Please report all spam threads, posts and suspicious members. We receive spam notifications and will take immediate action!
Page 3 of 4 FirstFirst 1234 LastLast
Results 21 to 30 of 33

Thread: Gigabyte GA-X48-DQ6 not running G.Skill PC2 8500 at rated timings




  1. #21
    Join Date
    May 2008
    Posts
    15

    Default Re: Gigabyte GA-X48-DQ6 not running G.Skill PC2 8500 at rated timings

    Thanks again for your help. I was only running the OCZ at 2.0V - the G.Skills were run at 2.1V. I was trying to reduce margins with the OCZ to see if I could get that same Memtest+ error, which would indicate a MB issue. Since I did not get any error from the OCZ when pushing them, I feel more confident that it is a bad pair of G.Skills.

    I will leave that fan though, as it will give the NB, RAM and maybe the GPU heatsink improved airflow. I will post an update when I get the replacement G.Skills.

    Thanks!

    Ira

  2. #22
    Lsdmeasap's Avatar
    Lsdmeasap is offline GIGABYTE Guru
    Join Date
    Jan 2008
    Location
    PCB Island
    Posts
    25,940

    Default Re: Gigabyte GA-X48-DQ6 not running G.Skill PC2 8500 at rated timings

    Ya, you should have no issues with a working pair of G.Skills. Both would likely run at 2.0 though ya never know without testing though.

    A extra fan can never hurt that is for sure, been thinking about adding one in mine

    And ya since you are not getting the error with the OCZ ones I would say that maybe you just got some slightly messed up G.Skills that is unless at the same voltage with the OCZ's you get the same error bit... then I think it would be the board

  3. #23
    Join Date
    May 2008
    Posts
    15

    Default Re: Gigabyte GA-X48-DQ6 not running G.Skill PC2 8500 at rated timings

    Quote Originally Posted by Lsdmeasap View Post
    And ya since you are not getting the error with the OCZ ones I would say that maybe you just got some slightly messed up G.Skills that is unless at the same voltage with the OCZ's you get the same error bit... then I think it would be the board
    Are you suggesting that I may not have repeated a MB problem because I lowered the voltages when I tested the OCZ memory? I thought raising the MCH, FSB and Vddr voltages all increased margin, so I lowered them and tightened the timing in hopes that a MB issue would reveal itself with the OCZ. However, before I came this forum I did do some Memtest+ runs with the G.Skills at nominal voltages and I got the same error, so I would say that the OCZ and G.Skills have been run under very similar conditions and the G.Skills failed.

    I will RMA the G.Skills today and let you know what happens with the new ones.

    Quick question. I have been reading that a lower RAM speed can give better performance than a higher RAM speed. My plan was to mildly OC my Q9450 to 3.2 GHz with the G.Skill at 1066 at the rated 5-5-5-15. Would I be better off with the RAM at 800 and timing of 4-4-4-12, letting me use a FSB:RAM ratio of 1:1? I am running the OCZ memory that way right now. I was planning on doing some memory benchmarks using Sandra to check this out.

    Thanks again,

    Ira

  4. #24
    Lsdmeasap's Avatar
    Lsdmeasap is offline GIGABYTE Guru
    Join Date
    Jan 2008
    Location
    PCB Island
    Posts
    25,940

    Default Re: Gigabyte GA-X48-DQ6 not running G.Skill PC2 8500 at rated timings

    Yeah, only you could test that and find out. A 1:1 ratio used to be best, but has since been tested and proven wrong many times with newer intel boards

    Running them at a slower speed with tighter timings would possibly be better, but again you would only know with your own tests as all sticks bandwidth is different at different speeds/timings.

    Here is the most recent discussion of the 1:1 ratio myth being tested and proven to be no longer a reality I have seen >>>
    The 1:1 myth that just won't die. - EXTREME Overclocking Forums

  5. #25
    Join Date
    May 2008
    Posts
    15

    Default Re: Gigabyte GA-X48-DQ6 not running G.Skill PC2 8500 at rated timings

    Lsdmeasap,<?xml:namespace prefix = o ns = "urn:schemas-microsoft-com:office:office" /><o:p></o:p>
    <o:p> </o:p>
    I got my replacement G.Skills and installed them last night. I ran Memtest+ Test #6 for over 8 hours for 64 passes and 0 errors at the following settings:<o:p></o:p>

    Robust Graphics Booster: Auto
    CPU Clock Ratio: 8X
    FSB: 400 (CPU @ 3.20 GHz)
    PCIe Frequency: 100
    Memory Multiplier: 2.66D (DDR2 1066)
    Performance Enhance: Standard
    RAM Timing: 5-5-5-15<o:p></o:p>

