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

ROG Crosshair VI overclocking thread - Page 1298

post #12971 of 17024
Quote:
Originally Posted by gupsterg View Post

I use VBOOT/VDIMM 1.35V in UEFI, ProbeIt point ~1.375V.

My thought is similar to The Stilt's, SOC does not need to be high as what many use. I believe it is currently masking an issue with FW.

I am not upset. Nor concerned. I am happy to use 3.8GHz/3200MHz with "Sleep/Resume" as I have 0 issues there for many days smile.gif . On a reboot it is rare issue. It is more of an issue when shutdown and boot (power active to PSU). Fully powered down rig and post is no go. My posts are just as info/feedback, etc.

I feel as if the 1.0.0.4a is the culprit when booting up (power active to PSU). I never had boot problems prior to that. I know they improved memory latencies and such with 1.0.0.4a so I assume that may be the cause. Further, confirming this theory is that I believe raising the SOC even further to 1.0v will solve the boot up issues.

Elmor confirmed that boot from no power (PSU is turned off/unplugged) is a Ryzen issue because it will not save the DDR Volt settings when PSU is off/unplugged.
Ryzen - Elmor
(13 items)
 
  
CPUMotherboardGraphicsRAM
AMD Ryzen 1700 @ 3.8Ghz ASUS ROG CROSSHAIR VI HERO GeForce GTX 1070 32GB DDR4-3466 16-15-15-35 
Hard DriveHard DriveHard DriveHard Drive
Samsung 960 EVO 1TB Samsung 830 SSD 256GB Hitachi 2TB HGST 4TB  
CoolingOSMonitorPower
Noctua NH-D15S Windows 10 Pro Acer Predator XB271HU SeaSonic PRIME Titanium 750W 
Case
Fractal Design Define R5 
  hide details  
Reply
Ryzen - Elmor
(13 items)
 
  
CPUMotherboardGraphicsRAM
AMD Ryzen 1700 @ 3.8Ghz ASUS ROG CROSSHAIR VI HERO GeForce GTX 1070 32GB DDR4-3466 16-15-15-35 
Hard DriveHard DriveHard DriveHard Drive
Samsung 960 EVO 1TB Samsung 830 SSD 256GB Hitachi 2TB HGST 4TB  
CoolingOSMonitorPower
Noctua NH-D15S Windows 10 Pro Acer Predator XB271HU SeaSonic PRIME Titanium 750W 
Case
Fractal Design Define R5 
  hide details  
Reply
post #12972 of 17024
Quote:
Originally Posted by rossctr View Post

Sorry for the stupid question but how do you guys test for memory stability? I've not tried memory OC before and the pics I see you guys post you have multiple instances of, what I assume is memtest, open.

Just updated BIOS to 1107, thought I would try out D.O.C.P.5 and it booted straight into windows

a4ijxo8.png

So I'm guessing I should make sure its stable.

Also, is it ok forcing PCIE back to Gen3?


I recommend Tech power up Memtest.

And yes been forcing G3 since day 1
post #12973 of 17024
Quote:
Originally Posted by Johan45 View Post

LN2 jumper is what it's named for. It raises and alters many voltages to aid in cold boot bug not cold booting at normal ambient temps. I really wouldn't advise using it 24/7 on normal ambient temps.

18°C isn't really normal ambient temps, if it was it would work. 2933 has been fine for the past month for me, on 3 different BIOSes and on my own 2 systems (i have customer builds that boot at 3200 since all i use on those are Samsung B-dies, even though its not guaranteed, they boot at 3200)

If i go from 2933 to 3200 in the same temps it wont boot, even though all the settings are IDENTICAL. Unless Auto prodoct is changing between the 2 (even though BIOS 1002 didnt have the option) i don't see what else it could be. It really makes NO sense to me, i thought upping SOC would work but that doesnt either. I can boot on 3200 at 1.0v when i couldnt before haha.
    
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
post #12974 of 17024
Quote:
Originally Posted by lordzed83 View Post

I recommend Tech power up Memtest.

