Page 23 of 23 FirstFirst ... 13 21 22 23
Results 221 to 229 of 229
  1. #221
    New ROGer Array
    Join Date
    Feb 2016
    Reputation
    10
    Posts
    3

    Help Me

    I have a excellent question for you. I have a Asus Rampage V Extreme which might run 64gb of DDR4 at 3200mhz. If I buy DDR4 3200 (PC4 25600) with a Timing of 16-18-18-36 a Cas Latency of 16 a Voltage of 1.35V. Lets say for example it won't run those clocks and I have to back it down to say 2800mhz. I was wandering if it would perform as well as DDR4 2800 (PC4 22400) with a Timing of 14-16-16-3 a Cas Latency of 14 and a Voltage of 1.35V? Would I be better to go for 2800 or 3200 and if I get the 3200 but it won't run at 3200 but only 2800mhz will it perform the same?

  2. #222
    ROG Guru: Orange Belt Array
    Join Date
    Aug 2015
    Reputation
    35
    Posts
    442

    Quote Originally Posted by Beastboy1989 View Post
    I have a excellent question for you. I have a Asus Rampage V Extreme which might run 64gb of DDR4 at 3200mhz. If I buy DDR4 3200 (PC4 25600) with a Timing of 16-18-18-36 a Cas Latency of 16 a Voltage of 1.35V. Lets say for example it won't run those clocks and I have to back it down to say 2800mhz. I was wandering if it would perform as well as DDR4 2800 (PC4 22400) with a Timing of 14-16-16-3 a Cas Latency of 14 and a Voltage of 1.35V? Would I be better to go for 2800 or 3200 and if I get the 3200 but it won't run at 3200 but only 2800mhz will it perform the same?
    Everything over 2133 is an "overclock" so nothing is guaranteed, BUT....

    Generally speaking any given memory binned to function at 3200 cas X will run at a lower speed AND lower cas Y. Whether the system actually performs the same or better at the lower speed, but better (also lower) CAS depends on:
    1. how much lower CAS?
    2. how much lower speed?
    3. Your application <- this ranges from marginal to HUGE. Some things simply don't get much better or worse no matter what you do with memory speeds beyond 2666.

    So, for example, I have 3000CAS14 (14-14-14-34-2T) 128G. If I run that same kit at 2800 12-12-12-32-1T (which it does), I get a 15% IMPROVEMENT in the same system with all else being equal on one particular application that involves large, randomly accessed memory image application using all 16 threads.

    I see far less improvement, but nothing slows down, in other applications. So, there is a case where even running slower, I can get as much if not more performance because of timing changes.

    Personally, provided your budget will allow and you are not paying a huge premium, I tend toward the highest clock and lowest CAS I can reasonably get and then I play around until I find the sweet spot that is:
    a. stable
    b. the fastest.

    Hope that helps, sorry there is no black or white answer. It depends on your chip(s) and application.

  3. #223
    ROG Guru: Orange Belt Array
    Join Date
    Aug 2015
    Reputation
    35
    Posts
    442

    Also, to add the to the specific to the application nature of this - realbench image editing benchmark for example shows improvement with DDR clock rate even at higher CAS. If I took 64G of that 128G kit out, I could run that same kit at 3200 CAS14 2T. I got a higher score with that, than I did with 2800 CAS14-1T, but it was a couple of percent.

  4. #224
    New ROGer Array
    Join Date
    Feb 2016
    Reputation
    10
    Posts
    3

    Cekim

    So to make a long story short get the 3200mhz and go for it as if it works I get great performance. However if it does not I could still possibly get better performance than the 2800mhz kit. Here are the links to the 2 kits I'm currently looking at the first being http://www.newegg.com/Product/Produc...ignorebbr=1and the second being http://www.newegg.com/Product/Produc...010&ignorebbr=. I'm all ears for any advise.

