Primarily based on the TU117 die the brand new GeForce GTX 1650 nonetheless contains all the further Turing shader improvements that enhance efficiency and effectivity, together with help for concurrent floating level and integer operations, a unified cache structure with bigger L1 cache, and adaptive shading.
The TDP for the brand new GPU merely is 75 watts and which means it doesn’t require an exterior energy connector, so it is going to be the quickest graphics card obtainable that gained’t want exterior PCIe energy. Then just like the GTX 1660 Ti and 1660, there received’t be a Founders Edition model, however no massive loss there.
The GTX 1650 packs 896 CUDA cores, that’s 36% fewer than what you get with the GTX 1660. That’s the greatest downgrade we’ve seen within the Turing product stack but. For virtually the most half we’ve seen cuts of around 20% when going from 2080 to 2070, 2070 to 2060, 2060 to the 1660 Ti after which just 8% from the 1660 Ti to the vanilla 1660.
The one different instance of over a ~30% downgrade was the RTX 2080 Ti to the RTX 2080, right here we noticed a 32% discount in core rely on and this result in a ~25% discount in efficiency at 4K. We count on to see an analogous efficiency discount for the GTX 1650 when in comparison with 1660 because the 36% discount in cores is backed by a 33% discount in reminiscence bandwidth.
Furthermore, whereas 1660 is armed with 6GB of GDDR5 reminiscence, the GTX 1650 will get 4GB utilizing a mere 128-bit vast reminiscence bus for peak reminiscence bandwidth of 128GB/s. The default clock speeds have been set to a base of 1485 MHz and a lift of 1665 MHz, a 7% clock velocity discount when in comparison with the GTX 1660.
Nvidia claims the GTX 1650 is twice as quick because of the GTX 950, and as much as 70% sooner than the GTX 1050 at 1080p, although the wording is a bit of tough right here as their declare is restricted to favorite video games with complicated shaders. So video games comparable to Shadow of the Tomb Raider and Wolfenstein II, for instance. Therefore the typical efficiency uplift might decrease relying on the title.
For testing, we’ll be trying carefully at 1080p efficiency in a dozen titles. Our check system is the standard GPU bench that features a Core i9-9900K clocked at 5 GHz with 32GB of DDR4-3200 reminiscence.