Introduction

A huge drop in discrete graphics market share and a wait of two years is what took AMD to finally launch their latest Radeon 300 series cards. Back with a full top-to-bottom lineup that has been aimed at consumers with some seriously competitive prices, the Radeon family is back with a bang. Consumers and AMD fans have long been awaiting this moment, the AMD Radeon 300 series will make up for users looking to upgrade below the $450 US range which NVIDIA has gained a huge market share from with their Maxwell based GeForce GTX 980, GeForce GTX 970 and GeForce GTX 960 series graphics cards.

The Radeon 300 series is an interesting lineup. It’s positioned aside from the Fiji based family which are part of the R9 Fury and R9 Nano series. The reason to keep the 300 and Fiji based parts separate is simply due to the fact that AMD is using existing chips to power their latest cards. GPUs such as Grenada, Antigua, Trinidad and Tobago have ever been heard but when we take a closer look at these chips, we find out that the Grenada GPU is actually based on Hawaii architecture, Antigua is based on Tonga architecture while Trinidad and Tobago are based on Pitcairn and Bonaire architecture. AMD find out during the launch of Radeon 200 series cards that the existing cores, which are stocked enough to be competitively placed against newer cores from competitors. It helps AMD save up cash from developing new chips and focus on new flagship grade GPUs instead. While this approach does have the benefits, it could lead to some complications in terms of feature support since not all older GPUs have the technology infused on architectural level to benefit from upcoming APIs and features. So today, we are going to look into the new Radeon R9 390 series card and see whether that card has enough gpu power to compete against the GTX 970 which is its direct competitor at $329 US price.

Meet the AMD GCN Based Grenada GPU

– Updated Hawaii With Twice The VRAM

The AMD Radeon 300 series might be the latest cards but they are using a architecture design that has existed in the market since January, 2012. The AMD GCN core architecture has already seen three revisions, GCN 1.0 was the original core designs featured on the Tahiti, Pitcairn and Cape Verde based cards, GCN 1.1 was fused on the Hawaii and Bonaire based cards while the latest GCN 1.2 core design is featured on Fiji and Tonga core based cards. As we had just mentioned, the AMD Grenada GPU is a new revision based on the Hawaii GPU that has been used on the Radeon R9 290X and Radeon R9 290 graphics cards.

The Grenada GPU architecture is a step forward from the GCN 1.0 architecture introduced in Radeon HD 7970. The Grenada GPU on the Radeon R9 390 features the GCN 1.1 architecture and measures at 438mm2. Grenada has been tweaked and organized by AMD in way to utilize more performance. In official details that AMD was able to reveal themselves, they did mention that Grenada features several manufacturing process optimization that allow better clock speeds to be obtained from the GPU under the same power envelope. The GPU also get a complete re-write of power management architecture which:

Under “worse case” power virus applications, the 390 and 390X have a similar power envelope to 290X
Under “typical” gaming loads, power is expected to be lower than 290X while performance is increased

The AMD Grenada GPU is composed of 6.2 Billion transistors on a 28nm die process. Due to an unfeasible 20nm node, high-performance GPU manufacturers have kept away from making a 20nm GPU and have the FinFET process node under sight for their next generation graphics plans.

Above, you can see the Grenada (Hawaii) GPU block diagram which shows that the GCN core consists of eight ACE’s or Asynchronous Compute Engines modules which is twice the amount featured on Tahiti, right in the middle of these modules is the main graphics command processor to allow global data share. The Grenada XT GPU consists of four Shader Engines each with its own parallel linked geometry processors and Rasterizers, each Shader Engine hold 11 Compute units with 64 Stream processor count, the full Radeon R9 390X Grenada XT GPU comes with 44 Compute units and a shader count of 2816 streaming processors, 176 Texture mapping units and 64 Render Back Ends. Later in the article, you will get to know how these new ACE units help leverage graphics performance under the DirectX 12 API. There’s one geometry process per shader engine which can balance load with other shader engines and has one primitive per cycle.

