AMD Trinity Preview - Gaming Performance: Part Deux

In a recent article, I shared a first look at the gaming performance of the AMD A10-5800K APU. Overall the new APU performed 1/3 faster than the previous flagship chip, the A8-3850. However, the A10 doesn't provide playable framerates with our normal stress test suite (which runs each game at max settings). However, with only slight drops in image quality the A10 is able to hit an average of 30fps in many popular titles. As we did with our previous APU reviews, I wanted to take a look at how the title performs in slightly more casual/mainstream titles.

Note: It's come to light that the RAM I pulled from my FX-8150 is underclocked for the A10-5800K. This is of particular impact, since the APU uses system RAM for VRAM. The performance using DDR3-1833 (versus the DDR3-1600 I used) is about 10% better.

Performance

In addition to looking at games from our normal stable of benchmarks, I wanted to look at some popular indie titles that are available. I thought these smaller games at lower price points would match up well to the smaller price point and form factor of something like an ultra-portable. I wouldn't expect to play the big shooters on this platform, but I do expect it perform well on a bevy of "indie" titles that are often at the forefront of gaming press.

DeathSpank
To test Deathspank, I played from the opening cinematic, up until the meeting the witch. I killed every monster and opened every chest along this short trek. There are no real graphical settings beyond V-sync, which I disabled.

I had the GPU configured for 2GB of RAM, and it certainly meets the other requirements. Considering almost no impact to the fps moving from 720p to 1080p, I'm left to believe the CPU power is the roadblock. The title plays quite well on the A10, with minimum framerates above 60fps.

Guild Wars
I realize that Guild Wars is not an "indie" title, but it has always been known as a very attractive game that ran well on a wide variety of hardware. I tested Guild Wars at 0xAA with the quality settings one step off of "max".

To test, I tried to get as controlled an environment as I could. I set up in Lion's Arch and ran from the last ship on the dock down the beach to the encampment there, and then back again. This provided lots of geometry to render in the docks and ships, along with a lot of horizon and water. Winding through the docks and the narrow passage that connects the dock to the beach required a lot of quick turns in rapid succession. This taxed the system in it's ability to quickly render new objects.

As with my testing on the A8, the A10 blew through this test. It did so well that I reran the tests adding 2xAA and then 4xAA. Even at 4xAA with the highest settings, the A10-5800K pushed past 120fps. It's pretty cool to think you could play Guild Wars at 120fps on a 120Hz panel, simply off an integrated GPU.

Guild Wars 2
As noted in the above discussion, Guild Wars was always considered to be both an attractive game and friendly to a wide range of systems. Obviously years have passed since the original's release. We've had two major releases of the DirectX technology (DX10 and DX11), and both games and the hardware they run on are significantly more sophisticated. So, how does Guild Wars 2 play on the A10 APU? We know the game looks great, but does that beauty come at the cost of performance on lower end systems.

The original Guild Wars had a quality slider that ranged offered a range of preset selections of the various quality settings. While you could customize individual settings, having the predefined selections made benchmarking fairly easy. Guild Wars 2 takes a different approach to quality selections. There are only three options - Best Performance, Auto-Detect, and Best Quality. These three settings will make different selections of all the quality options available. The problem in a consistent benchmark is that the settings will be different for different cards, making apples-to-apples comparisons difficult.

What I've done for this article is provide the fps results for each of the different selections: Performance, Quality and Auto-Detect. To help understand what performance looks like with these settings I've provided screenshots of the graphics options associated with each selection, and provided some in-game comparison screenshots for each selection. It clocked in at a 4-5 minutes run.

To test the performance I made a short run on the Divinity Reach map. I started at the Shaemoor Waypoint, went across the bridge, past the Fields Waypoint and into the farm at Shaymoor Fields. I fought a few worms in the fields, then went down to the river's edge where I fought some drakes and such, crossed the river, walked up the hill, hopped the fence on the western edge of the Village of Shaemoor and made my way back to the Waypoint where I started.

