Sort by
    Sort by:
    Best NVIDIA graphics card overall

    Compare

    nvidia-geforce-rtx-4090
    8.7
    Overall score
    8.7
    Technical Score
    9.0
    User score
    6.4
    Ratio quality/price

    Why buy?

    • Many compute units (128)
    • Many ray tracing cores (128)
    • Large VRAM capacity (24 GB)
    • High memory bandwidth (1008 GB/s)
    • High texture rate (1290 GTexel/s)
    • Many TMUs (512)
    • Wide memory bus (384 bit)
    • Many ROPs (176)

    Why avoid?

    • High TDP (450 W)
    • Slow VRAM clock (21 MHz)
    • High PSU requirement (850 W)
    • High board power limit (600 W)
    • Limited fan cooling (2)
    • Wide slot width (3 slot/s)
    • Long card length (304 mm)
    • Tall card height (137 mm)
    double-arrow
    • Avg. price: ~£2,030
    • VRAM: 24 GB
    • Memory bus width: 384 bit
    • Thermal Design Power (TDP): 450 W

    The Nvidia GeForce RTX 4090 is a flagship enthusiast-grade graphics card built on the Ada Lovelace architecture and TSMC’s 4N process, featuring 16,384 CUDA cores, 128 third-genera...Read more

    Best NVIDIA graphics card for video editing

    Compare

    nvidia-rtx-pro-6000-blackwell-workstation-edition
    8.7
    Overall score
    8.9
    Technical Score
    7.0
    User score
    6.4
    Ratio quality/price

    Why buy?

    • Many compute units (188)
    • Large VRAM capacity (96 GB)
    • Many ray tracing cores (188)
    • High boost clock (2617 MHz)
    • High memory bandwidth (1792 GB/s)
    • High texture rate (1968 GTexel/s)
    • Many TMUs (752)
    • Wide memory bus (512 bit)

    Why avoid?

    • High TDP (600 W)
    • High PSU requirement (1000 W)
    • Low base clock (1590 MHz)
    • High board power limit (600 W)
    • Few HDMI ports (0)
    • No fan stop
    • Limited fan cooling (2)
    • Tall card height (137.2 mm)
    double-arrow
    • Avg. price: ~£8,790
    • VRAM: 96 GB
    • Memory bus width: 512 bit
    • Thermal Design Power (TDP): 600 W

    The NVIDIA RTX Pro 6000 Blackwell Workstation Edition is a high-performance professional GPU built on the 5nm Blackwell architecture, featuring 24,064 CUDA cores, 752 fifth-generat...Read more

    Best value for money

    Compare

    nvidia-geforce-rtx-5080
    8.5
    Overall score
    8.6
    Technical Score
    8.3
    User score
    7.7
    Ratio quality/price

    Why buy?

    • Many compute units (84)
    • High boost clock (2617 MHz)
    • Many ray tracing cores (84)
    • High memory bandwidth (960 GB/s)
    • High texture rate (879.3 GTexel/s)
    • Many TMUs (336)
    • High memory speed (30000 MHz)
    • Many ROPs (112)

    Why avoid?

    • High TDP (360 W)
    • High PSU requirement (850 W)
    • Limited fan cooling (2)
    • Long card length (304 mm)
    • Tall card height (137 mm)
    • High board power limit (360 W)
    • No dual BIOS
    • High peak power draw (360 W)
    double-arrow
    • Avg. price: ~£930
    • VRAM: 16 GB
    • Memory bus width: 256 bit
    • Thermal Design Power (TDP): 360 W

    The Nvidia GeForce RTX 5080, launched in January 2025, is a high-end graphics card based on the Blackwell architecture (GB203 chip) featuring 10,752 CUDA cores, 84 fourth-generatio...Read more

    Compare

    nvidia-rtx-pro-5000-blackwell
    8.5
    Overall score
    8.6
    Technical Score
    7.3
    User score
    6.3
    Ratio quality/price

    Why buy?

    • Large VRAM capacity (48 GB)
    • Many compute units (110)
    • High boost clock (2617 MHz)
    • Many ray tracing cores (110)
    • High memory bandwidth (1344 GB/s)
    • High texture rate (1151 GTexel/s)
    • Many TMUs (440)
    • Wide memory bus (384 bit)

    Why avoid?

