Performing Regular System Checkups

Performing regular system checkups is important to sustaining a healthy PC. This isn’t true just for system builders, but for all PCs in general – component performance is affected by temperature, drivers, malware, and age. Regular checkups are easy and may alert you to impending failures soon enough that they’re preventable. In the very least, such a discovery would limit system down-time.

Let’s start with a baseline benchmark for the system.

Baseline Benchmark

After completing component and driver installation, a baseline benchmark provides a means to certify that (1) the drivers are functioning, and (2) the system is performing as expected given its component outfit.

Downloading software like Unigine will provide a free means to evaluate component performance with an assigned “score.” Benchmarks can fluctuate from test-to-test, but the scores should generally be within a few percentage points of each other.

We’d recommend running one of these utilities to evaluate system performance, then make a note of the score for future recall. It’s good practice to run the benchmarking utility every now and then – maybe once per year or once per major driver revision – just to make sure the system is equal to or exceeding previous benchmark scoring.

If the benchmark metric has significantly fallen, there’s a possibility that something’s wrong with the physical hardware (like dust causing a thermal build-up and subsequent throttling of the cores) or with the software (drivers, malware).

Temperature Checking

Temperature management and logging is fairly easy (and normally free for personal use). Tools like CAMHW Monitor+AIDA64, and SpeedFan can report component temperatures.

Check the idle temperature of your CPU and GPU after first build. Take a note of these somewhere for future recall – maybe with your benchmark score. Also take a note of loaded temperatures. A load can be generated by running 3DMark for CPU and GPU intensive tasks. Alternatively, playing a game that you’ll use again later (for the same test) serves as a real-world test.

Check the temperatures after a few minutes (we find that components tend to stabilize their max thermals after about 10-20 minutes). Log these numbers.

Software can be left running and monitored from time-to-time, or just executed at will. Check regularly to ensure that everything looks healthy. If there’s been a sudden spike in temperatures, it’s time to open up the system and inspect for dust or other issues (like a dead fan).

Hard Drive SMART Stats

SSDs and HDDs output a list of “SMART attributes” to the system, reporting device health and longevity. Tools like HD Tune can read storage device health, and will report how many failed reads and writes have occurred. More importantly, they’ll also predict if a failure is impending, making the suggestion to clone and replace a hard disk or SSD prior to its death.

If you begin hearing “clunking” or experiencing slow performance for normal tasks (or storage tasks), we’d recommend checking SMART attributes for errors. Most utilities will report the stats in a fashion which is user-friendly.

Fun fact: Enterprise drives that use helium will sometimes report their remaining helium levels through SMART attributes.

Malware Sweep

A final, semi-regular occurrence should be malware sweeping. Determining if a system is infected isn’t always trivial, but checking for basic and common issues will help prevent ongoing, long-term issues. Malware can impact a system in many ways, not the least of which includes potential for identity theft, account hacking, or just malicious performance reductions.

We’d recommend using a tool like Avast to regularly sweep for malware. Note that some files may be read as “false-positives,” or threats that aren’t actually a threat – keep an eye out for those.

Quarantine any infected files rather than deleting them. If it turns out the file was needed and the warning was a false-positive, the file can be restored from the quarantined ‘chest.’ – Steve Burke, GamersNexus