Temperature Analysis & Acoustics / Power Consumption
For all temperature testing, the cards were placed on an open test bench with a single 120mm 1200RPM fan placed ~8” away from the heatsink. The ambient temperature was kept at a constant 22°C (+/- 0.5°C). If the ambient temperatures rose above 23°C at any time throughout the test, all benchmarking was stopped. For this test we use the 3DMark Batch Size test at it highest triangle count with 4xAA and 16xAF enabled and looped it for one hour to determine the peak load temperature as measured by GPU-Z.
For Idle tests, we let the system idle at the Windows 7 desktop for 15 minutes and recorded the peak temperature.
With an all new heatsink design and high end thermal compound, the HD 6990 is able to keep both cores very cool for a dual chip card.
Noise does factor into the equation though and it’s not only from the fan. The high end Volterra VRMs are known for their efficiency but are also legendary for the loud hissing / whining they make when under high loads. Other areas of the PWM also contribute which actually makes the components louder than the fan at times. We found that enabling V-Sync reduced this noise but let’s be honest; you don’t buy an ultra high end card to play at a constant 60 FPS.
Meanwhile, the fan itself does make this card noisier than many single card setups but at subjectively equal acoustic levels to most high end dual card configurations. In our experience, the HD 6990’s fan noise is no louder than two HD 6970 cards in Crossfire but that doesn't mean it's anywhere near quiet either.
Aside from the racket the PWM on our card made, there was one other issue: the direction some of the heat travels.
When installed into a standard ATX case or even a unique product like the Silverston Raven series, the HD 6990 will be constantly pushing part of its hot exhaust air against the airflow direction of the enclosure. Considering we were seeing temperatures of nearly 130 degrees Fahrenheit, in-case temperatures could definitely be raised.
On the other hand, we actually prefer this design since it keep both cores’ heat well under control and most modern enclosures which are designed to handle this huge card should have no issue dispersing its heat.
For this test we hooked up our power supply to a UPM power meter that will log the power consumption of the whole system twice every second. In order to stress the GPU as much as possible we once again use the Batch Render test in 3DMark06 and let it run for 30 minutes to determine the peak power consumption while letting the card sit at a stable Windows desktop for 30 minutes to determine the peak idle power consumption. We have also included several other tests as well.
Please note that after extensive testing, we have found that simply plugging in a power meter to a wall outlet or UPS will NOT give you accurate power consumption numbers due to slight changes in the input voltage. Thus we use a Tripp-Lite 1800W line conditioner between the 120V outlet and the power meter.
The HD 6990’s power consumption is as expected very high for a single card but it was actually quite low considering the performance it brings to the table. By specifically binning the cores for the lowest leakage possible and then ensuring they stayed well under the temperature of the HD 6970, AMD ensured efficiency was about equal to that of two HD 6950 cards.
In the chart above we can see the effect higher temperatures, additional voltage, increased clock speeds and higher leakage cores have on the HD 6970 Crossfire solution.
Choosing a power supply for this card won’t be an inexpensive endeavor either. If you plan on keeping it at stock speeds, we are going to recommend a good 850W unit with a single or two large rails to ensure proper delivery. Overclocking brings in a whole new scenario and while this will be discussed in a later section, a good 1000W PSU plus an enthusiast motherboard with a separate Molex of FDD power connector for the PCI-E slots is recommended.
Temperature Analysis & Acoustics
For all temperature testing, the cards were placed on an open test bench with a single 120mm 1200RPM fan placed ~8” away from the heatsink. The ambient temperature was kept at a constant 22°C (+/- 0.5°C). If the ambient temperatures rose above 23°C at any time throughout the test, all benchmarking was stopped. For this test we use the 3DMark Batch Size test at it highest triangle count with 4xAA and 16xAF enabled and looped it for one hour to determine the peak load temperature as measured by GPU-Z.
For Idle tests, we let the system idle at the Windows 7 desktop for 15 minutes and recorded the peak temperature.
With an all new heatsink design and high end thermal compound, the HD 6990 is able to keep both cores very cool for a dual chip card.
Noise does factor into the equation though and it’s not only from the fan. The high end Volterra VRMs are known for their efficiency but are also legendary for the loud hissing / whining they make when under high loads. Other areas of the PWM also contribute which actually makes the components louder than the fan at times. We found that enabling V-Sync reduced this noise but let’s be honest; you don’t buy an ultra high end card to play at a constant 60 FPS.
Meanwhile, the fan itself does make this card noisier than many single card setups but at subjectively equal acoustic levels to most high end dual card configurations. In our experience, the HD 6990’s fan noise is no louder than two HD 6970 cards in Crossfire but that doesn't mean it's anywhere near quiet either.
Aside from the racket the PWM on our card made, there was one other issue: the direction some of the heat travels.
When installed into a standard ATX case or even a unique product like the Silverston Raven series, the HD 6990 will be constantly pushing part of its hot exhaust air against the airflow direction of the enclosure. Considering we were seeing temperatures of nearly 130 degrees Fahrenheit, in-case temperatures could definitely be raised.
On the other hand, we actually prefer this design since it keep both cores’ heat well under control and most modern enclosures which are designed to handle this huge card should have no issue dispersing its heat.
System Power Consumption
For this test we hooked up our power supply to a UPM power meter that will log the power consumption of the whole system twice every second. In order to stress the GPU as much as possible we once again use the Batch Render test in 3DMark06 and let it run for 30 minutes to determine the peak power consumption while letting the card sit at a stable Windows desktop for 30 minutes to determine the peak idle power consumption. We have also included several other tests as well.
Please note that after extensive testing, we have found that simply plugging in a power meter to a wall outlet or UPS will NOT give you accurate power consumption numbers due to slight changes in the input voltage. Thus we use a Tripp-Lite 1800W line conditioner between the 120V outlet and the power meter.
The HD 6990’s power consumption is as expected very high for a single card but it was actually quite low considering the performance it brings to the table. By specifically binning the cores for the lowest leakage possible and then ensuring they stayed well under the temperature of the HD 6970, AMD ensured efficiency was about equal to that of two HD 6950 cards.
In the chart above we can see the effect higher temperatures, additional voltage, increased clock speeds and higher leakage cores have on the HD 6970 Crossfire solution.
Choosing a power supply for this card won’t be an inexpensive endeavor either. If you plan on keeping it at stock speeds, we are going to recommend a good 850W unit with a single or two large rails to ensure proper delivery. Overclocking brings in a whole new scenario and while this will be discussed in a later section, a good 1000W PSU plus an enthusiast motherboard with a separate Molex of FDD power connector for the PCI-E slots is recommended.
Last edited: