ATI Radeon 5870 Eyefinity6 Review

Article Type: 
Review
- GPU Review
Teaser Icon: 
Custom
Image: 

This is the day all Eyefinity fans (current and potential) have been waiting for - the formal unveiling of the Radeon HD 5870 Eyefinity6 (E6) card. ATI first teased this card when it initially unveiled the Radeon 5000 series, and its Eyefinity technology. The Eyefinity6 card was seen by many members of the WSGF as the trifecta in multi-monitor gaming:

The E6 card is the last piece in the Radeon 5000 series lineup. In the time since the E6 announcement, the product landscape has changed and the E6 may not be the holy grail users once thought it to be:

Do these additional options hold the performance key for users with “just three screens” (irony full intended)? Do you need 2GB of RAM for 3x1 Eyefinity, or is it only relevant for more than three screens? Are any of the options for more than three screens usable for gaming?

We hope to answer all these questions (and more) over the course of this review.


Architecture & Specs

The HD Radeon Eyefinity6 carries almost the exact same specs as the original HD 5870. The only difference is the TDP and the power requirements. The need for more power could already be assumed given there is twice as much RAM to power, and now six ports to keep active. To ensure proper power is supplied to the card, ATI has increased the power requirements from 6+6 to 8+6 pin connections. The price also comes in slightly higher than the original HD 5870, with the E6 carrying an MSRP of $479.

Below is a spec block with comparing the original 5870, the 5970 and the E6.

Card GPUs Transistors Max Memory Shaders Clock (MHz) TDP (Watts) MSRP*
Core Mem Idle Max
ATI Radeon HD 5970 2 2 x 2.15B 2 x 1GB 2 x 1600 725 1000 51 294 $699
ATI Radeon HD 5870 E6 1 2.15B 2GB 1600 850 1200 34 228 $479
ATI Radeon HD 5870 1 2.15B 1GB 1600 850 1200 34 228 $399


A Tale of Two Markets

As referenced on the previous page, the Eyefinity6 brings two unique features over the existing HD 5870. These features are the support for more than three monitors, and the inclusion of 2GB of VRAM on the card. Beyond the productivity benefits, this card has appeal to two distinct gaming markets.


Beyond Three Screens

The first market is obviously those customers who wish to use more than three screens for productivity and/or gaming. Seeing as how the original HD 5870 with 1GB of VRAM was taxed in many games (using max settings), we’ll assume that the 2GB of VRAM is essential to move past three screens.

The original HD 5870 had four available connections, but only three could be used simultaneously. The E6 allows the user to set up six different screens in a multitude of configurations. This new array of options challenges our ideals of multi-monitor gaming and opens the door of multi-monitor gaming to more gamers.

But, will one GPU be enough? Will the gamer who pursues this display configuration have enough horsepower with a single GPU card, or is CrossFireX forgone assumption?


Better Performance on Three Screens

The second market is the current Eyefinity gamer, who probably cut his or her teeth in the world of the Matrox TripleHead2Go. They know first hand the demands of gaming up to 5040x1050, and the Eyefinity now allows for 5760x1200 and beyond.

The WSGF showed in previous benchmarking of the NVIDIA GTX275 that moving from 896MB to 1792MB of VRAM was beneficial in providing a “smoother” experience. Overall average fps didn’t noticeably improve, but “jitters” and “hiccups” from texture swaps were noticeably reduced resulting in smoother gameplay. We were never able to test a GTX285 1GB vs. a GTX285 2GB, so we couldn’t positively determine the “sweet spot.” Is it at 1GB, or just above? Or, is it all the way at 2GB?

Additionally, those tests were in the days of DX9 and a resolution limit of 5040x1050. We’re now in a world of far fewer limits. We easily game at 5760x1200, and fill our games with tessellations, ambient occlusions and “God rays.” Given all these changes, do the old assumptions and findings still stand?

ATI Radeon 5870 Eyefinity6 Review - Eyefinity Updated

Article Type: 
Review

Eyefinity Updated - Rethinking Immersion

Gamers visiting the WSGF can generally be grouped within two categories: those that are looking for information about gaming on widescreen displays (information we have been pioneering since 2003), and those looking for information about multi-monitor gaming (first introduced in 2006 with the release of the Matrox TripleHead2Go).

Matrox introduced gamers to “Surround Gaming” with the TH2Go, initially offering 3x1 support with 1280x1024 displays. Much discussion has taken place on the WSGF forums about the benefits and drawbacks of gaming across three “square” panels, when compared to a single (and possibly larger) widescreen display. These discussions had a slight change in direction and tone as Matrox released its updates to allow 3x1 support on widescreen displays. But through all of this time, multi-monitor and “Surround Gaming” ideals were confined to panels in a 3x1 landscape configuration. And the primary ideal was greater immersion through an increased aspect ratio and field of vision.

The release of the original HD 5870, and the current release of the HD 5870 Eyefinity6, continue to bring additional definitions of what can be considered “immersive gaming.” We have seen a shift on the WSGF since the release of the HD 5870 and the introduction of Eyefinity. We have seen more interest and dialogue about “traditional” 3x1 gaming, but we have also seen an interest in other variations such as 3x1 in portrait, 5x1 in portrait, 3x2 landscape and other configurations.


Immersion by Aspect Ratio

Traditionally, 3x1 landscape (3x1-L) has offered a clear sense of immersion over a single widescreen display. It is easy to see the increase aspect ratio offered by three displays, and the increased field of vision when a game properly supports that aspect ratio through “Hor+” or “Pixel based” behavior.

The one area lacking in a 3x1-L implementation is the physical height of the monitors. Most gamers end up with 3x22” or 3x24” monitors, which have an average height of 12” - 14”. While the 3x1-L setup extends to the users’ horizontal periphery, the vertical field of vision still includes objects and items other than the monitor.

The below graphic shows the difference in a single widescreen monitor, and a 3x1-L widescreen configuration. Additionally, most of the existing WSGF Surround gameplay videos on YouTube highlight this difference.

While doing a "final pass" review of the article, something struck me about the configuration options. You could set up 5x4:3 20" monitors, each with a resolution of 1600x1200. The horizontal resolution hits exactly 8000 pixels, which is under the "8k" limit. When compared the 5x1-P, the reduced height produces a wider FOV with fewer pixels to push. The trade-off would be a more pronounced edge distortion.

3x1-L Comp


Immersion by Physical Size

Many detractors of Eyefinity gaming would offer a statement such as, “I’d rather just buy a 50” HDTV and game off of that.” Point taken. However, your 50” TV only has a resolution of 1920x1080. And, that TV is probably sitting 12-18 feet away from you across your living room. If said user sat 2-3 feet from said HDTV, they could easily count the 2,073,000 pixels staring at them and be treated to a nice dose of the “screen door” effect.

The 3x1-P option offered by Eyefinity provides 6.2M pixels. And with a 22” monitor size, the three panels (interior bezels included) offer a diagonal of almost 39”. While this is less than the theoretical 50” HDTV mentioned in the statement above, the fact that the user is sitting less than four feet from these monitors means that your physical vision is fully immersed.

The 3x2-L option now available with the E6 card provides 12.4M pixels. Using six 22” monitors, the total panel size (again, including the interior bezels) is right at 56” on the diagonal. This value will vary slightly, depending on the angle between your center and outside monitors.

The 3x1-P mode already available with the original HD 5870 provides an aspect ratio of approximately 16:9.5 - dead in the middle between 16:9 and 16:10. The 3x2-L mode offered by the HD 5870 Eyefinity6 provides an aspect ratio of 16:6 - slightly wider than a normal widescreen aspect ratio.

Below is a table of current and future options (arranged by increased FOV). Using 16:10 monitors would produce slightly different results - such as slightly narrower aspect in 3x2, and slightly wider aspect in 5x1-P.