    • Limited fan cooling (1)
    • Low base clock (1590 MHz)
    • High TDP (300 W)
    • Few HDMI ports (0)
    • No XeSS support
    • High board power limit (300 W)
    • Large GPU die (750 mm²)
    • No dual BIOS
    double-arrow
    • Avg. price: ~£4,570
    • VRAM: 48 GB
    • Memory bus width: 384 bit
    • Thermal Design Power (TDP): 300 W

    The NVIDIA RTX Pro 5000 Blackwell is a high-end professional workstation GPU built on the Blackwell architecture, featuring 14,080 CUDA cores, 440 fifth-generation Tensor Cores, an...Read more

    Compare

    nvidia-geforce-rtx-4080-super
    8.3
    Overall score
    8.6
    Technical Score
    6.2
    User score
    7.1
    Ratio quality/price

    Why buy?

    • Many compute units (80)
    • Many ray tracing cores (80)
    • High boost clock (2550 MHz)
    • High texture rate (816 GTexel/s)
    • Many TMUs (320)
    • High memory bandwidth (736.3 GB/s)
    • Many ROPs (112)
    • High base clock (2295 MHz)

    Why avoid?

    • High TDP (320 W)
    • Slow VRAM clock (1438 MHz)
    • Limited fan cooling (2)
    • High PSU requirement (750 W)
    • Wide slot width (3 slot/s)
    • Long card length (304 mm)
    • High board power limit (320 W)
    • Single-GPU only (0)
    double-arrow
    • Avg. price: ~£1,200
    • VRAM: 16 GB
    • Memory bus width: 256 bit
    • Thermal Design Power (TDP): 320 W

    The Nvidia GeForce RTX 4080 Super is a high-end enthusiast graphics card built on the 5nm Ada Lovelace architecture, featuring 10,240 CUDA cores, 80 Ray Tracing cores, and 320 Tens...Read more

    Compare

    nvidia-geforce-rtx-4080-16gb
    8.2
    Overall score
    8.3
    Technical Score
    7.4
    User score
    7.0
    Ratio quality/price

    Why buy?

    • Many compute units (76)
    • Many ray tracing cores (76)
    • High texture rate (763.04 GTexel/s)
    • Many TMUs (304)
    • Many ROPs (112)
    • High boost clock (2505 MHz)
    • High memory bandwidth (716.8 GB/s)
    • Large VRAM capacity (16 GB)

    Why avoid?

    • High TDP (320 W)
    • Slow VRAM clock (1400 MHz)
    • Limited fan cooling (2)
    • High PSU requirement (750 W)
    • Wide slot width (3 slot/s)
    • Long card length (304 mm)
    • Tall card height (137 mm)
    • High board power limit (320 W)
    double-arrow
    • Avg. price: ~£1,210
    • VRAM: 16 GB
    • Memory bus width: 256 bit
    • Thermal Design Power (TDP): 320 W

    The Nvidia GeForce RTX 4080 16GB is a high-end enthusiast graphics card built on the 5nm Ada Lovelace architecture, featuring 9,728 CUDA cores, 76 third-gen ray tracing cores, and ...Read more

    Compare

    nvidia-rtx-pro-6000-blackwell-max-q-workstation-edition
    8.2
    Overall score
    8.2
    Technical Score
    7.5
    User score
    6.0
    Ratio quality/price

    Why buy?

    • Many compute units (188)
    • Large VRAM capacity (96 GB)
    • Many ray tracing cores (188)
    • High memory bandwidth (1792 GB/s)
    • Many TMUs (752)
    • Wide memory bus (512 bit)
    • Many ROPs (192)
    • High pixel rate (439.3 GPixel/s)

    Why avoid?

    • Limited fan cooling (1)
    • Low base clock (1590 MHz)
    • High TDP (300 W)
    • No FSR support
    • Few HDMI ports (0)
    • No fan stop
    • High board power limit (300 W)
    • High load temperature (88 °C)
    double-arrow
    • Avg. price: ~£8,570
    • VRAM: 96 GB
    • Memory bus width: 512 bit
    • Thermal Design Power (TDP): 300 W

    The Nvidia RTX Pro 6000 Blackwell Max-Q Workstation Edition is an enthusiast-class professional GPU built on the 5nm GB202 Blackwell architecture, featuring 24,064 CUDA cores, 752 ...Read more

    Compare

    nvidia-rtx-pro-4500-blackwell
    8.0
    Overall score
    8.4
    Technical Score
    4.6
    User score
    5.9
    Ratio quality/price

    Why buy?

    • Large VRAM capacity (32 GB)
    • Many compute units (82)
    • Many ray tracing cores (82)
    • High texture rate (858.4 GTexel/s)
    • High memory bandwidth (896 GB/s)
    • Many TMUs (328)
    • Many ROPs (112)
    • High pixel rate (293.1 GPixel/s)

    Why avoid?

