After having the chance to test the RX 460, RX 470, and RX 480 we now have a much better idea of how the current AMD lineup stands. There were a few special cases that I was curious about, though. Specifically the 4GB RX 480 and the 8GB RX 470. I was able to test the 4GB RX 480 by flashing our reference cards firmware but for the RX 470, MSI was nice enough to send a card for me to check out. I haven’t had a chance to see any of MSI’s current AMD lineup so the card also gave me a chance to see what they have going as well. So today I’m going to run the RX 470 Gaming X 8GB through all of our tests, dive into its features, and find out how it compares to the rest of the AMD lineup.
Product Name: MSI RX 470 Gaming X 8GB
Review Sample Provided by: MSI
Written by: Wes
Pictures by: Wes
Amazon Link: HERE
|Graphics Processing Unit||Radeon™ RX 470|
|Interface||PCI Express x16 3.0|
|Boost / Base Core Clock||
1254 MHz (OC Mode)
1242 MHz (Gaming Mode)
1206 MHz (Silent Mode)
|Memory Size (MB)||8192|
|Memory Clock Speed (MHz)||
6700 MHz (OC Mode)
6600 MHz (Gaming Mode)
6600 MHz (Silent Mode)
|Output||DisplayPort x 2 (version 1.4 Ready) / HDMI x 2(version 2.0b) / DL-DVI-D|
|Virtual Reality Ready||Y|
|DirectX Version Support||12|
|OpenGL Version Support||4.5|
|Multi-GPU Technology||Crossfire™, 2-Way|
|Card Dimension(mm)||276 x 140 x 37|
|Card Weight (g)||774|
|Power consumption (W)||110|
|Recommended PSU (W)||450|
|Power Connectors||8-pin x 1|
Before getting into the review I did test the RX 470 Gaming X 8GB in GPUz to confirm that our sample is running at the correct clock speeds and to also show what version of the driver all of our testing was done on. With a GPU clock speed of 1242 Mhz I can confirm that the review sample is running at the Gaming clock speed that all cards come with at default, the memory is at 6600 MHz, and the card has the 8 gigs of ram as listed.
The packaging for the RX 470 Gaming X 8GB really isn’t far off from the MSI packaging of the GTX 1070 that I took a look at. The box itself has a red background that fades into a black own at the bottom. The front has the MSI branding up to and the Gaming X down at the bottom. Over on the right is the red strip with the RX 470 8GB information that all AMD cards now have on the box, it also wraps around to the side. The best part though is the decision to just put a huge photo of the card right on the cover when all of the other manufacturers seem to go out of their way to avoid putting a photo of the video card on the front of the box. This shows everyone in the store exactly what they are getting.
The back of the box has three sections that feature the lighting, the included gaming app, and the noise level of the cooler. Each has a photo to go with the topic and a short explanation. There is also a small photo that shows the card goes great with MSI motherboard. Beyond that, I love that they include very short specification listing. They don’t include clock speeds, but it is still better than nothing. You also get a minimum system requirements listing and a basic feature list.
Inside the box is a black tray. Inside the card sits in a foam cutout with a foam panel on top. The card is also wrapped in a static protective bag to keep it free from any issues in shipping. On top of all of that is a thin box for the accessories. The RX 470 8GB doesn’t come with any power adapters or anything but you do get a driver/software disc and a quick users guide on the off chance you need instructions on getting everything up and running.
Also remember to remove the sticker that covers one of the fans. This is a reminder that the card does have the Zero Frozr functionality, meaning when you power your PC up the fans on the card won’t always be spinning, they power down when the card is under low load to keep the noise down.
Photos and Features
Like most other manufacturers, MSI’s design doesn’t really change very much from card to card. So the RX 470 8GB looks just like the 4GB model, the RX 480, and even the GTX 1070 Gaming X that I covered. What this means is the RX 470 8GB has the same black and red theme as the other MSI cards. They accomplish this with a large plastic fan shroud that is all red on the left and black on the right. The red section has a few groves cut into it and it has an octagon shaped cutout for the fan that it goes around. The black section also has lines, but here each line is filled with a backlit red fin. Altogether the shroud has a really aggressive styling that fits really well with the whole MSI product line.
The Twin Frozr cooler revolves around two large cooling fans that blow down over the heatsink. To pull the heat out into the heatsink MSI uses four heatpipes, two from the top and two from the bottom, to pull heat out from on top of the GPU out into the heatsink. From the bottom view, we get a good look at the heatsink design and the heatsink fits really well up against the caps on the PCB, meaning MSI doesn’t waste any potential cooling space. The open areas on the top, bottom, and end of the card do mean that a lot of the heat from the card will get vented into your case, so keep that in mind when picking a case. Unlike a reference card where the fan blows from right to left, the two large fans here blow down against the PCB and from there the airflow has to go somewhere.
While the card doesn’t have a full cover top shroud like Sapphires cards. MSI did wrap the shroud around the top a little over on the top left. In that space, they slipped in the MSI logo and a dragon. This is backlit with RGB and can be controlled using the gaming app.
For power, MSI faced the power plug up at the end of the card. The PCB is notched and the connection is upside down for a tighter fit and to avoid having to space out the heatsink for finger room. They went with an 8-pin connection as well so power shouldn’t be an issue at all.
The PCI slot end of the card has a very interesting air vent design. It doesn’t look like it would have as much airflow but given the cards design that shouldn’t be an issue. For display connections, they went with a setup similar to the Sapphire cards rather than the reference cards layout. This gets you one DVI port, two DisplayPorts, and two HDMI. The normal layout these days is three DisplayPorts and one HDMI but I think a lot of people will still need HDMI ports for a while. A lot of DisplayPort monitors still support HDMI but not the other way around so this works out well.
The back of the card shows off the all flat black PCB. That said I was surprised that MSI didn’t include a backplate. Sapphires RX 470 had one. The black PCB does still look good, I just get concerned with GPU sagging later in life on cards like this with a very large heatsink design.
Speaking of large, lets address the overall size of this card. For starters, the card is 1 1/3 inch taller than the top of the PCI slot, so you will need to be very careful on what case you pick. Some cases have issues with taller cards and this is one of the tallest designs yet. At about 11 inches long the card is longer than the reference card, but not really long enough to cause any concerns. The reason for the extremely tall card height is the two large fans that MSI went with. Normally video cards have 80 or at most 90mm fans but MSI went with 100mm fans. Larger fans normally does translate to better cooling and lower noise, but I am concerned that video cards are growing a little large in size.
Our Test Rig and Procedures
|Our Test Rig|
|CPU||Intel i7-5960X||Live Pricing|
|Memory||Kingston HyperX FURY Black 32GB Quad Channel Kit 2666 MHz||Live Pricing|
|Motherboard||Gigabyte X99-SOC Champion||Live Pricing|
|Cooling||Noctua NH-U12S Cooler||Live Pricing|
|Power Supply||Cooler Master V1000 Power Supply||Live Pricing|
|Storage||Kingston Hyper X Savage 960GB SSD||Live Pricing|
|Case||Dimastech Test Bench||Live Pricing|
|Our Testing Procedures|
|3DMark||The same goes for the most current version of 3DMark using the Fire Strike benchmark in normal, extreme, and ultra settings|
|Unigine Valley Benchmark 1.0||Using the Extreme HD preset to get an average FPS|
|Catzilla 4k||Default tests for 1080p, 1440p, and 4k resolutions using the overall score for each as our result|
|SteamVR||Default SteamVR test using Average Quality score|
|DOOM||Doom is tested on the Ultra quality setting. Tests are run at 1080p and 1440p using both OpenGL and Vulkan. The benchmark is a basic one using just the average FPS in the opening scene.|
|HITMAN 2016||Fullscreen with V-Sync turned off Detail, Texture Quality, Shadow Maps, and Shadow Resolution all set to their highest settings. We test using both DX11 and DX12 at both 1080p and 1440p resolutions.|
|Ashes of the Singularity||Built-in benchmark ran at 1080p and 1440p with graphics settings set to the “Crazy” setting with the exception of turning off V-Sync Mode. The benchmark scenario is set to GPU Focused and we use the Average Framerate for All Batches as the result. Tests are run both in DX11 and DX12|
|The Division||Built-in benchmark ran at 1080p and 1440p with graphics settings set to the default “Ultra” setting with the exception of turning off V-Sync Mode|
