Jump to content

850w PSU causing Vega 64 to crash?

trojanhorse

I have a Corsair AX860 with a Nitro+ Vega 64.

 

I uninstalled with DDU and reinstalled 19.17.5 drivers from AMD.

 

My memory is 2x dual rank Corsair 3200 AT e-die running at 3200cl16 based on Ryzen DRAM Safe calculations.

 

I also got a seperate PCIE cable for the 8pin GPU power so that both the power inputs are no longer daisy chained (this solved most of the errors).

 

The slight underclock and 150% power limit mod causes my system to crash after 10 minutes of RS6 Siege, which isn't even an intensive game.

 

The PSU is a few years old though, think its at least 5+ years. Should I replace it?

Link to comment
Share on other sites

Link to post
Share on other sites

I mean, it's a fine PSU, but it sounds like your issues are GPU related. I am un-familiar with the 150% power limit mod,  but if it is undo-able then try undoing it?

Brands I wholeheartedly reccomend (though do have flawed products): Apple, Razer, Corsair, Asus, Gigabyte, bequiet!, Noctua, Fractal, GSkill (RAM only)

Wall Of Fame (Informative people/People I like): @Glenwing @DrMacintosh @Schnoz @TempestCatto @LogicalDrm @Dan Castellaneta

Useful threads: 

How To Make Your Own Cloud Storage

Spoiler

 

Guide to Display Cables/Adapters

Spoiler

 

PSU Tier List (Latest)-

Spoiler

 

 

Main PC: See spoiler tag

Laptop: 2020 iPad Pro 12.9" with Magic Keyboard

Spoiler

PCPartPicker Part List: https://pcpartpicker.com/list/gKh8zN

CPU: AMD Ryzen 9 3900X 3.8 GHz 12-Core OEM/Tray Processor  (Purchased For $419.99) 
Motherboard: Asus ROG Crosshair VIII Formula ATX AM4 Motherboard  (Purchased For $356.99) 
Memory: G.Skill Trident Z RGB 32 GB (2 x 16 GB) DDR4-3000 Memory  (Purchased For $130.00) 
Storage: Kingston Predator 240 GB M.2-2280 NVME Solid State Drive  (Purchased For $40.00) 
Storage: Crucial MX300 1.05 TB 2.5" Solid State Drive  (Purchased For $100.00) 
Storage: Western Digital Red 8 TB 3.5" 5400RPM Internal Hard Drive  (Purchased For $180.00) 
Video Card: Gigabyte GeForce RTX 2070 8 GB WINDFORCE Video Card  (Purchased For $370.00) 
Case: Fractal Design Define R6 USB-C ATX Mid Tower Case  (Purchased For $100.00) 
Power Supply: Corsair RMi 1000 W 80+ Gold Certified Fully Modular ATX Power Supply  (Purchased For $120.00) 
Optical Drive: Asus DRW-24B1ST/BLK/B/AS DVD/CD Writer  (Purchased For $75.00) 
Total: $1891.98
Prices include shipping, taxes, and discounts when available
Generated by PCPartPicker 2020-04-02 19:59 EDT-0400

身のなわたしはる果てぞ  悲しわたしはかりけるわたしは

Link to comment
Share on other sites

Link to post
Share on other sites

10 minutes ago, SenKa said:

I mean, it's a fine PSU, but it sounds like your issues are GPU related. I am un-familiar with the 150% power limit mod,  but if it is undo-able then try undoing it?

yeah I mean I think its the power limit... any idea why it crashes? Is it the PSU causing unable to take 150w power?

Link to comment
Share on other sites

Link to post
Share on other sites

3 minutes ago, trojanhorse said:

yeah I mean I think its the power limit... any idea why it crashes? Is it the PSU causing unable to take 150w power?

I think your GPU just doesn't like the higher boost that the massive power target allows.

