Overclock.net › Forums › Graphics Cards › AMD/ATI › [Official] Vega Frontier / RX Vega Owners Thread
New Posts  All Forums:Forum Nav:

[Official] Vega Frontier / RX Vega Owners Thread - Page 279

post #2781 of 5164
Tried using clockblocker, it can put it at p7(actually a bit higher) but drops once superposition starts. Also, for some reason it keeps resetting the p6 state.
    
CPUMotherboardGraphicsRAM
e2140@3.2Ghz abit IP35-E HIS IceQ4 4850 4GB 667@800 5-5-5-15 
OSPower
win xp 32 bit Corsair 450VX@stock 
  hide details  
Reply
    
CPUMotherboardGraphicsRAM
e2140@3.2Ghz abit IP35-E HIS IceQ4 4850 4GB 667@800 5-5-5-15 
OSPower
win xp 32 bit Corsair 450VX@stock 
  hide details  
Reply
post #2782 of 5164
[
Quote:
Originally Posted by WannaBeOCer View Post

Also enabled HBCC and scores skyrocketed thumb.gif

How much ram did you add?
post #2783 of 5164
I see similar issues with Firestrike. Seems like more voltage is needed than in any other Game or bench. With 17.9.3 it seems more stable for me and I can run Firestrike with these settings: P7 1670 with 1150mv P6 with 1570\1085mv. HBM 1020@970mv. Core clock of 1622-1640mhz during bench, 290watt powerdraw. 24.400 Points graphics score.
@Trixx: LC Bios was not stable for me :/
post #2784 of 5164
I went ahead and purchased 3d mark advanced as I was running the free basic edition. The program runs better and more stable and I don't have to put up with the demo.
Was able to run Timespy @ lower volts but not Firestrike. Similiar results to elox and as per usual.
What I noticed with Superposition is that if you lower your p7 volts then it runs at lower mhz clock speed as if to compensate. Now I don't think it's Vega doing this but the Superposition program and if not then 3dmark certainly isn't doing the same thing.
Edited by raysheri - 10/7/17 at 4:22am
post #2785 of 5164
my best firestrike is 26 112, but superposition runs poor with the same clock not even 6567 4k test. I dont know why I cant hit those 7k score like you guys hit if I can get this much on firestrike. Something is weird with superposition benchmark.
I7 6700K
(15 items)
 
  
CPUMotherboardGraphicsRAM
I7 6700k Gigabyte Z170x-UD5 Rx Vega 64 G.Skill Ripjaws F4-3200C16-16GVK 
Hard DriveHard DriveCoolingOS
Samsung 256 830 serie Samsung 500 850 serie M-2 Custom Water Loop , 2 GPU block , 1 CPU block, ... Win7 64 / Win 10 64 
MonitorKeyboardPowerCase
3X LG 24 inch (2ms) Logitech G810 XFX 1250W  Bitfenix Shinobi XL modded 
MouseMouse PadAudio
Logitech G502 Logitech Hard Pad ( G440) Xonar dx to a 5.1 system 
  hide details  
Reply
I7 6700K
(15 items)
 
  
CPUMotherboardGraphicsRAM
I7 6700k Gigabyte Z170x-UD5 Rx Vega 64 G.Skill Ripjaws F4-3200C16-16GVK 
Hard DriveHard DriveCoolingOS
Samsung 256 830 serie Samsung 500 850 serie M-2 Custom Water Loop , 2 GPU block , 1 CPU block, ... Win7 64 / Win 10 64 
MonitorKeyboardPowerCase
3X LG 24 inch (2ms) Logitech G810 XFX 1250W  Bitfenix Shinobi XL modded 
MouseMouse PadAudio
Logitech G502 Logitech Hard Pad ( G440) Xonar dx to a 5.1 system 
  hide details  
Reply
post #2786 of 5164
Quote:
Originally Posted by GroupB View Post

my best firestrike is 26 112, but superposition runs poor with the same clock not even 6567 4k test. I dont know why I cant hit those 7k score like you guys hit if I can get this much on firestrike. Something is weird with superposition benchmark.
HBCC on for Superposition = ~10% extra score.

