F5F Stay Refreshed Hardware Desktop Check if your components handle 1440p at 120Hz smoothly.

Check if your components handle 1440p at 120Hz smoothly.

Check if your components handle 1440p at 120Hz smoothly.

J
JackSwordMan
Junior Member
21
10-21-2020, 10:23 PM
#1
I recently upgraded my system from an RTX 2080 XC EVGA to a RX7800XT Nitro with Sapphire, using 24.3.1 drivers. I started in safe mode with DDU before installing and enabled smart access memory and a resizable bar. The GPU and CPU were automatically overclocked via AMD Adrenalin. My Windows 10 hasn't been reinstalled.

My current setup includes a FlowerPower LEADEX III 850W Gold Mobo PSU, an ASUS ROG STRIX B550-F GAMING motherboard with Wi-Fi, an AM4 platform CPU (Ryzen 7 5800X), 32GB Corsair Vengeance RGB Pro C16 at 3200MHZ, and a 1TB Samsung 860 EVO plus a 2TB HDD.

After numerous tests in Hell Divers, Darktide, Phantasy Star Online NGS, and other games, crashes occurred frequently—especially when launching matches in Darktide. Despite checking the post, frame times and performance metrics remained inconsistent, sometimes improving after pressing Alt+Enter to switch between fullscreen and borderless windows. At high settings, FPS fluctuated widely between 100 and 30. Lowering settings helped stabilize but still showed micro-lags.

I’ve attached crash dumps and console logs for further analysis. I’ll continue testing games during troubleshooting. Please let me know if you can assist.
J
JackSwordMan
10-21-2020, 10:23 PM #1

I recently upgraded my system from an RTX 2080 XC EVGA to a RX7800XT Nitro with Sapphire, using 24.3.1 drivers. I started in safe mode with DDU before installing and enabled smart access memory and a resizable bar. The GPU and CPU were automatically overclocked via AMD Adrenalin. My Windows 10 hasn't been reinstalled.

My current setup includes a FlowerPower LEADEX III 850W Gold Mobo PSU, an ASUS ROG STRIX B550-F GAMING motherboard with Wi-Fi, an AM4 platform CPU (Ryzen 7 5800X), 32GB Corsair Vengeance RGB Pro C16 at 3200MHZ, and a 1TB Samsung 860 EVO plus a 2TB HDD.

After numerous tests in Hell Divers, Darktide, Phantasy Star Online NGS, and other games, crashes occurred frequently—especially when launching matches in Darktide. Despite checking the post, frame times and performance metrics remained inconsistent, sometimes improving after pressing Alt+Enter to switch between fullscreen and borderless windows. At high settings, FPS fluctuated widely between 100 and 30. Lowering settings helped stabilize but still showed micro-lags.

I’ve attached crash dumps and console logs for further analysis. I’ll continue testing games during troubleshooting. Please let me know if you can assist.

X
xTripleMinerx
Posting Freak
846
10-26-2020, 06:28 AM
#2
It seems the issue is driver-related, with limited options unless you adjust the renderer (as in some games like BG3). You could also eliminate OC and increase main and lower max clocks in Adrenalin (e.g., 1000/2000 instead of 500/2500), which often prevents crashes—something that helped me in earlier BG3 versions.
X
xTripleMinerx
10-26-2020, 06:28 AM #2

It seems the issue is driver-related, with limited options unless you adjust the renderer (as in some games like BG3). You could also eliminate OC and increase main and lower max clocks in Adrenalin (e.g., 1000/2000 instead of 500/2500), which often prevents crashes—something that helped me in earlier BG3 versions.

N
Nucleoside
Member
53
10-26-2020, 04:04 PM
#3
I can explain how adjusting the maximum clock speeds works. It affects the processor's performance and power usage.
N
Nucleoside
10-26-2020, 04:04 PM #3

I can explain how adjusting the maximum clock speeds works. It affects the processor's performance and power usage.

S
Shukokai
Member
231
10-26-2020, 04:36 PM
#4
Here is the information presented differently:

The item in question has a model number of 7900XTX.
S
Shukokai
10-26-2020, 04:36 PM #4

Here is the information presented differently:

The item in question has a model number of 7900XTX.