Brands I wholeheartedly reccomend (though do have flawed products): Apple, Razer, Corsair, Asus, Gigabyte, bequiet!, Noctua, Fractal, GSkill (RAM only)

Wall Of Fame (Informative people/People I like): @Glenwing @DrMacintosh @Schnoz @TempestCatto @LogicalDrm @Dan Castellaneta

Useful threads: 

How To Make Your Own Cloud Storage

Spoiler

 

Guide to Display Cables/Adapters

Spoiler

 

PSU Tier List (Latest)-

Spoiler

 

 

Main PC: See spoiler tag

Laptop: 2020 iPad Pro 12.9" with Magic Keyboard

Spoiler

PCPartPicker Part List: https://pcpartpicker.com/list/gKh8zN

CPU: AMD Ryzen 9 3900X 3.8 GHz 12-Core OEM/Tray Processor  (Purchased For $419.99) 
Motherboard: Asus ROG Crosshair VIII Formula ATX AM4 Motherboard  (Purchased For $356.99) 
Memory: G.Skill Trident Z RGB 32 GB (2 x 16 GB) DDR4-3000 Memory  (Purchased For $130.00) 
Storage: Kingston Predator 240 GB M.2-2280 NVME Solid State Drive  (Purchased For $40.00) 
Storage: Crucial MX300 1.05 TB 2.5" Solid State Drive  (Purchased For $100.00) 
Storage: Western Digital Red 8 TB 3.5" 5400RPM Internal Hard Drive  (Purchased For $180.00) 
Video Card: Gigabyte GeForce RTX 2070 8 GB WINDFORCE Video Card  (Purchased For $370.00) 
Case: Fractal Design Define R6 USB-C ATX Mid Tower Case  (Purchased For $100.00) 
Power Supply: Corsair RMi 1000 W 80+ Gold Certified Fully Modular ATX Power Supply  (Purchased For $120.00) 
Optical Drive: Asus DRW-24B1ST/BLK/B/AS DVD/CD Writer  (Purchased For $75.00) 
Total: $1891.98
Prices include shipping, taxes, and discounts when available
Generated by PCPartPicker 2020-04-02 19:59 EDT-0400

身のなわたしはる果てぞ  悲しわたしはかりけるわたしは

Link to comment
Share on other sites

Link to post
Share on other sites

21 minutes ago, SenKa said:

I think your GPU just doesn't like the higher boost that the massive power target allows.

 

It still crashes on 50% Power, I will run it stock now. What's the quickest way to test? Furmark/OCCT take too long. Playing actual games causes errors faster.

Link to comment
Share on other sites

Link to post
Share on other sites

30 minutes ago, SenKa said:

I think your GPU just doesn't like the higher boost that the massive power target allows.

Update: Crashes within 10 minutes of OCCT.

Link to comment
Share on other sites

Link to post
Share on other sites

150% power limit amd cards dont like it, even my 5700xt starts stuttering n unstable on 150% , stable on 125% to 135% tho   ,  then again    if u crash on somthing as weak as occt, its a very unstable pc

 

1 minute ago, trojanhorse said:

Update: Crashes within 10 minutes of OCCT.

 

a pc should be stable on even prime95 and thats much more stress heavy then occt so u for sure have problems

 

did u manually set ram timings to try once?  maybe auto is grabbing 1t command rate and ur sticks/setup can only handle 2t

(◑‿◐)

Link to comment
Share on other sites

Link to post
Share on other sites

11 minutes ago, Valkyrie Lenneth said:

150% power limit amd cards dont like it, even my 5700xt starts stuttering n unstable on 150% , stable on 125% to 135% tho   ,  then again    if u crash on somthing as weak as occt, its a very unstable pc

 

 

a pc should be stable on even prime95 and thats much more stress heavy then occt so u for sure have problems

 

did u manually set ram timings to try once?  maybe auto is grabbing 1t command rate and ur sticks/setup can only handle 2t

 

i was using occt power supply test which is basically just furmark and a cpu test in the bg. i did my own timings off the calc, i can try 2t but this is definitely a gpu problem as everything runs fine now with power save mode enabled (-25%).

 

its a massive loss in fps tho. do you know if that means the gpu is faulty?

