1. This site uses cookies. By continuing to use this site, you are agreeing to our use of cookies. Learn More.

The RX Vega 56 Owners Thread

Discussion in 'Graphics Cards' started by Kaapstad, Aug 14, 2017.

  1. Th0nt

    Soldato

    Joined: Jul 21, 2005

    Posts: 6,225

    Location: N.Ireland

    The power cable you using two different sources into that card? Do not daisy chain two from the same cable.
     
  2. Aurelijus1111

    Associate

    Joined: Jan 13, 2019

    Posts: 9

    I don't quite get what you're saying

    [​IMG]
     
  3. Kei

    Wise Guy

    Joined: Oct 24, 2008

    Posts: 2,383

    Location: South Wales

    You want to use two separate power supply cables and leave the loop out from the plug unused. Like in this picture, two separate cables and the loop outs from each are not connected.
    [​IMG]

    That said, it shouldn't really cause any issues having it the way you do until you stress the card as it should only pose a problem when the power draw is high.
     
    Last edited: Jan 13, 2019
  4. Aurelijus1111

    Associate

    Joined: Jan 13, 2019

    Posts: 9

    ehh, that will mean new psu, any recommendations, not fancy looking for high W, since im not a big fan of overclocking
     
  5. scrivz69

    Wise Guy

    Joined: Dec 6, 2008

    Posts: 2,013

    Location: York

    I will give it a try, thanks.
     
  6. Aurelijus1111

    Associate

    Joined: Jan 13, 2019

    Posts: 9

    Edit: I found my old psu box with all the accessories and there was a spear gpu power cable, i plugged one end into PSU, and other one into GPU as you guys showed, but I get the same results :/

    [​IMG]

    My PSU is EVGA 550GS supernova

    Pic from running on CPU gpu - (thats running on HDMI 1)
    [​IMG]

    And my GPU connected to Displayport to hit over 100hz -
    [​IMG]


    And after since the screen is black, it automatically goes back to HDMI
     
    Last edited: Jan 13, 2019
  7. Th0nt

    Soldato

    Joined: Jul 21, 2005

    Posts: 6,225

    Location: N.Ireland

    Have you disabled the igpu on the BIOS to force dedicated GPU?

    + check your BIOS version there may be an update to resolve issues.
     
  8. Aurelijus1111

    Associate

    Joined: Jan 13, 2019

    Posts: 9

    yea disabled igpu, bios updated, still nothing, spent all day trying to figure it out, had enough, gonna be sending the card back, and going for possibly rtx 2060
     
    Last edited: Jan 13, 2019
  9. welshrat

    Gangster

    Joined: Dec 31, 2011

    Posts: 210

    Well I have my card running pretty well now, undervolted to 1100mv at the top phase and 1075 on a P6 and P5, will try going lower over the next few months. Wattman seemed to have been affected by a setting in afterburner, which was

    (General Tab, Erase Autosaved Settings) I have now unchecked that and it seems to be holding values now.

    So all in all a nice upgrade over the 480 and will be a nice starting place when I build a Ryzen 3000 rig sometime after its release.

    Really nice to get back to PC gaming as I had an Xbox one X break 2 days before the warranty expired and rather than renew (already have a PS4 Pro and switch) I decided to upgrade the PC.

    Happy Days. :)
     
  10. Th0nt

    Soldato

    Joined: Jul 21, 2005

    Posts: 6,225

    Location: N.Ireland

    Shame that. I would try it in another rig before sending it back (if you got friend/relative who would help) but only concern would be even a quality 550w PSU, if its old and cannot provide the juice on the rail its never gonna play well.

    Did you flip the switch on the card to try it on the other BIOS?
     
  11. Aurelijus1111

    Associate

    Joined: Jan 13, 2019

    Posts: 9

    Yea, I'm thinking that perhaps the PSU may be the problem, but I can't figure it out how, it's a gold rated EVGA 550w(roughly 3 years old), pc is not overclocked, I'm running on 6600k, z170-m plus, couple ssd,1 hhd, 4 fans, the power should not be the problem, especially from the fact that I can run MSI r9 390 without any problem, which needs something like 280W, meanwhile vega 56 needs roughly what, 250? I wish I had a higher watt PSU to try it out.

