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

ROG Crosshair VI overclocking thread - Page 936

post #9351 of 17435
Quote:
Originally Posted by Raja@ASUS View Post

SOC @ 1.15V
DRAM Voltage @ 1.40V

DRAM timings as shown. 2666 memory ratio with 116.2 BCLK.


HTH

After setting BCLK did you attempt to reset the PCIe bus to version 3 mode? If so, did the GPU operate as expected?
     
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 #9352 of 17435
Quote:
Originally Posted by madweazl View Post

I've been messing around with 1002 since I shot you those results earlier; tonight while doing some stability testing I saw the temps reading in the 80s but the air coming off the HSF is fairly cool and I'm seriously doubting the temps actually being 80+. I hit the back of the socket with the temp gun and it's reading about 47° at the hottest area. Skew is disabled.

There are two possible interpretations for what happens with ITB AVX:

Three consecutive +10 C offsets are applied one after another, for a total of +30 C on top of the fixed +20 C for X CPUs. The first two happen in direct succession when memory allocation starts, the last one happens when Linpack AVX calculations start.

1. These could be pure offsets and the real temperature would be 50 C lower on a X CPU. This would be lower than the socket sensor, though.

2. Or they could be offsets only for the first few milliseconds to seconds until the real temperature finally catches up to the offset value and then takes over. If this was the case then these offsets could be seen as kind of "jump start" to ramp up fans before the real temps kick in.

Here is a screenshot of what happens when ITB AVX is started. Take a good look at the moment when Linpack calculations start! First it jumps by the +10 C offset and then it decreases at first, before catching up again.

Sorry for using a different scale on the socket sensor. Watch out when you compare the temps (0-100 socket, 20-100 Tctl/Tdie).


Edited by Timur Born - 4/14/17 at 5:01pm
post #9353 of 17435
Quote:
Originally Posted by goncalossilva View Post

Could someone remind me which USBs on the back of the C6H are from the CPU itself?

I have an inverted case......
Its the ones nearest the bios buttons.
Incidentally, they work well with an HTC Vive. Means you dont have to buy an extra Inateck USB3 pcie card with the Fresco chipset if you want your VR kit to work reliably.
The other USB ports are just as crap as usual but its a step in the right direction. Thanks AMD!
post #9354 of 17435
Quote:
Originally Posted by Timur Born View Post

There are two possible interpretations for what happens with ITB AVX:

Three consecutive +10 C offsets are applied one after another, for a total of +30 C on top of the fixed +20 C for X CPUs. The first two happen in direct succession when memory allocation starts, the last one happens when Linpack AVX calculations start.

1. These could be pure offsets and the real temperature would be 50 C lower on a X CPU. This would be lower than the socket sensor, though.

2. Or they could be offsets only for the first few milliseconds to seconds until the real temperature finally catches up to the offset value and then takes over. If this was the case then these offsets could be seen as kind of "jump start" to ramp up fans before the real temps kick in.

Here is a screenshot of what happens when ITB AVX is started. Take a good look at the moment when Linpack calculations start! First it jumps by the +10 C offset and then it decreases at first, before catching up again.

Sorry for using a different scale on the socket sensor. Watch out when you compare the temps (0-100 socket, 20-100 Tctl/Tdie).


I'm on a 1700 so I dont know what to make of the data. Did the stuff I posted regarding your question earlier (non-x with everything disabled) help you confirm anything?
Inferior FTW!
(22 items)
 
FrankenRyzen
(13 items)
 