Jet Set Radio HD
I tested JSR HD, but the results aren't worth putting into the table. The game is locked hard at 30fps with a framerate cap. I performed a few minute run on the first level, skating, grinding and tagging. The only graphical option is MSAA, which ranged from 0x - 8x. The A10-5800K kept the 30fps rate with no dips, all the way up to 8x MSAA.

Orcs Must Die!
To benchmark Orcs Must Die! I played the first level of the demo. I started FRAPS after I had gone through the initial setup of the spellbook and placing my first traps. I started FRAPS right as I released the horde of orcs, and stopped it as my character started his "winning" dance. Both settings tested were without MSAA enabled. The game offers Low and High settings with 0xMSAA, and then High with 2x, 4x and 8x MSAA.

Orcs Must Die! is a really fun title, that runs surprisingly well. The cartoony graphics and underground setting (no sky to render) surely help, and I was able to get 30fps+ out of the A10 even after addings 2xMSAA.

Orcs Must Die! 2
OMD2 performs similarly to the original title, with a few exceptions. At "Low" settings, the sequel runs at a significantly slower framerate than the original - 105fps vs. 87fps. Obviously both of these scores are extremely playable, so the hit really isn't a big deal. Also to note is that the MSAA setting is independent of the "quality" setting. In the original OMD, the quality settings were Low 0xAA - High 0xAA - High 2x - High 4xAA and High 8xAA. That's it. In OMD2, you can set MSAA independently of the "quality" selection of Low/High. And, there is no option for 0xAA - the minimum is 1xAA. Given this, it's good to see the baseline "High" performance to be similar to the original, even with the additional anti-aliasing being applies.

Below are three sets of screenshots that show the difference between the Low and High settings. The Low quality settings removes the metal textures and shading from the spikes (and metal objects on our hero), as well as much of the lighting. Additionally blood in the Low setting is completely opaque and looks very "thick", while the blood on the High setting disperses in a nice spray.

Torchlight
To test Torchlight I created a new character and played the first dungeon until my character hit 2nd level. I started FRAPS upon actually starting the dungeon (not on the initial cutscene in the city), and stopped it when my character leveled. This took about 4.5 minutes of gameplay.

Torchlight isn't a graphical powerhouse, and even offers a "Netbook" mode. I tested at max settings, and the game looks and plays great on the A10 APU. LIke Guild Wars, you could play this title on the integrated graphics of the 5800K and pull 120fps out of a 120Hz panel.

Torchlight II
Torchlight II is the sequel to the wildly popular Torchlight. Contrary to the first title, the opening sequence in TL2 takes place outdoors. To test the performance I gave a complete run-through of the first level. I explored every point on the map, vanquished every enemy, opened every chest and assigned every skill point and attribute I received from gaining levels. A lot of ratlings died to bring you this data. For the "High" test, Particle Detail was set at Medium, and Shadow Detail as set to High. For the "Very High" test, each attribute was bumped up to its max setting; High and Very High, respectively.

The title performs great. It almost hits 60fps at High, and stays well above 30fps at Very High. Dropping the resolutions to 1600x900 makes little if any difference. While I know the performance will certainly drop in later areas as the number of monsters increases (like it did in TL1), I'm certain the game will perform will at any quality configuration on the A10-5800K.

Performance Summary

The A10-5800K is a solid performer for single screen gaming. My experience here builds on my initial testing late last week - AMD's new APU can consistently deliver 30fps gaming, with good quality settings, at 1080p. As long as your have proper expectations, the A10 could be an economical centerpiece to your next gaming rig.

My final step in this review will be to run through our standard stress testing suite, with a Radeon HD 7970 GHz Edition installed. This will let me focus on the CPU power in the A10-5800K, and see how the top-end APU stacks up against high end CPUs from Intel and AMD.