Stories
Slash Boxes
Comments

SoylentNews is people

posted by martyb on Sunday April 02 2017, @06:28AM   Printer-friendly
from the is-that-really-such-a-bright-idea? dept.

Corsair has released its first Vengeance RGB DRAM modules, which come with heat spreaders that feature LED lighting that can be controlled by Corsair Link software or GIGABYTE'S RGB Fusion app:

Corsair last week started to sell its first Vengeance RGB memory modules, equipped with full sets of RGB LEDs that can change their colors using OS software. Right now Corsair offers DDR4-2666 and DDR4-3000 modules, but eventually we expect the Vengeance RGB lineup will be expanded.

Just like other Vengeance memory modules, the Vengeance RGB DIMMs are based on Corsair's custom PCBs as well as preselected ICs. The modules come with aluminum heat spreaders featuring RGB LEDs that can change colors dynamically using the Corsair Link software, allowing users to synchronize colors of RGB lighting of their DIMMs and specific motherboard brands. Lighting of each module can be controlled separately as well. The lighting of the Vengeance RGB can also be controlled using GIGABYTE's RGB Fusion app and Corsair states that eventually other producers of LED-controlling software are expected to follow.


Original Submission

 
This discussion has been archived. No new comments can be posted.
Display Options Threshold/Breakthrough Mark All as Read Mark All as Unread
The Fine Print: The following comments are owned by whoever posted them. We are not responsible for them in any way.
  • (Score: 3, Interesting) by Anonymous Coward on Sunday April 02 2017, @08:31AM (7 children)

    by Anonymous Coward on Sunday April 02 2017, @08:31AM (#487862)

    The light could actually be useful to identify modules with ECC errors: green for OK, yellow for correctable, red for uncorrectable.

    Once you have gigabytes of RAM, non-ECC is a huge mistake. You will have errors. Without ECC, your data will be corrupted. This messes up your filesystem, your family photos, your taxes, your bitcoins, your password keeper, your very best porn, etc.

    Starting Score:    0  points
    Moderation   +3  
       Insightful=1, Interesting=2, Total=3
    Extra 'Interesting' Modifier   0  

    Total Score:   3  
  • (Score: 3, Interesting) by kaszz on Sunday April 02 2017, @10:00AM (4 children)

    by kaszz (4211) on Sunday April 02 2017, @10:00AM (#487873) Journal

    I recall the spontaneous bit errors due to alpha emissions is on the order of 1 bit flip per gigabyte of DRAM per hour. And of course it gets worse with higher density. So ECC might be a very good idea.

    • (Score: 0) by Anonymous Coward on Sunday April 02 2017, @10:13AM (3 children)

      by Anonymous Coward on Sunday April 02 2017, @10:13AM (#487877)

      I can't recall the last time I've seen a real complication from bit flips causing data corruption or a broken webpage. Maybe if I created a million unsaved text files?

      Is there a virtual ECC? Unsaved small files could be mirrored in RAM several times.

      • (Score: 1) by Scruffy Beard 2 on Sunday April 02 2017, @11:15AM (1 child)

        by Scruffy Beard 2 (6030) on Sunday April 02 2017, @11:15AM (#487898)

        Even ECC done in hardware has a performance penalty.

        You might as well buy an old Xeon server with all the extra CPU cycles you are going to burn trying it.

        • (Score: 0) by Anonymous Coward on Sunday April 02 2017, @07:17PM

          by Anonymous Coward on Sunday April 02 2017, @07:17PM (#487999)

          At work, the machines with ECC do catch errors. On the non-ECC machines, we've seen weirdness like corrupt *.h files causing compile fails. Sometimes the corruption is only in memory; the copy on disk is good.

          We've gotten rid of most of the non-ECC machines. New developers get 32 gigabytes of ECC memory as of last year; it might be 64 now. It goes with a pair of Xeon chips. I think it can be a total of 12 to 32 cores depending on the chips. (some HP 820 or HP 840 workstation I think... an 8xx at least)

      • (Score: 2) by turgid on Monday April 03 2017, @08:47PM

        by turgid (4318) Subscriber Badge on Monday April 03 2017, @08:47PM (#488362) Journal

        I've got a (very old now) PC at home, my main system, which has an ASUS M4A77D motherboard and an AMD Phenom II X6 1045T CPU and 4GB of ECC DDR2 RAM (from crucial.com). Very occasionally I get an error on the console saying that an ECC error has been detected and fixed. For a while I was getting several a day. Someone suggested I try re-seating the RAM, which I did and the problem went away. I didn't collect logs to analyse the frequency of the errors.

  • (Score: 1) by Scruffy Beard 2 on Sunday April 02 2017, @11:11AM (1 child)

    by Scruffy Beard 2 (6030) on Sunday April 02 2017, @11:11AM (#487897)

    Once you have gigabytes of RAM, non-ECC is a huge mistake.

    Tell that to Intel! (They only support ECC with their Xeon line-up).