    All Memory Subtiming: Auto
    DDR OV: +0.20 (2.0V)
    PCIe: Normal
    FSB: Normal
    MCH: Normal
    Loadline Calibration: Disabled

    Vcore: 1.250V (<?xml:namespace prefix = st1 ns = "urn:schemas-microsoft-com:office:smarttags" /><st1:place w:st="on">Normal</st1:place> CPU Vcore = 1.18750V)<o:p></o:p>
    <o:p> </o:p>
    It would appear that the original problem was in fact a bad pair of G.Skills. I noticed that the price for the G.Skills has dropped quite a bit, so a 2<SUP>nd</SUP> pair is likely. Do you think I will need to boost the memory or MCH voltages or loosen any timing settings when I add the 2<SUP>nd</SUP> pair, or is it very possible that the 8GB would run at my current settings?<o:p></o:p>
    <o:p> </o:p>
    I really appreciate the time and help you have given me.<o:p></o:p>
    <o:p> </o:p>
    Thanks,<o:p></o:p>
    <o:p> </o:p>
    Ira<o:p></o:p>
    <o:p> </o:p>
    <o:p> </o:p>
    <o:p> </o:p>
    EQUIPMENT:<o:p></o:p>
    Gigabyte GA-X48-DQ6<o:p></o:p>
    Intel Q9450 Core 2 Quad OEM<o:p></o:p>
    G.Skill 4GB (2x2GB) DDR2 1066 (PC 8500)<o:p></o:p>
    OCZ Vendetta 2 CPU Heatsink + OCZ Freeze<o:p></o:p>
    ECS N8800GT-512MX+<o:p></o:p>
    AirLink101 AWLH6080 300N Wireless PCI Adapter<o:p></o:p>
    Antec P182 Gunmetal Black Case <o:p></o:p>
    2 x Nocuta NF-P12 fans (1 on OCZ V2, 1 on rear of case, both at 1400 rpm)<o:p></o:p>
    Corsair HX620W<o:p></o:p>
    BYTECC PCIe SATA II + PATA Controller card<o:p></o:p>
    1 x 300 GB Seagate SATA<o:p></o:p>
    1 x 500 GB Seagate SATA<o:p></o:p>
    2 x 1 TB WD SATA<o:p></o:p>
    2 x 750 GB Seagate PATA (connected via BYTECC card)<o:p></o:p>
    Pioneer 20X SATA DVR-215DBK (SATA)<o:p></o:p>
    Plextor PX-760 (PATA)<o:p></o:p>
    Lite-On 52x CD-RW (PATA)

  6. #26
    Lsdmeasap's Avatar
    Lsdmeasap is offline GIGABYTE Guru
    Join Date
    Jan 2008
    Location
    PCB Island
    Posts
    25,940

    Default Re: Gigabyte GA-X48-DQ6 not running G.Skill PC2 8500 at rated timings

    Nice! I thought we had too long of a thread going on here for G.Skills. I kinda thought they may have been bad!

    Glad to see you got a working pair!!!!

    Yeah, they are getting decent prices, $89 now with free shipping >>>
    Newegg.com - G.SKILL 4GB (2 x 2GB) 240-Pin DDR2 SDRAM DDR2 1066 (PC2 8500) Dual Channel Kit Desktop Memory - Desktop Memory


    Ahh, you may need to add a bit on the MCH which I am surprised you have not done already, and possibly loosen a timing or two. I can say, you will likely need/have to (And should already for best performance) set the memory subtimings manually

    Overall, I am very glad to hear you are back in action, I was beginning to wonder what happened here

  7. #27
    Join Date
    May 2008
    Posts
    15

    Default Re: Gigabyte GA-X48-DQ6 not running G.Skill PC2 8500 at rated timings

    I too was leaning toward a bad pair of G.Skills, since the OCZ had no issue, but swapping memory sockets (not channels) and getting the same Err Bit result I believed was reason to not completely rule out the motherboard.<?xml:namespace prefix = o ns = "urn:schemas-microsoft-com:office:office" /><o:p></o:p>
    <o:p> </o:p>
    Before I came to this forum I thought I had been running stable at 1066 at Normal MCH and 5-7-7-20 timing – Memtest+ and Prime95 appeared to be stable. However, longer Memtest+ Test #6 runs later proved otherwise. Since all the errors were isolated to bad G.Skills, I now expect 100% stability at Normal MCH but thought I would do some testing at the minimum DDR voltage (2.0V).<o:p></o:p>
    <o:p> </o:p>
    If further testing (Prime95, Memtest+ Tests #2, #5 and #6) proves stable, should I still increase the DDR voltage and MCH to add a bit of margin?<o:p></o:p>
    <o:p> </o:p>
    Now that I have resolved the memory errors, I may take a closer look at the overclocking thread to see what memory subtimings (they are currently all on Auto) and cross-strappings I should experiment with.<o:p></o:p>
    <o:p> </o:p>
    This is a HTPC connected to a 60” Sony KDS-60A3000 TV. It is on 24/7 so I want to keep the power, heat and noise to a minimum. My goal was 100% stability with only a modest OC of 3.2 GHz, since running the Q9450 at stock seemed like such a great waste of potential.<o:p></o:p>
    <o:p> </o:p>
    Thanks,<o:p></o:p>
    <o:p> </o:p>
    Ira
    Last edited by Ira; 06-27-2008 at 04:15 AM.