16:9 Monitor Config Resolution Aspect Ratio Normalized Aspect Ratio Field of Vision Total Pixels
3x1 Portrait 3240x1920 27:16 16:9.5 87 6.2M
Single Widescreen 1920x1080 16:9 16:9 90 2.1M
3x2 Landscape 5760x2160 16:6 16:6 113 12.4M
5x1 Portrait 5400x1920 45:16 16:5.7 115 10.4M
3x1 Landscape 5760x1080 16:3 16:3 143 6.2M
5 x 4:3 (1600x1200) 8000x1200 20:3 16:2.4 150 9.6M
5x1 Landscale (1600x900) 8000x900 80:9 16:1.8 157 7.2M


Balancing Aspect & Physical Size

ATI once again teases us with a future feature, in the form of 5x1-P. Windows itself handles 5x1-P with an extended Windows desktop (so productivity is covered), however Eyefinity can only group three displays into a portrait configuration. Physically, five monitors in portrait mode is right at the same width as three panels in landscape. Thus, your horizontal peripheral vision is filled. The portrait configuration also ensures that the vertical peripheral vision is filled as well.

I didn’t think I would be a fan of the portrait configuration, and was surprised to find that 3x1-P was Kyle Bennet’s (of HardOCP) preferred setup with the original HD 5870. I have five panels in portrait configuration, and since I’ve been benchmarking the 3x1-P configuration I can see why Kyle preferred this option. Though it isn’t to everyone’s taste, it does have its benefits.

Below is a comparison of the same scene as viewed through different Eyefinity options.

Aspect Comp

While less than the traditional 3x1-L aspect (of approximately 16:3.2), the 5x1-P implementation of 16:5.5 offers obvious enhancements over 16:9 or 16:10. Personally, I see 5x1-P as the perfect balance for physical and aspect immersion, for the following reasons:

The images and data for the table were pulled using the new WSGF FOV and Aspect Ratio Calculator created by forum member Delphium. The tool is a great update to our original FOV tool, and provides a wealth of information. Thanks for all of his hard work on this excellent tool.

ATI Radeon 5870 Eyefinity6 Review - Benchmarking

Article Type: 
Review

Setup & Installation

Cable connection is easier with Display Port cables. No longer are you required to plug in cables and twist tiny cramped thumb screws. DP cables click in and release with the press of a button. ATI does provide an array of adapters with the retail package, though it doesn’t offer enough to utilize Display Port connections for each monitor. Personally, I try to avoid adapters as much as possible, as I want to reduce the number of opportunities for failure and loose connections. Considering that all of my monitors have a DP connection, I purchased Mini-DP > DP cables from Monoprice.

Installation of the Catalyst Control Panel and configuring a 3x1-L or 3x2 setup follows the same steps as before. Setting up a 3x1-P configuration requires the user to first set the monitor rotation, and then create the Eyefinity group.

The one catch piece that is different comes when setting up a 3x1-P Eyefinity group while having five monitors attached. The group utilizes only active monitors, and will default to begin the group with the right-most monitor. To center the 3x1-P gaming experience, first disable the two outer monitors. You can then create the Eyefinity group without a program. In the future we’ll post a video on creating a 3x1-P group within the 5x1-P setup, and post on our YouTube Channel.

Bezel Compensation works as previously demonstrated. If you missed my original video on setting up the Bezel Compensation, you can view it below. With a 3x2 setup, you configure both horizontal and vertical bezels. In the final step, you end up with a triangle across each bezel to verify the settings. I didn’t use any Bezel Compensation in my review or benchmarking. One, we had already covered it with this YouTube video; and two, I wanted to focus on consistent benchmarking at the native resolutions.


System Specs

For all 16:10 benchmarks I used my existing three Dell U2410 monitors. For the 16:9 benchmarks, I used Dell P2210H monitors provided by ATI. The 10.3 preview driver was used on all cards except the HD 5870 Eyefinity6. The 10.3a preview driver was used on the Eyefinity6 card, as it offered a number of bug fixes and stability improvements.

My testing rig remains unchanged (except for driver updates and monitor configurations). It currently stands at:


Resolutions Tested

I initially tested the 5870 at 1920x1200 and 1680x1050, along with the respective 3x1-L iterations of 5760x1200 and 5040x1050. I continued this testing to allow for contrast and comparison across the ATI line of cards. Considering that the monitors ATI provided for the 3x2 setup were 1920x1080 monitors, adjustments were needed to the testing regimen.

I tested at 5760x1080 and 4800x900 to provide a comparison of 3x1-L using 16:10 and 16:9 panels. Previous testing showed minimal differences in these panels at since screen resolutions, so I did not test 1920x1080 or 1600x900. To provide an additional aspect comparison to 3x1-P, I tested 3x1-L using the 16:10 panels. This allows me to highlight any impact that aspect ratio has on framerates. These configurations offer an identical number of pixels (over 6M), but wildly different aspect ratios.

Finally, I tested the 3x2 configuration at 5760x2160 and 4800x900. This offers direct comparisons in moving from 3x1-L to 3x2.


Games Tested

I wanted to choose games that covered a variety of genres (action, FRP, RTS and racing), and a variety of technologies (DX9, 10 and 11). Some games are older and well known titles such as Half-Life 2 and Far Cry 2. Half-Life 2 chews through video cards at lower resolutions and even 3x1-L, but how does it scale to five and six monitors? Far Cry 2 is still tough on systems (at Ultra settings). Will it even be playable at these new configurations. I also wanted to test games that were new and demanding, so that we can begin "aging" them over time. I chose titles such as Battle Forge and the new S.T.A.L.K.E.R. demo for these reasons.

I chose games that had a built-in benchmark tool. This allows for repeatability and a relative "hands off" testing. Finally, all games must exhibit Hor+ behavior in widescreen and Eyefinity. The games I ended up testing were:

ATI Radeon 5870 Eyefinity6 Review - Batman: Arkham Asylum

Article Type: 
Review
Batman: Arkham Asylum
Eyefinity 5760x1200 (fps)
  1 GPU 2 GPU
1GB 2 4
2GB 39 55

Batman: Arkham Asylum is the well received action title by Rocksteady. The game offers very detailed environments with a great visual style and high quality. There are known issues with AA using ATI cards. You cannot set the AA level from within the game, and have to force it with the Catalyst Control panel. All settings within the game were maxed out. I used 4xAA with Adaptive Multi-Sampling (the balance between Quality and Performance) from within the CCC.

The AA implementation produces a greater than expected impact on the performance. I know there are ways to tweak the game to use the AA from within the game itself. However, I believe finding the information and implementing it is beyond the average gamer. While an enthusiast (i.e., target market of both the WSGF and this card) might go through the trouble, many mainstream gamers will not. My testing actually ran all the way down through the Radeon HD 5450, and using the "native" options provided the most consistent platform for testing.

The games runs like a demon in normal widescreen on any of these cards, with 4xAA. However, performance takes the expected hits and Eyefinity, and this is what we are really here to test. At the native resolutions of 5760x1200 (3x1920x1200), the game literally crawls with 4xAA on the 1GB card. Adding the additional VRAM provides playable experience, though it doesn't crack 60fps. In the end, you will have to make some adjustments to hit 60fps, and dropping to 2xAA gives massive speed improvements.

Batman:AA (pun actually not intended) is a unique title. It is made well, and plays well, but technical issues can really hamper performance. The addition of the extra 1GB shows how an increased frame buffer helps as AA increases. However, in this one instance the better choice is obviously to turn down the settings based on the shortcomings in development.

Our "Gold Standard" for testing is to use 4xAA/16xAF. In this once instance that just doesn't seem feasible. In dropped the settings to 2xAA we hit 42fps on a single Eyefinity6. Adjusting the AA type to MSAA (from Adaptive MSAA) made no difference.


Hitting 60fps

Turning off AA allows you to hit 60fps with a single Eyefinity6. I also hit 57fps with 2xAA and lowering the settings to "High" from "Very High." The only difference this makes is to turn off Ambient Occlusion.



Batman: AA

ATI Radeon 5870 Eyefinity6 Review - Battle Forge

Article Type: 
Review
Battle Forge
Eyefinity 5760x1200 (fps)
  1 GPU 2 GPU
1GB 18 23
2GB 22 38

Battle Forge is the free-to-play RTS from Electronic Arts. It offers a steampunk/fantasy RTS experience, where armies are build based on "decks" of cards similar to the Magic: The Gathering card game.

Battle Forge is one of ATI's spotlight (my terminology) games for the HD 5000 series cards, as it offers both DX11 and proper Eyefinity support. The game offers a number of DX11 features, and a wealth of options for tuning performance. Specifically, Battle Forge uses DX11 and Shader Model 5.0 to compute HighDefinition Ambient Occlusion (HDAO). For our tests we maxed out all of the settings and forced DX11 through the config.xml file.

