Overclock.net › Forums › AMD › AMD Motherboards › My experience with the Asus PRIME X370-Pro
New Posts  All Forums:Forum Nav:

My experience with the Asus PRIME X370-Pro - Page 547

post #5461 of 6766
Quote:
Originally Posted by Lermite View Post

It looks like so.

You can try a bit further with:

tRRDS: 4
tFAW: 28
tWRRD: 1

But I'm really not sure about the stability with such timings.

Ok :-)

Have you ever tried running the first six timings like in Stilt's fast timings for 3200?
14,13,13,13,28,42?
post #5462 of 6766
Quote:
Originally Posted by makatech View Post

If I understand you correctly 3333 isn't stable yet? When you have 3333 stable (hci memtest & prime95 blend or occt) then please let us know and we will try your settings. ;-)

I rather easily boot and run Windows 10 in both 3333 and 3466 running various benchmarks but getting it stable (in hci memtest, prime95) using good manual timings is a problem so far.

Now after realizing I can use really low SoC volt (currently 1.0) for 3200 I may be ready for another try soon though.

Using AIDA, it detected a error around 7hr30m ish, would like more, hence why i dont deem it "stable" but been using / working / gaming, in between fiddling, seemed ok, but do want 12 - 15 hours.

Obv ive cursed it.

SoC is 1.0v, ProcODT 60, RAM 1.4, (might be able to lower the RAM, i just set a forgot that tbf)

Ah, with 3466 didnt expect anything from it, maybe ill struggle then
Edited by inserf1 - 10/20/17 at 2:25pm
post #5463 of 6766
Quote:
Originally Posted by makatech View Post

Have you ever tried running the first six timings like in Stilt's fast timings for 3200?
14,13,13,13,28,42?

I've tried to many sets of timings that I don't remember all of them smile.gif
I gonna give a try to these ones tonight.
Multipurpose PC
(12 items)
 
  
MotherboardGraphicsRAMHard Drive
Asus Prime X370 Pro EVGA 1050 Ti HOF 3600 C17 Intel 600P 512MB 
Hard DriveCoolingOSMonitor
WD Red 3TB Alphacool Eisbaer 280 Windows 10 Enterprise BCC Acer G246HL 
KeyboardPowerCaseMouse
Microsoft Natural Ergonomic Keyboard 4000 v1.0 Seasonic Prime 650W Titanium SSR-650TD Antec P280 Logitech G502 
  hide details  
Reply
Multipurpose PC
(12 items)
 
  
MotherboardGraphicsRAMHard Drive
Asus Prime X370 Pro EVGA 1050 Ti HOF 3600 C17 Intel 600P 512MB 
Hard DriveCoolingOSMonitor
WD Red 3TB Alphacool Eisbaer 280 Windows 10 Enterprise BCC Acer G246HL 
KeyboardPowerCaseMouse
Microsoft Natural Ergonomic Keyboard 4000 v1.0 Seasonic Prime 650W Titanium SSR-650TD Antec P280 Logitech G502 
  hide details  
Reply
post #5464 of 6766
Quote:
Originally Posted by inserf1 View Post

Using AIDA, it detected a error around 7hr30m ish, would like more, hence why i dont deem it "stable" but been using / working / gaming, in between fiddling, seemed ok, but do want 12 - 15 hours.

Obv ive cursed it.

SoC is 1.0v, RAM 1.4 (might be able to lower the RAM, i just set a forgot that tbf)

Ok, I prefer running hci memtest for stability testing my memories (12 open instances for my 1600X), at least first. It's also nice because it doesn't push my cpu that hard thus focusing on memory testing. (tdie temp probably about 20 degrees lower comparing to a tough prime95 blend test.)

You are running SoC 1.0 for 3333MHz?? These low SoC values for our Asus Prime X370 board is really continuing to surprise me. Perrhaps this build to my theory that lower SoC will help me reaching stability on 3333 too (and perhaps even 3466).

Really interesting and only one single error after 7h30 isn't bad at all, it's good, you may be close running your system stable at 3333 which is kind of rare for our board (I think). I'm only aware of @Keith Myers right now on this forum.