Link to comment
Share on other sites

Link to post
Share on other sites

7 minutes ago, trojanhorse said:

 

i was using occt power supply test which is basically just furmark and a cpu test in the bg. i did my own timings off the calc, i can try 2t but this is definitely a gpu problem as everything runs fine now with power save mode enabled (-25%).

 

its a massive loss in fps tho. do you know if that means the gpu is faulty?

faulty gpu can make it stable on power save mode too so

 

try downgrading drivers tho, amd have alot of unstable drivers around, i had same problem

 

jus tkeep in mind my new 5700xt that clocks to 2.1ghz also crashes on 150% power limit, so dont run with 150% power limit itl crash guaranteed

(◑‿◐)

Link to comment
Share on other sites

Link to post
Share on other sites

39 minutes ago, Valkyrie Lenneth said:

faulty gpu can make it stable on power save mode too so

 

try downgrading drivers tho, amd have alot of unstable drivers around, i had same problem

 

jus tkeep in mind my new 5700xt that clocks to 2.1ghz also crashes on 150% power limit, so dont run with 150% power limit itl crash guaranteed

Even power save mode crashes now (1hr in OCCT).

 

Will downgrading help even now that it crashes at power save? How far should I downgrade? Try the previous 3 or go further?

Link to comment
Share on other sites

Link to post
Share on other sites

46 minutes ago, Valkyrie Lenneth said:

faulty gpu can make it stable on power save mode too so

 

try downgrading drivers tho, amd have alot of unstable drivers around, i had same problem

 

jus tkeep in mind my new 5700xt that clocks to 2.1ghz also crashes on 150% power limit, so dont run with 150% power limit itl crash guaranteed

People upped the limit to 150% with 5700 because thats how you oc'd vega.

Link to comment
Share on other sites

Link to post
Share on other sites

28 minutes ago, trojanhorse said:

Even power save mode crashes now (1hr in OCCT).

 

Will downgrading help even now that it crashes at power save? How far should I downgrade? Try the previous 3 or go further?

if occt already crashes thers a big problem, try running prime95 , on small fft, if thats stable cpu is fine, large fft for ram test, and if those pass both, its ur gpu or psu, however if i would have any problems related to a psu i would instantly take that thing out of my system, bad psu can blow up ur entire pc

(◑‿◐)

Link to comment
Share on other sites

Link to post
Share on other sites

2 hours ago, Valkyrie Lenneth said:

if occt already crashes thers a big problem, try running prime95 , on small fft, if thats stable cpu is fine, large fft for ram test, and if those pass both, its ur gpu or psu, however if i would have any problems related to a psu i would instantly take that thing out of my system, bad psu can blow up ur entire pc

 

I turned off my RAM OC now everything is stock, just ran optimised defualts in BIOS. I still crash under load, couldn't even finish a single game of Siege, and those matches are only 5-10mins long!

Link to comment
Share on other sites

Link to post
Share on other sites

11 minutes ago, trojanhorse said:

 

I turned off my RAM OC now everything is stock, just ran optimised defualts in BIOS. I still crash under load, couldn't even finish a single game of Siege, and those matches are only 5-10mins long!

its possible to be a bad stick also but then again i think u have gpu problem or psu, dont have a spare gpu to test?

(◑‿◐)

Link to comment
Share on other sites

Link to post
Share on other sites

47 minutes ago, Valkyrie Lenneth said:

its possible to be a bad stick also but then again i think u have gpu problem or psu, dont have a spare gpu to test?

Small FFT has been running for hours no fault - how can I a bad stick?

Link to comment
Share on other sites

Link to post
Share on other sites

2 minutes ago, trojanhorse said:

Small FFT has been running for hours no fault - how can I a bad stick?

because small fft = cpu only

 

it means ur cpu is stable,  but ram / gpu can be unstable ( or even mem controller in cpu )   but the cpu cores and cache itself are stable if small fft runs fine

(◑‿◐)

Link to comment
Share on other sites

Link to post
Share on other sites

31 minutes ago, Valkyrie Lenneth said:

because small fft = cpu only

 

it means ur cpu is stable,  but ram / gpu can be unstable ( or even mem controller in cpu )   but the cpu cores and cache itself are stable if small fft runs fine

 

ill run memtest overnight, what else cna i do to properly test my ram?

