That’s not cool: These top gaming phones with liquid cooling just got caught for benchmark cheating

REDMAGIC is one of the few gaming phone manufacturers still on the market, offering devices with capacitive shoulder triggers, active cooling, and a host of other additions. Unfortunately, it looks like the brand’s latest phones have been delisted from a series of benchmark rankings.

“Devices are delisted when the manufacturer has not complied with our benchmark rules,” reads an excerpt of a banner seen on the REDMAGIC 11 Pro series benchmark page.

So what exactly did REDMAGIC do that resulted in its delisting? Well, a Japanese-language YouTube video posted last month might have the answer. The channel apparently found a huge gulf between scores achieved with the standard version of the benchmark app and a disguised or stealth version. In fact, the video showed that the REDMAGIC 11 Pro series couldn’t complete its stress test via the standard app, while the stealth version ran to completion.

The YouTube channel also found that the stealth test achieved peak temperatures of roughly 40 degrees Celsius (104 degrees Fahrenheit) while the standard test topped out at a scorching 55 degrees Celsius (131 degrees Fahrenheit). It’s worth noting that this phone has a cooling fan and a liquid-cooling system. So even these radical measures weren’t enough to prevent the phone from getting ridiculously hot.

This is a familiar method of benchmark manipulation, though, as some manufacturers recognize benchmark apps and subsequently boost device performance to achieve maximum scores. This usually has the side effect of much higher device temperatures, which can eventually lead to the phone shutting down the benchmark app for safety reasons. This is clearly the case for the YouTube channel’s test. However, these same manufacturers don’t recognize a disguised version of the same benchmark app, so the phone defaults to a more conservative performance profile with less heating.

This type of behavior is against the rules, as UL Solutions notes on its website:

A platform must run the benchmark without modification as if it were any other application.

There’s nothing technically wrong with having an allowlist of apps and optimizing performance for them accordingly. This happens all the time for games and general apps, allowing your phone to strike the right balance between performance and efficiency for a given task. After all, you don’t need maximum performance for WhatsApp or Google Keep.

However, this practice is particularly problematic for benchmark apps, as the performance achieved in these benchmarks won’t reflect real-world usage for most people. These inflated scores can find their way into official marketing materials, as brands claim that their phone has the best performance. Oddly enough, some mobile gamers have used this practice to their advantage by running emulators disguised as benchmark apps to achieve maximum performance. But this approach comes with a huge rise in device temperature and dramatically shorter battery life.

This isn’t the first time we’ve heard of benchmark manipulation. The realme GT7 Pro previously engaged in similar behavior over a year ago, while there have been numerous other cases over the years. Nevertheless, I don’t think REDMAGIC will be the last phone maker accused of benchmark cheating.

We’ve asked REDMAGIC and UL Solutions about this apparent benchmark manipulation and will update the article as soon as the companies respond to our queries.