The ek vram blocks will work just fine.Anyone have recommendations on universal vrm blocks? Looking for the asrock taichi.
Ek doesn't have mosfet blocks for this or any x399 boards currently, only x99, which I'll have to verify hole placement before I'd try that. I didn't see vram blocks that could be adapted well to it on their site.The ek vram blocks will work just fine.
HPET on or off? With it off, I get 3382 @3.9GHz@1.225V.Got this stable so far at 4.1 @ 1.375.
Off increases bench scores, but makes OC utilities useless, and some versions of Windows suffer from the real time clock bug with it turned off. But, overall, it will increase your scores.On. Didn't know of this setting. Should I turn it off?
Cmd with adminOn. Didn't know of this setting. Should I turn it off?
I'm having issues with my RAM too and its G-skill TridentZ (Samsung b-die 3600@16-16-16-36) :Just got everything setup Had issues with my RAM though... (4x16GB Corsair Vengeance @ 3000 MHz). I've heard Corsair RAM is not the best to use with Ryzen but I had it from my prior build. Doesn't seem to want to boot at 3000 MHz. It's of course fine at the default (2133). I'll have to try other speeds tomorrow.
Aside from that, really happy with the build. The Noctua cooler is working great. With all cores running at 3.6-3.7GHz, it remained at around 62C when running a Handbrake encode.
Cmd with admin
Bcdedit /set useplatformclock false
Edit: also turn it off in bios.
I'm having issues with my RAM too and its G-skill TridentZ (Samsung b-die 3600@16-16-16-36) :
https://www.newegg.com/Product/Product.aspx?Item=N82E16820232491
A quick scan of all of the QVL lists show that 4 sticks is not generally supported @ 3600. The ram that managed to run 4 sticks at 3600 was rated at like 4133 and they're using CL times far above CL16 .. I see CL18/19 for 3600 4 sticks with 4000Mhz rated ram. Given this and my own personal observations about instability at the limits, I don't think the majority of people are really running 4x8GB Samsung B-die 3600Mhz CL16 ram sticks at the rating of 3600Mhz@CL16-16-16-36 timings.
So, people need to either run prime95 and find the instabilities that I and mobo manufacturers found at this setting or they're not being honest about their stats. I'll provide more info soon. Just had to yank an ill-suited AIO water cooler from my setup, chunk it in a box (returning it), and mount a Noctua NH-U14S which is suddenly and strangely no longer listed on Newegg.
This is what I want to here ddogg.My RAM is finicky at 3600 with this board. It ran flawlessly on my z170 board at 3600 16-16-16-36 but on this, it posts about half the time and the other half it just defaults to the stock timings of 2133. I just bumped it down to 3433 and it now runs without issues.
Prime 95 is not the only, nor the be all end all of memory stability testing. So wtf are you talking about? Memtest, linpack, or OCCT, AMONG OTHERS not enough for you?
Back from my Hiatus ...
Having issue getting :
https://www.newegg.com/Product/Product.aspx?Item=N82E16820232491
(2x8) x2 Kit running stable at 3600Mhz
I get boot loops which sometimes stabilize, lockups on prime95, and sometimes bit errors. Memory runs fine and stable at 3200Mhz CL14. I haven't tested the next bump 3433Mhz @CL16 but 3600 @CL16-16-16-36 is a no go as it is for most QVL lists from mobo manufacturers for 4sticks. I find it weird that all the QVL list state this and only show 2 sticks @ this speed yet there are claims here that it's running stable @ 3600 32GB (4x8) sticks @ CL16 (no modifications). So, for anyone claiming this, can you run prime95 and actually make sure your memory is stable at this rating. I found that it's stable in the sense that you can run Cinebench and open some web browser tabs. It's hardly stable when maxed.
Also, for anyone claiming successful OC's can you please state how you're doing the OC? AMD Ryzen Master or through Bios. I am finding when I do it with bios it actually doesn't take when checked by Hwinfo. It still runs @ a wattage ceiling lock of 180W.
Also, in case no one's noticed, if you crank the RAM clocks with the wattage ceiling, the CPU lowers the CPU clocks to preserve the wattage ceiling. Many reviewers also noted this (Cranking ram clocks lowers performance as the CPU uses more more for the memory controller and less is available to the cores with a 180W lock. So, the stats being posted here need more detail and clarification. It seems some of you are not stable at certain settings (haven't ran prime95) but are claiming you're stable.