|Bioshock Infinite||Using the Adrenaline Action Benchmark Tool we run Bioshock Infinite on the “Xtreme” quality setting. This has a resolution of 1920x1080, FXAA turned on, Ultra Texture detail, 16x Aniso Texture Filtering, Ultra Dynamic Shadows, Normal Postprocessing, Light Shafts on, Ambient Occlusion set to ultra, and the Level of Detail set to Ultra as well. We also run this same test at 2560x1440 using the same settings as mentioned above.|
|Tomb Raider||Using the Adrenaline Action Benchmark Tool we run Tomb Raider on the “Xtreme” quality setting. This has a resolution of 1920x1080, Exclusive Fullscreen turned on, Anti-Aliasing set to 2xSSAA, Texture Quality set to Ultra, Texture Aniso set to 16x Aniso, Hair Quality set to TressFX, Shadow set to Normal, Shadow Resolution on High, Ultra SSAO, Ultra Depth of Field, High Reflection quality, Ultra LOD scale, Post-Processing On, High Precision RT turned on, and Tessellation is also turned on. We also run this same test at 2560x1440 using the same settings as mentioned above.|
|Hitman: Absolution||Using the Adrenaline Action Benchmark Tool we run Hitman: Absolution on the “Xtreme” quality setting other than the MSAA setting is turned down from 8x to 2x. That setting puts the resolution at 1920x1080, MSAA is set to 2x, Texture Quality is set to High, Texture Aniso is set to 16x, Shadows are on Ultra, SSA is set to high, Global Illumination is turned on, Reflections are set to High, FXAA is on, Level of Detail is set to Ultra, Depth of Field is high, Tessellation is turned on, and Bloom is set to normal. We also run this same test at 2560x1440 using the same settings as mentioned above, except on the “high” setting.|
|Sleeping Dogs||Using the Adrenaline Action Benchmark Tool we run Sleeping Dogs on the “Xtreme” quality setting. That means our resolution is set to 1920x1080, Anti-Aliasing is set to Extreme, Texture Quality is set to High-Res, Shadow Quality is High, Shadow Filter is set to high, SSAO is set to High, Motion Blur Level is set to High, and World Density is set to Extreme. We also run this same test at 2560x1440 using the same settings as mentioned above.|
|Total War: ROME II||Ultra-setting tested at 1920x1080 and 2560x1440, built in forest benchmark|
|Middle-earth: Shadow of Mordor||Using the built-in benchmark we test with ultra settings at 1440p|
|Sniper Elite 3||Ultra-setting tested at 1920x1080 and 2560x1440, built in benchmark|
|Thief||Tested using the “Very High” setting at 1920x1080 and 2560x1440|
|Folding at Home 2.2||Using the Folding at Home benchmark 2.2.5 set to OpenCL, WU set to dhfr, and run length set to the default 60 seconds. We test at both double and single precision and use the score at the result|
|CompuBenchCL||Video Composition and Bitcoin tests|
|Unigine Valley Benchmark 1.0 heat testing||We run through Unigine Valley using the “Extreme” preset for 30 minutes to test in game cooling performance with the fan speed set to auto then again with the fan set to 100%.|
|Power Usage||Using Unreal Valley Benchmark 1.0, we get our “load” power usage number from the peak power usage during our test. We get our numbers from a Kill-A-Watt connected to the test benches power cord.|
|Noise Testing||Our Noise testing is done using a decibel meter 3 inches away from the video card on the bottom/fan side of the card. We test an idle noise level and then to get an idea of how loud the card will get if it warms all the way up we also turn the fan speed up to 50% and 100% and test both speeds as well. The 100% test isn’t a representation of typical in-game noise levels, but it will show you how loud a card can be if you run it at its highest setting or if it gets very hot.|
Before I jump into the in-game testing I do like to run through a few synthetic benchmarks. While they don’t give you an FPS like in-game benchmarks, they do help me compare from card to card consistently. So going in I wanted to see just how the RX 470 Gaming X 8GB would compare to other RX 470’s with less vRAM as well as the 4GB RX 480 that is also in a similar price range. I started my testing off with 3DMark using the Fire Strike benchmark as well as the new Time Spy benchmark for DX12 testing. In Fire Strike using the performance setting the RX 470 Gaming X 8GB falls in between the two other RX 470’s I have tested. The Sapphire 470 has a clock speed of 1260MHz compared to the MSI’s 1242 MHz Gaming Mode so I’m not shocked that it still ended up being a little faster even with the lower vRAM. The XFX 470, on the other hand, had a slightly lower clock speed of 1256MHz that is still higher than the MSI, but it came in lower in the charts. That shows me that the extra vRAM did make a difference, though the extremely high power draw of the Sapphire to keep the clocks stable helped more. In Time Spy I ended up with a similar result as well, with the 470 8GB coming in almost dead even with the GTX 980 Reference card.
