Increased hardware crashes

I’ve been getting increased hardware crashes, even after turning down my power calibration. It seems to be happening with previews more than exports, but honestly, I’m so embarrassed. Hardware crashes as in the whole computer forcibly restarts. (475.1 KB)

Do you know the wattage of your system’s power supply? Are there any overclocking or undervolting settings applied? Is XMP enabled in the motherboard’s BIOS?

+60 Mhz GPU

Revert back your CPU and GPU to the official clock frequencies and power targets told by the manufacturers. My 13900K CPU produces red cross errors even on default settings. I lowered the max clock ratio to the PCores to 5,4 Ghz in the BIOS and the errors are gone!

GPU and motherboard reset to default. Holding steady.

1 Like

It didn’t hold. Ya busted.
During an export, restarted and with no other programs running or tasks queued.
It did actually hold for almost the full export, but I realized I didn’t set the trim properly, so I restarted it. So, 1½ exports? (679.0 KB)

You know that my CPU causes those errors even when set to default settings in the BIOS. I had to lower the PCore ratios further to get it stable!

That’s a problem. This wasn’t happening on dynamic overclock settings, and it absolutely shouldn’t happen on default settings. Modern hardware is designed to protect itself by pulling back the limits, which I’ve taken advantage of to prevent crashes in the past.

Turn off the XMP and try it again. Another user had issues and resolved their crashes by turning that off.

Motherboard default settings disable XMP, so that didn’t solve it for 4.1.0. So far, holding steady with 4.1.1.

It’s been stable for longer, but my PC just crashed again. XMP and similar settings are disabled, but I must reiterate that this program used to run stable under overclock conditions. Hardware crashes really get on my nerves, because they destroy all progress. (1.3 MB)

I’ll just rephrase what Imo was trying to say; it’s not TVAI’s fault your system is crashing, it’s your system that is unstable. We really shouldn’t fault Topaz for making an extremely optimized and efficient program that stresses our hardware to the max. That’s what good engineers do; they take advantage of every ounce of capacity the hardware has in order to maximize the performance for the customers.

But it does mean we as customers need to ensure we have stable systems to begin with, systems that don’t crash when the hardware components are stressed.

Another user posted a similar issue just now, and I responded with some tips on how to verify it is indeed your own machine that’s unstable, and what you can do to fix that.

For now, just regard TVAI as an excellent stress test of a system’s components, one that stresses the machine in a way no single other benchmark program does. The demo version could actually be advertised for that purpose as well I think :slight_smile:

This topic was automatically closed 60 days after the last reply. New replies are no longer allowed.