Skip to main content

Long read: How TikTok's most intriguing geolocator makes a story out of a game

Where in the world is Josemonkey?

If you click on a link and make a purchase we may receive a small commission. Read our editorial policy.

Hades Canyon vs GTX 1060/ GTX 1050 Ti/ RX 570/ RX 580

The battle of the desktop GPUs.

Across the next couple of pages, we'll be comparing the Hades Canyon NUC with Radeon RX Vega M GH graphics against a range of mainstream desktop graphics cards from both AMD and Nvidia - specifically the GTX 1050 Ti, GTX 1060, RX 570 and RX 580. We used a reference model GTX 1060, but reference RX 570s and RX 580s aren't a thing. We used the Asus Strix version of the RX 570 with 4GB of memory, while our RX 580 results are a few percentage points higher than most other models as we're using Sapphire's extremely highly clocked Nitro model.

In all cases, the GPUs are benchmarked on a Core i7 6700K-based system running at 4.6GHz and paired with four 4GB modules of Corsair Vengeance LPX running at 3000MHz. Based on the results we got, we're fairly confident that across all of these benchmarks, we're definitely CPU-bound here - the Core i7 8809 really is rather potent.

And that's just as well as the Core i7 8809G in the NUC8i7HVK operates at an all-core turbo frequency of 3.9GHz - a good 700MHz slower than the desktop system that forms the basis of our test suite, and it's also paired with 2400MHz DDR4, again significantly slower than our usual benchmarking desktop system which runs with 3000MHz DDR4. We've also included overclocking results from the NUC, with Radeon WattMan used to push core frequency up to 1350MHz, along with increasing the HBM2 memory frequency from 800MHz to 900MHz, giving us a bandwidth boost of 230GB/s up from just over 205GB/s.

Here are the titles we're testing on this page, and here's a detailed look at how to use our benchmarking widget:

Assassin's Creed Unity

Starting as we mean to go, it's perhaps not surprising that the NUC's 24 compute units aren't really up to the challenge set by the GTX 1060. However, there is a handy 11 per cent advantage over the GTX 1050 Ti. The overclock is pretty noisy but it does see a decent 12 per cent uplift in performance. RX 570, RX 580 and GTX 1060 really are on another level here.

AC Unity: 1080p, Ultra High, FXAA

Ashes of the Singularity

It's a game that has caused real issues for Nvidia hardware over the years and that plays to the RX Vega M GH's strengths here, with a decent 21.6 per cent increase in performance over the GTX 1050 Ti. GTX 1060 can 'only' hand in a 24 per cent increase in performance, producing an uncannily similar performance level to the RX 570. The Nitro RX 580 here screams ahead in a game that does generally favour AMD hardware.

Ashes of the Singularity DX12: 1080p, Extreme, No AA

Battlefield 1

Running under DirectX 12, this is another game that flatters AMD hardware generally, with a particularly strong result from the Sapphire factory overclocked RX 580. Interestingly though, the NUC hands in only an average boost over the GTX 1050 Ti with a 14 per cent uplift. We got a decent return from our overclock here, with the NUC pumping out an additional 14 per cent of performance. Frame-time spikes on this one tend to come from random close-range explosions - there is no 'official' benchmark for this title.

Battlefield 1 DX12: 1080p, Ultra, TAA

Crysis 3

There's no official benchmark for Crysis 3 either, so we take a run through the end of the jungle stage in order to get our metrics. A pattern is definitely starting to form here, with an 11 per cent boost in performance over the desktop GTX 1050 Ti and a 14 per cent increase via overclocking. As expected though, GTX 1060 and the two 500 series AMD cards are on another level.

Crysis 3: 1080p, Very High, SMAA T2X

We've got a lot more data and analysis for you, so here's a breakdown of the entire content of this article:

Intel Hades Canyon NUC8i7HVK Introduction

Hades Canyon vs GTX 1050 Ti/ GTX 1060/ RX 570/ RX 580

Hades Canyon vs Mobile GTX 1060/ GTX 1060 Max-Q Benchmarks

Conclusion