GTA V shipped alongside an onslaught of graphics settings – none of which offer tool-tips – that can vastly control the fluidity of gameplay. In our recent and comprehensive GTA V benchmark, we tested multiple video cards for FPS at simple “max” and “high” settings, fluctuating resolution between 1080, 1440, and 4K along the way. That content now behind us, we took the opportunity to objectively benchmark various graphics settings for performance differences, then took a few screenshots for comparison of those settings.
This GTA V optimization guide assists in choosing the best graphics settings for frame-limited video cards, explaining the options along the way.
It's finally here.
Grand Theft Auto V took its time to migrate to PC, and from our preliminary overview and testing, it seems like the wait was worthwhile. GTA V's PC port exhibits unique PC features, like a VRAM consumption slider indicative of the maximum VRAM requirement of the current settings. The port also added first-person mode, complete with new 3D models and animations for the characters' arms, phone, guns, and what-have-you. As you'll find out in our benchmark results below, the game is also incredibly well-optimized across most graphics card configurations, something we can't say has been true for most games in recent history.
These things take time, and RockStar certainly took as much of that as it needed.
Using a suite of video cards spanning the Titan X, SLI GTX 980s, R9 290X and 270Xs, GTX 960s, 750 Ti cards, and more, we benchmarked GTA V in an intensive test. This GTA V PC benchmark compares FPS of various graphics cards at maximum settings in 1080p, 1440p, and 4K resolutions.
This article makes no intentions to comment on gameplay value.
We're revisiting our Evolve benchmark, now that the game has fully launched and (some) drivers have been updated. Our previous Evolve bench tested the game's beta, but disclaimed heavily that the beta meant a lack of driver support and software-side optimization. The return benchmark uses much of the same methodology and represents the same game as previously, so this article will be a bit shorter in length.
In Left 4 Dead-like form, Evolve reintroduces the concept of monster vs. humans multiplayer gameplay with high-fidelity graphics. 2K's soon-to-be released “Evolve” has already been analyzed by us a few times, but now we're returning to specifically benchmark the game's PC FPS performance.
This Evolve GPU FPS benchmark tests the game on Very High (max) and Medium settings, pitting some of the best graphics cards against one another. On our Evolve graphics bench, we tested the GTX 980 vs. the GTX 780, 770, 750 Ti, & R9 290X vs. the R9 285, 270X, R7 250X, & HD 7850. Once we got past the FPS limitations (resolved easily, as explained in an upcoming guide), testing Evolve was fairly easy and unrestrictive.
NOTE: This game is in BETA. Although it is near completion, results could be significantly improved prior to launch as GPU manufacturers move to finalize drivers specific to Evolve. The same is true as 2K continues to implement optimization patches.
Elite: Dangerous is one of the best-optimized games we've tested this year, right up there with GRID: Autosport. The game is a member of the impending cluster of space sim and space-flight combat games actively being developed. Like Star Citizen, Elite: Dangerous comes from the designers of a game that's decades-old, 1984's “Elite.”
With Elite: Dangerous' official launch, we've put the game on our GPU bench to test the FPS on various graphics card configurations, including an R7 250X, 270X, GTX 750 Ti, GTX 980, and more. In addition to our usual video settings tests, we ran Elite: Dangerous using AMD's VSR and nVidia's DSR (super resolution) to render output at 4K. These tests are representative of performance yield on a true 4K display. Our crash fix guide may be useful to those who are experiencing issues running Elite: Dangerous.
CIG's Star Citizen aims to revitalize the PC gaming space by fully utilizing every system component to its fullest potential, starting with multi-million poly-count objects that hit the GPU heavily. In our very first interview with CIG's Chris Roberts – a man who has managed to raise $60 million in crowd-funding – we discussed Star Citizen's emphasis on full system hardware utilization. Roberts told us “I've got eight cores – I want to use them.”
The game is currently in alpha, versioned at version 0.9.2.2. Star Citizen's persistent universe (the major multiplayer component) has yet to begin production and is still in the design phase, though the “Arena Commander” module is currently available for download to early backers. The next module in the release schedule will patch-in FPS elements, but the game's current alpha offers dog-fighting, free flight, hangar exploration, and “murray cup” racing.
Ubisoft launched all its AAA titles in one go for the holiday season, it seems. Only days after the buggy launch of Assassin's Creed Unity ($60) – a game we found to use nearly 4GB VRAM in GPU benchmarking – the company pushed Far Cry 4 ($60) into retail channels. Ubisoft continued its partnership with nVidia into Far Cry 4, featuring inclusion of soft shadows, HBAO+, fine-tuned god rays and lighting FX, and other GameWorks-enabled technologies. Perhaps in tow of this partnership, we found AMD cards suffered substantially with Far Cry 4 on PC.
Our Far Cry 4 GPU FPS benchmark analyzes the best video cards for playing Far Cry 4 at max (Ultra) settings. We tested lower settings for optimization on more modest GPU configurations. Our tests benchmarked framerates on the GTX 980 vs. GTX 780 Ti, 770, R9 290X, 270X, 7850, and more. RAM and VRAM consumption were both monitored during playtests, with CPU bottlenecking discovered on some configurations.
Update: For those interested in playing Far Cry 4 near max settings, we just put together this PC build guide for a DIY FC4 PC.
In a somewhat promising turn for the industry, Assassin's Creed Unity ($60) uses almost all of the VRAM we were able to throw at it. We'll get into that shortly. Regardless of the game's mechanics and value (reviewed here), there's no arguing that Assassin's Creed Unity has some of the most graphically-impressive visuals ever produced for a PC game. In coordination with nVidia and its GameWorks suite (detailed), Ubisoft implemented new Percentage Closer Soft Shadows, TXAA, and ShadowWorks technology to soften and blur lines between cast shadows. Not all graphics technologies require nVidia video cards.
In this GPU benchmark, we look at the best video cards for Assassin's Creed Unity for max (Ultra High) settings and other settings; our test pits the GTX 980 vs. GTX 780 Ti, 770, R9 290X, 270X, and more. Low settings tests are also included. Further, we checked RAM and VRAM consumption while playing ACU, hoping to further determine the game's most demanded resource.
As we tend to do with new game releases – GRID, Titanfall, and Watch_Dogs included – we decided to put Borderlands: The Pre-Sequel through its performance paces. We originally spoke about Borderlands: The Pre-Sequel at PAX, where we got hands-on with the game and discussed gameplay mechanics. Since then, the title has shipped at the now-normal $60 price-tag, complete with the usual nVidia partnership and a basis on Unreal Engine.
FRAPS has been in the gameplay capture business for over a decade now, inarguably serving as the best solution for early gameplay video recordings. The advent of casual streaming and competitive eSports has finally pushed recorded game content to widespread consumption. ISP-provided datarates have mostly stabilized to usable levels, which helps in production and consumption of high bit-rate content.
ShadowPlay was announced as a FRAPS alternative last year by nVidia, and is only compatible with nVidia devices. The tool uses an integrated H.264 video encoder on Kepler and Maxwell hardware, ensuring most the performance drag is loaded on the GPU rather than the CPU; moreover, it's being loaded on specific components of the GPU that are built for video encoding and largely unused while gaming.
FRAPS does no live encoding and only records raw data output, which theoretically means it will have the best quality (lossless), but also demands the most resources in storage and CPU cycles.
We moderate comments on a ~24~48 hour cycle. There will be some delay after submitting a comment.