I rarely run my tests that long (7h30) but I should! ;-)

Thanks for sharing.

When I try for 3333 again I may probably start with slightly higher SoC though, perhaps something between 1.02-1.05 to start with. Actually I had a stable setup for 3333 but with poor timings, only running the first six timings with manual values, the rest on auto. This time I am aiming for much better timings.

EDIT: Oh, you are running bios 0810, also very interesting. It is a possibility I will downgrade from 0902 soon. I have heard from another user in Sweden being more successful with 0810 for higher memory speeds (3333 & 3466).
Edited by makatech - 10/20/17 at 3:05pm
post #5465 of 6766
Okay so I am sure this was answered before but is there a fourth place to put fan headers. So far I have cpu, cpu opt, fan 1, and fan 2 connected. Does W_pump work as a fan header? are there any issues with it?
Main RIG
(8 items)
 
  
CPUMotherboardGraphicsRAM
R7 1700 ASUS PRIME X-370 PRO MSI gtx 1070 Gaming X G.SKILL 3200MHZ CL16 16gb@2933mhz 
Hard DriveCoolingOSMonitor
256 gb ssd/1tb hdd Cooler Master D92 Windows 10 Dell 1440p 165hz G-sync 
  hide details  
Reply
Main RIG
(8 items)
 
  
CPUMotherboardGraphicsRAM
R7 1700 ASUS PRIME X-370 PRO MSI gtx 1070 Gaming X G.SKILL 3200MHZ CL16 16gb@2933mhz 
Hard DriveCoolingOSMonitor
256 gb ssd/1tb hdd Cooler Master D92 Windows 10 Dell 1440p 165hz G-sync 
  hide details  
Reply
post #5466 of 6766
I guess you can try. But I bet the W pump header will make your fan run at top speed at all times. That's how it was here at least with the pump.
post #5467 of 6766
Maybe i could try HCI, think in the past (pre i7/X58 days) i tried to keep testing isolated, but then combining them, extra time, and extra annoying when you know they will work at such and such separately, by the end i'd use Blend anyway, but since then, CPU first, stock RAM, stress, then RAM, then a total CPU/IMC/RAM stress for a long run, then use, has served me ok.

Could be im impatient though.

For whatever reason, I've found 7-zip benchmark really quite good at finding errors with timings (may be in my head) it doesn't use much RAM, or max CPU, but it i found it really useful, and quite quick at finding a problem.

SoC, yea, slightly by accident, i got to a point when relaxing timing were having no effect on the stability, having started at 1.1, naturally i went up, still nothing, then went down, and while it didn't improve the stability, it didn't make it worse, just stopped at 1.0v thinking this seems low enough, excepting to up it again. But i haven't, and in this case, ProcODT to 60 got me over the hump (previously still using auto and thought ProcODT was useful for cold boots)

That would be sweet, kinda surprised myself, while i'm waiting, I've started to wonder if 3466 and beyond is worth it, and / or if 3200 with lower timing is better or even doable?

Yea, still on 0810, when i had the Hynix-M chips, 0902 gave me nothing but stock speeds

Now_setting.txt 4k .txt file
post #5468 of 6766
Quote:
Originally Posted by garretsw View Post

Okay so I am sure this was answered before but is there a fourth place to put fan headers. So far I have cpu, cpu opt, fan 1, and fan 2 connected. Does W_pump work as a fan header? are there any issues with it?

I'm using Water Pump header for fans since BIOS 504. The only (notable) difference from CHA1 and CHA2 fan headers is that you don't get the fan stop functionality and you can't set it to other sources than CPU temperature. You could set the first temperature interval to start at 0 RPM, but sometimes it doesn't work. I'm on BIOS 902 and I can swear I did managed to get PWM fans to stop and start on the water pump header, but now it doesn't work anymore...

My bet is that the AIO pump header acts the same, but as the manual says, only allows 1A of current draw whereas the Water Pump can work with 3A pumps.
Ryzen 7
(0 items)
  
Reply
Ryzen 7
(0 items)
  
Reply
post #5469 of 6766
Quote:
Originally Posted by inserf1 View Post

Maybe i could try HCI, think in the past (pre i7/X58 days) i tried to keep testing isolated, but then combining them, extra time, and extra annoying when you know they will work at such and such separately, by the end i'd use Blend anyway, but since then, CPU first, stock RAM, stress, then RAM, then a total CPU/IMC/RAM stress for a long run, then use, has served me ok.