  5. #225
    ROG Guru: Green Belt Array drop4205 PC Specs
    drop4205 PC Specs
    Laptop (Model)What are laptops
    MotherboardMaximus XI Formula
    ProcessorI9-9900K
    Memory (part number)Gskill Trident Z F4-3200C14-32GYZR
    Graphics Card #1Nvidia Reference RTX2080TI
    Graphics Card #2ha I wish
    Graphics Card #3ha I wish
    Graphics Card #4ha I wish
    Sound Cardha I wish
    MonitorAsus VN247H-P 23.6"
    Storage #1Samsung970 EVO Plus 1T
    CPU CoolerHeatkiller
    CasePhantex Evolove X
    Power SupplySessonic Prime Titanium 850
    Keyboard ROG Strix Flare
    Mouse ROG Gladius II
    Headset Astro A40
    Mouse Pad Corsair
    OS Windows 10 Home 64 Bit
    Accessory #1 Ek GPU Block
    Accessory #2 EK SE 360 Rad, EK PE 360 Rad
    Accessory #3 EK D5 Vario 140 Pump/Res
    drop4205's Avatar
    Join Date
    Mar 2013
    Reputation
    47
    Posts
    517

    Gskill ripjaw 5 running at 3200 no issue and able to get to cas 13 1t. I always see or read issues on ram and a lot are to do with corsair ram. Not sure why { maybe someone can chime in on this} but i have run gskill since my first build 5 years ago.
    Maximus XI Formula, I9-9900k, Phantex Evolove X, Seasonic Titanium 850W, Custom loop PE360+SE360 Rad, G.Skill Trident Z F4-3200C14 32g, Nvidia Reference RTX 2080 TI, Samsung 970 EVO Plus 1Tb, ROG Strix Flare keyboard, ROG Gladius II

  6. #226
    New ROGer Array
    Join Date
    Feb 2016
    Reputation
    10
    Posts
    3

    Drop4205

    I suppose it would have helped If I had mentioned in my last post that from what I have read getting this speed from 16gb and 32gb setups is easier than 64gb. I mean no offense by this but this is ultimately the reason I need help as I know there have been issues with higher memory kits and speed.

  7. #227
    ROG Guru: Orange Belt Array
    Join Date
    Aug 2015
    Reputation
    35
    Posts
    442

    Quote Originally Posted by Beastboy1989 View Post
    I suppose it would have helped If I had mentioned in my last post that from what I have read getting this speed from 16gb and 32gb setups is easier than 64gb. I mean no offense by this but this is ultimately the reason I need help as I know there have been issues with higher memory kits and speed.
    Yes, it gets increasingly difficult to get any given speed with more memory and more dimms. Hence why I can take 4 sticks of my 8 stick 3000 CAS 14 2T 128G kit and run it at 3200 CAS 14 1T, but all 8 dimms even struggles to reach 3000 CAS 14 on an X99Pro. It runs fine on an RVE at that speed.

    and yes Beastyboy, I have more than a couple x99 systems and corresponding memory kits and all have no issue running at lower than marked speed, but as good or better CAS. 2800 seems to be a bit of a wall for the X99Pro in that 2800 or less is easy - turn key. 3000 or more even if so marked requires tuning of SA, cache voltage, phase etc...
    Last edited by cekim; 05-29-2016 at 12:56 AM.

  8. #228
    ROG Enthusiast Array NIK111 PC Specs
    NIK111 PC Specs
    MotherboardZ270 MAXIMUS IX APEX
    ProcessorI7/7700K
    Memory (part number)TridentZ RGB F4-3600C16D-16GTZR
    Graphics Card #1Asus GeForce Rtx 2070 OC Edition
    MonitorAOC U3277PWQU--31.5 inch 4K screen
    Storage #1Samsung SSD 960 PRO 512GB
    Storage #2Force MP500 (PCI-E 3.0 x4
    CPU CoolerSwiftech H220
    CaseCooler Master Case 5 Pro
    Power SupplyCorsair 860 volt
    Keyboard Mad Catz Strike 5
    Mouse Mad Catz Rat ProX
    OS Windows 10 Pro x64
    NIK111's Avatar
    Join Date
    Nov 2013
    Reputation
    10
    Posts
    74

    GSkill Ripjaws V F4-3000C15D-16GVRB 2T to 1T

    I have enabled xmp on my RipjawsV 15-16-16-35 2T with 1.350 volts. So far I have dropped the timings to 14 15 15 35 2t at 1.36v and its stable.Just wondering what I need for voltages to try to get it to work with the stock xmp timings from 2T to 1T..Any suggestions appreciated...

  9. #229

Page 23 of 23 FirstFirst ... 13 21 22 23

Tags for this Thread

Posting Permissions

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