Some of you may have missed it but Monday the RX 460 officially launched. There wasn’t as much of a fuss about it because the RX 470 was taking most of the attention, but the RX 460 is also Polaris based. It is the first card on the Polaris 11 architecture where the RX 470 and RX 480 are on the Polaris 10. AMD is marketing the RX 460 as their e-sports card where the RX 470 was an HD card and the RX 480 was VR. This might be a little confusing, but the general idea is the RX 460 can’t handle everything at 1080p, but it is designed to work well with the popular e-sports titles like Dota 2 and League of Legends. So today I'm going to take a look at the Gigabyte RX 460 that AMD and Gigabyte sent over and then put it through our normal benchmark suite and see how it performs. That will round off a full week of AMD coverage here on LanOC as well!
Product Name: Gigabyte RX 460 Windforce OC 2GB
Review Sample Provided by: Gigabyte/AMD
Written by: Wes
Pictures by: Wes
Amazon Link: HERE
|Graphics Processing||Radeon RX460|
|Core Clock||Boost: 1212 MHz|
|Process Technology||14 nm|
|Memory Clock||7000 MHz|
|Memory Size||2 GB|
|Memory Bus||128 bit|
|Card Bus||PCI-E 3.0 x 8|
Dual-link DVI-D *1
HDMI-2.0b*1 (Max Resolution: 4096x2160 @60 Hz)
Display Port-1.4 *1 (Max Resolution: 7680x4320 @60 Hz)
|Digital max resolution||7680x4320@60Hz|
|Card size||H=37 L=191 W=111 mm|
Before diving into testing, I did want to include a copy of the GPUz from the card. With issues popping up this year with manufacturers sending review samples that are turned up to their optional OC clocks I want to make sure we keep everything transparent. The RX 460 does seem to be running at the 1212 MHz listed on their specifications listing so we are good to go there.
So what’s different on the RX 460? Well Unlike the RX 480 and RX 470, AMD doesn’t promote the 460 as a premium product. This is the meat and potatoes card for casual gaming and for people who play the popular games. AMD specifically notes 90+ FPS in titles like League, DOTA 2, Rocket League, and Overwatch. All very popular games in the e-sports market. They do note that this is at 1080p but just with high settings. This means no AA and not at ultra-settings like we normally test at.
The RX 460 has just 14 Compute Units compared to the 36 and 32 of the RX 480 and RX 470. That is less than half. It's clear that this isn’t just a small step down from the RX 470 like the 470 is to the 480. It has just 896 Stream Processors to 2048 on the RX 470 but it does at least run at a similar boost clock. The base clock, however, is actually higher on the RX 460. With a 128-bit memory interface, we only have 2 and 4 gig vRAM options available and our test card is actually the lower 2GB model. The much smaller GPU size does translate to a lower TDP. AMD is saying the board uses less than 75 watts compared to the 120 and 150 of the 470 and 480.
The box for the RX 460 is considerably smaller than the boxes for the RX 470 and RX 480’s as well as the GTX 1070 I recently covered from Gigabyte. It does however still have the Gigabyte eye on the cover like their other gaming products have. Other than that the front of the box is simple. It has the Radeon RX 460 logo wrapping around from the front to the side and in that they include the 2GB memory capacity of this card. Down at the bottom are Windforce and CO logos letting us know that this card does have a Windforce cooler and is overclocked. On the back of the box, they have a photo of the card in a box showing off how the Windforce cooler works. Beyond that, they also mention the durability and gigabytes one click overclocking software.
Inside, even with the smaller box, the card looks tiny inside of it. The have the card wrapped in a static bag and then sitting in a foam cutout to keep it safe. Up under the card and the foam, I also found a small quick guide, but beyond that there aren’t any other manuals or adapters included, not that you would need any adapters for a card without any power connections.
Photos and Features
The card is considerably smaller than most of the other Windforce cards. Normally Gigabyte goes longer to fit a three fan setup but for the RX 460, they have gone with two. This is till larger than the single fan design of the reference card that AMD had been showing off. The card has the same black and orange theme on the fan shroud as the other Gigabyte cards. The cooler doesn’t go taller than a standard PCIe card and isn’t far off from an ITX card in length. The shroud has a little shape to it but not enough to be distracting. The two fans have the Gigabyte logos on the centers and each blade has 5 small ribs on them. From the side profile, we can kind of get a peek at the cards cooler as well. Gigabyte went with a forged heatsink design rather than the standard heatpipes with sheet metal heatsink fins. This design is less efficient, but the RX460 isn’t exactly a high powered card, so with two fans this design should be more than enough still.
