GPUs Hardware Reviews

Reason instead of obesity: Gigabyte RX Vega64 Gaming OC in test

If there were enough Vega chips and thus a wide range of cards, the Gigabyte RX Vega64 in this form would probably be the ideal entry-level card, rather than a top-of-the-line model. But it really must always be the demonstration of what is technically feasible... Board layout As with the RX Vega56 Gaming OC, Gigabyte is extremely different from AMD's reference layout, which should also make the use of existing full-cover water coolers impossible. Also at the popular Raijintek Morpheus, the end of G... We have to put it in the first place that we actually had to double-benchmark all the games. At the last moment, Gigabyte incorporated AMD's new kernel from the source BIOS, which accommodates the custom models with a customized power table. D... Power consumption at different loads The power consumption in the gaming loop is at the measured approx. 261 watts in Balanced Mode about 20 watts below what could be measured at the reference in the default BIOS. This is all the more astonishing because... Overclocking and undervolting The conventional overclocking by means of an even higher power limit and more clock counteracts the current cooling concept, because the rather quiet cooler gets along quite well with what it has to dismount ex works. More on the other hand... Cooling system and backplate Of course, the generated waste heat is directly related to the recorded power, for which the cooling solution is responsible for optimum dissipation. And this is precisely where we are faced with various compromises, which are... Summary For the Gigabyte RX Vega56 Gaming OC we have already written a very detailed conclusion, which applies unreservedly to the bigger sister in the same forum. Butter-and-bread instead of lighthouse project and cost optimization for ...

If there were enough Vega chips and thus a wide range of cards, the Gigabyte RX Vega64 in this form would probably be the ideal entry-level card, rather than a top-of-the-line model. But does it really always have to be the demonstration of what is technically feasible?

Unlike Sapphire's heavyweight RX Vega64 Nitro+, the Gigabyte RX Vega64 Gaming OC, like its already tested smaller sister RX Vega 56 Gaming OC, is a fairly grey mouse with a healthy body-size index. Just don't stand out at all costs. And it almost seems as if it has even deliberately been designed to arouse as few desires as possible, so as not to trigger any demand.

It shows us once again what a normal "butter-and-bread card" with Vega10 could have looked like, or perhaps even should have looked like. It is thus the exact opposite of the Sapphire card and yet also has its own personal touch.

Market and availability

From internal sources, we know that the AIB received only a few chips in amD's mid-four-digit range, most of them Vega64. The general scarcity of packages thus prevents real mass production at the moment, because one cannot constantly drive and shut down a production line or even have it paused for longer. Costs and effort would simply be too high.

The fact that we were still able to get a copy is not due to the above-average capabilities of German importers, especially since, according to internal information, only cards in the double-digit range were available here, but on a certain Self-initiative in self-importing (including taxes).

Unboxing, optics and haptics

Weighing just 1006 grams, the Gigabyte card is almost 600 grams lighter than Sapphire's monstrous battleship. Gigabyte has also saved more significantly in dimensions and offers a relatively short map with only 27.8 cm installation length (outer surface slot panel until the end of the radiator cover). With a height of 13 cm (from the top of the motherboard slot to the top of the radiator cover, as well as a thickness of 4.7 cm from the board to the front of the radiator cover, the picture is a rather compact map.

The two counter-rotating fans with a rotor blade diameter of 9.5 cm each sit in an opening with a diameter of 10 cm. A total of 11 rotor blades per fan, with their special shape, mainly provide airflow and turbulence and less static pressure. This should be helpful to the cooling concept.

The backplate dispenses with the Aorus logo, which still featured on the pre-production model. The copper insert, which sits above the back of the package, thermally separates the backplate used for cooling from this area and also cools – a little. The integrated heatpipe is not (yet) seen here on these pictures. More on that later.

After all, one should plan at least 5 mm for the rear body, which can already lead to problems with some motherboards if heat sinks are sitting too close to the slot or the CPU cooler is extremely large.

It is already clear at the bottom of the card that Gigabyte has also worked a little on the shape of the vertically arranged slats. Due to the inclination and the waveform, the airflow should cool more efficiently. We will be questioning this at a later stage.

The top also clearly shows that Gigabyte has deliberately set the red pencil. No RGB effects and other LED playstuffs, but a somewhat jagged plastic desert of injection molding and printed orange color accents. The white gigabyte logo is also only printed. In addition, gigabytes of two 8-pin sockets are sufficient for external power supply.

The closed end of the card shows not much except the end of the two 8 mm and three 6 mm heatpipes, but only the connection cable of the fans. Other possible sockets were provided on the board (e.g. RGB output), which has been omitted in the course of development.

The slot panel is very interesting in that it has six instead of the five connectors of the reference card. Three HDMI 2.0 ports and three DisplayPorts 1.4 are available to the user. This makes the card suitable for connecting up to six monitors in eyefinity mode.