All Shader engines share the same pool of 16 64 KB L2 Cache with upto 1 TB/s L2/L1 bandwidth. The Grenada GPU has eight 64-bit MCs or memory controllers which result in the 512-bit bus interface with an abundant 8 GB GDDR5 memory. Multimedia processors are featured along side the main GEs that include VCE, UVD, DSPs for TrueAudio, Six of the New XDMA, Eyefinity Controllers and the DMA engine. Last of all, a small PCI-e 3.0 Bus interface is located which enables support with the latest PCI-e 3.0 enabled motherboards.

While Hawaii was the first graphics card in a while to feature a 512-bit bus that allowed for a 4 GB GDDR5 VRAM that pumped out 320 GB/s bandwidth, Grenada makes use of a more denser memory design and pushes the clock speeds further up from 1250 MHz to 1500 MHz that pumps out 384 GB/s bandwidth. The VRAM also doubles up to 8 GB GDDR5 which is enough to play games at higher resolutions with max details to texture and AA quality. The VRAM amount is even higher than Fiji that has 4 GB too but that is HBM (High-Bandwidth Mem) that makes use of a 4096-bit bus and pumps 512 GB/s bandwidth which is an unprecedented amount.

A GCN 1.1 architecture efficiency slide shows the main improvements between Tahiti and Hawaii and we can note that the geometry processing rate has been increased by 1.9 times with 4 Billion primitives per second on Hawaii compared to 2.1 Billion on its predecessor. Similarly the texture and pixel fill rate have been upped by 30 and 90 percent respectively while the bandwidth is rated at 320 GB/s compared to 264 GB/s on the HD 7970. Lastly, on the compute side users would get 12.8 peak GLops/mm2 compared to 12.2 on Tahiti. The Hawaii Radeon R9 290 series has a larger die size of 438mm2 so we are expecting upto 5.6 TFlops of compute power.

AMD Crossfire Direct Memory Access – XDMA Technology

In terms of technology incorporation, AMD is the first to say goodbye to Crossfire bridge with their latest XDMA technology that can open a direct channel for multiple GPUs to communicate through by using the PCI-Express bus interface. According to AMD, an external bridge like the Crossfire bridge which we have been using since a long time delivers just 900 MB/s bandwidth for the chips to communicate while the XDMA technology allows 32 GB/s transfer speeds which is a 35 times improvement. Some points of XDMA can be seen below:

XDMA is a unique solution in the graphics industry; no similar technologies presently exist for consumer GPUs.
In case you didn’t catch it, XDMA eliminates the need to install any bridge. Install matching GPUs and you’re set!
XDMA is designed for optimal performance with systems running PCI Express 2.0 x16 (16GB/s), PCI Express 3.0 x8 (16GB/s), or PCI Express 3.0 x16 (32GB/s).
Bandwidth of the data channel opened by XDMA is fully dynamic, intelligently scaling with the demands of the game being played, as well as adapting to advanced user settings such as vertical synchronization (vsync).
Designed for UltraHD via DisplayPort™, which permits for 2160p60 gaming on the AMD Radeon R9 290 Series.
XDMA fully supports the “frame pacing” algorithms implemented into the AMD Catalyst™ driver suite.

AMD Radeon R9 Series

– Enhanced for Gaming Beyond 1080P

AMD has yet again introduced their Radeon R9 and Radeon R7 series branding with the 300 series family cards. The card we are looking in specific is part of the Radeon R9 series which is meant for gamers who play beyond the resolution of 1920×1080 (Pixels). This lineup includes the Radeon R9 380, Radeon R9 390, Radeon R9 390X and the Radeon R9 Fury X graphics cards.