The test is actually quite strenuous with the number of objects, effects and particles on the screen at one time. There is a noticeable performance increase in moving to the 2GB frame buffer of the E6. The dual GPUs in the 5970 generally offer better performance that either single GPU card. However, pairing two GPUs each to 2GB of VRAM shows obvious performance improvements over the 5970.

Hitting 60fps

Setting AA to 2x, and setting Shadow Quality, SSAO and FX Quality at High (from Very High) will get you 30fps in Battle Forge. I could only get in the 40's if I set it to 4800x900, turned off AA, stopped forcing DX11, turned off SSAO and set everything to medium. Battle Forge is simply a 30fps game in Eyefinity.



Battleforge

ATI Radeon 5870 Eyefinity6 Review - DiRT 2

Article Type: 
Article
DiRT 2 (DX11)
Eyefinity 5760x1200 (fps)
  1 GPU 2 GPU
1GB 38 54
2GB 38 66

Dirt 2 is the latest iteration of the Dirt rally racing series from Codemasters. Like Battle Forge, Dirt 2 is a spotlight game for ATI with the HD 5000 series. Like Battle Forge it offers proper Hor+ gameplay in Eyefinity and DX11 support. Unless the user goes into the "hardware_settings_config.xml" file and forces DX9, Dirt 2 runs in DX11 mode. Unfortunately Dirt 2 does not offer a DX10 mode. This is unfortunate, as many games show improved performance when running in DX10 vs. DX9.

The true (noticeable) DX11 features come in to play based on the user settings in the in-game graphics options. Several key features are the "Hardware Tessellated Dynamic Water" (achieved through "Ultra" quality water), "Hardware Tessellated Dynamic Cloth" (achieved through "High" quality cloth), and DX11 Accelerated HDAO (through "High" quality HDAO).

The DX11 water and cloth offer more realistic geometry and movement. The DX11 water produces actual waves in deep puddles (as the player drives through), rather than simple "swirls" in the texture surface. The DX11 cloth offers more realistic ripples and waves in the cloth material over the DX9 version. On the other hand, the DX11 HD Ambient Occlusion (HDAO) offers an accelerated computation path.

DX11 doesn't necessarily provide earth-shaking changes to gameplay. But, it provides more realistic "movement" in the world's objects - cloth, water, grass, etc. While a DX9 or DX10 game is perfectly enjoyable, the DX11 technology offers better immersion by making the "little things" more lifelike. Additionally, it offers better computation paths through increased parallelism (and better computation paths for DX10), much like DX10 offered better performance (over DX9) in games such as Far Cry 2.

Looking at the below graph, we are starting to hit the CPU limits of a stock Core i7-920. I arrive at this conclusion by seeing how the E6 CFX maxes out at about 74fps with 1680x1050, 1920x1200 and 4800x900 (3x1600x900).

Hitting 60fps

I cracked 60fps at 4800x900 @ 2xAA. I took everything set at "Ultra" down to "High", except for Water. This allowed me to maintain the tessellated water in DX11. I also lowered Post Processing from High to Low. With plenty of options, there are a number of variations available to suit your taste.



DiRT 2

ATI Radeon 5870 Eyefinity6 Review - Far Cry 2

Article Type: 
Review
Far Cry 2 (DX10)
Eyefinity 5760x1200 (fps)
  1 GPU 2 GPU
1GB 38 54
2GB 38 66

Far Cry 2 (and the whole Cry/Crysis series) has long been considered a system killer. If not a killer, then at least a good strong test. As always, we run our test at max settings with 4xAA. The benchmark tool within Far Cry 2 offers settings for High, Very High and Ultra. We chose Ultra with 4xAA. Like we saw with our previous tests of the GTX275, the additional VRAM provides a smoother gameplay experience. While the overall average fps changes very little, there are far more "stops" and "stutters" in the gameplay with a 2GB framebuffer.

As quality and AA increases, the need for a larger framebuffer increases (there is more data to hold in the VRAM). The Far Cry 2 benchmark tool is unique in that it offers a realtime graph of the fps count, and allows us to see the reduction in these stutters (as evidences by the reduction in momentary dips in fps). With the increased framebuffer the game is having to swap textures and other graphic information from the HDD to the video memory on fewer occasions, offering a smoother experience.


Hitting 60fps

In Far Cry 2, 60fps can be attained with 0xAA on Very High quality settings.



Far Cry 2

Far Cry 2 1GB vs. 2GB

ATI Radeon 5870 Eyefinity6 Review - Far Cry 2 Anti-Aliasing Analysis

Article Type: 
Review
Far Cry 2 (DX10)
Eyefinity 5760x1200 (fps)
  1 GPU 2 GPU
1GB 38 54
2GB 38 66

This is a detailed analysis of the different combinations of AA and Quality settings in Far Cry 2. The Eyefinity6 outperforms both the original 5870 and the 5970 at 8xAA. The problem is that no one really plays at that setting, and performance at that point is still not "playable." One thing to note is that the AA setting consistently has a higher impact on performance, rather than the quality (High, Very High, Ultra High) quality setting.


Hitting 60fps

In Far Cry 2, 60fps can be attained with 0xAA on Very High quality settings.



Far Cry 2 AA-1

Far Cry 2 AA-2

ATI Radeon 5870 Eyefinity6 Review - Grand Theft Auto IV

Article Type: 
Review
Grand Theft Auto IV
Eyefinity 5760x1200 (fps)
  1 GPU 2 GPU
1GB 28 32
2GB 27 32

I initially chose GTA IV based on the fact that it had a built-in benchmark tool, and based on the fact that it was considered a system killer at max settings. Realistically, I don't believe the GTA IV benchmark tool offers a good representation of the actual gameplay experience. The benchmark is very much a "corridor" run on a dense city street, while the game itself is open world and many locations offer variety architecture and/or a distant horizon. Additionally, the benchmark is set during a nighttime setting. While this is potentially good for showing off the game's neon lighting effects, it offers no sun, clouds, atmospheric coloring, lens flare or texture variety in the sky.

Additionally, I found that the game hit a CPU limit of 45/46fps at 1680x1050 on a Radeon HD 5830 - not exactly cutting edge hardware. The one benefit of the game is that the graphics option screen offers a reading of your system VRAM and calculates how much is needed based on your selections. By default it won't let you go over the limit of your video card.

With a couple of command line switches, the game allows you to max everything out well and beyond what your system is "capable of." Down side it that overloading the video card seems to have little effect. I'm not certain if the non-existent limit is due to the horsepower of the Radeon 5000 series, or limitations of the benchmark tool.

The only time I found the limit to be an actual impediment is when I tried to max out the settings on a the Radeon 56/44/54xx series with 512MB of RAM. At that low of a framebuffer, the benchmark wouldn't load. Otherwise, I could overload the 1GB VRAM all I wanted, and the game performed reasonably well. Considering the issue of overloading the VRAM, the addition of the second GB of VRAM didn't effect performance.

Hitting 60fps

I worked for a while to get an fps improvement on this benchmark tool. Even cutting the settings to 1/2 or 1/3 of max, and I had only increased 2fps to 30fps total. I'm sure that lower settings would make an impact in the real game, but I'm not seeing it here. I'm also do for a re-format and re-install. Don't think I will be carrying over GTA IV into the benchmark lineup, given its limited impact.



GTA IV

ATI Radeon 5870 Eyefinity6 Review - H.A.W.X.

Article Type: 
Review
H.A.W.X. (DX10)
Eyefinity 5760x1200 (fps)
  1 GPU 2 GPU
1GB 48 64
2GB 48 70

H.A.W.X. is one last title that ATI has been showing off with regards to its Radeon 5000 line. While the other titles are Hor+ and offer cutting-edge DX11 features, HAWX is quite the opposite. While it is a Hor+ title, it is only a DX10 title and runs quite well on a wide variety of hardware. While you can't hit 60fps on a 5700 or 5600 card with any great detail, 30fps is rather easily attainable.

The well running and scalable title makes it a natural fit when showcasing both lower-end hardware pushing three panels, and high-end hardware pushing six panels. It also comes as no surprise that the 2nd GB of VRAM largely goes unused, considering the title runs capably on lower hardware.