    Also while surfing google looking for a solutions I've seen people saying that they use even a 400w psu to run vega 56, 500-550 is fine...
     
    Last edited: Jan 13, 2019
  12. welshrat

    Gangster

    Joined: Dec 31, 2011

    Posts: 210

    Seems unlikely, I just had a wattmeter on mine as I use a BeQuiet 630 Watt on mine, System was drawing 310 watts full load with 4790k at 4.3 ghz and 16gb mem, ssd etc etc,

    I guess it worth ruling out just in case though.
     
  13. nitram100

    Mobster

    Joined: Aug 7, 2009

    Posts: 4,270

    Location: London

    What's your system spec? It just about runs on my brothers 650 ocz modxstream.
     
  14. Aurelijus1111

    Associate

    Joined: Jan 13, 2019

    Posts: 9

    [​IMG]
     
  15. Th0nt

    Soldato

    Joined: Jul 21, 2005

    Posts: 6,225

    Location: N.Ireland

    Its not the wattage overall its can the PSU maintain a healthy voltage when stressed, I heard the Vega's are a bit fussy and dont like ripple - google the Seasonic Focus issue. However Aurelijus' problem is different as it doesnt yield a screen at all?

    Only time I had that was when the cable from GPU to monitor was bad. Swapped that and instantly got display so another thing to rule out.
     
  16. creesten

    Associate

    Joined: Nov 14, 2018

    Posts: 28

    Hi, I downgraded my AMD drivers to get the old fan curves back and stuff and decided to undervolt overclock my Sapphire Pulse 56.

    I'm testing P6 at 1577MHz and 1040mV and P7 at 1632MHz and 1070mV. Memory at 900MHz/1000mV. Power limit +50%. I test my clocks using fire strike with 3DMark. GPU-Z logs my power draw from the GPU. It seems to fluctuate around 205-210W at 99% usage, with very ocassional spikes to 220W. Few sensor readings are showing 1500+ W!!!!!! Is this even possible? At the same time, the temperature of VR SOC/ VR MEM shows at 1500C too. Is something broken, or are sensors just unreliable like that?

    I also noticed a slight coil whine. Doesn't bother me too much.
     
  17. DragonQ

    Soldato

    Joined: Jun 13, 2009

    Posts: 6,588

    Yes, something is broken. I noticed over the weekend my VRAM temperature was also spiking up to some ridiculous value in the thousands and also spiking down to 0 sometimes.

    P.S. You'll want to set your "memory voltage" to the same as your P6 voltage, since it's actually a voltage floor and nothing to do with the VRAM.
     
  18. Aurelijus1111

    Associate

    Joined: Jan 13, 2019

    Posts: 9

    ordered a 750w aerocool psu from amazon for 55 pounds for one last try, to see if PSU is actually the problem, in case it won't work, i'll just return it, fingers crossed I guess, since I got this vid card for so cheap comparing to its actual price
     
  19. nitram100

    Mobster

    Joined: Aug 7, 2009

    Posts: 4,270

    Location: London

    Definitely worth a shot mate, cracking card it is.
     
  20. creesten

    Associate

    Joined: Nov 14, 2018

    Posts: 28

    Hi peeps. Bad news. My second card started artifacting today, I only see it in rocket league so far. But that was the main game that showed faulty VRAM in my first card too.

    It only happens when I press "Training" and the dark background shows up. I get a quick blink of green and white'ish squares across the whole bottom of my screen. It's like a strip that always seems to be in a similar place. Happens every time I enter or exit a menu option with a dark background. I don't have many other games to test with yet.

    Oh no. Does anyone know how I can test the VRAM myself? So I don't have to go through that dreadful RMA process once again?