The Radeon R9 380 is the entry level card in this lineup that has a starting price of $199 US. The Radeon R9 380 is priced to compete against the GeForce GTX 960 and also advertised to be a 1440P graphics card. The Radeon R9 390 (the one we are testing today) is priced at $329 US and is directly placed against the GeForce GTX 970 from NVIDIA which is the single most selling Maxwell based graphics card. The Radeon R9 390 has some benefit over the competitor which include a 8 GB VRAM, 512-bit bus but we will see how these features translate into gaming performance. Next up, we have the Radeon R9 390X which is priced at $429 US and is competing with the GeForce GTX 980 since there’s no other graphics cards in this price range. The GeForce GTX 980 retails at a official price of $499 US and performs faster than the 8 GB R9 390X.

Last up, we have the Fiji GPU based flag ship Radeon R9 Fury X which is placed against the GeForce GTX 980 Ti. Both cards are priced at $649 US and meant to be positioned as 4K Gaming cards but we have seen that Fury X is a bit slower than the competition and consumes a lot of power.

Overall, the Radeon R9 series lineup is decent but doesn’t get aggressive or competitive pricing schemes as the Radeon 200 series. AMD has yet to launch three more Radeon R9 cards which include the Radeon R9 Fury, R9 Nano and the Radeon R9 Fury X2 (Dual-chip graphics card). It will show how well AMD could do in the market place in 2H of 2015.

AMD Radeon R9 Series:


GPU Codename	Antigua Pro	Grenada Pro	Grenada XT	Fiji Pro/LE?	Fiji Pro	Fiji XT	Fiji XT X 2
GPU Process	28nm	28nm	28nm	28nm	28nm	28nm	28nm
Transistors	5.0 Billion	6.2 Billion	6.2 Billion	8.9 Billion	8.9 Billion	8.9 Billion	8.9 Billion X 2
GPU Shaders	1792	2560	2816	TBD	TBD	4096	4096 X 2
TMUs/ROPs	112 / 32	160 / 64	176 / 64	TBD	TBD	256 / 64	256 / 64 X 2
Core Clock	970 MHz	1000 MHz	1050 MHz	TBD	TBD	1050 MHz	~1000 MHz
VRAM	2-4 GB GDDR5	8 GB GDDR5	8 GB GDDR5	4 GB HBM	4 GB HBM	4 GB HBM	4 GB HBM X 2
Memory Bus	256-bit	512-bit	512-bit	4096-bit	4096-bit	4096-bit	4096-bit X 2
Memory Clock	5.7 GHz	6.0 GHz	6.0 GHz	500 MHz	500 MHz	500 MHz	500 MHz
Bandwidth	182.4 GB/s	384 GB/s	384 GB/s	512 GB/s	512 GB/s	512 GB/s	1024 GB/s (cumulative)
TDP	190W	275W	275W	175W	275W	275W	300W
Power Connector	2 x 6-Pin	8+6 Pin	8+6 Pin	8-Pin	8+8 Pin	8+8 Pin	8+8 Pin
Launch Date	18th June 2015	18th June 2015	18th June 2015	Summer 2015	14th July 2015	24th June 2015	Autumn 2015
Launch Price	$199 US+	$329 US	$429 US	$499 US	$549 US	$649 US	TBD

AMD Radeon 300 Series Feature Set

– DirectX 12, Vulkan, Liquid VR

The AMD Radeon 300 series is based on existing chips but GCN did incorporate architectural enhancements to benefit from the latest APIs. Future APIs such as DirectX 12, Vulkan and Mantle are shaping up to be one of the most incremental update for application and game developers as they allow further optimization of the graphic capabilities and further performance to be harnessed from hardware utilizing each feature implemented on the core design.