The fan shroud does wrap around all three internal edges but each has multiple vents for air circulation. This is because the cooler design pushes air down onto the heatsink, not across the heatsink like a blower cooler. So the heated air will vent out all of the edges of the card. One thing I noticed when looking at the top edge of the fan shroud was that the cooler design is a fairly universal one. Here we can see cutouts for Crossfire/Sli bridges and also a hole for a power connection. Both of which the RX 460 doesn’t have. That right, this card doesn’t have to be plugged into any extra power cables when hooking it up. It runs completely off the PCIe slot.
On the back, we get a better look at the black PCB of the RX 460. The PCB is about an inch shorter than the cooler so we can also see part of the heatsink and the fan shroud sticking out over on the left. With this being a budget focused card, there, of course, isn’t a backplate covering anything up. Not to mention with the shorter length it’s a lot less likely to be needed.
On the back PCI slot, the RX 460 has a few display connection options. They give you a DVI, HDMI, and DisplayPort. There is only one of each though so people with multiple monitors will have to keep that in mind. Having only three connections does mean the second half of the slot is completely open for ventilation. Gigabyte slipped in this V-shaped vent design. Considering a lot of the card still vents inside of the card, I’m not sure how much air will be blowing out the back, but at least they didn’t waste the extra space from the missing extra DisplayPort connections.
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.|
To start off my testing of the RX 460 I ran it through our entire collection of synthetic benchmarks. Just to be clear, while I did test it at 1440p and 4k in a few of these it's not really designed for those resolutions at all, especially in the 2GB configuration that I am testing today. I just test every card in all of our tests so we can see now and in the future how the RX 460 compares to other cards. Anyhow, I started my testing off with 3DMarks Fire Strike benchmarks and in the 1080p “performance” benchmark the RX 460 had a GPU score of 5582. While this looks low compared to all of the higher end cards we test, it is actually a good number for a lower end card. It is only a touch below the 4GB R7 470 from last generation and it is actually ahead by a large margin of the 2GB R7 370. Considering that is one step up last generation I would say the RX 460 is on point for where it should be in the product line. The result was similar I the newer Time Spy benchmark only the only older card that I have tested on that new benchmark was the GTX 780. I would say the RX 460 coming close to the GTX 780 in this test is actually really good.
In my next benchmark, the Unigine Valley Benchmark, I put the RX 460 through the 1080p HD benchmark. Here it came down in near the bottom of the chart once again but only a few FPS away from the two R7 370’s.
In Catzilla I was surprised the RX 460 even ran the 4k test but as you can see the number was so low it wouldn’t even fit in the bar. At 1080p the RX 460 still put up respectable numbers, being about 3k behind the R9 380.
Like I said in the opening, the RX 480 is the VR card, the RX 470 is an HD card, and the RX 460 is for the less demanding e-sports titles at 1080p. So it’s no surprised that the RX 460 put up a 0.3 score in the SteamVR benchmark, this card isn’t going to handle anything VR, but that’s okay.
As always it’s the in-game performance that I’m most concerned about. So I ran the RX 460 through our full in-game benchmark suite. This is testing the card at both 1080p and 1440p in a wide variety of games with each game turned all the way up to its highest quality preset. In other words, this is a worst case for most games. I went into this knowing the RX 460 wasn’t designed at all for the 1440p resolution and even at 1080p it's really only designed for the e-sports focused games that aren’t as demanding graphically. That said let's take a look at the results. At 1080p the RX 460 only had one game in the 60+ sweet spot but overall half of the games tested were still more than playable. This fits what AMD has marketed the card as. At 1440p the only surprise was the RX 460 was able to be playable in three of the games tested.
All in all, we have to take a closer look at the individual results to really get a better idea of what's going on. I found that games like Doom and Ashes of the Singularity that are always demanding weren’t playable at all. Those two up four of the 7 unplayable results. The Doom result was especially interesting to me because we know that typically playing using Vulkan in Doom helps AMD cards a LOT but the RX 460 actually dropped in performance. **Edit- retesting later actually did show an improvement, 30 in 1080p and 23 in 1440p. So Vulkan Doom at 1080p is actually in the playable range** Most of our slightly older test games were the games that were very playable. What this tells me is that the RX 460 can still handle any of today's modern games, but you are going to have to bump the settings down a little from the Ultra settings we are testing at. The 2GB vRAM is especially holding the card back here. The gap between the RX 460 and the RX 470 is the most surprising thing, though, it really feels like there should be a card or two between them, especially with the RX 470 4 and 8 gig models coming in above some of the 4 GB RX 480 cards.
Most people wouldn’t consider the RX 460 to be a contender for overall compute performance but I did want to still run it through our tests. Specifically, I'm curious how well it performs in the Bitcoin mining given its low TDP and lack of any power connection. I started first with Folding at Home though to test the single and double precision performance. In the single precision test the RX 460 performed about where I expected it to be, it has about half the performance of the reference RX 480, this is in line with its core count. In the double precision test as expected all of the Nvidia cards drop off here so the RX 460 ends up coming in closer to the GTX 980 than you would ever normally expect. We are a little less than half the performance of the RX 480 here, though.
