The NVIDIA GeForce GTX 1070 Review



Analyzing Temperatures & Frequencies Over Time

Modern graphics card designs make use of several advanced hardware and software facing algorithms in an effort to hit an optimal balance between performance, acoustics, voltage, power and heat output. Traditionally this leads to maximized clock speeds within a given set of parameters. Conversely, if one of those last two metrics (those being heat and power consumption) steps into the equation in a negative manner it is quite likely that voltages and resulting core clocks will be reduced to insure the GPU remains within design specifications. We’ve seen this happen quite aggressively on some AMD cards while NVIDIA’s reference cards also tend to fluctuate their frequencies. To be clear, this is a feature by design rather than a problem in most situations.

In many cases clock speeds won’t be touched until the card in question reaches a preset temperature, whereupon the software and onboard hardware will work in tandem to carefully regulate other areas such as fan speeds and voltages to insure maximum frequency output without an overly loud fan. Since this algorithm typically doesn’t kick into full force in the first few minutes of gaming, the “true” performance of many graphics cards won’t be realized through a typical 1-3 minute benchmarking run. Hence why we use a 10-minute warm up period before all of our benchmarks.

For now, let’s see how these new algorithms are used when the card is running at default speeds.

One of the main setbacks we encounters with the GTX 1080 was the stock heatsink causing temperatures to climb to unacceptable levels, leading to lower than expected clock speeds over time. The GTX 1070 on the other hand is made of slightly stouter stuff since its cut-down GP104 core essentially produces less heat. This means much lower temperatures from the stock heatsink even though it doesn’t utilize a high end vapor chamber.

Oddly enough, while the GTX 1070 starts off from a slightly higher temperature from near-idle, its more efficient core remains substantially cooler than the GTX 1080 throughout our testing.

Fan speeds for this card are extremely lethargic when compared with some other NVIDIA GPUs and that’s something which will lead directly to lower acoustic results. More importantly, the fan curves scales in an almost linear curve vis a vis the GTX 1080 so –based on the temperatures above- the heatsink is able to properly cool the GP104 core without excessive noise.

When we get down into the heart of this particular situation, it becomes evident that the GTX 1070’s core output, heatsink and fan speeds are able to work together to insure throttling doesn’t occur. This is actually in marked contrast to the GTX 1080 which tended to lower its core frequencies by a pretty substantial amount in an effort to hit a given TDP.

Amongst other reference-based solutions, the GTX 1070 Founders Edition is among the most consistent we’ve seen. That means relatively “flat” performance numbers without any evidence of throttling or clock gating. It should be interesting to see how this setup fares in overclocked situations.

Posted in

Latest Reviews