The slides go off to detail what we already know about the current state of multi-core processors (AMD’s FX CPUs in specific) which even though have a higher core count, don’t tend to deliver much enhanced performance as compared to Intel parts due to applications not featuring multi-core support which leads them to deliver better performance on CPUs with higher single core IPC. AMD’s Bulldozer and Piledriver CPUs didn’t manage to deliver a huge gain on IPC hence the processors pretty much slacked down against three generations of Intel CPUs. The other factor is that modern CPUs are technically unable to keep up with performance growth of graphics cards which is usually in double digits compared to a single digit growth on the CPU side. There’s also the fact that current API/Driver overhead are restricting CPUs with multiple cores to utilize only 1 core to talk to the GPU to execute a process. So even with 4,6 or 8 core CPUs, you are in the same league as a person who owns a dual core processor unless or until the application can guarantee more brute multi-processor communication. With DirectX 12 API, that scenario is going to change and following are just a couple of benefits to name:

Better use of multi-core CPUs
More on-screen detail
Higher Min/Max/Avg frame-rates
Smoother gameplay
More efficient use of GPU hardware
Reduced System power draw
Allows for new game designs previously considered impossible due to restriction by older APIs

Multi-threaded Command Buffer Recording:

With DirectX 12, AMD has three key features to incorporate in their hardware. The first is Multi-threaded command buffer recording, with command buffer being a list of commands issued to the CPU to execute while running a game such as Lightning, Effects and will be used to improve multi-core CPU performance. The fact about this technology is that it won’t only work for AMD CPUs, both also for other processors which include Intel and several ARM based processors. It allows higher FPS gain by utilizing several more CPU cores and allowing them to communicate with the GPU cores simultaneously. This also allows much better CPU utilization and delivers better performance per watt since previously, you were wasting most of the CPU power on doing nothing. The technology will allow the CPU to become less of a bottleneck then it previously was and the majority of the performance will be determined by the GPU since the CPU will be properly utilized.

A chart made by AMD perfectly illustrates how the API is going to work in the benefit of CPUs. On DirectX 11, most of the work is handled by a single core which is represented by “Core 1″. This core is processing all the tasks currently issued to it and the majority of the bottleneck is caused by the DirectX work which takes even more time than the game itself. At the same time, other cores are working but some cores remain unused while the CPU is sipping the same amount of performance as any normal workload. The total time it takes to complete a workload on DirectX 11 is 29ms or 34 FPS.

The difference with DirectX 12 is that all CPU cores are being utilized with load distributed across all cores of an 8 core processor. The DirectX work is significantly reduced which allows the game to quickly present the execution to the user in just 15ms or 66 FPS compared to 29ms or 34 FPS with DirectX 11. This technology also allows CPU to deliver higher draw calls than previously with an AMD A10-7850K pumping out 2,739,266 draw calls at 84W compared to 521,221 draw calls at 91W.

AMD GCN Asynchronous Compute Engines:

The next feature from AMD is Fine-Grain Asynchronous Compute Scheduling and Execution which will deliver on three promises, higher FPS, Greater VR Support and better image quality. To gain higher FPS, DirectX 12 will shift from several complex serial workloads to several parallel workloads allowing several way execution rather than a complex one-way execution on serial loads. Aside from that, if there’s additional workload, the idle GPU resources will be put to task to work instead of waiting for their turn which results in a speedy execution process that will deliver better performance through multi-threading improvements on the GPU side.

This great parallelism also means that GPUs will allow lower latency on virtual reality headsets and make them more responsive. Effects such as Physics, Lightning and memory utilize different GPU resources but the execution is serial only, a parallel path on DirectX 12 allows the total render time to be reduced to deliver lower latency and higher FPS. It enables a genuine multi-threaded approach to graphics. It allows for tasks to be simultaneously processed independently of one another. So that each one of the multiple thousand shader units inside a modern GPU can be put to as much use as possible to improve performance.

AMD Explicit Multiadapter Technology:

DirectX 12 Mutliadapter technology sounds great since it will allow PC users to actually utilize iGPU which is currently the least important piece of hardware in any enthusiast and mainstream PC. CPUs, APUs with these iGPUs will be able to be utilized in a way that they can actually add performance benefits to gaming titles. DX12 can further allow several Multi-GPU configuration such as allowing CrossFire and SLI cards to work in harmony and even the VRAM of each GPU available on a PC to be combined and fully utilized. Only time will tell whether game developers and software makers will actually take benefit from the additional and useful technologies which DirectX 12 API will feature.

