AMD’s Radeon RX 9070 series GPUs are a welcome addition to the GPU market, assuming you can get one at MSRP. AMD has faced fewer complications than Nvidia’s lackluster GPU launch, marred by a shortage of supplies. UserBenchmark has not taken lightly to AMD’s and the tech media’s supposed antics. In its purported review of the RX 9070 XT, one of the best graphics cards, the website claims that Radeon GPUs fall short in real-world performance while failing to mention the GPU in question even once.
For the uninitiated, UserBenchmark (UB) is infamous in the tech landscape for its radical perspectives versus AMD, which it commonly refers to as “Advanced Marketing Devices.” For context, it once recommended readers purchase a Core i5-13600K over the Ryzen 7 9800X3D, asserting, and I quote, “Spending more on a gaming CPU is often pointless.”
Userbenchmark is only good for comparing component performance with people with the exact same SKU (to see if something is really wrong).
You should never use them for comparison across companies or even SKU families.
I’d argue there are MUCH better alternatives (cine, pass, 3Dmark, hwbot, cpuz, et al) for that that are equally if not easier to use. Furthermore, installing anything on a machine from these nutties is begging to be hacked (best case scenario).
One useful thing about userbench is that you can benchmark all components at the same time to identify a massive discrepancy between median performance for your component and what you’re seeing in your build.
I treat userbench less as a benchmark and more as a high level validation system. Just to make sure there are no critical issues that are messing up my new build.
Passmark has this feature as well, and is much more reputable.
Ooh, nice, I’ve been looking for a replacement ever since UserBenchmark locked the benchmark behind a paywall and a ridiculous timewaster captcha.
It was a handy way to just make sure everything was working properly - like my BIOS is being a dumbass and deciding to run my RAM at half speed for some unknown reason, and not letting the new CPU boost to full potential.
M8, c’mon, people are taking their time to recommend you actual benchmarks that are somewhat transparent, reputable and provide all the functions that userbenchmark is supposed to have. Shit, even Aida64 is better than userbench. The least you could do is not spread complete asinine takes and have people install literal spyware from a bunch of loons that clearly make a buck out of fooling idiots to install their software and selling their data. FFS