:2, Insightful)

    by blueZhift ( 652272 )
    Call me a geek, but this is the scariest thing I've read all day! Sooner or later these cheap untested and mismarked components are going to make it into crucial safety or health systems and a lot of people are going to die who didn't have to.

    This is just another example of sacrificing quality and ultimately safety in the name of a few more dollars (or a lot more dollars if you're really dirty and unethical). Over the last few years, I've been paying more for strong brands I can trust, but with so much cou
  • How is this a Managers Journal topic? This is an anandtech/toms-hardware topic. My manager has no idea what type of ram is in our machines. Heck *I* have no idea what type of ram is in our machines. Its whatever Dell/IBM/Sun provided.

  • The End User can't properly test DRAM (Score:4, Interesting)

    by Detritus ( 11846 ) on Tuesday April 19, 2005 @10:44AM (#12281518) Homepage
    The end user does not have the equipment, or the device data, to properly test DRAM. You can run memtest86 and find the gross problems, but it will not find the more subtle problems, like sensitivity to timing, temperature and supply voltages.

    The greedy assholes running the PC industry should be shot. They are the ones that said that end-users didn't need parity memory anymore because RAM quality was so good. They say end-users don't need ECC. All along, they've been more concerned about their profit margins than the reliability of their products. They aren't the ones who get stuck with a flakey computer that crashes every day, or silently corrupts the user's data, with no indication of the true cause of the problem. They just pocket the money and pass the costs to the end-user. If untested DRAM floods the market, the problem will just get worse.

    The cost of ECC memory is trivial in comparison to the time and cost involved in dealing with the consequences of flakey memory.

  • Memtest is not a memory tester (Score:3, Informative)

    by panurge ( 573432 ) on Tuesday April 19, 2005 @10:46AM (#12281541)
    An in-board memory test is not the same as a proper memory test using a dedicated test set. The in-board approach will not be able to reproduce the range of voltage levels, speeds and timings of a test set. It may provide the equivalent of an infant-mortality test by exercising the DRAM through the descending leg of its bathtub curve, but it cannot tell you what the allowance for degradation with time and temperature is at your chosen settings.

    Of course, if this is for your games machine or something you upgrade every few months anyway, doesn't matter. But if you think that memory might stick around for a while and get used in a business critical application...well, I wouldn't, that's all.

    And yes, I do buy Crucial memory. Given my dislike of rebuilding things late into the night or being stuck without working hardware, it is extremely cheap insurance.

  • The solution, obviously... (Score:3, Informative)

    by Wakko Warner ( 324 ) * on Tuesday April 19, 2005 @10:57AM (#12281697) Homepage Journal
    ...is to not buy cheap-ass no-name RAM. Spend the extra 30 bucks and get some damn Crucial or Mushkin, ffs.

    - A.P.

  • Vendor Identification Through Data Scramble? (Score:3, Informative)

    by Inode Jones ( 1598 ) on Tuesday April 19, 2005 @12:40PM (#12282857) Homepage
    Are there any products on the market that can identify a DRAM vendor by using the data scramble patterns? Such a tool might be useful to flush out crappy DRAM.

    DRAM is just a bunch of capacitors on a chip. When the chip is powered down for a while, the voltage on the caps leaks towards ground. When the DRAM is powered up, all caps are at ground. Discharge to this state can be accelerated by exposing the die to light.

    Here's where it gets interesting: just because a cap is at ground on the die does not mean that you will read a zero out of the chip. With modern folded bitline architectures, half of the cells will read out as zero, and the other half as one. The pattern of 1/0 forms a definite pattern, called the "data scramble" which is a function of the chip architecture, and which will differ from vendor to vendor. Provided that few cells have been overwritten by the PC bootup, you can recover the scramble pattern and possibly identify the vendor.

    Remember your old Commodore 64? Power it up, cold, and POKE 53265,59. That will slam the video chip into graphics mode. See the pattern? It's not random. That's the data scramble.

    Two DRAM chips having different data scrambles are definitely not the same design. The converse is not true: two DRAM chips having identical data scrambles might be made by the same vendor, but there is a slight chance that two different vendors just happened on the same pattern. I don't know how much variation there is in scramble patterns, but this might be a useful way to trace chips to vendors.

    The more technical explanation for scramble patterns: the sense amplifiers in a DRAM chip are essentially differential. The inputs to the sense amp are two bitlines. Each bitline is connected to a different physical column in the memory array. Between cycles, the bitlines are pre-charged to VDD/2. When a row of DRAM is read, one bitline is connected to the cell capacitor and receives an offset charge while the other bitline is held at the reference. The sense amp then "pulls apart" the bitlines, driving the higher one to VDD and the lower one to ground. Depending on which bitline a zero-charged capacitor is connected to, the sense amp can swing one way or the other. The exact connection depends highly on the cell geometry and fabrication process.

    Past the sense amp, more fun happens. DRAMs are so dense that the signal from the sense amp requires one or two more levels of amplification before being suitable to drive to pins. To diminish crosstalk effects, the data buses are "twisted" like twisted-pair, which creates further address-dependent inversions in the pattern.

    The combination of cell geometry and data bus twist create a vendor-unique pattern. It's unlikely that two vendors with two different designs will happen on the same scramble pattern.

Slashdot Top Deals

"No matter where you go, there you are..." -- Buckaroo Banzai

Close