DirectX 12 Parallelizes command list building and execution across CPUs
Coordinates workload across multiple GPU engines
Improves memory efficiency and reduces memory fragmentation
Improves frame rate stability with object caching
Enables direct hardware control and more game-specific optimizations

Talking more on the Explicit Multiadapter technology, it is mentioned that it can leverage all the hardware in a system through parallel command generation and execution, independent memory management and can allow multiplie GPU topologies such as Multiple discrete GPUs (Crossfire/SLI) and integrated plus discrete GPUs. This allows more control, more capability, more performance and supports custom load balancing off work offering two distinct API patterns, a Linked GPU pattern and an Unlinked GPU pattern.

HIS Radeon R9 390 IceQ X2 8 GB Graphics Card

The HIS Radeon R9 390 IceQ X2 is a custom built and factory overclocked graphics card. As detailed in the GPU specifications, the card holds 2560 stream processors, 160 texture mapping units and 64 raster back ends. The Radeon R9 390 IceQ X2 features core clock speed of 1020 MHz (20 MHz factory overclock) and the memory is clocked at 6000 MHz. The card itself features 8 GB of GDDR5 memory that is featured on a 512-bit bus that pumps out 384 GB/s bandwidth. As far as the cooling design is concerned, HIS makes use of their iconic IceQ X2 cooler, the only visual difference would be the sticker that they have placed on the shroud aside of which the cooler is pretty much the same thing we saw on the previous 200 series cards with its gold and blue color scheme which you will see in detail in the unboxing section below.

HIS Radeon R9 390 IceQ X2 Product Gallery:

HIS Radeon R9 390 IceQ X2 8 GB – Unboxing

The HIS Radeon R9 390 IceQ X2 ships inside a large cardboard package which is hidden inside a cool plastic cover. Their are tons of labels and technical details regarding the HIS IceQ X2 cooler which is featured on the card. Like some of the high-end cards, HIS has equipped their non-reference graphic card with a dual-bios to revert back to the stock settings incase some mishap happens while overclocking. Other than this, we can see logos for Turbo Boost, PCI-e 3.0, 3 GB GDDR5, 4K x 2K Ultra HD and HDMI support.

The back side provides brief details on features such as the IceQ X2 design and the normal marketing material which tells that the card is cooler and quieter than the reference models thanks to a 5 Heatpipe heatsink that delivers more cooling performance for stability and overclocking. There’s also a mention of the 6 Dynamic Phase Control PWM IC, a part of the high-quality components used on the PCB. Rest of the logos include AMD Eyefinity, AMD Crossfire, AMD GCN Architecture and AMD App Acceleration Technology.

Out of the box, we can see the HIS Radeon R9 390 IceQ X2 graphic card which is massive in size with the PCB measuring 29cm long and the cooler measuring 30cm long. It will be slight bit of an issue to equip the card in some small cases so you better check the max graphic card length your casing can support beforeBUYING

this graphic card. The cooler being a revised version of the IceQ X2 cooler features a Beige / Metallic Gold color scheme which is a new refreshing color scheme which hasn’t been previously used on a graphics card.

Display outputs on the card include Dual DVI, full length HDMI and a single display port which can be used with the multi-stream technology and Eyefinity multi-monitor technology. The Radeon R9 390 is fully compatible on AMD Freesync monitors. A rather small vent is placed right on top of the ports to push heat out side of the card so it stays cool and gets a fresh supply of cool air.

For cooling, the HIS Radeon R9 390 IceQ x2 uses two 89mm dual axial fans that tend to get loud under load but are specifically designed to push lots of air into the heatsink so that the card stays cool even under massive loads and overclocked conditions. The Q&C fan blade increases air velocity make cooling more efficiently while keeping noise level at minimum.

