NVIDIA GeForce GTX 1070 vs NVIDIA GeForce RTX 3090 Ti

差异规格基准褒奖

差异

需要考虑的原因
NVIDIA GeForce GTX 1070

报告一个错误

更高的时钟速度

左右 28% 更好的时钟速度

2 GHz 1.56 GHz

需要考虑的原因
NVIDIA GeForce RTX 3090 Ti

报告一个错误

更多内存

200更多内存百分比

24 GB 8 GB

所有排名中的位置

共同立场 NVIDIA GeForce RTX 3090 Ti 在流行的基准测试中的GPU，用于与其他模型进行比较.

Passmark Performance Test GPU

1 place

规格

技术数据

NVIDIA GeForce GTX 1070

NVIDIA GeForce RTX 3090 Ti

一般信息

显卡的基本技术数据比较 NVIDIA GeForce GTX 1070 和 NVIDIA GeForce RTX 3090 Ti, 芯片，信息处理单元.

建筑学

Pascal Ampere
Execution Units

1920 -
GPU芯片

GP104 GA102-350-A1
L1 Cache

48 KB 128 KB
L2 Cache

2 MB 6 MB
渲染输出单位

64 112
纹理单位

120 336
基于

- GA102
CUDA Cores

- 10752
着色器

- 10752

内存设置

Video memory capacity, type, bus width and bandwidth describe different properties. These values alone do not determine graphics performance.

Memory Bandwidth

256.3 GB/s 1,008 GB/s
Memory Bus

256 bit 384 bit
Memory Clock

2000 MHz -
Memory Data Rate

8 Gbps -
内存大小

8 GB 24 GB
存储器类型

GDDR5 GDDR6X
Memory Speed (legacy frequency label)

- 1300 MHz

时钟频率

Published GPU clock frequencies describe different operating modes. Clock speed alone does not determine graphics performance.

基准频率

1506 MHz 1560 MHz
最大频率

1683 MHz 1860 MHz
超频

- 没有

能源消耗

Graphics card power limits and recommended system power supply describe different specifications.

System Power Supply

450 W (PSU recommendation) 850 W (System power requirement)
TDP

150 W 450 W
Graphics Card Power

- 450 W
Maximum GPU Temperature

- 92 °C

接口

连通性和连接.

DisplayPort Ports

3 × DisplayPort 1.4a 3 × DisplayPort 1.4a
DVI端口

1 × DVI -
HDMI端口

1 × HDMI 2.0 1 × HDMI 2.1
最大。显示

5 4
PCIe Interface

PCIe 3.0 x16 PCIe 4.0 x16
Power Connectors

1x 8-pin 3x PCIe 8-pin cables (adapter in box) OR 450W or greater PCIe Gen 5 cable
HDCP Version

- 2.3
Other Display Outputs (unparsed)

- HDMI

技术数据

在电脑游戏中被充分使用的技术数据.

CUDA

6.1 -
DirectX

12.1 12_2
最大分辨率

3840 × 2160 7680 × 4320
OpenCL

3.0 3.0
OpenGL

4.6 4.6
Shader Model

6.7 (6.4) 6.6
Vulkan

1.3 1.3

尺寸

所比较的设备在尺寸、重量和插槽上的差异

高度

40 mm -
长度

267 mm -
Slot Size

Dual-slot -
Width

112 mm -

更多信息

GPU的接口和发布日期的比较 NVIDIA GeForce RTX 3090 Ti 和 NVIDIA GeForce GTX 1070.

Process Technology

16 nm 8 nm
发布日期

Q3/2016 -
Manufacturer specification scope

- NVIDIA Founders Edition/reference design. Clock specifications apply to gaming use. AIB shipping specifications and complete-system power needs may differ.

Theoretical Performance

FP16 (theoretical)

101.0 GFLOPS (1:64) 40 TFLOPS
FP32 (theoretical)

6.463 TFLOPS 40 TFLOPS
FP64 (theoretical)

202.0 GFLOPS (1:32) 0.625 TFLOPS
Pixel Rate

107.7 GPixel/s 208.3 GPixel/s
Texture Rate

202.0 GTexel/s 625.0 GTexel/s

基准

性能测试 GPUs

Shared tests with results for both models: 14

Blender Cycles

Each row uses one Blender version, operating system and compute backend.

Test and settings	NVIDIA GeForce GTX 1070	NVIDIA GeForce RTX 3090 Ti	Result difference¹
Blender 5.2.0 / OPTIX / Windows	481.6157 samples/min Submissions: 20	5250.819731 samples/min Submissions: 11	+990.3 %
Blender 5.1.1 / OPTIX / Windows	444.806088 samples/min Submissions: 11	5459.157697 samples/min Submissions: 5	+1127.3 %
Blender 4.5.0 / OPTIX / Linux	525.482173 samples/min Submissions: 16	5309.43807 samples/min Submissions: 24	+910.4 %
Blender 4.5.0 / OPTIX / Windows	520.511331 samples/min Submissions: 62	5585.151795 samples/min Submissions: 115	+973.0 %
Blender 4.4.0 / OPTIX / Linux	544.297956 samples/min Submissions: 15	4979.851473 samples/min Submissions: 8	+814.9 %
Blender 4.4.0 / OPTIX / Windows	522.08629 samples/min Submissions: 62	5700.575741 samples/min Submissions: 51	+991.9 %
Blender 4.3.0 / OPTIX / Linux	531.884683 samples/min Submissions: 7	5840.23445 samples/min Submissions: 8	+998.0 %
Blender 4.3.0 / OPTIX / Windows	521.037577 samples/min Submissions: 56	5490.159484 samples/min Submissions: 35	+953.7 %
Blender 4.2.0 / OPTIX / Linux	551.555412 samples/min Submissions: 11	5972.585335 samples/min Submissions: 15	+982.9 %
Blender 4.2.0 / OPTIX / Windows	532.275463 samples/min Submissions: 81	5673.136912 samples/min Submissions: 51	+965.8 %
Blender 4.1.0 / OPTIX / Linux	555.845146 samples/min Submissions: 9	6034.022292 samples/min Submissions: 7	+985.6 %
Blender 4.1.0 / OPTIX / Windows	543.479512 samples/min Submissions: 41	5769.869986 samples/min Submissions: 32	+961.7 %

More tests: 2

Test and settings	NVIDIA GeForce GTX 1070	NVIDIA GeForce RTX 3090 Ti	Result difference¹
Blender 4.0.0 / OPTIX / Linux	544.324065 samples/min Submissions: 14	6168.026652 samples/min Submissions: 16	+1033.2 %
Blender 4.0.0 / OPTIX / Windows	538.994218 samples/min Submissions: 115	5124.064923 samples/min Submissions: 111	+850.7 %

¹ Difference of the second result relative to the first in that row. No percentage is calculated for different upscaling algorithms. This compares published results, not a universal performance index.

Individual model tests

These results have no shared test of both models with the same version and settings. They are shown separately without a percentage difference.