 
CPUMotherboardGraphicsRAM
i5 6600K 4.5@1.320v Gigabyte Z170X-Gaming 7 Founder's GTX 1070 200/800 Corsair Vengeance LED 3200@4000 
Hard DriveCoolingCoolingCooling
MyDigitalSSD BPX 120 (RAID 0) EK-FB GA Z170X Monoblock XSPC RX360 (White) XSPC RX240 (White) 
CoolingCoolingCoolingCooling
EK Vardar F2s (x5) EK-FC1070 GTX1070 (nickel/plexi)  EK GTX1070 FE backplate EK Double Bay PWM D5 MX 
CoolingCoolingOSMonitor
Bitspower 250 Z-Multi Res EK HDC fittings (12mm) Windows 10 Samsung C32F391 
KeyboardPowerCaseMouse
Razer Blackwidow Chroma EVGA Supernova G1 1000w Phanteks Enthoo Primo Razer Naga Epic 
Mouse PadAudio
Razer Vespula Sennheiser HD598 
CPUMotherboardGraphicsRAM
Ryzen R7 1700 @3.9 Asus Crosshair VI Hero Radeon HD 6950 G.skill RGB 3200 @3500 C16 
Hard DriveCoolingOSMonitor
Samsung 960 Evo CM 212 Evo (push/pull) Windows 10 Pro 46" Sharp Aquos 
KeyboardPowerCaseMouse
Razer Anansi 750w Keian CM ATCS 840 Logitech G300 
Mouse Pad
My desk... 
  hide details  
Reply
Inferior FTW!
(22 items)
 
FrankenRyzen
(13 items)
 
 
CPUMotherboardGraphicsRAM
i5 6600K 4.5@1.320v Gigabyte Z170X-Gaming 7 Founder's GTX 1070 200/800 Corsair Vengeance LED 3200@4000 
Hard DriveCoolingCoolingCooling
MyDigitalSSD BPX 120 (RAID 0) EK-FB GA Z170X Monoblock XSPC RX360 (White) XSPC RX240 (White) 
CoolingCoolingCoolingCooling
EK Vardar F2s (x5) EK-FC1070 GTX1070 (nickel/plexi)  EK GTX1070 FE backplate EK Double Bay PWM D5 MX 
CoolingCoolingOSMonitor
Bitspower 250 Z-Multi Res EK HDC fittings (12mm) Windows 10 Samsung C32F391 
KeyboardPowerCaseMouse
Razer Blackwidow Chroma EVGA Supernova G1 1000w Phanteks Enthoo Primo Razer Naga Epic 
Mouse PadAudio
Razer Vespula Sennheiser HD598 
CPUMotherboardGraphicsRAM
Ryzen R7 1700 @3.9 Asus Crosshair VI Hero Radeon HD 6950 G.skill RGB 3200 @3500 C16 
Hard DriveCoolingOSMonitor
Samsung 960 Evo CM 212 Evo (push/pull) Windows 10 Pro 46" Sharp Aquos 
KeyboardPowerCaseMouse
Razer Anansi 750w Keian CM ATCS 840 Logitech G300 
Mouse Pad
My desk... 
  hide details  
Reply
post #9355 of 17435
Quote:
Originally Posted by Timur Born View Post

Tctl vs. Windows power profiles:



Some nice +10 C spikes during (mostly) idle times there. Assuming that the reported temps are indeed +20 C higher than real due to a fixed offset, this means that there is another +10 C offset dynamically applied. Since the load did neither vary by amount nor type (as in CPU instructions) I wonder why power profiles have such a big impact on the Tctl behavior!?

It's noteworthy that CPU (Socket) temps don't budge more than 1 C bit over the whole time-span.

Besides the conjecture that the 10C addition is due to some algorithm, there could be another reason for the seemingly constant jump; the measurement being displayed is from a sensing diode that is in the silicon where there is (locally) low thermal mass, and it responds immediately to some particular adjacent circuit that is briefly exercised intermittently. When the circuit activity ends, the temperature decays back to the baseline as the heat spreads out to less local thermal mass.

This alternate conjecture somewhat depends on the time scale of the plots vs. silicon chip thermal characteristics we won't have. One would want to look at the thermal decay time from a deliberately larger and twice as long processing burst. But it might be impossible to determine anything from such a test if the displayed temperature diode source jumps around as the processing uses different cores. Seven core parking may be needed.
     