  8. #28
    Lsdmeasap's Avatar
    Lsdmeasap is offline GIGABYTE Guru
    Join Date
    Jan 2008
    Location
    PCB Island
    Posts
    25,940

    Default Re: Gigabyte GA-X48-DQ6 not running G.Skill PC2 8500 at rated timings

    Normal MCH is likely not good, for two reasons. You need a bit more for 2x2, and possibly a bit more for 1066 as well.

    That way you dont have to run those loose Arse 577-20 timings and can run 555-15

    So ya, I would say add it in there, and when you get time do some reading and start lowering those timings....ONE AT A TIME

    Good luck, and again glad to see you have some working memory.

    I hate DOA ram, ya never know what is causing the errors

  9. #29
    Join Date
    May 2008
    Posts
    15

    Default Re: Gigabyte GA-X48-DQ6 not running G.Skill PC2 8500 at rated timings

    But I am running at 5-5-5-15. That is why I was asking if you thought there was still any reason to boost the voltages. These are my current settings and so far they appear stable (I will do some overnight runs of Prime95 and Memtest+):

    Robust Graphics Booster: Auto
    CPU Clock Ratio: 8X
    FSB: 400 (CPU @ 3.20 GHz)
    PCIe Frequency: 100
    Memory Multiplier: 2.66D (DDR2 1066)
    Performance Enhance: Standard
    RAM Timing: 5-5-5-15
    All Memory Subtiming: Auto
    DDR OV: +0.20 (2.0V)
    PCIe: Normal
    FSB: Normal
    MCH: Normal
    Loadline Calibration: Disabled
    Vcore: 1.250V (Normal CPU Vcore = 1.18750V)

    I understand that boosting the MCH and DDR voltage would add some margin, and one would definitely do this to help reach a desired OC especially at 1066 and 2x2 RAM, but I think I am already there. I don't know how much margin I currently have but assume if Memtest+ and Prime95 are stable, then I have sufficient margin.

    I am probably being too analytical here, but I hesitate to boost MCH voltage to increase margin when I may already have enough margin -- that may be unnecessarily adding to power/heat. Another way to ensure margin I am considering is just boosting the FSB a bit and leaving everything else the same. That way if Prime95 and Memtest+ show that I am still stable I can back down to a 400 FSB and know there is margin.

    I am an Electrical Engineer, if you didn’t already figure that out.

    Thanks!
    Last edited by Ira; 06-27-2008 at 05:27 AM.

  10. #30
    Lsdmeasap's Avatar
    Lsdmeasap is offline GIGABYTE Guru
    Join Date
    Jan 2008
    Location
    PCB Island
    Posts
    25,940

    Default Re: Gigabyte GA-X48-DQ6 not running G.Skill PC2 8500 at rated timings

    Ahh well two things, sorry I thought you was running 5-7-7-20 is why I suggested that and yeah, us overclockers are always shooting for that 24 Hours Orthos blend test to pass is why I suggested to add some mch as well. Orthos or Prime95 on the blend test can appear stable for a long time, I have even ran Orthos (Which is what I use instead of Prime95) and it failed at 19 hours so I tried again.

    Just really depends on what you feel is stable though I suppose. Some people only do 12 hours runs and leave it at that, it just a personal choice anyway. Stable is what a person uses a computer for and it be stable so whatever is stable is really stable.

    Plus NO ONE ever uses their PC in any real way at 100% load anyway for X Number of hours like we do stress tests at anyway so stableness it likely in reality reached before us (Overclockers) even say it is anyway

    Well Then I guess all you have to do when you have some free time is tweak those subtimings, as Auto is not going to give you the fastest/best settings.

Page 3 of 4 FirstFirst 1234 LastLast

Thread Information

Users Browsing this Thread

There are currently 1 users browsing this thread. (0 members and 1 guests)

Posting Permissions

  • You may not post new threads
  • You may not post replies
  • You may not post attachments
  • You may not edit your posts
  •