I'll grab the paid version of Timespy/Firestrike soon and give that a good run to see if it's more stable. TBH though I don't think Superposition is odd, but I think Firestrike is outdated. Timespy seems to run smoother than FS and ultimately more stable with the demo version at least.
    
CPUMotherboardGraphicsRAM
AMD Ryzen ThreadRipper 1950X ASUS ROG ZENITH EXTREME AMD Radeon RX Vega64 Corsair 32GB 3200MHz DDR4 
Hard DriveCoolingOSMonitor
Samsung 850 EVO SSD Noctua U14S ThreadRipper Edition Win 10 Enterprise Samsung BX2450 x2 (soon to change) 
KeyboardPowerCaseMouse
Roccat RYOS MK Pro Corsair AX1200i In-Win 909 Black Swiftpoint Z-Mouse 
Mouse PadAudio
Razer Ironclad Creative Sound Blaster X AE-5 
  hide details  
Reply
    
CPUMotherboardGraphicsRAM
AMD Ryzen ThreadRipper 1950X ASUS ROG ZENITH EXTREME AMD Radeon RX Vega64 Corsair 32GB 3200MHz DDR4 
Hard DriveCoolingOSMonitor
Samsung 850 EVO SSD Noctua U14S ThreadRipper Edition Win 10 Enterprise Samsung BX2450 x2 (soon to change) 
KeyboardPowerCaseMouse
Roccat RYOS MK Pro Corsair AX1200i In-Win 909 Black Swiftpoint Z-Mouse 
Mouse PadAudio
Razer Ironclad Creative Sound Blaster X AE-5 
  hide details  
Reply
post #2787 of 5164
Superposition seems to stress the card more, but firestrike ultra stressed it even more. The best test of stability I use is one of the levels in crysis 3 which digitalfoundry use in their reviews, it crashes pretty quickly there even if it has passed those two benchmarks easily.

The card probably gets stuck at lower p-states if it doesn't get enough power for the clocks at the given voltage, I get better results with auto settings on clocks and voltage once that happens. This is almost always the case with superposition once you get past the initial scene.
    
CPUMotherboardGraphicsRAM
e2140@3.2Ghz abit IP35-E HIS IceQ4 4850 4GB 667@800 5-5-5-15 
OSPower
win xp 32 bit Corsair 450VX@stock 
  hide details  
Reply
    
CPUMotherboardGraphicsRAM
e2140@3.2Ghz abit IP35-E HIS IceQ4 4850 4GB 667@800 5-5-5-15 
OSPower
win xp 32 bit Corsair 450VX@stock 
  hide details  
Reply
post #2788 of 5164
Quote:
Originally Posted by gamervivek View Post

Superposition seems to stress the card more, but firestrike ultra stressed it even more. The best test of stability I use is one of the levels in crysis 3 which digitalfoundry use in their reviews, it crashes pretty quickly there even if it has passed those two benchmarks easily.

The card probably gets stuck at lower p-states if it doesn't get enough power for the clocks at the given voltage, I get better results with auto settings on clocks and voltage once that happens. This is almost always the case with superposition once you get past the initial scene.

The P5 sticking problem is annoying, I've resorted to using ClockBlocker to force a P7 state in testing. Gives me more accurate results to work with.

What level of Crysis 3 is it that gives you the crash? I'd be interested to test that out smile.gif
    
CPUMotherboardGraphicsRAM
AMD Ryzen ThreadRipper 1950X ASUS ROG ZENITH EXTREME AMD Radeon RX Vega64 Corsair 32GB 3200MHz DDR4 
Hard DriveCoolingOSMonitor
Samsung 850 EVO SSD Noctua U14S ThreadRipper Edition Win 10 Enterprise Samsung BX2450 x2 (soon to change) 
KeyboardPowerCaseMouse
Roccat RYOS MK Pro Corsair AX1200i In-Win 909 Black Swiftpoint Z-Mouse 
Mouse PadAudio
Razer Ironclad Creative Sound Blaster X AE-5 
  hide details  
Reply
    