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 #9356 of 17435
Quote:
Originally Posted by madweazl View Post

I'm on a 1700 so I dont know what to make of the data. Did the stuff I posted regarding your question earlier (non-x with everything disabled) help you confirm anything?
Yes, thank you. If you are idling at around 20 C then I am idling at around 20 C, so the +20 C offset is always present. The other "offsets" or maybe "jump starts" happen on top of that.

The question is why you end up at around the same 80 C with IBT AVX as I do, since you should be 20 C lower on Tctl. But I am currently running my AIO at 100%, so maybe yours just runs lower atm? I suggest you turn on the graphs, set polling to 100 - 200 ms and watch for those jumps.

Here is a better comparison of Tdie (Tctl - 20) vs. Socket. As you can see socket temps stabilize and remain about 15 C lower than Tdie. Asus uses Socker + 30 C for their SIO CPU calculation - once Tctl increases to the moon - so I wonder about their reasoning for that big offset?!

post #9357 of 17435
Quote:
Originally Posted by Timur Born View Post

There are two possible interpretations for what happens with ITB AVX:

Three consecutive +10 C offsets are applied one after another, for a total of +30 C on top of the fixed +20 C for X CPUs. The first two happen in direct succession when memory allocation starts, the last one happens when Linpack AVX calculations start.

1. These could be pure offsets and the real temperature would be 50 C lower on a X CPU. This would be lower than the socket sensor, though.

2. Or they could be offsets only for the first few milliseconds to seconds until the real temperature finally catches up to the offset value and then takes over. If this was the case then these offsets could be seen as kind of "jump start" to ramp up fans before the real temps kick in.

Here is a screenshot of what happens when ITB AVX is started. Take a good look at the moment when Linpack calculations start! First it jumps by the +10 C offset and then it decreases at first, before catching up again.

Sorry for using a different scale on the socket sensor. Watch out when you compare the temps (0-100 socket, 20-100 Tctl/Tdie).


Notice that the CPU socket behaves as expected with an exponential decay, while the other plots have a linear decay, not typical of REAL materials. So one is forced to conclude that some function is deliberately messing with the data, perhaps to keep the fan speed up longer as suggested.
     
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 #9358 of 17435
Quote:
Originally Posted by Timur Born View Post

Yes, thank you. If you are idling at around 20 C then I am idling at around 20 C, so the +20 C offset is always present. The other "offsets" or maybe "jump starts" happen on top of that.

The question is why you end up at around the same 80 C with IBT AVX as I do, since you should be 20 C lower on Tctl. But I am currently running my AIO at 100%, so maybe yours just runs lower atm? I suggest you turn on the graphs, set polling to 100 - 200 ms and watch for those jumps.

Here is a better comparison of Tdie (Tctl - 20) vs. Socket. As you can see socket temps stabilize and remain about 15 C lower than Tdie. Asus uses Socker + 30 C for their SIO CPU calculation - once Tctl increases to the moon - so I wonder about their reasoning for that big offset?!


I'm on air so there is likely a large delta between what your cooler is doing compared to mine.

Edit: and I have no idea which IBT I'm running LOL.
Inferior FTW!
(22 items)
 
FrankenRyzen
(13 items)
 
 
CPUMotherboardGraphicsRAM
i5 6600K 4.5@1.320v Gigabyte Z170X-Gaming 7 Founder's GTX 1070 200/800 Corsair Vengeance LED 3200@4000 
Hard DriveCoolingCoolingCooling
MyDigitalSSD BPX 120 (RAID 0) EK-FB GA Z170X Monoblock XSPC RX360 (White) XSPC RX240 (White) 
CoolingCoolingCoolingCooling
EK Vardar F2s (x5) EK-FC1070 GTX1070 (nickel/plexi)  EK GTX1070 FE backplate EK Double Bay PWM D5 MX 
CoolingCoolingOSMonitor
Bitspower 250 Z-Multi Res EK HDC fittings (12mm) Windows 10 Samsung C32F391 
KeyboardPowerCaseMouse
Razer Blackwidow Chroma EVGA Supernova G1 1000w Phanteks Enthoo Primo Razer Naga Epic 
Mouse PadAudio
Razer Vespula Sennheiser HD598 
CPUMotherboardGraphicsRAM
Ryzen R7 1700 @3.9 Asus Crosshair VI Hero Radeon HD 6950 G.skill RGB 3200 @3500 C16 
Hard DriveCoolingOSMonitor
Samsung 960 Evo CM 212 Evo (push/pull) Windows 10 Pro 46" Sharp Aquos 
KeyboardPowerCaseMouse
Razer Anansi 750w Keian CM ATCS 840 Logitech G300 
Mouse Pad
My desk... 
  hide details  
Reply
Inferior FTW!
(22 items)
 
