Notebookcheck Logo
, , , , , ,
search relation.
, , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , ,
 

NVIDIA GeForce RTX 3070 Ti Laptop GPU

NVIDIA GeForce RTX 3070 Ti Laptop GPU

The Nvidia GeForce RTX 3070 Ti Laptop GPU (mobile, GN20-E6, Max-P, Max-Q) is a mobile graphics card for notebooks and the refresh of the RTX 3070. It was announced early 2021 at CES and is based on the same GA104 Ampere chip as the old RTX 3070 Laptop. Compared to the old card, the newer one offers 15% more shaders but less clock speed. It will still feature the 256 Bit memory bus and 8 GB GDDR6. The RTX 3070ti is available in different power variants ranging from 80 Watt TGP with a Boost clock of 1035 MHz up to 125 Watt models with a Boost clock of 1485 MHz.

There is no more Max-Q variant (formerly used for the low power variants) but every OEM can choose to implement Max-Q technologies (Dynamic Boost, Whispermode).

The performance depends on the TGP setting and cooling of the laptop. On average it should be somewhere between the old RTX 3070 Mobile and RTX 3080 Mobile (at similar TDP settings). Therefore, the card is best suited for QHD gaming in max details (for demanding games like Cyberpunk 2077). Less demanding games like the Farming Simulator 22 or older games should also run in 4K with max. detail settings.

The GA104 chip offers 6,144 FP32 ALUs of which half can also execute INT32 instructions (i.e. 3,072 INT32 ALUs). The RTX 3070 Ti however won't use all ALUs on the chip. With Turing all shaders could still execute FP32 or INT32 instructions. The raytracing and tensor cores on the chip were also improved according to Nvidia. The RTX 3070 Ti mobile series uses 5,888 of the 6,144 cores. The Ampere chips also include an improved 5th generation video encoder (NVENC for H.264 and H.265) and a 7th generation decoder (for various formats now including AV1).

The GA104 chip is manufactured by Samsung in 8nm (8N), which is not quite able to keep up with the 7nm node at TSMC (e.g. used by AMD and also for the professional GA100 Ampere chip).

GeForce RTX 3000 Series
NVIDIA GeForce RTX 3090 Ti compare 10752 @ 1.56 - 1.86 GHz384 Bit @ 21000 MHz
NVIDIA GeForce RTX 3090 compare 10496 @ 1.4 - 1.7 GHz384 Bit @ 19500 MHz
NVIDIA GeForce RTX 3080 Ti compare 10240 @ 1.37 - 1.67 GHz384 Bit @ 19000 MHz
NVIDIA GeForce RTX 3080 compare 8704 @ 1.45 - 1.71 GHz320 Bit @ 19000 MHz
GeForce RTX 3080 Ti Laptop GPU compare 7424 @ 0.98 - 1.59 GHz256 Bit @ 14000 MHz
NVIDIA GeForce RTX 3070 compare 5888 @ 1.5 - 1.73 GHz256 Bit @ 14000 MHz
GeForce RTX 3080 Laptop GPU compare 6144 @ 0.78 - 1.25 GHz256 Bit @ 14000 MHz
NVIDIA GeForce RTX 3060 Ti compare 4864 @ 1.41 - 1.67 GHz256 Bit @ 14000 MHz
GeForce RTX 3070 Ti Laptop GPU 5888 @ 0.92 - 1.48 GHz256 Bit @ 14000 MHz
GeForce RTX 3070 Laptop GPU compare 5120 @ 0.78 - 1.29 GHz256 Bit @ 14000 MHz
NVIDIA GeForce RTX 3060 compare 3584 @ 1.32 - 1.78 GHz192 Bit @ 15000 MHz
GeForce RTX 3080 Max-Q compare 6144 @ 0.78 - 1.25 GHz256 Bit @ 6000 MHz
GeForce RTX 3070 Max-Q compare 5120 @ 1.29 GHz256 Bit
GeForce RTX 3060 Laptop GPU compare 3840 @ 0.82 - 1.7 GHz192 Bit @ 14000 MHz
GeForce RTX 3060 Max-Q compare 3840 @ 1.28 GHz192 Bit
NVIDIA GeForce RTX 3050 compare 2560 @ 1.55 - 1.78 GHz128 Bit Bit @ 1750 MHz
GeForce RTX 3050 Ti Laptop GPU compare 2560 @ 1.22 - 1.49 GHz128 Bit @ 12000 MHz
GeForce RTX 3050 Laptop GPU compare 2048 @ 1.24 - 1.5 GHz128 Bit @ 12000 MHz
CodenameGN20-E6
ArchitectureAmpere
Pipelines5888 - unified
TMUs184
ROPs96
Raytracing Cores46
Tensor / AI Cores184
Core Speed915 - 1480 (Boost) MHz
Memory Speed14000 effective = 1750 MHz
Memory Bus Width256 Bit
Memory TypeGDDR6
Max. Amount of Memory8 GB
Shared Memoryno
Memory Bandwidth448 GB/s
APIDirectX 12_2, Shader 7.0, OpenGL 4.6
Power Consumption125 Watt (80 - 125 Watt TGP)
Transistor Count17.4 Billion
Die Size 392 mm²
technology8 nm
PCIe4.0
DisplaysHDMI 2.1, DisplayPort 1.4a
FeaturesNVIDIA DLSS, NVIDIA Reflex, Resizable BAR, NVIDIA Broadcast, NVIDIA Ansel, NVIDIA ShadowPlay®, NVIDIA G-SYNC®, Advanced Optimus, Nvidia Max-Q, Dynamic Boost, VR-ready, Nvidia Omniverse
Notebook Sizelarge
Date of Announcement04.01.2022