One note in the HAWX benchmark. Above 1920x1200 the game simply would not allow 4xAA. While the scores provide that the title could handle it, the option simply is not available. Though it isn't our norm, we chose to accept 2xAA so that we could get consistent readings across the spectrum of hardware.


Hitting 60fps

I was able to get 65fps by turning each of the DX10 effects down a notch. This means Shadows and Sun Shafts at Low, and SSAO at Medium.



HAWX

ATI Radeon 5870 Eyefinity6 Review - Half-Life 2

Article Type: 
Review
Half-Life 2: Episode 2
Eyefinity 5760x1200 (fps)
  1 GPU 2 GPU
1GB 77 95
2GB 77 93


 

Half-Life 2: Lost Coast
Eyefinity 5760x1200 (fps)
  1 GPU 2 GPU
1GB 122 160
2GB 121 148

Half-Life 2 is the classic first person shooter. It is a DX9 title. At this point DX9 doesn't really tax graphics hardware with any of its "features," and any benchmarking comes down to a pushing raw pixels. Half-Life 2 does not like the extra video RAM provided by the Eyefinity6, and until you get to Eyefinity and beyond it doesn't like CrossFireX.

Considering that the single Eyefinity6 pushes past 60fps on the six-panel setup, and the single HD 5870 pushes past 120fps in 3x1-L Eyefinity, it is probably time to retire this benchmark.

Hitting 60fps

The Source Engine hits 60fps in HL2 at max settings on one Eyefinity6 card.



HL2 Ep2

HL2: LC

ATI Radeon 5870 Eyefinity6 Review - Heaven Demo

Article Type: 
Review
Heaven (DX9)
Eyefinity 5760x1200 (fps)
  1 GPU 2 GPU
1GB 17 22
2GB 18 22

 

Heaven (DX10)
Eyefinity 5760x1200 (fps)
  1 GPU 2 GPU
1GB 1 2
2GB 15 26

 

Heaven (DX11)
Eyefinity 5760x1200 (fps)
  1 GPU 2 GPU
1GB 1 2
2GB 12 21

The Unigine Heaven Demo is unique in that it is the only demo which allows for the following components in one package:

  • Synthetic Demo (i.e., a demo designed to "test" a system)
  • Comparable tests of DX9, DX10 and DX11
  • Is Hor+ (rather than limited to a few predefined aspect ratios)

The ability to compare DX9, DX10 and DX11 in the same environment allows for the unique ability to see when and if additional VRAM comes into play. For the DX9 test, the additional VRAM of the Eyefinity6 card offers no improvement over the original Radeon HD 5870. However, the additional frame buffer does provide additional benefits in a multi-GPU scenario.

The real interesting comparisons start at DX10. The additional framebuffer offers no help at normal widescreen aspect ratios. However, the 2GB of VRAM offers substantial improvement over 1GB in Eyefinity resolutions. The difference moves between a veritable slideshow and something bordering on playability. While additional tweaks would be needed to hit 30fps+ with the E6 card, it simply does not appear possible to do so with the original 5870 and maintain the DX10 path.

The DX10 trend continues with the DX11 code, and the extra VRAM shows its value. In both DX10 and DX11, the additional framebuffer completely outpaced the horsepower of the second GPU in the Radeon 5970.


Hitting 60fps

It is not possible to hit 60fps with this benchmark on a single GPU at Eyefinity resolutions. Considering it is meant to test/tax your system, that is to be expected. 40fps+ can be hit with 0xAA and 1xAF, but that looks downright horrible. Here is what I was able to achieve with a reasonable quality:

  • DX9: 4800x900, 2xAA, 4xAF, low shaders - 34fps, med shaders - 28fps
  • DX10: 4800x900, 2xAA, 4xAF, low shaders - 32fps, med shaders - 26fps
  • DX11: 4800x900, 2xAA, 4xAF, tessellation enabled, low shaders - 23fps, med shaders - 20fps

ATI Radeon 5870 Eyefinity6 Review - Heaven Demo DX9

Article Type: 
Review
Heaven (DX9)
Eyefinity 5760x1200 (fps)
  1 GPU 2 GPU
1GB 17 22
2GB 18 22

The Unigine Heaven Demo is unique in that it is the only demo which allows for the following components in one package:

  • Synthetic Demo (i.e., a demo designed to "test" a system)
  • Comparable tests of DX9, DX10 and DX11
  • Is Hor+ (rather than limited to a few predefined aspect ratios)

The ability to compare DX9, DX10 and DX11 in the same environment allows for the unique ability to see when and if additional VRAM comes into play. For the DX9 test, the additional VRAM of the Eyefinity6 card offers no improvement over the original Radeon HD 5870. However, the additional frame buffer does provide additional benefits in a multi-GPU scenario.

The real interesting comparisons start at DX10. The additional framebuffer offers no help at normal widescreen aspect ratios. However, the 2GB of VRAM offers substantial improvement over 1GB in Eyefinity resolutions. The difference moves between a veritable slideshow and something bordering on playability. While additional tweaks would be needed to hit 30fps+ with the E6 card, it simply does not appear possible to do so with the original 5870 and maintain the DX10 path.

The DX10 trend continues with the DX11 code, and the extra VRAM shows its value. In both DX10 and DX11, the additional framebuffer completely outpaced the horsepower of the second GPU in the Radeon 5970.


Hitting 60fps

It is not possible to hit 60fps with this benchmark on a single GPU at Eyefinity resolutions. Considering it is meant to test/tax your system, that is to be expected. 40fps+ can be hit with 0xAA and 1xAF, but that looks downright horrible. Here is what I was able to achieve with a reasonable quality:

  • DX9: 4800x900, 2xAA, 4xAF, low shaders - 34fps, med shaders - 28fps



Heaven DX9

ATI Radeon 5870 Eyefinity6 Review - Heaven Demo DX10

Article Type: 
Review
Heaven (DX10)
Eyefinity 5760x1200 (fps)
  1 GPU 2 GPU
1GB 1 2
2GB 15 26

The Unigine Heaven Demo is unique in that it is the only demo which allows for the following components in one package:

  • Synthetic Demo (i.e., a demo designed to "test" a system)
  • Comparable tests of DX9, DX10 and DX11
  • Is Hor+ (rather than limited to a few predefined aspect ratios)

The ability to compare DX9, DX10 and DX11 in the same environment allows for the unique ability to see when and if additional VRAM comes into play. For the DX9 test, the additional VRAM of the Eyefinity6 card offers no improvement over the original Radeon HD 5870. However, the additional frame buffer does provide additional benefits in a multi-GPU scenario.

The real interesting comparisons start at DX10. The additional framebuffer offers no help at normal widescreen aspect ratios. However, the 2GB of VRAM offers substantial improvement over 1GB in Eyefinity resolutions. The difference moves between a veritable slideshow and something bordering on playability. While additional tweaks would be needed to hit 30fps+ with the E6 card, it simply does not appear possible to do so with the original 5870 and maintain the DX10 path.

The DX10 trend continues with the DX11 code, and the extra VRAM shows its value. In both DX10 and DX11, the additional framebuffer completely outpaced the horsepower of the second GPU in the Radeon 5970.


Hitting 60fps

It is not possible to hit 60fps with this benchmark on a single GPU at Eyefinity resolutions. Considering it is meant to test/tax your system, that is to be expected. 40fps+ can be hit with 0xAA and 1xAF, but that looks downright horrible. Here is what I was able to achieve with a reasonable quality:

  • DX10: 4800x900, 2xAA, 4xAF, low shaders - 32fps, med shaders - 26fps



Heaven DX10

ATI Radeon 5870 Eyefinity6 Review - Heaven Demo DX11

Article Type: 
Review
Heaven (DX11)
Eyefinity 5760x1200 (fps)
  1 GPU 2 GPU
1GB 1 2
2GB 12 21

The Unigine Heaven Demo is unique in that it is the only demo which allows for the following components in one package:

  • Synthetic Demo (i.e., a demo designed to "test" a system)
  • Comparable tests of DX9, DX10 and DX11
  • Is Hor+ (rather than limited to a few predefined aspect ratios)

The ability to compare DX9, DX10 and DX11 in the same environment allows for the unique ability to see when and if additional VRAM comes into play. For the DX9 test, the additional VRAM of the Eyefinity6 card offers no improvement over the original Radeon HD 5870. However, the additional frame buffer does provide additional benefits in a multi-GPU scenario.