And yes been forcing G3 since day 1

It failed memtest as soon as I pushed start haha. Trying to get 3200 to boot up but cant seem to get it to work, just reverts itself back to 2933
post #12975 of 17024
Quote:
Originally Posted by gupsterg View Post

I use VBOOT/VDIMM 1.35V in UEFI, ProbeIt point ~1.375V.

My thought is similar to The Stilt's, SOC does not need to be high as what many use. I believe it is currently masking an issue with FW.

I am not upset. Nor concerned. I am happy to use 3.8GHz/3200MHz with "Sleep/Resume" as I have 0 issues there for many days smile.gif . On a reboot it is rare issue. It is more of an issue when shutdown and boot (power active to PSU). Fully powered down rig and post is no go. My posts are just as info/feedback, etc.


I think it depends on IMC and Memory. In my case anything lower than 1.137 is not 100% stable not for long rendering periods. I had pc crash 3 times when rendering same 3 hour long project around 88% at 1.130 bumped to 1.137 rendered 3 hour project and another 2 projects NO PROBLEMS. And on 1.130 i passed memtests ibt aida and like everything tongue.gif

Power Director is still best stress test biggrin.gif


The more time i spend tweekeing more im sure that one cant really recommend any settings for this platform. Everyone needs different volts and settings ATM. With more bioses and microcodes it will change. But not at this stage.
Edited by lordzed83 - 5/1/17 at 11:58am
post #12976 of 17024
Quote:
Originally Posted by lordzed83 View Post

I think it depends on IMC and Memory. In my case anything lower than 1.137 is not 100% stable not for long rendering periods. I had pc crash 3 times when rendering same 3 hour long project around 88% at 1.130 bumped to 1.137 rendered 3 hour project and another 2 projects NO PROBLEMS. And on 1.130 i passed memtests ibt aida and like everything tongue.gif

Power Director is still best stress test biggrin.gif


The more time i spend tweekeing more im sure that one cant really recommend any settings for this platform. Everyone needs different volts and settings ATM. With more bioses and microcodes it will change. But not at this stage.

Isn't that a part of the fun of a motherboard like the Crosshair VI Hero, all the adjustments and settings we have access to? If we didn't enjoy tweaking, most of us would probably be happy with a B350 based motherboard.
post #12977 of 17024
Quote:
Originally Posted by Targonis View Post

Quote:
Originally Posted by gupsterg View Post

Yep aware of that setting smile.gif . Fail_CNT, posted about it only today earlier in thread wink.gif .

I figured you did, but for those who have not been keeping track of these things, if they read your posts and tried to repeat what worked without increasing Fail_CNT, they might run into trouble. Nice job on documenting your work btw, now we just need Elmor to put this stuff in the top post, complete with all the things to set for things to run well. biggrin.gif Elmor does enough for us, so I wouldn't expect a fully written guide, but it WOULD be nice. There are a lot of places to tweak settings, voltages, etc for the newbies.

And public thanks to gupsterg for his earlier clear description of procODT operation.

I am close to adopting the belief that not only does Asus believe in overclocking, they believe (and this is easy to do given this thread alone) that overclocking is pursued as an end in itself. With that assumption followed by asking themselves how to make it more fun, they chose to make overclocking like a hyperdimensional dungeon game. Each room (parameter) has to be "felt out" by the blind adventurer to find out what is in it, and what trolls are waiting for the unwary. To add complexity, what the adventurer does in one room may be communicated to other unspecified rooms, including those already passed through, changing their contents (functions). And the Final Boss (Asus) gets to change the room details every couple of weeks -- sometimes making accessible rooms that previously weren't but which must be traversed.