Could be im impatient though.

For whatever reason, I've found 7-zip benchmark really quite good at finding errors with timings (may be in my head) it doesn't use much RAM, or max CPU, but it i found it really useful, and quite quick at finding a problem.

SoC, yea, slightly by accident, i got to a point when relaxing timing were having no effect on the stability, having started at 1.1, naturally i went up, still nothing, then went down, and while it didn't improve the stability, it didn't make it worse, just stopped at 1.0v thinking this seems low enough, excepting to up it again. But i haven't, and in this case, ProcODT to 60 got me over the hump (previously still using auto and thought ProcODT was useful for cold boots)

That would be sweet, kinda surprised myself, while i'm waiting, I've started to wonder if 3466 and beyond is worth it, and / or if 3200 with lower timing is better or even doable?

Yea, still on 0810, when i had the Hynix-M chips, 0902 gave me nothing but stock speeds

Now_setting.txt 4k .txt file

From a performance point of view 3333 may be a sweetspot because it should give great performance without having to raise voltage etc too high but 3466 should still give you even slightly better performance. (Most probably a very small difference from 3333 though.) A lot having to do with the timings you set. If being able to run 3466 stable using good timings on our board you will probably be the only one here doing so right now with our board.

You will have to run tests both for stability and benchmarks comparing your diffferent setups.

For me lowering SoC to 1.0 (from 1.11875) was the key for reaching stability for my current 3200 timings & configuration. Of course the other values are important for stability too though. Remains to see if a low SoC value will be the key for reaching 3333 stability too.

If I understand your timings correct for your 3333 setup your first four timings are 14,14,14,14? This is good, I know from the forum in Sweden a couple of guys managing to run 3333 stable on other boards using 14,15,15,15 instead (slightly looser) but this is probably depending on your specific memory modules too (you may have really good ones).
Edited by makatech - 10/20/17 at 11:44pm
post #5470 of 6766
Quote:
Originally Posted by makatech View Post

Ok :-)

Have you ever tried running the first six timings like in Stilt's fast timings for 3200?
14,13,13,13,28,42?

I tried 14-13-13-13-28-41 because the minimum value of tRC is tRP+tRAS
But it's far from stable.

Then I tested the timings I just posted.
I got only one error with 16 instances of HCI Memtest with 1370% as completion.
Perhaps I'll manage to get rid of this flaw by playing again with CLDO_VDDP.

My IMC are lazy as **** sozo.gif
Multipurpose PC
(12 items)
 
  
MotherboardGraphicsRAMHard Drive
Asus Prime X370 Pro EVGA 1050 Ti HOF 3600 C17 Intel 600P 512MB 
Hard DriveCoolingOSMonitor
WD Red 3TB Alphacool Eisbaer 280 Windows 10 Enterprise BCC Acer G246HL 
KeyboardPowerCaseMouse
Microsoft Natural Ergonomic Keyboard 4000 v1.0 Seasonic Prime 650W Titanium SSR-650TD Antec P280 Logitech G502 
  hide details  
Reply
Multipurpose PC
(12 items)
 
  
MotherboardGraphicsRAMHard Drive
Asus Prime X370 Pro EVGA 1050 Ti HOF 3600 C17 Intel 600P 512MB 
Hard DriveCoolingOSMonitor
WD Red 3TB Alphacool Eisbaer 280 Windows 10 Enterprise BCC Acer G246HL 
KeyboardPowerCaseMouse
Microsoft Natural Ergonomic Keyboard 4000 v1.0 Seasonic Prime 650W Titanium SSR-650TD Antec P280 Logitech G502 
  hide details  
Reply
New Posts  All Forums:Forum Nav:
  Return Home
  Back to Forum: AMD Motherboards
Overclock.net › Forums › AMD › AMD Motherboards › My experience with the Asus PRIME X370-Pro