I have essentially the same ram, and 4 sticks, and its stable@3600, EXCEPT mine is not the RGB variety.I'm having issues with my RAM too and its G-skill TridentZ (Samsung b-die 3600@16-16-16-36) :
https://www.newegg.com/Product/Product.aspx?Item=N82E16820232491
A quick scan of all of the QVL lists show that 4 sticks is not generally supported @ 3600. The ram that managed to run 4 sticks at 3600 was rated at like 4133 and they're using CL times far above CL16 .. I see CL18/19 for 3600 4 sticks with 4000Mhz rated ram. Given this and my own personal observations about instability at the limits, I don't think the majority of people are really running 4x8GB Samsung B-die 3600Mhz CL16 ram sticks at the rating of 3600Mhz@CL16-16-16-36 timings.
So, people need to either run prime95 and find the instabilities that I and mobo manufacturers found at this setting or they're not being honest about their stats. I'll provide more info soon. Just had to yank an ill-suited AIO water cooler from my setup, chunk it in a box (returning it), and mount a Noctua NH-U14S which is suddenly and strangely no longer listed on Newegg.
I did 16 18 18 on my sticks and boot stable, but I don't have time to tighten and qualify that OC. Did 3333 which failed right in and tested overnight. Might do the 3600 overnight with memtest86 tonight, followed by a couple others overnight if that passes, although I actually get better performance with the 3333@CL14, so no real point unless I can get a lower CL or better timings stable. In a couple weeks, I can truly put it through its paces.I have essentially the same ram, and 4 sticks, and its stable@3600, EXCEPT mine is not the RGB variety.
Also note: I have the Taichi motherboard, what are you running ?
Prime 95 is not the only, nor the be all end all of memory stability testing. So wtf are you talking about? Memtest, linpack, or OCCT, AMONG OTHERS not enough for you?
I have the 4133 Trident z (F4-4133C19D kits). This is based on the older sticks, still b-die, but were clocked at 3600 CL16 18-18-38. When I tried that, plus 2T, it was performing well. Also, if you are using memtest to qualify, it needs set to parallel, not single or round robin, otherwise you don't work the ram hard enough for certain errors to appear. And when I qualify my ram, I use multiple stability stress techniques, not just one. But, your insane obsession saying p95 alone is what I take issue with. I've passed p95 and failed on other tests before. This alone shows you need a wider test array. That you said and I agree with. But there is no one way, nor should you be over-dependent on p95. Also, knowing whether you are getting row-hammer bit flips is NOT something p95 can tell you.
It isn't the end all be all.. relax buddy. Why do you think no Motherboard vender states that you can run 3600Mhz rated ram at CL16-16-16-36 with a 4x8GB configuration? I'll wait for the answer. What you're forgetting here is that Memtest doesn't stress the CPU simultaneously while pushing the RAM. Why do you need to do both? because you have a wattage manager on the CPU, you have two independent dies and you have a cross die infinity fabric that is clocked to the RAM settings in addition to the on-die infinity fabric. So, no motherboard QVL list states that 3600Mhz rated ram is stable at CL16 settings (samsung b-die) even, yet you guys are claiming you have magic sticks that is somehow stable? Simply.. No. It's not stable. Ddog just confirmed my exact observations and this is on the Asrock Taichi X399. So, unless you have a different motherboard, I'd love to hear your extended comments? I'd rather we be critical about this and clear than sit here and get upset at somebody pointing this out and have someone spend $50-100 more on ram that won't run at the rated speeds or we allow for an issue to be raised for a specific mobo manufacturer to look into. As for your'e tone, I'm not getting dragged into another one of those exchanges.
2 people have confirmed 3600Mhz@CL16 (4x8GB) settings on samsung b-dies is unstable on Asrock X399 taichi. This is inline w/ Mobo manufr's QVL list. Period.
I have the 4 sticks of the same ram (RGB) doesn't matter. It's the same samsung b-dies. I have a Taichi motherboard. Essentially, I'm running the same thing you are. Did you run Prime95, do you get boot loops, and can you post your DRAM timing configuration.I have essentially the same ram, and 4 sticks, and its stable@3600, EXCEPT mine is not the RGB variety.
Also note: I have the Taichi motherboard, what are you running ?
Yeah, I did too but saw boot loop issues just like you did which indicates its not truly stable. Then I re-ran prime95 and got hardware errors (memory bit errors) indicating this is not stable. Memtest doesn't do what prime95 does w.r.t to CPU stress. So, memtest doesnt guarantee you're stable. I have no doubt that it ran fine on a 6700k. The biggest difference here is that you have two dies w/ two independent memory controllers + infinity fabric per die + infinity fabric across dies all tied to your ram speed and timings. Far more can go wrong at these ram speeds on timings on Threadripper than 6700k. I thank you for being honest/detailed about your observations. This is the detail I was honestly looking for as I began pulling my hair out wondering if it was faulty ram (which it is not).When it did post at 3600, I ran Prime95 for about an hour and nothing failed. Booting would be an issue though. The RAM is definitely capable of running at 3600 CL16 as I had it running for about a week on my 6700k with no problems. I prefer having a higher stable OC on the CPU than RAM.
