Slashdot Mirror

← Back to Stories (view on slashdot.org)

Firms Get Away with Selling Untested DRAM

Posted by timothy on 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.)

5 of 344 comments (clear)

  1. If you're stuck with one of these... by Anonymous Coward · · Score: 5, Informative

    Here is the obligitory Memtest86 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 that allow you to avoid the bad bits if they are isolated enough.

    1. Re:If you're stuck with one of these... by Anonymous Coward · · Score: 5, Informative

      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/

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

    2. Re:If you're stuck with one of these... by gravygraphics · · Score: 5, Informative

      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.

  2. Re:unmarked and untested == pirated? by yknott · · Score: 5, Informative

    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.

  3. Re:Also try Prime95 by gkitty · · Score: 5, Informative
    I agree that memtest86 is useful but not sufficient and that prime95 is much more throrough. Memtest confirms that patterns that have been set hold their state briefly, which is a good test against gross failures (and I have seen these).

    But Prime95 confirms that no bit anywhere in nearly the complete memory space ever spuriously changes. I have seen plenty of memory that passes metest86 that fails prime95.

    Based on my experience, Corsair will replace memory that fails prime95. Mushkin will NOT (despite a "lifetime" warranty); they basically told me that memory can't be expected to be 100% perfect all the time and that prime95 was too strenuous; if it passes memtest86 there will be no replacement. My other modules (from Geil, Samsung, and a few old no-name sticks) have always been perfect. IMO it's unconscionable to sell untested ram given how hard it is to return.