NVIDIA GeForce GTX 980 SLI (Laptop) vs NVIDIA GeForce RTX 4070 Ti SUPER

NVIDIA GeForce GTX 980 SLI (Laptop)

The Nvidia GeForce GTX 980 SLI is a high-end DirectX 12-compatible graphics solution for laptops, consisting of two GeForce GTX 980 cards. Each card has a 28 nm GM204 GPU based on the Maxwell architecture with similar specifications as the desktop GeForce GTX 980. The clocks for the SLI-setup are usually identical to a single GTX 980.

With SLI, each card renders a single frame (AFR mode). Therefore, it may suffer from micro-stuttering in low fps ranges of 30 fps. This happens because of different time-spans between two frames (e.g., irregular delays between sequential frames).

The performance is on a level with the two desktop GTX 980 cards in SLI. Due to the different cooling solutions, the long-term performance may be slightly lower compared to a well cooled desktop, but Nvidia assures there will only be small degrades in performance. As with all SLI combinations, the performance depends on the driver support for the used benchmark and game. In our tests with an MSI GT80S, the 980 SLI was the fastest graphics solution for laptops. The 980M SLI was 17% slower, a single 980 26% slower. Therefore, the 980 SLI for laptops offers enough performance for all demanding games from 2015 in high details and 4K resolution.

The core features are also the same as the desktop 980, so expect support for DirectX 12 (FL 12_1). There will be no support for the new H.265 video engine as found on the desktop GTX 950.

The power consumption of a single card is a lot higher than other mobile graphics cards like the GeForce GTX 980M (100 Watt TDP). According to Nvidia, the TDP of the reference design is still 165 Watt like in the desktop version. So, the TDP of the SLI combination may be as high as 330 Watt. Therefore, only big laptops with efficient cooling methods can house the GeForce GTX980 for laptops.

NVIDIA GeForce RTX 4070 Ti SUPER

The Nvidia GeForce RTX 4070 Ti SUPER is a high-end desktop graphics card based on the Ada Lovelace architecture. It uses the AD103 chip and offers 8.448 cores and 16 GB GDDR6X graphics memory (256-bit bus with 22.4 Gbits).

The gaming performance is still sufficient for 4k with ultra settings, even in demanding games such as Cyberpunk, but 60 fps is no longer always achieved. With ray tracing, the use of DLSS and possibly reducing the resolution is recommended. In a competitive comparison, the SUPER variant ranks, as expected, between the RTX 4080 and the old 4070 Ti in a competitive comparison.

The AD103 chip is produced at TSMC in 5nm (4N with EUV) and is specified with a TGP (Total Graphics Power) of 285 watts.