HIS Radeon R9 390 IceQ X2 8 GB – Close-Up

Before taking apart the cooler shroud and heatsink, we take a last look at the graphic card from the top. The black shroud coupled with the silver mesh looks great and does a great job hiding the PCB components. The X mesh in the middle represents the IceQ X2 cooler and we can see an embedded logo on the top left corner of the card. The silver X shroud goes well with the Beige metallic gold color the shroud is embedded in. HIS has used stickers on the front and back of the shroud which give a nice touch to the card.

The Radeon R9 390 features the latest XDMA Crossfire technology which rids the PCB of goldfinger connectors and instead users the PCI-Express bus to communicate with up to four graphics cards that can be used parallel to one another for enhanced performance in games. We can also spot the Dual bios switch placed right next to the Crossfire connectors.

For cooling, HIS went with an absolutely gorgeous yet massive IceQ X2 cooler which features an enormous amount of aluminum heatsink fins crammed beneath the cooler shroud. The heatsink is a revised version of original IceQ X2 cooler offering superior cooling performance and thermal dissipation.

The main heatsink block features a 40 x 47.5 mm copper heatsink base which is equipped with two 8mm and three 6mm pure copper heatpipes which dissipate heat from the copper base to the aluminum fin array. The heatsink is cooled off with the dual 89mm axil fans which pushes the heat out of the card from the sides and the front exhaust panel.

The main copper contact based for the GPU is equipped with five dense heatpipes which run across the large densely packed aluminum fins array which are cooled off by the dual axial fans located at the top of the cooling shroud.

The HIS Radeon R9 390 IceQ X2 OC is powered by a single 8 Pin + 6 Pin power configuration that feeds 275W of power to the GPU Core, memory and the rest of the components of the PCB. It should be noted that the non-reference design allows for more stability and much improved overclocking support than the reference variant.

Removing the cooler reveals a secondary PCB heatsink that is placed to provide additional cooling the PLL, memory chips, VRM, PWM and various other voltage and power revaluation chips. There’s a large support bracket on the side of the PCB to firmly hold the card in place to avoid any bends due to the heavy weight.

Next up, we have the PCB itself with several components which we will detail in a short moment but the first thing we notice is the blue PCB. HIS Digital has been using blue colors for PCB for some time now since it suits the IceQ X2 design scheme but most people would disagree and a black PCB would be better off.

HIS has equipped their Radeon R9 390 IceQ X2 OC with 6+1+1 (6 Phase Core, 1 Phase Memory, 1 Phase PLL) phase power supply with DirectFET MOSFETs which delivers monstrous amount of power and voltage to the board resulting in higher stability and better overclocking over the reference models. In addition to this, the card comes with solid state capacitors, CHiL 8228G VRM controller and solid state chokes. HIS Radeon R9 390 IceQ X2_VRM 1

Below, we can see the AMD Hawaii Pro (Grenada Pro) core which has spanned a lifetime of almost 2 years in the graphics industry and still going strong. The Hawaii chip features 6200 Million transistors crammed inside the 438mm2 die. You can see a IHS around the die which improves overall heat dissipation off the card.

For memory, HIS equipped their card with the SK Hynix made H5GC4H24AJR-T2C memory modules (16 chips in total) which provide a VRAM of 8GB along a 512-bit memory interface. The modules are supposed to run at 6.00 Gbps with a stock voltage of 1.5V.

All games were tested on 2560×1440 (16:9) resolution.

Image Quality and graphics configurations have been provided in the screenshots below.
Games with PhysX were benchmarked with the settings on Low or Off for fair comparisons.

HIS Radeon R9 390 IceQ X2 8 GB – 1440P Graphics Performance

3DMark Firestrike:

Assassins Creed Unity:

Battlefield 4:

Crysis 3:

Civilization: Beyond Earth:

Dragon Age: Inquisition:

Far Cry 4:

GTA V:

Project Cars:

The Witcher 3: Wild Hunt:

Watch Dogs:

HIS Radeon R9 390 IceQ X2 @ 1100 MHz Overclocked Performance:

HIS Radeon R9 390 IceQ X2 8 GB – Thermals and Acoustics

No graphic card review is complete without evaluating its temperatures and thermal load. The HIS Radeon R9 390 IceQ X² OC we were shipped features the revised IceQ X² cooler which delivers much better cooling and stability over the reference model. This coupled with a custom designed PCB which improves overall stability and power rating of the board.