Dude, I'm so not sharing as soon as I get my timings stable later this month because of your attitude. First, IMC varies, meaning having a better chip can change this ad is. Second, p95 doesn't tell anything if your OC isn't stable in the first place, not can you always narrow the ram testing to the ram alone being the cause. Third, considering my ram have ran, granted on an Intel setup, 3733@CL14 for about a year, it is down to timings and IMC. Fourth, that is because those two kits are a quad kit. So they didn't grab two of the 2x8 kits. Big surprise.I have the 4 sticks of the same ram (RGB) doesn't matter. It's the same samsung b-dies. I have a Taichi motherboard. Essentially, I'm running the same thing you are. Did you run Prime95, do you get boot loops, and can you post your DRAM timing configuration.
The QVL for this motherboard clearly shows the configuration you're running isn't likely stable :
https://www.asrock.com/mb/AMD/X399 Taichi/index.us.asp#Memory
DDR4 3600 8GB G.Skill F4-4266C19D-16GTZA Samsung B-die SS 2pcs v
DDR4 3600 8GB G.Skill F4-4133C19D-16GTZA SS 2pcs v
DDR4 3600 8GB G.Skill F4-4000C19D-16GTZSW SS 2pcs v
DDR4 3600 4GB Avexir AVD4UZ136001704G-4BZ1RR SS 2pcs v
DDR4 3600 8GB G.Skill F4-3733C18Q-32GTZSW SS 2pcs 4pcs v
DDR4 3600 8GB G.Skill F4-4133C19D-16GTZKW SS 2pcs v
The only thing approved for 4 pieces @ 3600 is 3733 rated ram @CL18.
Here's Gigabytes QVL :
http://download.gigabyte.us/FileList/Memory/mb_memory_x399-aorus-gaming7.pdf
Nothing in here shows any ramstick being able to run in a 4x8GB configuration @3600MHz @16-16-16-36 timings. Not even 4133Mhz downclocked ram can do this.
So, either you guys have magical ram or you haven't probed stability enough. Stability in my mind is a situation in which the CPU is strained to its max in both compute and memory operations. I think you guys are forgetting that this not only clocks up the infinity fabric on each die but there is one connecting the dies too that gets clocked up with these settings. I never thought you could run 3600Mhz ram CL16 given all of this. The QVLs, ddog, and my observations prove it...
So, please indicate if you ran Prime95 and also a screen cap of your DRAM timings from bios.
Now were talking. This is the detail that needs to be discussed !I have the 4133 Trident z (F4-4133C19D kits). This is based on the older sticks, still b-die, but were clocked at 3600 CL16 18-18-38. When I tried that, plus 2T, it was performing well. Also, if you are using memtest to qualify, it needs set to parallel, not single or round robin, otherwise you don't work the ram hard enough for certain errors to appear. And when I qualify my ram, I use multiple stability stress techniques, not just one. But, your insane obsession saying p95 alone is what I take issue with. I've passed p95 and failed on other tests before. This alone shows you need a wider test array. That you said and I agree with. But there is no one way, nor should you be over-dependent on p95. Also, knowing whether you are getting row-hammer bit flips is NOT something p95 can tell you.
Lots of assumptions and an over-pronounced tone ... I have 4-5 usb sticks of stability testing suites and tools. Memtest is one of them and I know exactly how to probe memory for issues. I just got my setup rebuilt last night (crap AIO cooler that I didn't want to do stress testing on). So, of the tests that I ran, was prime95. Prime95 showed hardware error memory bit flip issues @3600Mhz CL16 (the rating of my ram and the XMP setting). So, I knew right away what was happening. I can confirm this with Memtest but I dont need to. Instead, I thought, first thing : let me post on the thread and see if I am the only one. Now, I could have gone and testing this further myself, done memory training, found a sweet spot that worked for me, and not disclose the settings and then say.. Yeah, hey my stuff is stable at 3600Mhz @CL16. But, what would that have done for everyone else? Were on the same page here AJC, were all trying to stabilize our systems. Lets drop the tone and stick to the technical details.And when I qualify my ram, I use multiple stability stress techniques, not just one. But, your insane obsession saying p95 alone is what I take issue with. I've passed p95 and failed on other tests before. This alone shows you need a wider test array. That you said and I agree with. But there is no one way, nor should you be over-dependent on p95. Also, knowing whether you are getting row-hammer bit flips is NOT something p95 can tell you.