What is the worst that can happen when installing incompatible RAM?

Discussion in 'hardware' started by BlackScarletLove, Apr 21, 2015.

  1. BlackScarletLove

    BlackScarletLove Registered Member

    Joined:
    Jul 12, 2007
    Posts:
    66
    I was harvesting parts from a computer, and I found a couple gigs of RAM. I could always use a couple more gigs, but I don't know if they are compatible -- and I do know that if they are incompatible, it won't work. But, exactly what do I risk by trying it out? Will the system just not boot, so I take it out and it's as if nothing ever happened? Or will attempting such an installation actually result in permanent damage to my system if it's not compatible?
     
  2. StillBorn

    StillBorn Registered Member

    Joined:
    Nov 19, 2014
    Posts:
    297
    You see pink elephants and dream of sheep (and far from the proper manner) at night. And dear lud, may the bed bugs not bite to the point where my hard drive becomes unrecognizable.
     
    Last edited: Apr 21, 2015
  3. whitedragon551

    whitedragon551 Registered Member

    Joined:
    Sep 30, 2008
    Posts:
    3,264
    Location:
    USA
    System wont boot.
     
  4. Bill_Bright

    Bill_Bright Registered Member

    Joined:
    Jun 29, 2007
    Posts:
    4,041
    Location:
    Nebraska, USA
    There are two kind of compatibility - physical and electrical. DDR2 and DDR3 are physically incompatible. The notches in the contacts are in different locations so they will not even fit in the slots - even though (without careful examination) they look like they will. You can force them in with excessive force, but you will then permanently damage the motherboard slots, and possibly the RAM too.

    So before even thinking about trying them, you need to make sure they are the same type of RAM.

    As for electrically compatible, you should research your motherboard and see what it supports. Then use the labels on the RAM to see what it is, and if compatible.

    Other than physical damage if the wrong type RAM, I agree the worst that can happen is the system just won't boot.
     
  5. hawki

    hawki Registered Member

    Joined:
    Dec 17, 2008
    Posts:
    6,065
    Location:
    DC Metro Area
    Amnesia
     
  6. Tyrizian

    Tyrizian Registered Member

    Joined:
    Apr 26, 2012
    Posts:
    2,839
    My advice would be, just don't do it.

    Research your system and then purchase the appropriate RAM that is needed, saves you a lot of headaches.
     
  7. zfactor

    zfactor Registered Member

    Joined:
    Mar 10, 2005
    Posts:
    6,102
    Location:
    on my zx10-r
    lol if its not compatible the system will not boot. now if you are talking about speed differences then thats a whole 'nother story
     
  8. mirimir

    mirimir Registered Member

    Joined:
    Oct 1, 2011
    Posts:
    9,252
    Even if they fit and the machine boots, there may be speed issues, as zfactor notes. The speed is part of the RAM's SKU. If the speeds are different, at worst (if it boots) all RAM will run at the slowest speed. Maybe some of the RAM will run faster, depending on design. But hey, just google it :)
     
  9. Bill_Bright

    Bill_Bright Registered Member

    Joined:
    Jun 29, 2007
    Posts:
    4,041
    Location:
    Nebraska, USA
    Depending on the design of the motherboard, yeah, that's true. But more RAM still trumps slower RAM. That is, I would rather have more slow RAM than less fast RAM.
     
  10. BlackScarletLove

    BlackScarletLove Registered Member

    Joined:
    Jul 12, 2007
    Posts:
    66
    o drat. it is ddr2 and i have ddr3. o well. thanks :)
     
  11. Bill_Bright

    Bill_Bright Registered Member

    Joined:
    Jun 29, 2007
    Posts:
    4,041
    Location:
    Nebraska, USA
    Well, most of us who have been actively involved in doing computer tech support for others professionally, or as amateurs/enthusiasts for our own computers for any length of time have a box or two (or a shelf, or a storeroom full of shelves! :() full of legacy computer components - whether it be legacy RAM, EIDE hard and optical drives, AGP graphics cards, CPUs, monitors, motherboards, or whatever. All perfectly good stuff that have outlived their usefulness. In other words, join the club.
     
  12. crawfish

    crawfish Registered Member

    Joined:
    Jul 2, 2014
    Posts:
    25
    The worst that can happen is that the system will seem to operate normally, pass Memtest86+, yet fail when under load such as when running Prime95, or occasionally spontaneously reboot for no apparent reason. This was a common problem for the Intel P55 chipset, which required some very specific RAM to work properly. See for example this page, which is where I ended up after encountering the problem myself:

    https://alternativerecursion.wordpress.com/2009/10/09/intel-p55-lga1156-chipset-and-ddr3-memory/
     
  13. Bill_Bright

    Bill_Bright Registered Member

    Joined:
    Jun 29, 2007
    Posts:
    4,041
    Location:
    Nebraska, USA
    How is that the worst thing? At least the system partially worked.