In Unigine’s Valley Benchmark the 470 8GB pulled 45.2 FPS but came in a little lower than both of the cards with higher clock speeds. The same happened in Catzilla as well across all three benchmarks. I was a little surprised the extra frame buffer didn’t help in the 4k results, though.
My last test was the SteamVR benchmark. Here the MSI pulled back ahead of the XFX but was still behind the Sapphire. The extra vRAM helped a little more here than the extra clock speed. This kept the RX 470 Gaming X 8GB in the green that shows that it is capable of a good VR experience.
Now that we have an idea of how the cards stand compared to each other, it’s finally time to get down to the in-game performance numbers. To do that I ran through our entire in-game benchmark test suite. The suite has 12 different games tested at both 1080p and 1440p at their highest settings. Three of the games are also tested in more than one configuration to see the difference between DX11 and DX12 performance or in the case of the most recent addition, DOOM I test out both OpenGL and Vulkan performance. The problem is, with so many results it can be a little too much to take in. To help with that I have condensed our results into two graphs, one for 1080p and the other for 1440p. All of our games are run at their max settings and we use the average FPS as the end result. The graphs below are broken down into three FPS ranges to represent unplayable (below 30), playable but not ideal (30 to 60), and ideal (over 60 FPS). So what did I find from those results?
Well at 1080p the RX 470 Gaming X 8GB can play basically anything you throw at it at the highest settings. 11 of the games tested were in the 60+ range and all but one were playable. The single outlier was the normal Ashes of the Singularity result that is almost always down there. This puts the MSI right with the Sapphire 470 as far as results go. The XFX had one more in the 30-60 range than the other two RX 470’s but was close as well.
At 1440p, the RX 470 was never really designed to game at this resolution but I still wanted to test It out. What I found was that the card could still play most everything, but there was, of course, a big shift down from everything being 60+ down into the playable but not perfect range. So if you want to game at this resolution with the RX 470 Gaming X 8GB you should plan on turning the settings down slightly from the ultra-settings I test with to stay in the 60+ range.
As always I also include all of the detailed results down below for those of you (like me) who want to see the exact results. Almost across the board in all of the tests, the Sapphire did come out ahead due to its higher clock speed in combination with its high power profile. The XFX, on the other hand, came in lower across the board even though it does have a higher clock speed. This also happens in all of the games tested at 1440p as well, the 8GB frame buffer does give a slight edge but not enough to be the fastest RX 470 tested. Beyond that, there aren’t any big shockers. The DX12 performance of the RX 470 Gaming X 8GB is great just like the other AMD cards. Vulkan testing in Doom is, even more, mind blowing. The RX 480 with a 4GB vRAM does still outperform the RX 470 with 8GB in all of the tests as well.
While it doesn’t play a big role in gaming performance, I do like to also take a look at compute performance on all of the cards that come into the office. This is simply to cover every possible aspect that you might be using a card. In this case, there are three typical uses for a video card outside of gaming. You use it to farm cryptocurrencies, you use it when editing videos and photos, and you might let your PC run folding while it is idle to help research. So I have tested each of these situations using the folding at home benchmark and then CompubenchCL for the other two. Starting with Folding at Home I test both single and double precision. In the single precision benchmark, the extra vRAM seemed to help and edged the MSI up past the two higher clock speed 470’s tested. This ends up being extremely close to both reference RX 480’s as well. In the double precision benchmark, like all of the other AMD cards the 470 8GB jumped up higher in the charts because of Nvidia’s lack of interest in including FP64 performance in their consumer cards. Oddly enough, here both of the higher clock speed 470’s were a little faster.
