• Competitor rules

    Please remember that any mention of competitors, hinting at competitors or offering to provide details of competitors will result in an account suspension. The full rules can be found under the 'Terms and Rules' link in the bottom right corner of your screen. Just don't mention competitors in any way, shape or form and you'll be OK.

The RX Vega 64 Owners Thread

I'm quite happy with my undervolted Vega64 with mem at 1050. CP2077 at 1440p with most stuff set to high, few things on medium and i'm getting between 30-50fps. it'd be ideal to be pushing more towards 60fps but I'm content enough to not want to rush out an buy a 3080 or whatever just yet. Havent tried the FidelityFX stuff just yet so might give that a whirl
 
I'm quite happy with my undervolted Vega64 with mem at 1050. CP2077 at 1440p with most stuff set to high, few things on medium and i'm getting between 30-50fps. it'd be ideal to be pushing more towards 60fps but I'm content enough to not want to rush out an buy a 3080 or whatever just yet. Havent tried the FidelityFX stuff just yet so might give that a whirl
Same here quite happy. 1700/1100 for myself 1440p medium settings 50-60fps chugging along nicely though had to restart after a fresh windows install,so only just restarted
 
Been using 1652/1180 which has been fine with some reduced settings. I did try resolution at 85% but found it looked a tad to soft to me.
 
I took one off but I can't identify it. The thread diameter is 1.95mm, and the pitch is about 0.6mm (hard to measure exactly because there are so few threads). It's too coarse to be a metric size.
 
I took one off but I can't identify it. The thread diameter is 1.95mm, and the pitch is about 0.6mm (hard to measure exactly because there are so few threads). It's too coarse to be a metric size.
cheers, guess its take it to a pc shop when open see if got one, as thought broke me fan as blade come away from housing, but after investigating its fine:cool: but still leaving off for time being as thermals better
 
which side is which for the bios switch on the nitro, atm its in furest away from gpu ports, i want to try flash a 56 onto it

thx

At least for reference cards I believe it's the switch towards the ports, for the 'faster' bios.
afaik it only changes the board power limit, so you can use something like GPUZ to see which position gives the highest reading.
Under Advanced tab, general.
 
Last edited:
Can anyone help with settings for undervolt please? I had a nice 975mV that I got from this thread once but now can't find it. It was faster than stock and only pulled 170/200W.
 
Can anyone help with settings for undervolt please? I had a nice 975mV that I got from this thread once but now can't find it. It was faster than stock and only pulled 170/200W.

Until AMD fix the clock discrepancy between Windows Fast Start Up being enabled and disabled, which translates to 25 MHz higher clocks when disabled (or just after a system restart), it's pretty difficult trying to provide accurate and stable under volt setting.

However, assuming Windows is in the default state, i.e. Fast Start UP enabled, the following should provide good results.
Vega-undervolt.jpg

Also, note that Vega voltages actually increase in 6.25 mV increments, and entered voltages just revert to the closest actual number at or above the figure entered. In the above scenario, actual voltages would be 950 mV, 956.25 mV, 962.5 mV, 968.75 mV, 975 mV etc., so 970 mV translates to 975 mV actual.
 
Last edited:
Morning,

Up until last week (before it got hot again) My Sapphire Radeon RX VEGA 64 Nitro+ 8GB HBM2 has run fine since I got the system in December 2018.

I don't overclock and the system is well placed for ventilation.

I am now getting black screen crashes playing RB6 and also on Titanfall2.

I assumed it was heat related but after getting the temps down (found the fan settings in radeon app) and also cold booted and launched game and it crashed straight away, these black screen crashes happen when it feels like but only in games.

It seems that degradation of the HYNIX HBMA memory could be the cause.

I need to ask OCUK if the card is still under warranty.

Is there anyway to get the card stable (undervolting) or is it on it's last legs?

Luckily, I kept my R9 290 card so that runs OK, no issues.

P.S.

System spec if that helps.

OcUK Gaming Zenith Gaming PC - AMD Ryzen 7 2700X

Sapphire Radeon RX VEGA 64 Nitro+ 8GB HBM2 PCI-Express Graphics Card
Intel 660P 512GB M.2-2280 PCI-e 3.0 x 4 NVMe QLC 3D NAND Solid State Drive (SSDPEKNW512G8X1)
Phanteks Eclipse P400S Glass Midi Tower Case - Noise Dampened Black/Red
Gigabyte Aorus RGB 16GB (2x8GB) DDR4 PC4-25600C16 3200MHz Dual/Quad Channel Kit

The PSU was replaced with an 850w after having some initial power issues after receiving the system.
 
It seems that degradation of the HYNIX HBMA memory could be the cause.
That is a bold pinpoint claim to make from these symptoms. Need specific tests that load memory and core separately to confirm.

But something has degraded. Could be as trivial as thermal pad obetween vrm and heatsink, could be some capacitors on the board. Could be the chip.

Not much you can do.
Check temps in GpuZ or HWInfo64, maybe spot something obvious.
For diagnostics also try lowering power slider in radeon app. That would achieve two goals, lower load on power delivery circuits and some headroom for temperatures to perhaps crash more slowly.

But in the end, prepare to spend a while with your old trusted friend R9 290
 
Back
Top Bottom