The real interesting comparisons start at DX10. The additional framebuffer offers no help at normal widescreen aspect ratios. However, the 2GB of VRAM offers substantial improvement over 1GB in Eyefinity resolutions. The difference moves between a veritable slideshow and something bordering on playability. While additional tweaks would be needed to hit 30fps+ with the E6 card, it simply does not appear possible to do so with the original 5870 and maintain the DX10 path.

The DX10 trend continues with the DX11 code, and the extra VRAM shows its value. In both DX10 and DX11, the additional framebuffer completely outpaced the horsepower of the second GPU in the Radeon 5970.


Hitting 60fps

It is not possible to hit 60fps with this benchmark on a single GPU at Eyefinity resolutions. Considering it is meant to test/tax your system, that is to be expected. 40fps+ can be hit with 0xAA and 1xAF, but that looks downright horrible. Here is what I was able to achieve with a reasonable quality:

  • DX11: 4800x900, 2xAA, 4xAF, tessellation enabled, low shaders - 23fps, med shaders - 20fps



Heaven DX11

ATI Radeon 5870 Eyefinity6 Review - S.T.A.L.K.E.R.: Call of Pripyat

Article Type: 
Review
Pripyat: Daylight (DX11)
Eyefinity 5760x1200 (fps)
  1 GPU 2 GPU
1GB 1 1
2GB 14 24

 

Pripyat: Night (DX11)
Eyefinity 5760x1200 (fps)
  1 GPU 2 GPU
1GB 1 1
2GB 12 20

 

Pripyat: Rain (DX11)
Eyefinity 5760x1200 (fps)
  1 GPU 2 GPU
1GB 1 1
2GB 12 21

 

Pripyat: Sun Shafts (DX11)
Eyefinity 5760x1200 (fps)
  1 GPU 2 GPU
1GB 1 1
2GB 9 15

S.T.A.L.K.E.R. - Call of Pripyat is the new Crysis.

Prior to the game's release, the developer put out a benchmarking tool to test your system configuration. It offers a number of different options for utilizing DX9, 10 or 11 code paths. It also offers options for varying levels of HDAO and Shadow Quality. The demo itself isn't very pretty to look at (lots of dirt and dirt-colors), but it does put a beating on your system.

The additional VRAM of the Eyefinity6 comes into play at any Eyefinity resolution. At 5760x1200 and above the Radeon HD 5970 comes to the same crawl as the original HD 5870. While there are plenty of options to dial back, the additional frame buffer of the Eyefinity6 allows the user more options on how they configure their exprience.


Hitting 60fps

60fps will be tough to achieve and still utilize all the DX10/11 goodies they are tossing at you. 30fps may be a better goal (on a single card), like with Crysis before it. Two cards and you might get the lofty 60fps, but not with max settings and certainly not with Eyefinity. I was able to achieve this balance of performance and quality, basically doubling fps. You can get a few more fps by dropping to DX10 and reducing SSAO even more, but it's only 1-2fps for a lot of quality loss.

High Preset, DX11, 4800x900, 2xAA, DX10.1 style MSAA, HDAO SSAO, Ultra SSAO, Use DX10.1, Enable Tessellation, Contact Hardening Shadows:

  • Day - 25fps
  • Night - 23fps
  • Rain - 25fps
  • Sun Shafts - 20fps

ATI Radeon 5870 Eyefinity6 Review - S.T.A.L.K.E.R.: Call of Pripyat (Daylight)

Article Type: 
Review
Pripyat: Daylight (DX11)
Eyefinity 5760x1200 (fps)
  1 GPU 2 GPU
1GB 1 1
2GB 14 24

S.T.A.L.K.E.R. - Call of Pripyat is the new Crysis.

Prior to the game's release, the developer put out a benchmarking tool to test your system configuration. It offers a number of different options for utilizing DX9, 10 or 11 code paths. It also offers options for varying levels of HDAO and Shadow Quality. The demo itself isn't very pretty to look at (lots of dirt and dirt-colors), but it does put a beating on your system.

The additional VRAM of the Eyefinity6 comes into play at any Eyefinity resolution. At 5760x1200 and above the Radeon HD 5970 comes to the same crawl as the original HD 5870. While there are plenty of options to dial back, the additional frame buffer of the Eyefinity6 allows the user more options on how they configure their exprience.


Hitting 60fps

60fps will be tough to achieve and still utilize all the DX10/11 goodies they are tossing at you. 30fps may be a better goal (on a single card), like with Crysis before it. Two cards and you might get the lofty 60fps, but not with max settings and certainly not with Eyefinity. I was able to achieve this balance of performance and quality, basically doubling fps. You can get a few more fps by dropping to DX10 and reducing SSAO even more, but it's only 1-2fps for a lot of quality loss.

High Preset, DX11, 4800x900, 2xAA, DX10.1 style MSAA, HDAO SSAO, Ultra SSAO, Use DX10.1, Enable Tessellation, Contact Hardening Shadows:

  • Day - 25fps
  • Night - 23fps
  • Rain - 25fps
  • Sun Shafts - 20fps



STALKER - Daylight

ATI Radeon 5870 Eyefinity6 Review - S.T.A.L.K.E.R.: Call of Pripyat (Night)

Article Type: 
Review
Pripyat: Night (DX11)
Eyefinity 5760x1200 (fps)
  1 GPU 2 GPU
1GB 1 1
2GB 12 20

S.T.A.L.K.E.R. - Call of Pripyat is the new Crysis.

Prior to the game's release, the developer put out a benchmarking tool to test your system configuration. It offers a number of different options for utilizing DX9, 10 or 11 code paths. It also offers options for varying levels of HDAO and Shadow Quality. The demo itself isn't very pretty to look at (lots of dirt and dirt-colors), but it does put a beating on your system.

The additional VRAM of the Eyefinity6 comes into play at any Eyefinity resolution. At 5760x1200 and above the Radeon HD 5970 comes to the same crawl as the original HD 5870. While there are plenty of options to dial back, the additional frame buffer of the Eyefinity6 allows the user more options on how they configure their exprience.


Hitting 60fps

60fps will be tough to achieve and still utilize all the DX10/11 goodies they are tossing at you. 30fps may be a better goal (on a single card), like with Crysis before it. Two cards and you might get the lofty 60fps, but not with max settings and certainly not with Eyefinity. I was able to achieve this balance of performance and quality, basically doubling fps. You can get a few more fps by dropping to DX10 and reducing SSAO even more, but it's only 1-2fps for a lot of quality loss.

High Preset, DX11, 4800x900, 2xAA, DX10.1 style MSAA, HDAO SSAO, Ultra SSAO, Use DX10.1, Enable Tessellation, Contact Hardening Shadows:

  • Day - 25fps
  • Night - 23fps
  • Rain - 25fps
  • Sun Shafts - 20fps



STALKER - Night

ATI Radeon 5870 Eyefinity6 Review - S.T.A.L.K.E.R.: Call of Pripyat (Rain)

Article Type: 
Review
Pripyat: Rain (DX11)
Eyefinity 5760x1200 (fps)
  1 GPU 2 GPU
1GB 1 1
2GB 12 21

S.T.A.L.K.E.R. - Call of Pripyat is the new Crysis.

Prior to the game's release, the developer put out a benchmarking tool to test your system configuration. It offers a number of different options for utilizing DX9, 10 or 11 code paths. It also offers options for varying levels of HDAO and Shadow Quality. The demo itself isn't very pretty to look at (lots of dirt and dirt-colors), but it does put a beating on your system.

The additional VRAM of the Eyefinity6 comes into play at any Eyefinity resolution. At 5760x1200 and above the Radeon HD 5970 comes to the same crawl as the original HD 5870. While there are plenty of options to dial back, the additional frame buffer of the Eyefinity6 allows the user more options on how they configure their exprience.


Hitting 60fps

60fps will be tough to achieve and still utilize all the DX10/11 goodies they are tossing at you. 30fps may be a better goal (on a single card), like with Crysis before it. Two cards and you might get the lofty 60fps, but not with max settings and certainly not with Eyefinity. I was able to achieve this balance of performance and quality, basically doubling fps. You can get a few more fps by dropping to DX10 and reducing SSAO even more, but it's only 1-2fps for a lot of quality loss.