    It is not uncommon for bad RAM to pass software based RAM testers like Memtest86+, MemTest86, or Windows own integrated tester. No software based tester is conclusive. If they report the RAM is bad, it is bad. But they cannot always determine if RAM is bad.

    Note to conclusively test your RAM, you need to use sophisticated and very expensive test equipment, like this $4,000 Eureka Express DDR3 DIMM Memory Module Tester.

    I no longer use or recommend Memtest86+ as it is no longer current and has not been updated since Aug 2013 (before Windows 8/8.1 :(). MemTest86 (now from PassMark Software :)), is under continual development with the most recent update as recently as Feb 2015.
     
  14. CrusherW9

    CrusherW9 Registered Member

    Joined:
    Dec 27, 2012
    Posts:
    517
    Location:
    United States
    Can confirm. I bought a new kit a few months ago and I started getting blue screens. I checked the ram with numerous overclocking stress tests and Memtest86 and they all reported the ram was fine. Popped in my old kit and the BSOD's disappeared.
     
  15. crawfish

    crawfish Registered Member

    Joined:
    Jul 2, 2014
    Posts:
    25
    I value the integrity of my data, and I would much rather a system fail immediately than go on "partially working" with no overt errors for long lengths of time. I replied because the consensus "System won't boot" situation represents one of the better things, not the worst.

    Which is why I mentioned Prime95. The AIDA64 stress tester also reported errors. They too are not going to "conclusively" prove a system is sound. However, I now consider it necessary to run them after adding or changing RAM, even in a system that previously passed those tests, even after Memtest86+ or the like has found no problems with the RAM change.

    Unfortunately, the PassMark software requires UEFI, else it will boot into like a 10 year old version of MemTest. Not sure what the relevance of the publication date of Windows 8 is. I looked at PassMark recently because it's been updated with rowhammer tests, which Memtest86+ lacks. Aside from that, Memtest86+ is still relevant and useful. I built the machine I'm using now in May 2013, and it recognizes all the hardware fine.
     
  16. Bill_Bright

    Bill_Bright Registered Member

    Joined:
    Jun 29, 2007
    Posts:
    4,041
    Location:
    Nebraska, USA
    I agree and for me, my data is worth much more than all my hardware. So I keep current backups.

    I still like Memtest86+ too. I was just noting that development appears to have ceased. And while MemTest86 "supports" UEFI, it certainly does not require it and definitely does not revert back to a 10 year old version. :(
     
  17. crawfish

    crawfish Registered Member

    Joined:
    Jul 2, 2014
    Posts:
    25
    I would certainly hope so.

    http://www.memtest86.com/download.htm
    IMPORTANT: MemTest86 V5 (and later) images supports dual booting (UEFI/BIOS). On machines that don't support UEFI, the older V4 BIOS release of MemTest86 will be automatically booted.

    It might as well be 10 y/o, because it's no better than Memtest86+ for my purposes and in particular doesn't support the new rowhammer tests they implemented in v 6. Looking at the changelogs, their v4 has just had a few bugfixes since Memtest86+ 5.01 came out in Sep 2013, at which point, my understanding is they were about equal in capabilities. But yeah, sorry for the hyperbole on the timeframe. When I booted it, I was so disappointed to see v4 that I just wrote it off as outdated and irrelevant.
     
  18. Brian K

    Brian K Imaging Specialist

    Joined:
    Jan 28, 2005
    Posts:
    12,113
    Location:
    NSW, Australia
    I just looked at Memtest86+ 5.01. Booted from a UFD.

    Computer has UEFI/BIOS. Memtest boots in MBR mode (so does my Win8 as I prefer a MBR boot) and the caption at the top left says...

    Memtest86+ 5.01
     
  19. Bill_Bright

    Bill_Bright Registered Member

    Joined:
    Jun 29, 2007
    Posts:
    4,041
    Location:
    Nebraska, USA
    :( You are just biased. The latest V4 version is from May 2014 so hardly 10 years old. And the point is, it is still under development while 86+ does not appear to be.

    I am NOT and never have said 86+ is bad or should be avoided. The two programs are closely related, after all - coming out of the same familial roots. Heck, even the integrated Windows memory test works.
     