CPUMotherboardGraphicsRAM
AMD Ryzen ThreadRipper 1950X ASUS ROG ZENITH EXTREME AMD Radeon RX Vega64 Corsair 32GB 3200MHz DDR4 
Hard DriveCoolingOSMonitor
Samsung 850 EVO SSD Noctua U14S ThreadRipper Edition Win 10 Enterprise Samsung BX2450 x2 (soon to change) 
KeyboardPowerCaseMouse
Roccat RYOS MK Pro Corsair AX1200i In-Win 909 Black Swiftpoint Z-Mouse 
Mouse PadAudio
Razer Ironclad Creative Sound Blaster X AE-5 
  hide details  
Reply
post #2789 of 5164
It's the jungle stage where you ram the train through a wall, it loads the card well enough, earlier stages are more cpu limited for me. Another reason I think for the instability is varying loads, superposition loads the card constantly but games are more variable and likely to crash.
    
CPUMotherboardGraphicsRAM
e2140@3.2Ghz abit IP35-E HIS IceQ4 4850 4GB 667@800 5-5-5-15 
OSPower
win xp 32 bit Corsair 450VX@stock 
  hide details  
Reply
    
CPUMotherboardGraphicsRAM
e2140@3.2Ghz abit IP35-E HIS IceQ4 4850 4GB 667@800 5-5-5-15 
OSPower
win xp 32 bit Corsair 450VX@stock 
  hide details  
Reply
post #2790 of 5164
Quote:
Originally Posted by gamervivek View Post

It's the jungle stage where you ram the train through a wall, it loads the card well enough, earlier stages are more cpu limited for me. Another reason I think for the instability is varying loads, superposition loads the card constantly but games are more variable and likely to crash.
Well after pissing about a bit for a couple of hours, can't get that to crash mine with my current settings. Though after watching DF's video on Vega64 and having done tons of testing I think either my card is just insane or an outlier but I'm not seeing the low performance they did. Admittedly I'm not running stock clocks and running a very undervolted 24/7 setup as well as clockblocker to eliminate the P5 locking and scene switch frame dropping (very noticeable when switching to scene 10 in Superposition).

Running Clockblocker maybe why I'm not seeing the instability with P state switches too.
    
CPUMotherboardGraphicsRAM
AMD Ryzen ThreadRipper 1950X ASUS ROG ZENITH EXTREME AMD Radeon RX Vega64 Corsair 32GB 3200MHz DDR4 
Hard DriveCoolingOSMonitor
Samsung 850 EVO SSD Noctua U14S ThreadRipper Edition Win 10 Enterprise Samsung BX2450 x2 (soon to change) 
KeyboardPowerCaseMouse
Roccat RYOS MK Pro Corsair AX1200i In-Win 909 Black Swiftpoint Z-Mouse 
Mouse PadAudio
Razer Ironclad Creative Sound Blaster X AE-5 
  hide details  
Reply
    
CPUMotherboardGraphicsRAM
AMD Ryzen ThreadRipper 1950X ASUS ROG ZENITH EXTREME AMD Radeon RX Vega64 Corsair 32GB 3200MHz DDR4 
Hard DriveCoolingOSMonitor
Samsung 850 EVO SSD Noctua U14S ThreadRipper Edition Win 10 Enterprise Samsung BX2450 x2 (soon to change) 
KeyboardPowerCaseMouse
Roccat RYOS MK Pro Corsair AX1200i In-Win 909 Black Swiftpoint Z-Mouse 
Mouse PadAudio
Razer Ironclad Creative Sound Blaster X AE-5 
  hide details  
Reply
New Posts  All Forums:Forum Nav:
  Return Home
  Back to Forum: AMD/ATI
Overclock.net › Forums › Graphics Cards › AMD/ATI › [Official] Vega Frontier / RX Vega Owners Thread