Next, I tested in CompuBenchCL to take a look at Video Composition performance and then, of course, the Bitcoin test that I’m most curious about. In the Video Composition test, the RX 460 was close in performance to the older GTX 780 and well above half of the RX 480. In the Bitcoin benchmark, the RX 460 pulled 235.05 MHash/s. A few generations back this was an impressive number as shown by the GTX 780 sitting behind it. But this is about 1/3 of the performance of the RX 480. Given the prices, it doesn’t really make the RX 460 a hidden gem, even when taking into account the lower power usage.
Cooling, Noise, and Power
For my last tests, I always focus on things like power, cooling performance, and noise. These are all performance metrics that not everyone considers, but they are all very important when picking components to go with your video card. You don’t want to get a card that runs too hot without a case to help keep it cool and you wouldn’t want to not have enough power to power everything. Speaking of power, the power usage benchmark was the first test I ran the RX 460 through. Given its low TDP of 75 watts and the lack of any power connection, I knew it would be the lowest wattage card I have tested on the current test configuration. Even then I was still wasn’t sure what I would see. Remember this is our full X99 powered test bench, not just the individual wattage of the card. So at 130 watts for everything while running in game I was surprised. The Polaris architecture when scaled down seems to be more power efficient than the RX 470 and RX 480 that both pulled more than I expected.
Next, I pulled out the decibel meter and tested the RX 460 Windforce’s noise when running the fan at 100% and at 50%. This gives us a good idea of how loud the card can go and what we should expect when it's in normal use. It also turns the fan completely off when you aren’t putting it under load as well. The dual fan cooler ended up being a little on the high side with the fan turned all the way up but I think this is partially related to Gigabyte using slightly shorter fans than everyone else, helping them have normal height cards. At the slower speed, the card was in line with all of the other cards and frankly 64.2 decibels isn’t going to be noticeable at all in an actual case. I just wouldn’t recommend running the fan on high all of the time.
The last benchmark was taking a look at the cooling performance of the Windforce RX 460. Given the low power draw, you would think it would automatically be very cool, but I was a little concerned when I saw Gigabyte went with a simpler heatsink design behind the fan shroud. With the card set to the stock settings, it ended up running at 65 degrees when being looped in Valley Benchmark. To test out the overall cooling capacity I did the same test but with the fan turned all the way up to see what it is capable of and it ran at an impressive 45 degrees. Basically, the lower cost heatsink design is offset by the lower wattage but the dual fan cooler does a good job with the stock settings and is capable of doing a great job if you crank the fans all the way up.
Overall and Final Verdict
With less than half of the processing power of the RX 470, the RX 460 is a completely different product than AMDs next step up. The RX 460 stands out against all of the cards I have tested over the last few months because it's not designed to be anything like the high-end cards on the market. AMD went considerably smaller on the Polaris 11 architecture, especially after cutting things down even more beyond that. This puts it in a class of cards that frankly doesn’t fit very well in our normal GPU testing scheme. This is because the RX 460 is designed to run the most popular games at 1080p, but not with the settings cranked up and all of our tests have things like AA turned up and textures that push the limits of even the most high-end cards.
So where does that leave the RX 460? Well, unlike the other Polaris cards I have tested, it uses very little power and doesn’t require a power connection at all. It is very similar to the GTX 950 in both power and performance though I don’t have direct numbers to compare with the exception of our single 3DMark benchmark that didn’t have to get cleared when we moved to the new test bench. In that bench, the GTX 950’s are actually up a little ahead of the RX 460. This is a little concerning considering that is a last generation card and the 1000 series equivalent has yet to be launched.
The Windforce model that I tested ended up having good cooling, a good look, and it doesn’t take up a ton of room. It did test a little loud in our noise testing when the fan was cranked up, but the lower TDP of the RX 460 really shouldn’t ever be getting that hot.
In the end, it really feels like we are actually taking a look at the RX 450. There is such a huge gap between the RX 460 and RX 470 feels out of place in this product lineup. But AMD did at least put out a nice power efficient card for people who just need to play their basic games. The RX 460 has a suggested price of $109 so the performance does fit the price point in comparison to rest of the current lineup. The Gigabyte card I tested has an MSRP of $119.99 so it is a little higher and given the low 2GB frame buffer I do wonder if it shouldn’t be at that opening price. Anyhow, things get a little interesting when we compare the RX 460 to the current GTX 950’s on the market. They bottom out at $125 but a lot of them also have rebates that push the card below the RX 460 pricing. Basically, the RX 460 is a great card, but Nvidia has given them a lot of competition at the same price point. I don’t know if this is just due to them clearing out older stock or if it is a permanent price point, but AMD and Nvidia fans both have good options for playing today's popular titles at decent settings without breaking the bank. It really just comes down to your preference.
Live pricing: HERE