Overclock.net › Forums › AMD › AMD Motherboards › ROG Crosshair VI overclocking thread
New Posts  All Forums:Forum Nav:

ROG Crosshair VI overclocking thread - Page 825

post #8241 of 17133
Quote:
Originally Posted by DDR4 View Post

KHX2666C15D4/8G Can not start the standard profile of 2400 and 2666 MHz.
Started with the help of another one ranked DDR4 2133MHz.
A month has passed and the Asus can not do it. madsmiley.png

A lot of this depends on the exact memory. Many are running fine, but Ryzen is NOT an Intel platform, so the way your RAM negotiates won't automatically work. With that said, you also need to pick the correct BIOS version, 1T or 2T.
post #8242 of 17133
Quote:
Originally Posted by Targonis View Post

0083 is a 2T BIOS, 0081 and 0082 are 1T.

Sorry I wasn't very clear....I cannot boot at 3200 with either 081/082, it requires 083 to boot. The same memory was fine at 3200 on 1002

AKA 1T was working, now I need 2T to be stable at 3200
post #8243 of 17133
Quote:
Originally Posted by hotstocks View Post

Guys, coming from multiple Intel builds over the years and a stringent rule that stable is only if passes all of Intel Burn Test levels. Normally this will increase cpu temps about 20C, but previous i5 and i7s all pass with same or minor voltage bump. On my Ryzen 1700, I can pass every other benchmark (realbench, prime, etc.) at 3900mhz and about 1.3875v LLC3. 3.9 is my wall basically. And can even pass the quick IBT standard, However to pass IBT at Maximum it requires a downclock to 3850mhz and a large increase in volts to about 1.43v. So my question is IBT puts an ungodly demand on a cpu, but Intel chips always pass it at same voltage or close and at same speed. This Ryzen needs to lower clock 50mhz and increase volts about .05! So is this test valid or am I being too hard on what I consider 100% stability? For a daily driver where all important information is stored and used, does Ryzen need to pass IBT at maximum or is there something that is causing this test to not work with Ryzen or be brutal on it, since no real world scenario will ever be like IBT maximum. Thanks

This is why there is a 1700, 1700X, and 1800X. You can't expect to take a 3.4GHz chip that is at the bottom of a QA bin for what the chip QAs at and expect to overclock as far as the other Ryzen 7 chips.
post #8244 of 17133
Quote:
Originally Posted by Targonis View Post

This is why there is a 1700, 1700X, and 1800X. You can't expect to take a 3.4GHz chip that is at the bottom of a QA bin for what the chip QAs at and expect to overclock as far as the other Ryzen 7 chips.

I realize that, unless you get a golden chip. But 3850-3900 on a 1700 is pretty average. That is not my question, my question is on whether a Ryzen must pass Intel Burn Test on maximum to be considered stable, or is their an issue with that old stress test and Ryzen where it only needs to pass on standard or high? Like I said it passes every other stress test.
post #8245 of 17133
Quote:
Originally Posted by madweazl View Post

I have been, just not sure how far I can go headscratch.gif

That's exactly where I am also at wink.gif Though I want to try underclocking later to make sure vcore is adequate.
main
(10 items)
 
  
Reply
main
(10 items)
 
  
Reply
post #8246 of 17133
Hi @elmor
I have just tried the c6h_p0vidfix program and it does not fully work for me. Whenever I execute it, clock goes up to 4GHz (my P0 config), but that's only until an idle period makes it drop to another Pstate. After that it doesn't raise to P0 unless I execute the program again, with the same results.

Something I am doing wrong or have misunderstood?
post #8247 of 17133
Quote:
Originally Posted by y0bailey View Post

Sorry I wasn't very clear....I cannot boot at 3200 with either 081/082, it requires 083 to boot. The same memory was fine at 3200 on 1002

AKA 1T was working, now I need 2T to be stable at 3200

Check the ProcODT setting, as well as the other setting to tell it to test up to 5 times before failing(new options with 0079-0083). I was able to get my RAM to be unstable but work at 2933 with some tweaks, and I had NO luck previously.
post #8248 of 17133
Quote:
Originally Posted by hotstocks View Post

I realize that, unless you get a golden chip. But 3850-3900 on a 1700 is pretty average. That is not my question, my question is on whether a Ryzen must pass Intel Burn Test on maximum to be considered stable, or is their an issue with that old stress test and Ryzen where it only needs to pass on standard or high? Like I said it passes every other stress test.

