Today's review will focus on how the official increased clock frequencies of the 7970 GHz edition aka Tahiti XT2 – 1050MHz on the core (13.5% increase from 925MHz) and 1500MHz on the memory (9% increase from 1375MHz), help peg back or increase the performance delta over Nvidia's GTX 680 Kepler. We run a plethora of benchmarks on the latest beta drivers (Catalyst 12.7 and 304.48) from both camps, sometimes with interesting results to report.
HD 7970 Tahiti XT
Tahiti B0 CR XT C38601 GDDR5 3GB
300MHz / 150MHz : 0.850V
925MHz / 1.38GHz : 1.170V
HD 7970 GHz Edition Tahiti TX2
Tahiti B0 XT2 C38620 GDDR5 3GB
300MHz / 150MHz : 0.950V
1.05GHz / 1.5GHz : 1.256V
…Unfortunately, our efforts to find out more from the horse mouth about the exact scenarios when and how the boost will kick in was met with silence.
DTE works as a deterministic model of temperature in a worst case environment, as to give us a better estimate of how much current the ASIC is leaking at any point in time. As a first order approximation, ASIC power is roughly a function of: dynamic_power(voltage, frequency) + static_power(temperature, voltage, leakage).
Traditional PowerTune implementations assume that the ASIC is running at a worst case junction temperature, and as such always overestimates the power contribution of leaked current. In reality, even at a worst case ambient temp (45C inlet to the fansink), the GPU will not be working at a worst case junction temperature. By using an estimation engine to better calculate the junction temp, we can reduce this overestimation in a deterministic manner, and hence allow the PowerTune architecture to deliver more of the budget towards dynamic power (i.e. frequency) which results in higher performance. As an end result, DTE is responsible for about 3-4% performance uplift vs the HD7970 GHz Edition with DTE disabled.
The DTE mechanism itself is an iterative differential model which works in the following manner. Starting from a set of initial conditions, the DTE algorithm calculates dTemp_ti/dt based on the inferred power consumption over a previous timeslice (is a function of voltage, workload/capacitance, freq, temp, leakage, etc), and the thermal capacitance of the fansink (function of fansink and T_delta). Simply put, we estimate the heat into the chip and the heat out of the chip at any given moment. Based on this differential relation, it’s easy to work back from your initial conditions and estimate Temp_ti, which is the temperature at any timeslice. A lot of work goes into tuning the parameters around thermal capacitance and power flux, but in the end, you have an algorithmic way to always provide benefit over the previous worst-case assumption, but also to guarantee that it will be representative of the entire population of parts in the market.
We could have easily done this through diode measurements, and used real temperature instead of digital temperature estimates…. But that would not be deterministic. Our current method with DTE guarantees that two parts off the shelf will perform the same way, and we enable end users to take advantage of their extra OC headroom on their parts through tools like Overdrive.
Spot the differences between the regular 7970 and 7970 GHz
Nope, the old and "new" cards are completely identical hardware wise. We've confirmed this fact with AMD and several AIB partners. Whats allegedly different is the hand-picked better ASICs that goes into the GHz edition cards.
ASIC Quality and Overclocking
Out of the three cards we've laid our hands on, the ASIC quality readings ranged from 59.8% to 61% – lower than the 70-80% that we've had on our bunch of other older 7970s. That said, we could run the new cards at 1.3V with MSI Afterburner without tripping the overvoltage protection and attained core frequencies of between 1260MHz and 1290MHz, slightly higher than the older cards.
How much will the new cards cost?
Starts from US$499. This places the 7970 GHz edition head on with the GTX 680's official MSRP.