The RX 6000 series Owners' Club

[Nordic]

I am considering an upgrade from my 1060 3gb to a 6750xt for $400. Nearly 3x the performance for $400 seems decent.

I found an even better deal with a 6750 xt for $390 after rebate. I put the order in. This will be my first AMD GPU since my beloved 7970. Is there anything I should know about the Ryzen 6000 series?


My main rig has had the following in the last decade.
AMD Radeon HD 6950
AMD Radeon HD 7970
NV GTX 970
NV 1060 3gb (sidegrade)
AMD RX 6750 xt.

 

[gulkor]

Hi Everyone, Just joined forums and rx 6000 series club!

Bought on Amazon for $510 XFX Speedster QICK319 AMD Radeon RX 6800 Core Gaming Graphics Card with 16GB GDDR6 HDMI 3xDP, AMD RDNA 2 RX-68XLALFD9

I would like do WC on GPU, is it possible? Should i return Gpu should i need to, what card would allow me WC and be same perf?

GPU-Z says cant find video card in data base. Checking Aida64 shows as Video Adapter XFX RX 6800 Merc319 Black.

Never had this happen to me until now.

Thank you!

 

[AusWolf]

I found an even better deal with a 6750 xt for $390 after rebate. I put the order in. This will be my first AMD GPU since my beloved 7970. Is there anything I should know about the Ryzen 6000 series?


My main rig has had the following in the last decade.
AMD Radeon HD 6950
AMD Radeon HD 7970
NV GTX 970
NV 1060 3gb (sidegrade)
AMD RX 6750 xt.

Congrats! That's a really awesome deal!

What brand is it? If it's the reference model, what you need to know is that it's hot. But it's hot by design, so as long as you have good case airflow, you don't need to worry about it. But you do need good case airflow. Also, it's pretty much an overclocked and overvolted 6700 XT, so don't expect to push it a lot further without hitting thermal boundaries.

Other than that, RDNA 2 is a pretty sound series.

 

[Nordic]

Congrats! That's a really awesome deal!

What brand is it? If it's the reference model, what you need to know is that it's hot. But it's hot by design, so as long as you have good case airflow, you don't need to worry about it. But you do need good case airflow. Also, it's pretty much an overclocked and overvolted 6700 XT, so don't expect to push it a lot further without hitting thermal boundaries.

Other than that, RDNA 2 is a pretty sound series.

Today's GPU's are a mouthful, but it is the MSI RX 6750 XT MECH 2X 12G OC. It looks like it has dropped another $20 since yesterday.

 

[BatRastard]

The MSI RX 6600 Mech 2 I bought about two or three weeks ago for $239?!? It just dropped to $189 ...

 

[AusWolf]

Today's GPU's are a mouthful, but it is the MSI RX 6750 XT MECH 2X 12G OC. It looks like it has dropped another $20 since yesterday.

MSi's Mech is kind of their "bare minimum" series, just like the reference design. It'll work, but don't expect to overclock it to the Moon.

The MSI RX 6600 Mech 2 I bought about two or three weeks ago for $239?!? It just dropped to $189 ...

I know that feeling. AMD started bundling The Callisto Project just a week after I bought my 6750 XT.

 

[Kabouter Plop]

I think something is failing in my system getting system freezes and gpu driver crashes while playing world of warcraft, couple of months ago if had problems with memory but i fixed it and im unable to find errors so far which is frustating getting new memory kit later today altho bit slower im worried my psu may be faulty
im seeing it switch from 11.980v to 11.884v every couple of seconds and sometimes dip to 11.788v in rare case and peek at 12.076v
Perhaps my memory is fine but ram slot is having issues.

my 4e stick bassicly had 3 pins that where bit smudged dirty and not as glossy at that time couple of months ago that stick refused to post but after i cleaned it up it took it again no problem, if tested multiple times overnight without errors yet i get system freezes while playing wow or a tdr mostly a system freeze which very frustating, running furmark for 30 minutes is no problem at all no problems so it can't be my gpu right ?

occt 3d test also passes without errors as well as vram test, this is confusing surely system freezes are caused by hardware problem right even memory won't error.

 

[AusWolf]

I think something is failing in my system getting system freezes and gpu driver crashes while playing world of warcraft, couple of months ago if had problems with memory but i fixed it and im unable to find errors so far which is frustating getting new memory kit later today altho bit slower im worried my psu may be faulty
im seeing it switch from 11.980v to 11.884v every couple of seconds and sometimes dip to 11.788v in rare case and peek at 12.076v
Perhaps my memory is fine but ram slot is having issues.

my 4e stick bassicly had 3 pins that where bit smudged dirty and not as glossy at that time couple of months ago that stick refused to post but after i cleaned it up it took it again no problem, if tested multiple times overnight without errors yet i get system freezes while playing wow or a tdr mostly a system freeze which very frustating, running furmark for 30 minutes is no problem at all no problems so it can't be my gpu right ?

11.78 V is a bit low, but still within the ATX spec. I would swap that PSU anyway, just to be safe.

 

[Kabouter Plop]

11.78 V is a bit low, but still within the ATX spec. I would swap that PSU anyway, just to be safe.

I thought so to appearntly 11.4 is also within spec some reviews mention 11.788 as well its rarely dropping that much tho if disabled c states now see if it has any impact as well as soc oc mode enabled, im starting to think that all my problems with freezes are chipset driver related cos they been fixing some issues with them as well or perhaps my board is failing
I think the odds of my psu failing is quite low but i still have my evga 1300 g2 laying around i could plug in the bare minimum and test with that psu but not looking forward to it.

Got my new memory kit now hynix instead of samsung b die timings are awful lol 18 22 22 42 64 i just wanna confirm it gets rid of the system freezes if it does im rma'ng my memory kit even tho i get no errors obviously its to blame if this fixes it, it has failed posting before on 1 stick that i fixed.

 

[BatRastard]

I ran a PSU with a +12v rail that once it dropped to 11.44, it was all downhill from there. I knew age, wear, and tear would degrade it farther. I replaced with a Corsair TX650 in 2008 and it's still going. The fan died, but that PSU feeds my Linux box today. I'd replace yours ASAP. Sure, 11.44 is still in spec, but it's probably dealing with dirty power (depending on the wiring).

 

[Kabouter Plop]

Just froze with new memory kit either my gpu is dying or motherboard or psu which is only a year old with 10 year warranty, running with just 1 stick now cos the memory bank that i had a stick fail in may be faulty as well so im running only 16 gb in a2 now to see how that effects system freezes last thing i can try is replace pci-e cable of my mainboard and gpu and reseat mainboard cable
Seems to happen more often on agesa 1203c which lets set edc at 190 when im capped to 140 seems to happen less often

If not mentioned this yet but only have freezes in wow never in other games altho that matters nothing yet since i only played wow lately, im testing with heaven running in background right now see if i can reproduce freezes in heaven and then test other 3d loads as well preferably those that i can leave running in background while not overloading gpu as crazy like furmark would but enough load to make watts go up and down like normal game loads would.

If put my only stick now in the slot as well i suspect to be faulty while testing in hopes i can trigger it faster.

 

[gulkor]

Just froze with new memory kit either my gpu is dying or motherboard or psu which is only a year old with 10 year warranty, running with just 1 stick now cos the memory bank that i had a stick fail in may be faulty as well so im running only 16 gb in a2 now to see how that effects system freezes last thing i can try is replace pci-e cable of my mainboard and gpu and reseat mainboard cable
Seems to happen more often on agesa 1203c which lets set edc at 190 when im capped to 140 seems to happen less often

If not mentioned this yet but only have freezes in wow never in other games altho that matters nothing yet since i only played wow lately, im testing with heaven running in background right now see if i can reproduce freezes in heaven and then test other 3d loads as well preferably those that i can leave running in background while not overloading gpu as crazy like furmark would but enough load to make watts go up and down like normal game loads would.

If put my only stick now in the slot as well i suspect to be faulty while testing in hopes i can trigger it faster.

Run memtest test your memory.

MemTest86 - Official Site of the x86 Memory Testing Tool

MemTest86 is the original self booting memory testing software for x86 and ARM computers. Supporting both BIOS and UEFI, with options to boot from USB.

www.memtest86.com

 

[Kabouter Plop]

Run memtest test your memory.

MemTest86 - Official Site of the x86 Memory Testing Tool

MemTest86 is the original self booting memory testing software for x86 and ARM computers. Supporting both BIOS and UEFI, with options to boot from USB.

www.memtest86.com

Not even gonna bother since i had system freezes with brand new kit, im testing different loads now to confirm it it exclusively happens in wow.
Normally i would tho, if i start having freezes in other 3d loads or tdr's im gonna test my spare PSU that is 1300w

 

[gulkor]

Not even gonna bother since i had system freezes with brand new kit, im testing different loads now to confirm it it exclusively happens in wow.
Normally i would tho, if i start having freezes in other 3d loads or tdr's im gonna test my spare PSU that is 1300w

I found this on blizzard forum. https://us.forums.blizzard.com/en/wow/t/extreme-freezing-and-performance-issues-in-sepulcher/1197851


I would try uninstall your GFX drivers, then run safe mode and run DDU, then reinstall fresh GFX drivers,

Some members on wow forum said it helped freezing and locking up,

wow forum member said

Update from my end:

Cleaning my case (to improve GPU temps) and moving the WoW install to the C drive didn’t resolve the issue. My GPU temps did improve by a degree or two, but it’s still in the 75-80 range, which should be perfectly acceptable for an Nvidia card.

So, I tried to remove the graphics driver again, but this time I used the DDU (Display Driver Uninstaller) tool to fully remove the driver and reinstall the latest from Nvidia. Much to my surprise…so far this has worked. I’ve been able to play for longer periods without a freeze thus far. I did dailies last night and this morning along with a couple bosses in LFR as test cases. Tonight I have raid with my guild, so we’ll see if it holds. If it does, then the issue is either resolved or at least back to a manageable level.

To comment on Karsh’s experience. Removing my addons, uninstalling and reinstalling, and moving the install were all unsuccessful for me, and in fact made my issue much worse to the point where I could no longer log in to a character without forcing DirectX 11. Also, after looking at your dxdiag Karsh, we both have a GeForce GTX 1070 card, which could be a common element to the issue, possibly.

 

[Nordic]

MSi's Mech is kind of their "bare minimum" series, just like the reference design. It'll work, but don't expect to overclock it to the Moon.

I bought it for the performance per dollar value. It's not like a 6750 xt was ever going to set any overclocking records.

 

[AusWolf]

I bought it for the performance per dollar value. It's not like a 6750 xt was ever going to set any overclocking records.

Agreed. Just like my AMD reference card. At 100 quid cheaper than AIB ones, I'm really not complaining.

 

[Kabouter Plop]

I rather not buy from a vendor that does shady stuff like void warranty if remove stickers on their coolers even tho illigal in some places in the world, it just proofs they seek a way out of warranty at any cost possible even for something as simple as removing a sticker.

 

[HD64G]

Glad to have joined this club as I got the Sapphire Pulse RX6750 XT for €500 a few days ago and just yesterday installed it. Just testing a few undervolting profiles as my 144Hz 1080P monitor (which now works @144Hz with VRAM downclocking properly opposed to my RX5700 not doing so) doesn't need so much graphics power for most games. The autoundervolting setting gives 2,7GHz that work stable at 1,175V instead of the default 1,2V. I already tested and almost surely the 2GHz@1V profile is stable (80% of the performance in 3DMark Timespy for 90W less power draw - 13600 to 11000 points in Timespy) and will test a few more. Any suggestion for a good such profile from an RX6750XT or RX6700XT owner?

 

[Krajlevic]

Personally, I wouldn't rely on a Molex adapter long-term, but it'll probably do until you change your PSU to a unit with proper 8-pin PCI-e cables - which should be your next upgrade, imo.

After checking deeply, the BIOS of the card mentionned that it's a 143W as you can see below so nothing is strange...
And I've changed my cable for a proper PCI/E 8 pins

 

[AusWolf]

Glad to have joined this club as I got the Sapphire Pulse RX6750 XT for €500 a few days ago and just yesterday installed it. Just testing a few undervolting profiles as my 144Hz 1080P monitor (which now works @144Hz with VRAM downclocking properly opposed to my RX5700 not doing so) doesn't need so much graphics power for most games. The autoundervolting setting gives that the 2,62GHz works stabe at 1,175V instead of the defeult 1,2V. I already tested and almost surely the 2GHz@1V profile is stable (80% of the performance in 3DMark Timespy for 90W less power draw - 13600 to 11000 points in Timespy) and will test a few more. Any suggestion for a good such profile from an RX6750XT or RX6700XT owner?

Congrats and welcome!

Personally, I don't play with profiles - I just set a 60 FPS limit and call it a day. This may or may not be a useful tip depending on what display you have.

 

[HD64G]

I just tested some tunes and observed that while I set the GPU voltage at 1V in the AMD suite, the GPUz reported 0.825V. And for another tune the AMD suite was adjusted to give 1.175V and the GPUz reported 1.043V. Any info on that deviation @W1zzard ?

 

[outpt]

Does anyone know how to turn down the temperature in msi or amd suites
For a 6800 non X

 

[HD64G]

Does anyone know how to turn down the temperature in msi or amd suites
For a 6800 non X

In the AMD driver you can just set the fans to speed up sooner and keep temps lower. There is a curve with points you can move.

 

[AusWolf]

I just tested some tunes and observed that while I set the GPU voltage at 1V in the AMD suite, the GPUz reported 0.825V. And for another tune the AMD suite was adjusted to give 1.175V and the GPUz reported 1.043V. Any info on that deviation @W1zzard ?

As far as I've seen, even though the voltage bar has set values, the actual voltage you see in the end isn't a set value. It's more of a curve offset kind of thing - same with the frequency bar.

 

[HD64G]

Tested another tune with 2,3GHz@1,125V in the driver suite and resulted in 2,28GHz@0,87V. Great efficiency with max 110W in power draw. I can just suppose that the firmware has algoriths and the die auto-adjust the voltage needed to keep those clocks no matter what max limit is set. I have no problem to that, instead I find it innovative and very efficient without losing any considerable performance (~90% of stock performance for just above half the power).