“How many frames does it get in Valheim?”
“Will it run COD at high FPS? What about Tarkov?”
“Can I play Cyberpunk on it? I have a 4K 120Hz Ultrawide monitor and I want to maximize that”
These are the sort of questions that come into our inbox constantly. They’re not necessarily difficult questions to ask, but they’re very hard to answer. It seems rather reasonable to ask what sort of performance a PC will have in your favorite games. What causes such a reasonable question to have generally unfulfilling answers is the fact that there are hundreds upon hundreds of PC games that our clientele love to play.
There are about one hundred triple A games released from major studios every year. Additionally, more than ten thousand games from various indie studios launch in that same timespan, any one of which has the potential to go viral and be played by millions. So, when it comes to the question of performance for so many games that have the entire industry’s eyes on them, finding a solution is a nothing short of a towering task.
Plenty of other System Integrators have results up on their website for multiple games at various resolutions with different configurations. When a company publishes these estimates, they are just that: estimates. Of course, driver changes, Windows updates and game patches can all lead to slightly different performance for a given title, but these estimates are generally update agnostic. They’re multiples of 5 or don’t change at different resolutions or even are just as vague as 60 FPS +. The reason for that is that many of these games are never tested by these other companies. There are various services that will benchmark a PC with one engine through a series of test passes and use that information to guess what performance would be like in a completely different game. While efficient, this is hardly the most transparent practice in the PC gaming industry.
We conceived of our benchmarking videos as a way to gauge performance and detail what a PC from Starforge Systems is capable of. The nature of these videos is obviously promotional, but we aspire for them to be as transparent and comprehensive as possible. They take a great deal of testing, re-testing and even more testing. A single video involves hundreds of data points collected from dozens of test passes. The data that we collect for each PC is compiled into a spreadsheet, averaged and inspected for any outliers. At times, certain games run significantly better for the first two or three test passes. When this happens, we run additional testing for that particular game. While this may hurt the numbers that we publish, it leads to a much more accurate framerate. Very few people who play their favorite first-person shooter play for only fifteen, twenty minutes at a time.
While there are many titles that have a huge following that are not a part of our benchmarking suit, we’re very happy with the options that we currently use. Each operates on a slightly different engine, has different developers and, most importantly for our purposes, has an in-game benchmark. The ability to run a game over and over and control for all of the poorly optimized parts of a game by putting the system through a uniform test bench. Comparing two of our systems side-by-side in slightly different in-game environments could create results that don’t actually represent the differences between them. An in-game benchmark pass ensures that there’s no bias introduced to the test results.
Reporting back frame rates to potential buyers can entice them towards certain units, but the way in which we record gameplay for each title is part of an effort to take transparency to the next level. It would be easy to simply tell our audience what they want to hear when it comes to running a game like Modern Warfare. But showing the end results of the benchmark pass keeps us honest. There are times when a computer has rather competitive average frame rates while playing that game, but any fan of Battle Royale games knows that lows can be just as important. The results screen will not only show off our system’s average, but also just where dips in framerate might end up. An informed decision is a good decision.
There’s also something to be said about the accessibility that comes with both visual and described representations of data points. Gamers who may be deaf/hard of hearing will always be able to benefit from the results screen at the end of a loop, as well as our summary section after each game. Those who may be watching from a phone screen may benefit from hearing specific performance and details called out during each pass. There are always improvements that can be made in how data is presented, but the way that we showcase system performance is a step above our competition.
High performance is only half of the battle for some users. For many, the purchase of a PC is a serious investment. We take the time to discuss the parts that make up the PC, albeit briefly, in an ongoing effort to educate our clientele on the responsibilities of components within a computer. We partner with companies to provide our coolers, motherboards, power supplied etc., because they have proven themselves to be reliable manufacturers. A quick mention of each part and its job inside of the PC can be informative in guiding a prospective buyer towards the PC that’s right for their use-case.
Editing video can be incredibly memory intensive. Three-dimensional design might require a powerful graphics card. Certain professional software benefit from higher clock speeds or more cores on a CPU. We highlight our components in each of our benchmarking videos to create reference points for those who want to be able to do more than just game on their Starforge computer.
How can we improve these benchmarking videos going forward? As mentioned, there are always things that can be done better with projects like this. For starters, when it comes to systems directly marketed to creators, our Voyager line, we want to supplement gaming performance with benchmarks on popularly used creative tools. Capturing data on these programs presents some small challenge, but what is perhaps more important is how we present that performance as well as double check our results. A slight deviation in frame rates due to a driver update is one thing, but render time in a professional environment can mean meeting a deadline or finishing just behind. The process is one we’re actively exploring and hope to start introducing in the near future.
On the gaming side, we also are eager to expand and update our portfolio of tested games. Astute viewers will notice that we publish some of our testing results directly on the website. These help greatly in system-to-system comparison. But not every gamer plays those three titles. In fact, many games in our suite have been out since before Starforge launched in August of 2022. Finding a way to present newer releases, call out new features while still allowing gamers to make direct comparisons between our PCs is another step in improving these benchmarking videos.
We’re incredibly proud of the PCs that we build here at Starforge Systems. The benchmarking videos are our way of summarizing the capabilities of each computer model and presenting it in a transparent, professional way to folks who are interested in buying from us. They are a labor of love, and at the time of writing are doing something that no other system integrator seems to be doing. But as in all things, we’re constantly seeking to improve our process, the way we present data and ultimately provide the best possible experience for those who take an interest in our PCs.