Finally, let us take a first look at the technical data. The value displayed as the maximum boost clock, on the other hand, is rather a wishful thinking, which is unlikely to be achieved within the set power limit or at all.

Compared to the relevant reference cards, this looks like this:

Model Radeon
RX Vega64
RX Vega64
Gaming OC
RX Vega56
GTX 1070 Ti
GTX 1080
Gpu Vega 10 Vega 10 Vega 10 GP104 GP104
Chip size 484 mm2 484 mm2 484 mm2 314 mm2 314 mm2
Transistors 12.5 billion 12.5 billion 12.5 billion 7.2 billion 7.2 billion
GPU base clock/
Boost clock
1274 MHz
1546 MHz
1274 MHz
1630 MHz
1156 MHz
1471 MHz
1607 MHz
1683 MHz
1607 MHz
1733 MHz
Shader/SIMD 4096/64 4096/64 3585/56 2432/19 2560/20
Texture Units/
Pixel fill rate
99 GPix/s 104 GPix/s 94 GPix/s 108 GPix/s 114 GPix/s
Texture Fill Rate 396 GTex/s 417 GTex/s 330 GTex/s 244 GTex/s 257 GTex/s
Memory connection 2048 Bit 2048 Bit 2048 Bit 256 bits 256 bits
Storage type HBM2 HBM2 HBM2 GDDR5 GDDR5X
Memory bandwidth
484 GB/s 484 GB/s 410 GB/s 256 GB/s 320 GB/s
Memory clock
1.9 Gbps 1.9 Gbps 1.6 Gbps 8 Gbps 10 Gbps
Storage expansion
8 GB 8 GB 8 GB 8 GB 8 GB
DX Feature Level 12_1 12_1 12_1 12_1 12_1
PCIe sockets 2 x 8-pin 2x 8 Pin 2 x 8-pin 1x 8-pin 1x 8-pin
Tbp 295 watts 295 watts 210 watts 180 watts 180 watts

Test system and measurement methods

The new test system and the methodology have already been described in great detail in the basic article "How We Test Graphics Cards" (English: "How We Test Graphics Cards") and therefore, for the sake of simplicity, we now only refer to this detailed Description. So if you want to read everything again, you are welcome to do so. However, we have again improved CPU and cooling to largely exclude possible CPU bottlenecks for this fast card.

If you are interested, the summary in table form quickly provides a brief overview:

Test systems and measuring rooms
Intel Core i7-6900K -4.3GHz
MSI X99S XPower Gaming Titanium
G.Skill TridentZ DDR4 3600
1x 1 TByte Toshiba OCZ RD400 (M.2, System SSD)
2x 960 GByte Toshiba OCZ TR150 (Storage, Images)
Be Quiet Dark Power Pro 11, 850-watt power supply
Alphacool Ice Block XPX
5x Be Quiet! Silent Wings 3 PWM (Closed Case Simulation)
Thermal Grizzly Kryonaut (for cooler change)
Lian Li PC-T70 with expansion kit and modifications
Modes: Open Benchtable, Closed Case
Monitor: Eizo EV3237-BK
Power consumption:
non-contact DC measurement on the PCIe slot (Riser-Card)
non-contact DC measurement on the external PCIe power supply
Direct voltage measurement on the respective feeders and on the power supply
2x Rohde & Schwarz HMO 3054, 500 MHz multi-channel oscillograph with memory function
4x Rohde & Schwarz HZO50, current togor adapter (1 mA to 30 A, 100 KHz, DC)
4x Rohde & Schwarz HZ355, touch divider (10:1, 500 MHz)
1x Rohde & Schwarz HMC 8012, digital multimeter with storage function
Optris PI640, infrared camera
PI Connect evaluation software with profiles
NTI Audio M2211 (with calibration file)
Steinberg UR12 (with phantom power for the microphones)
Creative X7, Smaart v.7
own low-reflection measuring room, 3.5 x 1.8 x 2.2 m (LxTxH)
Axial measurements, perpendicular to the center of the sound source(s), measuring distance 50 cm
Noise in dBA (Slow) as RTA measurement
Frequency spectrum as a graph
Operating system Windows 10 Pro (Creators Update, all updates)
Radeon Software Adrenaline (17.12.2, status December 2017)

Danke für die Spende

Du fandest, der Beitrag war interessant und möchtest uns unterstützen? Klasse!

Hier erfährst Du, wie: Hier spenden.

Hier kannst Du per PayPal spenden.

About the author

Igor Wallossek

Editor-in-chief and name-giver of igor'sLAB as the content successor of Tom's Hardware Germany, whose license was returned in June 2019 in order to better meet the qualitative demands of web content and challenges of new media such as YouTube with its own channel.

Computer nerd since 1983, audio freak since 1979 and pretty much open to anything with a plug or battery for over 50 years.

Follow Igor:
YouTube Facebook Instagram Twitter