High Preset, DX11, 4800x900, 2xAA, DX10.1 style MSAA, HDAO SSAO, Ultra SSAO, Use DX10.1, Enable Tessellation, Contact Hardening Shadows:

  • Day - 25fps
  • Night - 23fps
  • Rain - 25fps
  • Sun Shafts - 20fps



STALKER - Rain

ATI Radeon 5870 Eyefinity6 Review - S.T.A.L.K.E.R.: Call of Pripyat (Sun Shafts)

Article Type: 
Review
Pripyat: Sun Shafts (DX11)
Eyefinity 5760x1200 (fps)
  1 GPU 2 GPU
1GB 1 1
2GB 9 15

S.T.A.L.K.E.R. - Call of Pripyat is the new Crysis.

Prior to the game's release, the developer put out a benchmarking tool to test your system configuration. It offers a number of different options for utilizing DX9, 10 or 11 code paths. It also offers options for varying levels of HDAO and Shadow Quality. The demo itself isn't very pretty to look at (lots of dirt and dirt-colors), but it does put a beating on your system.

The additional VRAM of the Eyefinity6 comes into play at any Eyefinity resolution. At 5760x1200 and above the Radeon HD 5970 comes to the same crawl as the original HD 5870. While there are plenty of options to dial back, the additional frame buffer of the Eyefinity6 allows the user more options on how they configure their exprience.


Hitting 60fps

60fps will be tough to achieve and still utilize all the DX10/11 goodies they are tossing at you. 30fps may be a better goal (on a single card), like with Crysis before it. Two cards and you might get the lofty 60fps, but not with max settings and certainly not with Eyefinity. I was able to achieve this balance of performance and quality, basically doubling fps. You can get a few more fps by dropping to DX10 and reducing SSAO even more, but it's only 1-2fps for a lot of quality loss.

High Preset, DX11, 4800x900, 2xAA, DX10.1 style MSAA, HDAO SSAO, Ultra SSAO, Use DX10.1, Enable Tessellation, Contact Hardening Shadows:

  • Day - 25fps
  • Night - 23fps
  • Rain - 25fps
  • Sun Shafts - 20fps



STALKER - Sun

ATI Radeon 5870 Eyefinity6 Review - Analysis

Article Type: 
Review

This section of the review compiles all of the data and puts together several pieces of analysis.

Aspect Scaling

If you look at some of our early benchmarking, where I tested every step from 1024x768 to 5040x1050, you can see a number of instances where aspect ratio made an impact on performance. In one particular article I looked at how 1280x720 compared to 1920x480. The first is an "HDTV" widescreen that was (and still is) popular on laptops. The second was the lowest possible Surround resolution from the TripleHead2Go - 3x640x480. The interesting thing is that both resolutions have 921,600 pixels - exactly. We did find that moving from widescreen to Surround made a noticeable impact on performance.

Those were the days of DX9 when the biggest impact to performance seemed to be resolution and (based on our finding) aspect ratio. Now we are in a DX10/11 world where SSAO and Tessellation bring their own impacts. Do these impacts overshadow any issues with aspect ratio? Are today's graphics cards so power that aspect ratio is no longer an issue?

I used 3x1920x1200 (16:10 Landscape Eyefinity) as a baseline, and then tested two variations. They are 3x1200x1920 (16:10 Portrait Eyefinity) and 3x1920x1080 (16:9 Landscape Eyefinity).


CrossFireX Scaling

Forum Moderator Paradigm Shifter asked if I could do some scaling charts for the Eyefinity6 vs. E6 CFX performance. I've set them up into three groups. They are scaling with widescreen, 3x1-L Eyefinity and 3x2 Eyefinity. Hopefully these will be a couple more pieces of analysis to help you make buying decisions. I looked at CFX scaling in three different configurations:


Hitting 60fps

While our testing is designed to stress the card as much as possible, we all realize that games have quality settings for a reason. So, beyond seeing where the breaking point is for any individual card, I wanted to provide some insight into what image quality you can expect and get 60fps at 5760x1080 (3x1920x1080) Eyefinity.

Note, it is simply not possible to hit 60fps in some games at this resolution. In those instances we shot for 30fps.

ATI Radeon 5870 Eyefinity6 Review - Aspect Ratio Analysis

Article Type: 
Review

If you look at some of our early benchmarking, where I tested every step from 1024x768 to 5040x1050, you can see a number of instances where aspect ratio made an impact on performance. In one particular article I looked at how 1280x720 compared to 1920x480. The first is an "HDTV" widescreen that was (and still is) popular on laptops. The second was the lowest possible Surround resolution from the TripleHead2Go - 3x640x480. The interesting thing is that both resolutions have 921,600 pixels - exactly. We did find that moving from widescreen to Surround made a noticeable impact on performance.

Those were the days of DX9 when the biggest impact to performance seemed to be resolution and (based on our finding) aspect ratio. Now we are in a DX10/11 world where SSAO and Tessellation bring their own impacts. Do these impacts overshadow any issues with aspect ratio? Are today's graphics cards so power that aspect ratio is no longer an issue?

I used 3x1920x1200 (16:10 Landscape Eyefinity) as a baseline, and then tested two variations. They are 3x1200x1920 (16:10 Portrait Eyefinity) and 3x1920x1080 (16:9 Landscape Eyefinity). You can see the data below.

In general it ends up being a mixed bag, and largely based on the game. For the most part, the differences are within 10% of the baseline and tend to show improvements in both variations. The first five titles do show a drop in performance in 3x1-P. I have a couple of theories on this:

  • Batman - Even though the game is Hor+, the screen caps I took from the demo prove it to be Vert-. So, the 3x1-P is actually showing more than the 3x1-L in 16:9.
  • Battle Forge - In this demo, the action is concentrated on the center of the screen(s). The outer edges are the non-descript and non-textured walls of a chasm. There isn't much to tax the system there and it almost appear that those areas help to boost the scores (much like staring at a wall in an FPS helps your fps). When you set the screen to 3x1-P, which is a very large 16:9.5, you end up with the action focused on a super high resolution with very large models and set pieces.

If the game isn't providing a great deal of imagery (and high-impact imagery at that) in the outer screens, then you will probably see a performance drop when moving to a 3x1-P mode.

One thing was confirmed out of this - there is an issue with CFX scaling in Portrait mode in DX9. You will notice how the DX9 Heaven takes a huge hit to performance as CFX is disengaged. Also, we see minimal boots in Half-Life 2 (another DX9 title) with CFX, when compared to the single card performance. This issue is being discussed in the review discussion thread.



AspectScaling1

AspectScaling2

ATI Radeon 5870 Eyefinity6 Review - CrossFireX Scaling: 3x1 Eyefinity

Article Type: 
Review

Forum Moderator Paradigm Shifter asked if I could do some scaling charts for the Eyefinity6 vs. E6 CFX performance. I've set them up into three groups. They are scaling with widescreen, 3x1-L Eyefinity and 3x2 Eyefinity. Hopefully these will be a couple more pieces of analysis to help you make buying decisions.

This is the 3x1 Eyefinity scaling data. At Eyefinity resolutions you start to see more scaling. The scaling in many games is quite impressive.



CFX-EF1

CFX-EF2

ATI Radeon 5870 Eyefinity6 Review - CrossFireX Scaling: 3x2 Eyefinity

Article Type: 
Review

Forum Moderator Paradigm Shifter asked if I could do some scaling charts for the Eyefinity6 vs. E6 CFX performance. I've set them up into three groups. They are scaling with widescreen, 3x1-L Eyefinity and 3x2 Eyefinity. Hopefully these will be a couple more pieces of analysis to help you make buying decisions.

This is the 3x2 Eyefinity scaling data. At Eyefinity resolutions you start to see more scaling. The scaling in many games is quite impressive.



CFX-2EF1

CFX-2EF2

ATI Radeon 5870 Eyefinity6 Review - CrossFireX Scaling: Widescreen

Article Type: 
Review

Forum Moderator Paradigm Shifter asked if I could do some scaling charts for the Eyefinity6 vs. E6 CFX performance. I've set them up into three groups. They are scaling with widescreen, 3x1-L Eyefinity and 3x2 Eyefinity. Hopefully these will be a couple more pieces of analysis to help you make buying decisions.