Link to comment
Share on other sites

Link to post
Share on other sites

13 minutes ago, trojanhorse said:

 

ill run memtest overnight, what else cna i do to properly test my ram?

run custom setup in prime filling up ur ram with large fft or blend , need minimum 12hour stable for ram check

 

+ memtest also stable for 4pass or more

(◑‿◐)

Link to comment
Share on other sites

Link to post
Share on other sites

22 hours ago, Valkyrie Lenneth said:

run custom setup in prime filling up ur ram with large fft or blend , need minimum 12hour stable for ram check

 

+ memtest also stable for 4pass or more

 

definitely the gpu man is there no way i can get this fixed? even running -150% power i still had crashes everythign running stocfk

Link to comment
Share on other sites

Link to post
Share on other sites

10 minutes ago, trojanhorse said:

 

definitely the gpu man is there no way i can get this fixed? even running -150% power i still had crashes everythign running stocfk

try older drivers, or different psu ,  try different pci-e slot, maybe pci-e slot is fucked ( saw it happen before )

 

if its rly based on gpu hardware defect its just screwed, prolly clocked it to death, its like 0.001% chance to blow it up but its possible

(◑‿◐)

Link to comment
Share on other sites

Link to post
Share on other sites

On 8/18/2019 at 10:52 PM, Valkyrie Lenneth said:

try older drivers, or different psu ,  try different pci-e slot, maybe pci-e slot is fucked ( saw it happen before )

 

if its rly based on gpu hardware defect its just screwed, prolly clocked it to death, its like 0.001% chance to blow it up but its possible

 

Fixed the issue. The second PCIE 8pin was on a different 12V line to the other 8pin. This was causing the power to trip or something, setting of overcurrent protection.

 

I'm gonna go balls to the walls OCing now. Do you have a guide for OCing the Nitro+ Vega 64? Currently voltage sits around 1000mv with 1125mhz mem and 1645core on 150% power. 

 

I am also trying to OC my RAM (Crucial AT 3200Mhz CL16). CL16 works fine from the DRAM calc, using those timings... but I can't get CL14 stable in HCI MemTest. It's mostly stable so do you know whats best to tweak? Something like raising/lowering tFAW might help the OC right?

Link to comment
Share on other sites

Link to post
Share on other sites

1 hour ago, trojanhorse said:

 

Fixed the issue. The second PCIE 8pin was on a different 12V line to the other 8pin. This was causing the power to trip or something, setting of overcurrent protection.

 

I'm gonna go balls to the walls OCing now. Do you have a guide for OCing the Nitro+ Vega 64? Currently voltage sits around 1000mv with 1125mhz mem and 1645core on 150% power. 

 

I am also trying to OC my RAM (Crucial AT 3200Mhz CL16). CL16 works fine from the DRAM calc, using those timings... but I can't get CL14 stable in HCI MemTest. It's mostly stable so do you know whats best to tweak? Something like raising/lowering tFAW might help the OC right?

dont oc ram if u dont even know that btw, its really easy to have ram unstable and clocking it for mhz cancels it out by cas, and tightening the kit requires proper knowledge of all timings, including secondary timings , ocing / tightening ram will only give unstability if u dont know all those ;f

 

if u rly want to tighten ram that badly get a better kit in the first place, coz 3200 cl16 is pretty basic / bad already

 

anyway good that u found the prob with the gpu :F

 

(◑‿◐)

Link to comment
Share on other sites

Link to post
Share on other sites

Create an account or sign in to comment

You need to be a member in order to leave a comment

Create an account

Sign up for a new account in our community. It's easy!

Register a new account

Sign in

Already have an account? Sign in here.

Sign In Now

×