  20. crawfish

    crawfish Registered Member

    Joined:
    Jul 2, 2014
    Posts:
    25
    What crawfish actually said:

    It might help you respond more constructively if you were not to clip posts after the first few words and instead respond to the entirety of what was written, which included an apology for exaggerating the timeframe, which I had done when I wrote earlier, "...boot into like a 10 year old version of MemTest" (bold added) along with the (IMO) reasonable explanation why I was exasperated after downloading, setting it up on USB stick, and booting my computer into it, only to find it was the old v4. That's not "bias." That's frustration at having wasted my time. I would've happily used MemTest86 if v6, which implements the feature I was looking for, were not UEFI-only.

    Do you have any evidence that 86 v4 is "still under development", because the changelog and feature comparison page indicate it has been left way behind v6. Again, I don't know that 86 v4 is any more capable than Memtest86+, and a handful of bugfix releases, the last two in Nov 2013 and May 2014, don't constitute "under development" in any meaningful way. I'm not having problems with Memtest86+ so don't care that the other program has had a few bugfixes. What matters to me is that 86 v4 doesn't implement the feature I was looking for, which is v6 and UEFI-only.

    OK. I haven't suggested you said that "86+ is bad or should be avoided," so I don't know why you need to call it out, but OK.
     
  21. crawfish

    crawfish Registered Member

    Joined:
    Jul 2, 2014
    Posts:
    25
    Yes, that's right. These are two different programs we're talking about.
     
  22. Bill_Bright

    Bill_Bright Registered Member

    Joined:
    Jun 29, 2007
    Posts:
    4,041
    Location:
    Nebraska, USA
    I didn't "call it out". My intent in that first post was merely to inform readers, first, that no software based tester is conclusive!

    Secondary to that, my intent was to inform users that MemTest86+ is no longer under development and has not been updated since Aug 2013 while MemTest86 is still under continual development with recent updates. And so MemTest86 is the one "I" recommend and use. I did not "call out" MemTest86+, say it was bad or that it should be avoided! In fact I said later I still like MemTest86.

    So I am afraid it was you who had the cow and went on an OT multi-post rampage as if I had insulted your religion or favorite color or choice of beer.

    So be it. Now the facts are BlackScarletLove has determined it is DDR2 which, of course, is incompatible with DDR3. So whether it be "what's the worst that can happen", or what's the better RAM tester, it's all moot. So have a good weekend. I'm outa here.
     
  23. crawfish

    crawfish Registered Member

    Joined:
    Jul 2, 2014
    Posts:
    25
    And there you go again clipping and answering non-responsively, right after I asked you to stop doing that. As for informing readers that "no software based test is conclusive", you did that in reply to my first post here, where I described exactly that situation.

    Repeating yourself doesn't make your wrong statement any more true. AFAICT, MemTest86 is not appreciably more current than Memtest86+ for non-UEFI systems, and development of v4 has stopped compared to the UEFI-only v6, which is abundantly clear from Passmark's changelog and feature comparison pages.

    You are too much. Not only do you reply to things that haven't been suggested, as with your "I am NOT and never have said 86+ is bad or should be avoided," you make equally bizarre characterizations about what has been said. Seek help.
     
  24. Bill_Bright

    Bill_Bright Registered Member

    Joined:
    Jun 29, 2007
    Posts:
    4,041
    Location:
    Nebraska, USA
    Whatever! I didn't make any false statements about 86+. It was not non-response, it was you ignoring the facts.

    I said development on 86+ has ceased. That is true.
    I said development on 86 is recent and ongoing. That is true.

    You said 86 requires UEFI. That is false.
    You said 86 reverts to something like 10 years old. That is also false.

    So bizarre characterizations? Yeah right.

    None of this OT waste of time would have happened if you didn't have a cow over true facts.

    I am done wasting my time and apologize to BlackScarletLove for my part in this mess.
     
  25. crawfish

    crawfish Registered Member

    Joined:
    Jul 2, 2014
    Posts:
    25
    The mess is entirely yours, and it is due to ignoring what's been said, with the effect being carrying on a conversation mainly with yourself. You've been clipping the first few words and then writing a reply of non sequiturs, and now you've graduated to not quoting at all. I'm not going to try to correct your repeated misstatements yet again, because the result will just be more nonsense from you. It's a waste of time.
     
  1. This site uses cookies to help personalise content, tailor your experience and to keep you logged in if you register.
    By continuing to use this site, you are consenting to our use of cookies.