This is the widescreen scaling data, along with a few notes.

Half-Life 2 doesn't "scale" well, but it's already pushing 150-200fps at these resolutions. Not much more room for any scaling.



CFX-WS1

CFX-WS2

ATI Radeon 5870 Eyefinity6 Review - Hitting 60fps

Article Type: 
Review

While our testing is designed to stress the card as much as possible, we all realize that games have quality settings for a reason. So, beyond seeing where the breaking point is for any individual card, I wanted to provide some insight into what image quality you can expect and get 60fps at 5760x1080 (3x1920x1080) Eyefinity.

Note, it is simply not possible to hit 60fps in some games at this resolution. In those instances we shot for 30fps.

Batman: Arkham Asylum - Turning off AA allows you to hit 60fps with a single Eyefinity6. I also hit 57fps with 2xAA and lowering the settings to "High" from "Very High." The only difference this makes is to turn off Ambient Occlusion.

Battle Forge - etting AA to 2x, and setting Shadow Quality, SSAO and FX Quality at High (from Very High) will get you 30fps in Battle Forge. I could only get in the mid 50's if I set it to 4800x900, turned off AA, stopped forcing DX11, turned off SSAO and set everything to Medium. Battle Forge is simply a 30fps game (with any amount of eye candy).

Dirt 2 - I cracked 60fps at 4800x900 @ 2xAA. I took everything set at "Ultra" down to "High", except for Water. This allowed me to maintain the tessellated water in DX11. I also lowered Post Processing from High to Low. With plenty of options, there are a number of variations available to suit your taste.

Far Cry 2 - 60fps can be attained with 0xAA on Very High quality settings.

GTA IV - I worked for a while to get an fps improvement on this benchmark tool. Even cutting the settings to 1/2 or 1/3 of max, and I had only increased 2fps to 30fps total. I'm sure that lower settings would make an impact in the real game, but I'm not seeing it here. I'm also do for a re-format and re-install. Don't think I will be carrying over GTA IV into the benchmark lineup, given its limited impact.

HAWX - I was able to get 65fps by turning each of the DX10 effects down a notch. This means Shadows and Sun Shafts at Low, and SSAO at Medium.

Heaven Demo - It is not possible to hit 60fps with this benchmark on a single GPU at Eyefinity resolutions. Considering it is meant to test/tax your system, that is to be expected. 40fps+ can be hit with 0xAA and 1xAF, but that looks downright horrible. Here is what I was able to achieve with a reasonable quality:

  • DX9: 4800x900, 2xAA, 4xAF, low shaders - 34fps, med shaders - 28fps
  • DX10: 4800x900, 2xAA, 4xAF, low shaders - 32fps, med shaders - 26fps
  • DX11: 4800x900, 2xAA, 4xAF, tessellation enabled, low shaders - 23fps, med shaders - 20fps

Half-Life 2 - The Source Engine hits 60fps in HL2 at max settings on one Eyefinity6 card.

STALKER - 60fps will be tough to achieve and still utilize all the DX10/11 goodies they are tossing at you. 30fps may be a better goal (on a single card), like with Crysis before it. Two cards and you might get the lofty 60fps, but not with max settings and certainly not with Eyefinity. I was able to achieve this balance of performance and quality, basically doubling fps. You can get a few more fps by dropping to DX10 and reducing SSAO even more, but it's only 1-2fps for a lot of quality loss. High Preset, DX11, 4800x900, 2xAA, DX10.1 style MSAA, HDAO SSAO, Ultra SSAO, Use DX10.1, Enable Tessellation, Contact Hardening Shadows:

  • Day - 25fps
  • Night - 23fps
  • Rain - 25fps
  • Sun Shafts - 20fps

ATI Radeon 5870 Eyefinity6 Review - Productivity & Entertainment

Article Type: 
Review

Productivity & Multitasking

While it's not a direct benefit of Eyefinity, one of the greatest benefits I find of a multi-monitor system is the ability to create a customized media hub. Most monitors will let you attach more than one device, so even with a single monitor system it's possible to hook up both your PC (DVI) and your PS3 (HDMI). However, I find that the multitude of connections available in a multi-monitor configuration opens up a wealth of possibilities.

These benefits shouldn't be discounted when you are considering and investment in an Eyefinity system. And, the additional options available with moving to a five or six monitor setup should be considered as well.

I do need to point out one issue with multi-tasking on DisplayPort connections. DP is an active connection, and communication runs both ways. If you disconnect the cable, Windows will know. If you have an Eyefinity grouping with that monitor, it will be broken. This could prove problematic if you are in-game and change monitor inputs to multi-task for some reason. You may be able to get around this by using the DP>DVI adapters, but I'm not sure.

I was surprised the first time I experienced this on the Dell P2210H monitors, as it hadn't occurred with the Dell U2410's. My best guess is that the Picture-In-Picture feature on the U2410 keeps all the connections active all the time. There may be another reason, but do keep this in mind.


Moving from Landscape to Portrait

Moving to portrait does have some drawbacks, but 3x1-L and 3x2 have theirs as well. The portrait configuration requires you to look at certain documents or programs in a different light. It's helped me see the WSGF as many other members do. Most other websites are rather narrow as well (fitting within the 1080px) and browsing is actually improved with the longer length.

Multiple spreadsheets or PowerPoint style documents fit better one above the other on a portrait display, versus side by side on a landscape. Dreamweaver offers a similar benefit with allowing code and page preview to both show extensively on over the other. The only real productivity issue I've found is Photoshop, but I'm learning to live with that. Finally, DVDs and Netflix streaming still plays at full resolution within the 1080px width of the monitor.

My main issue with 3x1 is "tennis neck" from looking left and right. On both 3x1 and 3x2 I lost my cursor quite a bit. The landscape offering seems to be better for keeping track of the cursor. Make no mistake, all of the configurations are massive surfaces and can sometimes be a bit much to deal with in a Windows desktop.


5x1 Portrait with Eyefinity6

My big complaint with 3x2 is that neither set of monitors was at the right height. If you centered the overall monitor matrix to the center of your vision, then you were looking at the center bezels. Do to the fact that I have a shallow desk (which backs into a corner), I was sitting too close to the installation. I was craning my neck up or down to see each row of monitors. While backup and focusing on the overall image is fine for gaming, it put productivity tasks a bit out of eyesight.

Given all of that, I was more excited about the possibility of the 5x1-Portrait as it would reduce the overall height and eliminate the center bezel. I found that I could center the monitors and see the top and bottom edges just with moving my eyes (rather than my whole head). Games currently play on the center three monitors (until ATI releases the 5x1-P driver update), and that is where non-Eyefinity games will continue to play in the future. The 5x1-P setup allows for the outer items to multi-task while gaming on the center three.

I set up the five monitors in a similar manner as the 3x2. The Eyefinity6 is connected to all the monitors. The work laptop is connected to the left monitor via VGA, and the Mac Pro is connected to the other four monitors via DVI. Gaming on the three center monitors (and disabling the outer two monitors) allows the outside monitors to be used without breaking the Eyefinity group. In the last shot below, you can see Guild Wars running on the center three monitors while I have the Guild Wars Wiki up on the Mac Pro on the right-hand monitor.

If you have multiple computers, this configuration allows for true multitasking while gaming (with multiple mice and keyboards) without the need for Alt-Tabbing from inside the game.



5x1a
Win7 with 5x1 Extended Desktop

5x1b
Work laptop on left. Mac Pro on remaining four screens

5x1c
Guild Wars in 3x1-P in center. GW Wiki on Mac Pro on right.

ATI Radeon 5870 Eyefinity6 Review - Productivity & Entertainment (Continued)

Article Type: 
Review

3x1 Landscape

Since my first TripleHead2Go, I have attempted to use virtually every available port in a traditional three monitor setup. My last 3x1 setup was three Dell U2410's. I specifically chose them for their multitude of connections: DP, DVIx2, VGA, HDMI and Component. At the time of my initial setup, all existing Eyefinity cards offered a variety of ports versus the all-DisplayPort configuration of the Eyefinity6 card.

