Review — Nvidia GeForce RTX 2070 Founders Edition – Complicated Equation with Many Unknowns | Launch review

1 - Vorbemerkung und Einführung 2 - Tear Down - Platinenanalyse 3 - Ashes of the Singularity 4 - Battlefield 1 5 - Destiny 2 6 - The Division 7 - Doom (Vulkan) 8 - Far Cry 5 9 - Forza Motorsport 7 10 - Ghost Recon 11 - GTA V 12 - Metro Last Light 13 - Tomb Raider 14 - The Witcher 3 15 - World of Warcraft 16 - Leistungsaufnahme, Temperaturen, Lautstärke 17 - Zusammenfassung und Fazit

Power consumption

This time, too, I have to improvise a bit and use the data of my US colleague. That is why I have to refer to some sub-areas, such as CAD, Multimonitor and others do not and unfortunately cannot check the quality of the power supply including the spikes. Nevertheless, the available values of a USB riser card should be accurate enough to get a first picture.

With just under 12 watts in the Idle, you're also well in the race. In the gaming loop and the Torture loop, almost 188 and 189 watts, only marginally more than Nvidia indicates as TBP. This is therefore a kind of point landing and is therefore approx. 15 to 17 watts above what I was able to measure for the GeForce GTX 1080 FE at the time.

However, my test sample at the time was a chip with very high quality and below-average power consumption, so the comparison may seem a little bit lagging. But the GeForce RTX 2070 still does not offer a real improvement.

Chris' measurements with TLAA vs. DLSS also produce only small differences in power consumption that are less than 2 watts. Which would also clarify that it doesn't really matter whether the tensor cores are included or not, because the TBP affects the entire board and is regulated accordingly. The shader clock was then approx. 15 to 30 MHz lower, which is also rather negligible.

Compliance with the specifications for the PCI-Express motherboard slot is also given, even within the scope of possible measurement inaccuracies. So you can leave it calm:

Temperatures, fan speeds, noise

Since measurements in open construction are rather non-practical, this time I can only represent half of the truth and have to be content with the few available values. According to Chris' measurements in the gaming and torture loop 73°C, the card gets hot and the fan speeds are just below 1900 rpm in the open setup, which should be clearly audible. The measurements in the acoustic laboratory and the sound spectrum must also be omitted due to a lack of sample.