Metal Gear Solid V: The Phantom Pain Benchmarked, Performance Review

[Steve]

I must embarrassingly admit that I've never played a Metal Gear game before, though that's partly because the series has focused exclusively on console and portable gaming devices for much of its existence, until recently with Ground Zeroes -- which hit PCs nine months after being released on console. Now, almost a year later we have a new Metal Gear Solid V title and this time around it shipped simultaneously for PC and console players on September 1.

Metal Gear Solid V: The Phantom Pain was built using the same Fox Engine as Ground Zeroes so you can expect to see subsurface scattering, physically-based rendering and other impressive technologies.

We expect stunning visuals as Konami recommends an Intel Core i7 and GeForce GTX 760 -- interestingly, no AMD hardware is mentioned and you can probably thank Nvidia's influence for that. Using the latest AMD and Nvidia drivers, we tested 26 DirectX 11 graphics cards covering most price ranges.

Read the complete review.

 

[dividebyzero]

Thanks Steve for another timely review.

Sad state of affairs having to play pass-the-parcel with the Fury cards. Hopefully someone at AMD didn't think the cost of the card outweighed the marketing benefit of having it benchmarked by a high profile tech site.

 

[Cryio]

Great review. Unfortunate that Fury cards are still so rare. The fps cap is unfortunate too. Hopefully they'll lift it in a future patch.

As for the game itself, the horrible "The Way It's Meant to be Played" manifests again with AMD GPUs having less than ideal performance. For AMD owners, this game will be worth visiting some months down the line after the Red Team had the proper time to release optimized drivers.

A 780 delivering better performance than a 390X at 1440p and same performance at 4K ? Preposterous !

 

[noel24]

I think it's a first benchmark that shows GTX960 more or less halfway between 950 and 970. Probably new driver. And probably the reason to not look at cards' release benchmarks when buying a new one. Still, Maxwell minimum framerates in 1080p seems kinda off? 10frames or 25% between brotherly 950 and 960, and only 5 frames (or 10%) between cousins of 960 and 970? It's like 970 is held back, maybe by frame limiter messing with chip's utilization? No such thing in 1440p.

 

[damnthereaper]

I would like to say it's nice to see a site still willing to test older cards like hd 7xxx series, a lot of places don't put them in their benchmarks and it make sit hard to estimate what your performance will be when you have to compare your card to one of the benchmark cards and try to determine play ability that way. You guys just made my decision much easier to get this game!!!

 

[rozatarra]

Very nice review, but, how many users can use this info if you have this build? most of us have an intel core i3- i5 with 4-8gb ram and a 200$ video card. I would like to see this review on a normal build, not on a dreaming machine.

 

[Steve]

Very nice review, but, how many users can use this info if you have this build? most of us have an intel core i3- i5 with 4-8gb ram and a 200$ video card. I would like to see this review on a normal build, not on a dreaming machine.

The CPU choice simply ensures that the GPU’s aren’t being limited. Testing on a low-end CPU would be pointless, especially if it inhibits performance to say 40fps for example. Typically we provide CPU testing in these articles but with the fps cap there wasn’t any point.

FYI system memory (RAM) has no impact on performance, at least when talking about capacity so it doesn’t matter if we tested with 8GB or 128GB the results would be the same.

 

[kuroiei]

Hm... I've been an MGS fan for a long time, remember playing the demo of MGS1 (on PC) and going through MGS2 on a GeForce 2 MMX (again, on PC...), so while not stunning, they were very playable. Now even if I'm not meeting the requirements for MGSV / MGS:GZ I've downloaded and am currently playing through Grond Zeroes. On AMD Phenom II X4 965 (~3.5Ghz), 8GB of DDR2 RAM and GeForce GTX 560. Yup. Lower than the minimum requiremetns. And guess what? It's speedy, playable and smooth. True, I didn't crank the setting to the max, but still the game looks superb. I'm wondering if MGSV would run similarly or just that the one level in Ground Zeroes is optimized to hell and back.

 

[Lionvibez]

How to remove FPS cap.

http://steamcommunity.com/app/311340/discussions/0/619574421273582659/

 

[lripplinger]

Thank you for including the GTX 660ti in the benchmarks. Nice to see my card is (MSI version) is still plenty capable of 1080p high settings, even on new titles.

 

[Steve]

How to remove FPS cap.

http://steamcommunity.com/app/311340/discussions/0/619574421273582659/

So does that work for The Phantom Pain?