And like the hapless adventurer in the ancestral DecSystem 10 DND game, every bad troll encounter leads to a restart from scratch back at the mailbox (every bad parameter combination leads to a reboot). Nice work Asus! biggrin.gif
     
CPUMotherboardGraphicsRAM
AMD Ryzen 1800X ASUS ROG CROSSHAIR VI HERO X370 Asus ROG STRIX nVidia GeForce GTX1080Ti-O11G G. Skill TridentZ F4-3200C14D-32GTZ 
Hard DriveOptical DriveCoolingOS
Samsung SSD 850 EVO 256GB Asus UltraDrive 8X Noctua NH-D15 SE-AM4  Linux Mint 18.1 MATE 64-bit  
MonitorKeyboardPowerCase
LG 4K TV OLED55B6P Logitech MK235 RF Wireless Keyboard & Mouse PC Power & Cooling Turbo-Cool 860i Nanoxia Project S 
Mouse
See Keyboard 
CPUGraphicsRAMHard Drive
147,456 cores all registers exposed at the console 4096 36-bit words rotary drum 
Optical DriveMonitor
6 magnetic tape drives neon lites for every register; oscilloscope for... 
  hide details  
Reply
     
CPUMotherboardGraphicsRAM
AMD Ryzen 1800X ASUS ROG CROSSHAIR VI HERO X370 Asus ROG STRIX nVidia GeForce GTX1080Ti-O11G G. Skill TridentZ F4-3200C14D-32GTZ 
Hard DriveOptical DriveCoolingOS
Samsung SSD 850 EVO 256GB Asus UltraDrive 8X Noctua NH-D15 SE-AM4  Linux Mint 18.1 MATE 64-bit  
MonitorKeyboardPowerCase
LG 4K TV OLED55B6P Logitech MK235 RF Wireless Keyboard & Mouse PC Power & Cooling Turbo-Cool 860i Nanoxia Project S 
Mouse
See Keyboard 
CPUGraphicsRAMHard Drive
147,456 cores all registers exposed at the console 4096 36-bit words rotary drum 
Optical DriveMonitor
6 magnetic tape drives neon lites for every register; oscilloscope for... 
  hide details  
Reply
post #12978 of 17024
Quote:
Originally Posted by skullbringer View Post

So is the 1107 worth the upgrade, or is it just a fix for niche issues?

Apart from the fixed bricking, I could not really tell any differences in the bios revisions tbh. I know this may sound ignorant, but I was able to run this ever since and could not get more performance out of it since launch.
- daily:
CPU: 4.0GHz @ 1.4V
RAM: 3200 MHz 14-14-14-34 1.45V
SOC: 1.1V
- benching:
CPU: 4.15GHz @ 1.5V
RAM: 3615 MHz 14-14-14-34 1.6V
SOC: 1.2V

Is there any information on the roadmap ahead for really big improvements, like tunable subtimings and ram speeds of 3866+ with new May 1.05 agesa?

Still, even if there are no perf gains, it is remarkable how almost 13000 posts in this thread alone of community members sharing their experiences and industry partners like Elmor and Raja have made this new platform work. Ironing out all the little bugs of all the edge usecases really could not have been done so quickly without such a large "beta testing crowd" and it really makes me proud to be a part of this community. Thanks to all of you.
1.6V on ram? Isn't that going to fry it?
post #12979 of 17024
Quote:
Originally Posted by finalheaven View Post

I am really worried that people (we) might be getting our hopes up too much. People seem to expect that 1.0.0.5 will fix everything memory related and that everyone will be able to easily run 3200 at C14 @ 1.35v no matter the chip/brand.

I wish that this all becomes true but again AMD has never promised specifics and their official support is a lot lower...

My hopes are only being able to run my 3600 rated modules at 1.35V, 4000 MHz, 12-12-12-12-32. Glad you were not referring to me.
post #12980 of 17024
Hmm, so for getting the highest gaming framerates at 1080p with 1080/Titan, I understand that getting the highest memory bandwith is way more important than tight timings because what you really want is the infinity fabric as high as possible (1/2 speed of the ddr). What no one still can answer is if you are at 3600 ddr4, what method gets the infinity fabric the fastest and sub timings the fastest? Does it not matter which strap or which bclk you use to get to 3600 at all? Are they all the same, or is their a best strap and bclk to use to get the highest infinity fabric or I should say lowest latency for the infinity fabric?
New Posts  All Forums:Forum Nav:
  Return Home
  Back to Forum: AMD Motherboards
Overclock.net › Forums › AMD › AMD Motherboards › ROG Crosshair VI overclocking thread