Since I couldn't set up the Eyefinity with all DP connections, I chose to use the DisplayPort and DVI-1 connections. Two DVI-2 connections were hooked up to my Mac Pro. The HDMI ports were paired with the HDTV cable box, Xbox 360 and PS3 Slim (aka the Blu-Ray player). The VGA connections were assigned to the external VGA port on my work (day job) laptop, Nintendo Wii and Sega Dreamcast. Yes, VGA cables are available for both the Wii and Dreamcast (and yes they help the picture quality).

I had originally intended to connect the Wii through a set of component cables, and use two VGA connections with my TripleHead2Go and my work laptop. However, the old chipset on the laptop wouldn't accommodate the TH2Go. I purchased the available Sound Bars for each monitor as well. The HDMI devices played directly through. I ran cables from the PC audio into the three cables, hooking up the three bars as Left-Center-Right.



3x1a

3x1b


3x2 with Eyefinity6

Knowing that I wasn't going to keep the 3x2 configuration as my final option, I didn't spend a great deal of time working on a perfect alignment on the monitors. I got them relatively close for testing, but I would do a better job if they were being installed permanently.

In testing the Eyefinity6, my next "new" configuration was the "three over three," or 3x2 configuration. The Dell P2210H monitors that ATI provided have DisplayPort, DVI and VGA connections. They do not have an HDMI connection. I will probably end up getting an HDTV to for all the HDMI devices. However, the additional monitors allowed me to add a 2nd video card to my Mac Pro and connect a total of four monitors. Below are pictures of three different configurations.

Six displays offer a wealth of options. The three I am highlighting here are:

The "three over three" configuration would work well if your monitors had HDMI connections. You could game on the lower three (if that aspect ratio was your preference for that game), and put a movie or TV on in a top screen. I'm not a WoW player, but I do hear there can be a lot of down time while waiting for team and raid groups to gather.



3x2a
3x2 Eyefinity Configuration

3x2b
Work laptop connected in the lower left. Mac Pro on four screens.)

3x3b
3x1 Eyefinity on top three screens. Work laptop and Mac on bottom three.

ATI Radeon 5870 Eyefinity6 Review - Conclusions

Article Type: 
Review

Conclusions

I would recommend the HD Radeon 5870 Eyefinity6 to someone looking to create and Eyefinity setup. I think the incremental cost of $80 is quite reasonable considering the extra 1GB of frame buffer and the additional Eyefinity configurations the E6 card brings. In my mind the real issue of investment is not between the two 5870 cards, but between the Eyefinity6 and the HD 5970.

Based on current demand and initial supplies, the E6 is retailing for about $499. The HD 5970 is still running $699 to $749. From what I've seen in my benchmarks, and in reading the forums, the HD 5970 looks to be a great card - but it is a dead end. With the 5970 card you get blazing fast speed, but you are locked in to three screens with no real options for expansion.

Forum members have not been able to get the 5970 to run in CFX with another 5970 (Quad-Fire) or a 5870 (Tri-Fire). Additionally, my benchmarking proves that the 1GB framebuffer becomes a bottleneck when paired with multiple-GPUs. I do realize the 5970 is clocked slightly lower than the 5870, but I don't believe that solely accounts for all the difference.

On the other hand, a single Radeon 5870 Eyefinity6 gives you 2GB of frame buffer, additional display options, and the ability to run two E6 cards in CrossFireX. My CFX setup ran first time without a hitch.


1GB vs. 2GB

Make no mistake about it - 2GB of VRAM is not needed to play any game today in 3x1 Eyefinity. However, you won't be able to do it at max settings and consistently hit 60fps. You will have to reduce quality, sometimes to the point of virtually abandoning the DX11 code path and feature set all together.

Extra VRAM used to be coveted for things such as bigger textures (those are still out there), bigger resolutions (and those are getting bigger), and anti-aliasing (though 2x is probably fine, really). However, in todays gaming landscape there are many new things vying for your framebuffer - HDAO, Hard Shadows, Tessellation and Full-Screen Post Processing to name a few.

Two 2GB is not a requirement today. Most DX9 titles and less than Eyefinity resolutions simply won't take advantage of the second gigabyte. However, if you want to take advantage of these newer technologies being introduced in DX11 and coming into the maturity from DX10, then 2GB will be your friend. It can smooth out existing gameplay experiences, and open doors to experiencing new DirectX features. This will be even more crucial with 3x1 Eyefinity and Beyond.

For the $80 price difference between the original HD 5870 and the Eyefinity 6 (which also brings six monitor outputs), I would consider the investment a hedge against speedier obsolescence. Even if you don't believe me, or think ATI is offering something you don't need - take a look at their competition. NVIDIA is launching the first reference cards in the new GTX400 (Fermi) line with 1.25GB and 1.5GB. Apparently they see the current or future need for greater than 1GB of frame buffer as well.


New & Future Eyefinity Options

The 3x2 Eyefinity setup is great in creating a huge single mass of displays. Putting six 1080p displays in one location provides over 12M pixels at your disposal, with great options for multi-tasking. For me the middle bezels are a gaming deal breaker, and the physical height of the panels can be a hinderance.

Like the initial release of the Radeon HD 5870, ATI leaves me wanting more. Last time it was Bezel Compensation and better Intermediate Resolutions. They delivered those in Catalyst 10.3 and 10.2, respectively. This time they leave me wanting the 5x1-Portrait configuration. I know that an internal driver is floating around, and hopefully it won't be long before we start to hear more about it. I won't even mention that 7x1-Portrait would still get under the "8k limit" in Windows with 1920x1080 panels (7560x1920 with a 16:4 aspect ratio ^_^)

There is still the lingering issues of Profiles not working consistently. I had them working on the 10.3 Preview driver, but lost them with 10.3a. Granted by that time I had layered about a half dozen drivers on top of each other over the life of this install, and run through almost a dozen video card combinations.

The 10.3a Preview driver was sent out to fix some blanking and sync issues with 5+ panel installs. The driver initially left me with some issues. Once I finished my benchmarking and had time to set up a new clean install, the issues were indeed addressed. I hope for the same thing with the Profiles and Hotkeys.

What's Next?

I will probably go back and re-examine Batman:Arkham Asylum with 2xAA. And, I would like to get in an RTS and MMO. I got World in Conflict, but couldn't get the benchmark properly due to some missing texture issues. I'm hoping the clean install will fix that as well. I have figured out a horse ride I can take through Lord of the Rings Online that will give me a repeatable event. It will also allow me to test DX9, DX10 and the upcoming DX11 patch.

I also want to go back through and figure out what settings you need to hit 30fps or 60fps at 3x1 Eyefinity with each of the cards that supports Eyefinity. I did that for the 5400, 5500 and 5600 for widescreen and it was an interesting experience.

I have tested every card that ATI offers from the 5450 through the 5970. This review basically covers the benchmarks on what they consider "Ultra Enthusiast" (the Eyefinity6 and the 5970, along with CrossFireX), as well as the top "Enthusiast" card (the original HD 5870). I plan to put together reviews for each of the remaining card markets: Mainstream (5400, 5500 and 5600), Performance (5700) and Enthusiast (5800).

With all that data, and the information on "Hitting 30/60", I should have all the data points needed for a complete Eyefinity Buyers Guide. The goal will be to help you pick a GPU, and monitor configuration based on the games you play and the performance you want.

I've been working with the folks at Ergotech about a stand for 5x1-Landscape. They sent me a longer center bar, so that I could fit all three middle monitors on the flat surface. From there, the outer monitors sit on the wings and angle toward the center. There are a few more details I'd like to look at, but I think we have something cool in the works. Once we get it all put together, Erogotech has said this will be their first WSGF-branded SKU. Cool, eh?


Final Thoughts

ATI has produced a compelling product with the Radeon HD 5870 Eyefinity6. It offers great options for configuration and expandability. Unless you're looking to go for 5x1-P or 3x2, I don't see the immediate need for any existing Radeon 5870 owner to rush out and buy the card. Though, if you are playing the DX11 games the 2nd GB helps, I'm sure you're original HD 5870 would fetch a good price on eBay. This would be a good time for one of ATI's partners to offer a trade-up program.

However, if you are looking for a bigger upgrade or to start fresh with Eyefinity, I don't see a reason to choose the original HD 5870 over the Eyefinity6. The extra features and options are well worth the price differential.

If you are certain you will only ever be running 3x1 panels and want the fastest card today, then the HD 5970 is still for you. But, I do see the Eyefinity6 as the better option for the long term with its ability to use CrossFireX.