 

[Cryio]

Hm... I've been an MGS fan for a long time, remember playing the demo of MGS1 (on PC) and going through MGS2 on a GeForce 2 MMX (again, on PC...), so while not stunning, they were very playable. Now even if I'm not meeting the requirements for MGSV / MGS:GZ I've downloaded and am currently playing through Grond Zeroes. On AMD Phenom II X4 965 (~3.5Ghz), 8GB of DDR2 RAM and GeForce GTX 560. Yup. Lower than the minimum requiremetns. And guess what? It's speedy, playable and smooth. True, I didn't crank the setting to the max, but still the game looks superb. I'm wondering if MGSV would run similarly or just that the one level in Ground Zeroes is optimized to hell and back.

MGSV wouldn't work at all for you at the moment, if you haven't read on the issues on AMD CPUs older than current FX CPUs and Athlon x4 760/860. So basically you.

 

[kuroiei]

MGSV wouldn't work at all for you at the moment, if you haven't read on the issues on AMD CPUs older than current FX CPUs and Athlon x4 760/860. So basically you.

Didn't see any news on that, no... Care to share a link or something? Because the minimum spec requirements for both MGSV:GZ and PP seems to be the same, and I didn't encounter any problems at all, but would want to know if that's the case with the full game.

Edit: apparently there's some issues with SSE 4.1 and older AMDs, so I won't risk buying it now. Maybe when it's on a sale and I have my rig updated. Thanks for the info anyway!

 

[buglenuge]

Very nice review, but, how many users can use this info if you have this build? most of us have an intel core i3- i5 with 4-8gb ram and a 200$ video card. I would like to see this review on a normal build, not on a dreaming machine.

I5 3570k @ 4.2ghz, msi gtx 670 power edition, 16 gb ram, win 10, 1680 x 1050 native monitor resolution, everything set to extra high, depth of field and volumetric clouds on. Looks great, no noticeable slow down or stuttering in about 10 hours of gameplay. I haven't put a frame rate monitor on it, but I haven't felt the need to.

 

[robb213]

How to remove FPS cap.

http://steamcommunity.com/app/311340/discussions/0/619574421273582659/

So does that work for The Phantom Pain?

I can't say about what Lionvibez said since I don't own the game yet, but the PCGamingWiki page has an entry on making the FPS variable vs fixed, at the bottom. Not sure if the modding tools have been updated too, or just re-uploaded to the linked dropbox account on the page.

 

[Cruarc]

Just FYI this game works fine on an I7 920 at stock speeds with a GTX 980. Playing at max settings and no framerate drops.

 

[plonk420]

Will this get a CPU benchmark? please? pretty please??

 

[Steve]

Will this get a CPU benchmark? please? pretty please??

Sorry no, we listed the reasons why we didn't test CPU performance.

 

[fimbles]

Great review. Unfortunate that Fury cards are still so rare. The fps cap is unfortunate too. Hopefully they'll lift it in a future patch.

As for the game itself, the horrible "The Way It's Meant to be Played" manifests again with AMD GPUs having less than ideal performance. For AMD owners, this game will be worth visiting some months down the line after the Red Team had the proper time to release optimized drivers.

A 780 delivering better performance than a 390X at 1440p and same performance at 4K ? Preposterous !

Is a 390x just a rebranded 290x? In which case the 780 performs at a very similar level. Maybe even slightly better in most games.

 

[Lionvibez]

So does that work for The Phantom Pain?

I don't have this game yet so haven't tested it

 

[unknownvirus]

HIS Radeon R9 380 (4096MB) is listed in the test system specs however the one used in the benchmark is 2GB. Is there any performance difference between 2GB vs 4GB? I am considering buying this or the GTX 960 2GB which is the same price.

 

[Erik91]

Would this run on an i3-4160 with an R9-380 gpu? I know the gpu can, but I'm wondering if the CPU would be a bottleneck

 

[Erik91]

HIS Radeon R9 380 (4096MB) is listed in the test system specs however the one used in the benchmark is 2GB. Is there any performance difference between 2GB vs 4GB? I am considering buying this or the GTX 960 2GB which is the same price.

Unless you're playing on resolutions higher than 1080p, the difference of a 2GB and a 4GB card should be minimal

 

[robb213]

Would this run on an i3-4160 with an R9-380 gpu? I know the gpu can, but I'm wondering if the CPU would be a bottleneck

I'm positive it would be fine. As Steve pointed out, and it's really true, the game isn't hard on CPUs. I have my doubts if quad-cores are even necessary at all.

 

[Cryio]

Is a 390x just a rebranded 290x? In which case the 780 performs at a very similar level. Maybe even slightly better in most games.

Ehm, no. 290X is equal to 780 Ti level of performance and usually as fast or faster than a 970. 390X is miles faster than a 780, 970 and as fast as a 980 or slightly faster than a 980, if the benchmarks from Guru3d's recent review of Devils 390X are anything to go by.