Stable is where you don't run into problems with whatever load you want to throw at it. For computer sales, you want to be able to throw EVERYTHING at it, full load for 24 hours without any crashing or other problems. Intel Burn Test is one test for that task. I consider stable anything that lets me run my machine at high/max loads for 18 hours a day and then go for two to three weeks without any sort of crash. Now, that may or may not work if Windows drivers or other programs cause problems, and then you need to figure out if your overclock is the cause of the problem, or the other programs are the key.

For many, the way I do it is probably "stable enough".
post #8249 of 17133
Quote:
Originally Posted by Targonis View Post

Check the ProcODT setting, as well as the other setting to tell it to test up to 5 times before failing(new options with 0079-0083). I was able to get my RAM to be unstable but work at 2933 with some tweaks, and I had NO luck previously.

What is the "safe" range for ProcODT settings? There isn't any real documentation on it, so I'm hesitant to tweak it without understanding. I've read people putting it in the 80's, but I don't know if that is safe and what that is actually doing?
post #8250 of 17133
Quote:
Originally Posted by Targonis View Post

Stable is where you don't run into problems with whatever load you want to throw at it. For computer sales, you want to be able to throw EVERYTHING at it, full load for 24 hours without any crashing or other problems. Intel Burn Test is one test for that task. I consider stable anything that lets me run my machine at high/max loads for 18 hours a day and then go for two to three weeks without any sort of crash. Now, that may or may not work if Windows drivers or other programs cause problems, and then you need to figure out if your overclock is the cause of the problem, or the other programs are the key.

For many, the way I do it is probably "stable enough".

Its going to depend on every situation.

Around here you're not stable unless you run intelburntest at maximum for a million hours. Ive done all my rigs and customer rigs the same way. 15-30min run of realbench/intelutility (still uses linpack btw) then i will run it all day just on to check for idle/background task stability, in the process ill check for WHEA errors then change offset/voltage accordingly. After all thats done, ill game on it solid for a couple days and run photoshop to edit some raw/jpeg files. Thats my baseline, when i change OC i repeat the process.

My 1700x at 3.8ghz is doing it fine at 1.199v under load, hasn't crashed once, hasn't 8 coded once, ZERO WHEA errors in my final setup so for me in my case its stable.

The people running IBT for 24hrs (and unless your PC is actually maxing out all cores at 100% for 24/7 then yea) all you are doing is slowly degrading your chip. I know musmus will argue with me to the death but i really don't care. Ive been doing it this way for 15years ive never had a customer or my pc have an unstable OC. Then again its also based on usage, if someone is folding then ill run the pc folding for 36hrs and change voltage accordingly. Its all dependent, just like audio and video are.
    
CPUMotherboardGraphicsRAM
R7 1700X asus crosshair 6 Sapphire R9 390 Nitro CMK16GX4M2B3200C16 
Hard DriveHard DriveCoolingCooling
Kingston HyperX 3K Seagate ST2000DM006 EKWB Supremacy Evo EK-XRES 100 Revo D5 - Acetal 
CoolingCoolingCoolingCooling
Alphacool GPX r9 390 m01 alphacool xt45 240 Alphacool xt45 360 Alphacool D5 PWM 
OSMonitorKeyboardPower
Windows 10 LG 29UM68 Corsair Strafe Corsair RM1000 
CaseMouseMouse Pad
Thermaltake X5 Corsair M65 Pro RGB Razer Goliathus Speed 
  hide details  
Reply
    
CPUMotherboardGraphicsRAM
R7 1700X asus crosshair 6 Sapphire R9 390 Nitro CMK16GX4M2B3200C16 
Hard DriveHard DriveCoolingCooling
Kingston HyperX 3K Seagate ST2000DM006 EKWB Supremacy Evo EK-XRES 100 Revo D5 - Acetal 
CoolingCoolingCoolingCooling
Alphacool GPX r9 390 m01 alphacool xt45 240 Alphacool xt45 360 Alphacool D5 PWM 
OSMonitorKeyboardPower
Windows 10 LG 29UM68 Corsair Strafe Corsair RM1000 
CaseMouseMouse Pad
Thermaltake X5 Corsair M65 Pro RGB Razer Goliathus Speed 
  hide details  
Reply
New Posts  All Forums:Forum Nav:
  Return Home
  Back to Forum: AMD Motherboards
Overclock.net › Forums › AMD › AMD Motherboards › ROG Crosshair VI overclocking thread