Next, I ran CompubenchCL but focused on just two results. The video composition benchmark and the bitcoin test. The extra vRAM in the video composition test made a big difference, putting the card up in the middle of the GTX 1060’s. For Bitcoin mining, the MSI pulled a respectable 548.48 MHash per second. This was a little less than the Sapphire but well above the XFX and a good representation of the performance I saw in most of our other tests as well.
Cooling, Noise, and Power
For my last batch of testing, I like to focus on all of the aspects that aren’t really just due to the clock speed set and the GPU of the card. Having tested a few different RX 470’s its really nice to be able to test the MSI to see how it compares to XFX and Sapphires cards in things like cooling performance, power usage, and noise. My first test is power usage and this is the one I was the most curious about. When I tested the Sapphire RX 470 I was completely blown away at how much power it pulled, the power draw of the card does, however, play a big role in how it is consistently quick in a lot of the benchmarks even with a similar clock speed to the XFX. I put the card under load using Valley Benchmark and then using a Kill A Watt I document the highest wattage pulled from our entire test bench. In this case, it was 238 watts, significantly less than the Sapphire and even 11 watts less than the XFX 470 as well making this the lowest power usage 470 I have tested. This put it a lot closer to the GTX 1060’s.
Next, I test overall fan noise. Please keep in mind that our results aren’t comparable to any other website, our office has a high overall noise level so we test with the meter closer to the card than you would ever have your ear, especially if the card is in a case. So these aren’t numbers to show what you will hear, but they do help show the overall noise difference between cards. The MSI RX 470 Gaming X 8GB with its two HUGE fans doesn’t have to run the fans as fast as everyone else, so both my 50% and 100% fan speed results show the card being nearly the quietest tested. This is one of the big perks to the large card size.
Last but not least I take a look at the cooling performance of each card. This can vary due to manufacturer fan profiles, so I test once with the stock settings to see what you can expect out of the box. Then I test again with the fans turned all the way up to 100% fan speed to see what the cooler is capable of. Given the huge cooler and fans, it’s no surprised that right out of the box the MSI is the coolest running card tested at 59 degrees. When I turned the fans up it was even crazier with a peak temperature of 42 degrees.
Overall and Final Verdict
Alright, so when going into this review I had already taken a look at a few different RX 470’s. So I wasn’t as focused on how the MSI RX 470 Gaming X 8GB compared to cards from Nvidia. I was more curious if having double the vRAM would help performance at all and I also wanted to see how MSI’s RX 470’s compares to the competition, especially in things like cooling, power usage, and noise. All of the areas that the manufacturer design decisions play a big role in. So first off I can say that the additional vRAM wasn’t as helpful as I would have hoped. I was expecting to see a bump in performance in things like 4k and 1440p testing, but I didn’t see it there at all. It did help in a few of the compute benchmarks and a few of the games did see a small improvement. So I wouldn’t focus too much on the additional vRAM, but the rest of our testing should also represent the 4GB MSI RX 470’s as well. Overall the card was consistently faster than the XFX RX 470 that had a higher clock speed but slower than the Sapphire. Sapphire managed this by really cranking up the power on their card and they have a little more consistent clock speeds to show for it. It’s not a big gap, but there is a difference. However, I’m not sure if that difference is worth the 50 extra watts that is required to do it or not.
That is where MSI really jumped ahead. It has the lowest power draw of any of the RX 470’s I have tested, putting it within 27 watts of the GTX 1060’s. This also translated into extremely low temperatures, but with the fans turned up and with the stock profile. At the same time, the card is also the quietest tested. They managed the cooling and noise performance, however, due to going with two huge fans. This isn’t a downside to everyone, but you will have to keep in mind the overall card height when shopping for a case as it is over an inch taller than a standard card. Some cases have room for that, but others don’t, especially as you get into smaller cases.
In the end, MSI has put together a great RX 470 and If I were on the market for one I would be looking at both the MSI and the Sapphire. They both have unique downsides and upsides. Basically, if you don’t want your card to run hot and pull a lot of extra power the MSI is the way to go, but it is a little tall.
Live Pricing: HERE