FrankenRyzen
(13 items)
 
 
CPUMotherboardGraphicsRAM
i5 6600K 4.5@1.320v Gigabyte Z170X-Gaming 7 Founder's GTX 1070 200/800 Corsair Vengeance LED 3200@4000 
Hard DriveCoolingCoolingCooling
MyDigitalSSD BPX 120 (RAID 0) EK-FB GA Z170X Monoblock XSPC RX360 (White) XSPC RX240 (White) 
CoolingCoolingCoolingCooling
EK Vardar F2s (x5) EK-FC1070 GTX1070 (nickel/plexi)  EK GTX1070 FE backplate EK Double Bay PWM D5 MX 
CoolingCoolingOSMonitor
Bitspower 250 Z-Multi Res EK HDC fittings (12mm) Windows 10 Samsung C32F391 
KeyboardPowerCaseMouse
Razer Blackwidow Chroma EVGA Supernova G1 1000w Phanteks Enthoo Primo Razer Naga Epic 
Mouse PadAudio
Razer Vespula Sennheiser HD598 
CPUMotherboardGraphicsRAM
Ryzen R7 1700 @3.9 Asus Crosshair VI Hero Radeon HD 6950 G.skill RGB 3200 @3500 C16 
Hard DriveCoolingOSMonitor
Samsung 960 Evo CM 212 Evo (push/pull) Windows 10 Pro 46" Sharp Aquos 
KeyboardPowerCaseMouse
Razer Anansi 750w Keian CM ATCS 840 Logitech G300 
Mouse Pad
My desk... 
  hide details  
Reply
post #9359 of 17435
Quote:
Originally Posted by kaseki View Post

Besides the conjecture that the 10C addition is due to some algorithm, there could be another reason for the seemingly constant jump; the measurement being displayed is from a sensing diode that is in the silicon where there is (locally) low thermal mass, and it responds immediately to some particular adjacent circuit that is briefly exercised intermittently. When the circuit activity ends, the temperature decays back to the baseline as the heat spreads out to less local thermal mass.
I can follow that argument, but still doubt that this would lead to quite exact jumps of +10 C steps, regardless of what temperature the jump was based on (starting at 20 C looks the same as starting at 60 C) and how the cooling is set (100% cooling looks the same as 0%).

And why would there be no jumps for idle load using the Power Safer profile versus all the other profiles? It's still the same load happening at the same places inside the silicon. Do also note that the two spikes in the first Power Safe segment are slightly larger jumps than +10 C, because there was real CPU load happening. This likely was some +2 C due to load and then the +10 offset happened on top of that.
Edited by Timur Born - 4/14/17 at 5:38pm
post #9360 of 17435
Quote:
Originally Posted by Raja@ASUS View Post

SOC @ 1.15V
DRAM Voltage @ 1.40V

DRAM timings as shown. 2666 memory ratio with 116.2 BCLK.


HTH
cool, with your settings got my 2x16 Corsair cl16 at 14-14-14-34 stable (ibt standard and high)
New Posts  All Forums:Forum Nav:
  Return Home
  Back to Forum: AMD Motherboards
Overclock.net › Forums › AMD › AMD Motherboards › ROG Crosshair VI overclocking thread