S
sheepbottle5
Junior Member
44
10-26-2020, 05:28 PM
#5
Thanks for the update. I realized something a bit odd—I think you're running all your games in fullscreen. You have a dual monitor configuration: a 1440P LG display at 165Hz and a 1080P Samsung screen at 60Hz. Previously, using borderless windows gave great performance, but now with this new card, switching to fullscreen helps resolve many of the slowdowns at low and high performance levels. I'm currently testing this setup.
S
sheepbottle5
10-26-2020, 05:28 PM #5

Thanks for the update. I realized something a bit odd—I think you're running all your games in fullscreen. You have a dual monitor configuration: a 1440P LG display at 165Hz and a 1080P Samsung screen at 60Hz. Previously, using borderless windows gave great performance, but now with this new card, switching to fullscreen helps resolve many of the slowdowns at low and high performance levels. I'm currently testing this setup.

S
SoyDash
Posting Freak
859
10-27-2020, 12:18 AM
#6
I only had one monitor. There were many problems with the 7000 series when using multiple monitors. It might help to disconnect the secondary to see if that improves things.
S
SoyDash
10-27-2020, 12:18 AM #6

I only had one monitor. There were many problems with the 7000 series when using multiple monitors. It might help to disconnect the secondary to see if that improves things.

0
0Slender0
Member
211
10-28-2020, 04:06 PM
#7
It's disappointing, but I'm open to giving it a shot.
0
0Slender0
10-28-2020, 04:06 PM #7

It's disappointing, but I'm open to giving it a shot.

L
Lizy5
Junior Member
20
10-29-2020, 12:48 AM
#8
Begin uninstalling GPU drivers and chipset using ddu. Next, fetch the most recent chipset driver compatible with your motherboard and install it first. After that, restart the system and then reinstall the GPU drivers, then restart once more. Avoid auto-overclocking; observe stability. If default settings still cause crashes, it may point to a driver problem. It could also relate to Windows issues—remember, when switching from NVIDIA to AMD GPUs years ago, I experienced similar problems while others didn't. After a fresh Windows install, the issues disappeared, though I haven't tried that yet. Possibly the driver is the root cause.

I'm uncertain about auto-overclocking. Increasing core frequency usually doesn't lower voltage. My previous GPU performed well at full utilization, but the frequency dropped when games demanded maximum cores. This often occurs at higher resolutions due to power limits being reached. To maintain higher frequencies, check your GPU's hotspot temperature. If it stays within specs, raise the power limit and verify temperatures again. Manufacturers sometimes cap increases for safety, especially on lower-tier coolers, which can cause higher temps. Raising the power limit may help, but only if the cooler can handle it.

Increasing fan speed might be necessary if hotspots exceed limits, leading to noise. Otherwise, simply boosting frequency isn't enough—temperatures will rise until the GPU throttles. In that case, you'll need to adjust fan speeds or consider a cooler upgrade. The best approach is usually to raise the power limit rather than just tweaking frequency.

For example, if your GPU maxes at 1400MHz and drops to 1200MHz in heavy games, increasing the limit to 1500MHz might keep performance stable. This way, the GPU runs at its intended speed without hitting limits. Download the latest stable BIOS for your motherboard if you're upgrading, as newer chips often support better performance. I don't have all the details, but this seems like a solid strategy.
L
Lizy5
10-29-2020, 12:48 AM #8

Begin uninstalling GPU drivers and chipset using ddu. Next, fetch the most recent chipset driver compatible with your motherboard and install it first. After that, restart the system and then reinstall the GPU drivers, then restart once more. Avoid auto-overclocking; observe stability. If default settings still cause crashes, it may point to a driver problem. It could also relate to Windows issues—remember, when switching from NVIDIA to AMD GPUs years ago, I experienced similar problems while others didn't. After a fresh Windows install, the issues disappeared, though I haven't tried that yet. Possibly the driver is the root cause.

I'm uncertain about auto-overclocking. Increasing core frequency usually doesn't lower voltage. My previous GPU performed well at full utilization, but the frequency dropped when games demanded maximum cores. This often occurs at higher resolutions due to power limits being reached. To maintain higher frequencies, check your GPU's hotspot temperature. If it stays within specs, raise the power limit and verify temperatures again. Manufacturers sometimes cap increases for safety, especially on lower-tier coolers, which can cause higher temps. Raising the power limit may help, but only if the cooler can handle it.

Increasing fan speed might be necessary if hotspots exceed limits, leading to noise. Otherwise, simply boosting frequency isn't enough—temperatures will rise until the GPU throttles. In that case, you'll need to adjust fan speeds or consider a cooler upgrade. The best approach is usually to raise the power limit rather than just tweaking frequency.

For example, if your GPU maxes at 1400MHz and drops to 1200MHz in heavy games, increasing the limit to 1500MHz might keep performance stable. This way, the GPU runs at its intended speed without hitting limits. Download the latest stable BIOS for your motherboard if you're upgrading, as newer chips often support better performance. I don't have all the details, but this seems like a solid strategy.