Benchmarks

Performance Rating - 3DMark 11 + Fire Strike + Time Spy
41.5 pt (45%)
3DMark - 3DMark Time Spy Score
min: 7369     avg: 10406     median: 10530 (55%)     max: 11959 Points
3DMark - 3DMark Time Spy Graphics
min: 9452     avg: 10409     median: 10221 (47%)     max: 11739 Points
3DMark - 3DMark Cloud Gate Score
min: 19528     avg: 51532     median: 58095 (71%)     max: 65396 Points
3DMark - 3DMark Cloud Gate Graphics
min: 33126     avg: 124349     median: 140844 (57%)     max: 168092 Points
3DMark - 3DMark Fire Strike Standard Score
min: 13606     avg: 23312     median: 23918 (60%)     max: 26257 Points
3DMark - 3DMark Fire Strike Standard Graphics
min: 22612     avg: 27116     median: 27275 (51%)     max: 29729 Points
3DMark 11 - 3DM11 Performance Score
min: 14168     avg: 28011     median: 30768 (62%)     max: 34126 Points
3DMark 11 - 3DM11 Performance GPU
min: 13799     avg: 33791     median: 35097 (51%)     max: 40880 Points
3DMark Vantage
3DM Vant. Perf. total +
3DM Vant. Perf. GPU no PhysX +
3DMark 06 3DMark 06 - Standard 1280x1024 +
Cinebench R10 Cinebench R10 Shading (32bit) +
Cinebench R11.5 Cinebench R11.5 OpenGL 64 Bit +
Cinebench R15
Cinebench R15 OpenGL 64 Bit +
Cinebench R15 - Cinebench R15 OpenGL 64 Bit
min: 98.7     avg: 195.8     median: 207 (56%)     max: 223 fps
Cinebench R15 OpenGL Ref. Match 64 Bit +
Cinebench R15 - Cinebench R15 OpenGL Ref. Match 64 Bit
99.6 % (100%)
Power Consumption - Furmark Stress Test Power Consumption - external Monitor *
min: 133     avg: 174.9     median: 181.7 (28%)     max: 211 Watt
0153045607590105120135150165180195210Tooltip
Power Consumption - The Witcher 3 Power Consumption - external Monitor *
min: 115     avg: 167.8     median: 156.7 (22%)     max: 223 Watt
0153045607590105120135150165180195210225Tooltip
Power Consumption - Witcher 3 Power Consumption *
min: 149     avg: 186.6     median: 189.6 (42%)     max: 227 Watt