Note – We tested load with Kombuster which is known as ‘Power viruses’ and can permanently damage hardware. Use such software at your own risk!

IceQ X² is one of the quietest coolers, making the card quieter than the reference cooler. The card is below 28dB when watching movies, surfing Facebook, working. The card remains quiet whether you are gaming, online socializing, entertaining or working.

HIS IceQ X2 Technical Details:

The super large 2x 89mm dual axial fans draws massive amount of cool air to cool the GPU directly.
The Q&C fan blade increases air velocity make cooling more efficiently while keeping noise level at minimum.
The fan is extremely durable with a life span of up to 50,000 hours.
The 2x 8mm and 3x 6mm wide heatpipes optimize cooling performance by removing heat from the core area.
The extra large 40 x 47.5 mm copper heatsink provides a large area to dissipate heat efficiently.
The card also carries memory and mosfet heat sinks to provide extra heat dissipation from these areas.
The card also features fan control and monitoring, you are able to read fan speed & Asic temperature from iTurbo or other softwares.

Conclusion

The HIS Radeon R9 390 IceQ X2 features an impressive build design which comes with their iconic cooler. The Radeon R9 390 IceQ X2 is crafted with the Grenada Pro core which is simply a more refined Hawaii. The last time we saw Hawaii in the 200 series family, it was impressive but it featured significantly higher temperatures and consumed a lot of power. This has been slightly improved with the Grenada or Hawaii 1.1 revision since it has lower temperatures and new power management features which took AMD a year to perfect. The lower temperatures mean that AMD could boost the clock speeds to deliver better performance while keeping the power envelope around the same as Hawaii 1.0. While the improvement may not seem a lot, this allows HIS and other AIB partners of AMD to churn better cards that are factory overclocked out of box.

A second and most notable improvement of the HIS Radeon R9 390 IceQ X2 is that it comes with 8 GB of VRAM which is twice the memory buffer featured on Radeon R9 290 series. The more complex and highly dense vram chips mean that AMD can boost the clock speeds for the memory too, delivering 384 GB/s bandwidth. The aim of AMD is to deliver 4K gaming performance ready cards and they have the designs ready but when it comes to performance, we simply don’t have any new title that is playable at 4K with the cards hence the 8 GB VRAM goes off to a waste. The 2560x1440P resolution is the most suited resolution for the cards unless you are going to play older titles at higher resolutions which can run fine at 4K (3840×2160) or you can run the latest titles with reduced image quality to make them at least playable on 4K but that defeats the purpose of owning a high end graphics card. When we compare 4GB to 8GB on 2K resolution, we see very minimal improvements.

The HIS IceQ X2 cooler does a great job as usual, cooling the card and keeping it stable under full load. The only issue with the card is its odd color scheme which I have seen on several of HIS cards. The blue PCB doesn’t look that great while the gold shroud does look good but HIS should replace the PCB color to black otherwise its a really odd color scheme. The power consumption on the other hand is something that HIS can’t control as AMD designed the chip and the usage could be out of proportion most of the times under full load. This shows that the cooler keeps the Grenada tame under its hood which would otherwise be struggling on lower end coolers.

HIS R9 390 IceQ X2 is priced at $329 US and enters in the same market which has been conquered by the GTX 970 since September 2014. Both cards have their benefits, GCN makes an aim for a totally high performance oriented design while Maxwell keeps everything arranged in an efficient manner. Both cards trade blows with each other and at $329, its hard to recommend either one but for users who really demand 4K ready graphics performance, then the Radeon R9 390 8 GB IceQ X2 is a perfect card to own with huge amounts of memory, regardless of the power demand.