NVIDIA GeForce GTX 1660 SUPER Appears to Be Identical to Base Model Aside from Memory

[Tsing]

The specifications for MAXSUN's upcoming GeForce GTX 1660 SUPER GPU has been published online, courtesy of Chinese retailer jd.com. Apparently, the card will be largely identical to the non-SUPER model aside from the inclusion of GDDR6 memory (vs. GDDR5) and higher speed/bandwidth (14 Gbps, 336 GB/s vs. 8 Gbps, 192 GB/s). CUDA cores and base/boost clocks remain at 1408 and 1530 MHz/1785 MHz, respectively.

These specifications...show us that the GTX 1660 Super isn't as Super as Nvidia's existing RTX Super series, lacking GPU core upgrades over today's GTX 1660 (non-Super). That's right; the GTX 1660 Super offers users the same 1,408 CUDA cores and the same core clock speeds. All that changes is the graphics card's memory, which has been upgraded to GDDR6.

 

[Dudhunter]

No surprise here. Nvidia will continue to short value until threatened.

 

[AntiQuark]

I agree with you in general but this might be just one mfrs take on it.
And 50% increase in memory bandwidth is a heck of a lot, it must have some use.
It will be interesting to see test results.

Edit the memory is actually 75% faster!

 

[Riccochet]

I don't think it's going to have much of an impact without an increase in CUDA cores.

 

[Zarathustra]

Maxsun? Are they new?