    • Limited fan cooling (1)
    • Low base clock (1635 MHz)
    • Few HDMI ports (0)
    • No XeSS support
    • No dual BIOS
    • No RGB lighting
    • Older FSR support (N/A)
    • No HDMI output
    double-arrow
    • Avg. price: ~£2,420
    • VRAM: 32 GB
    • Memory bus width: 256 bit
    • Thermal Design Power (TDP): 200 W

    The NVIDIA RTX PRO 4500 Blackwell is a high-end professional workstation graphics card built on the 5nm Blackwell architecture, featuring 10,496 CUDA cores, 328 5th Gen Tensor Core...Read more

    Compare

    nvidia-geforce-rtx-3090-ti
    8.0
    Overall score
    8.0
    Technical Score
    7.9
    User score
    6.2
    Ratio quality/price

    Why buy?

    • Many compute units (84)
    • Many ray tracing cores (84)
    • Large VRAM capacity (24 GB)
    • High memory bandwidth (1008 GB/s)
    • Many TMUs (336)
    • Wide memory bus (384 bit)
    • Many ROPs (112)
    • Many AI cores (336)

    Why avoid?

    • Low boost clock (1860 MHz)
    • High TDP (450 W)
    • Low base clock (1560 MHz)
    • High PSU requirement (850 W)
    • Slow VRAM clock (1313 MHz)
    • High board power limit (450 W)
    • Small L2 cache (6 MB)
    • Long card length (313 mm)
    double-arrow
    • Avg. price: ~£1,580
    • VRAM: 24 GB
    • Memory bus width: 384 bit
    • Thermal Design Power (TDP): 450 W

    The NVIDIA GeForce RTX 3090 Ti is an enthusiast-class flagship GPU built on the Ampere architecture, featuring a fully unlocked GA102 chip with 10,752 CUDA cores, 84 Ray Tracing co...Read more

    Compare

    nvidia-rtx-5000-ada-generation
    8.0
    Overall score
    8.7
    Technical Score
    1.7
    User score
    5.9
    Ratio quality/price

    Why buy?

    • Large VRAM capacity (32 GB)
    • Many compute units (100)
    • Many ray tracing cores (100)
    • High texture rate (1020 GTexel/s)
    • Many TMUs (400)
    • High boost clock (2550 MHz)
    • Many ROPs (176)
    • High pixel rate (448.8 GPixel/s)

    Why avoid?

    • Low base clock (1155 MHz)
    • Limited fan cooling (1)
    • Few HDMI ports (0)
    • No fan stop
    • High TDP (250 W)
    • High idle noise (40 dB)
    • High load noise (47.5 dB)
    • No dual BIOS
    double-arrow
    • Avg. price: ~£3,760
    • VRAM: 32 GB
    • Memory bus width: 256 bit
    • Thermal Design Power (TDP): 250 W

    The NVIDIA RTX 5000 Ada Generation is a high-end professional workstation GPU built on the 4nm Ada Lovelace architecture, featuring 12,800 CUDA cores, 400 4th-generation Tensor Cor...Read more

    Compare

    nvidia-rtx-a6000
    8.0
    Overall score
    8.1
    Technical Score
    7.0
    User score
    5.9
    Ratio quality/price

    Why buy?

    • Large VRAM capacity (48 GB)
    • Many compute units (84)
    • Many ray tracing cores (84)
    • Many TMUs (336)
    • Wide memory bus (384 bit)
    • High memory bandwidth (768 GB/s)
    • Many ROPs (112)
    • Many AI cores (336)

    Why avoid?

    • Low boost clock (1800 MHz)
    • Limited fan cooling (1)
    • Low base clock (1410 MHz)
    • Low memory speed (16000 MHz)
    • High TDP (300 W)
    • Few HDMI ports (0)
    • No fan stop (No)
    • Small L2 cache (6 MB)
    double-arrow
    • Avg. price: ~£4,000
    • VRAM: 48 GB
    • Memory bus width: 384 bit
    • Thermal Design Power (TDP): 300 W

    The Nvidia RTX A6000 is a high-end professional workstation graphics card built on the 8nm Ampere architecture, featuring 10,752 CUDA cores, 336 third-generation Tensor cores, and ...Read more

    Compare

    nvidia-geforce-rtx-3090
    7.9
    Overall score
    7.7
    Technical Score
    9.1
    User score
    6.0
    Ratio quality/price

    Why buy?

    • Many compute units (82)
    • Large VRAM capacity (24 GB)
    • Many ray tracing cores (82)
    • High memory bandwidth (936.2 GB/s)
    • Wide memory bus (384 bit)
    • Many TMUs (328)
    • Many ROPs (112)
    • Many AI cores (328)

    Why avoid?

    • Low boost clock (1695 MHz)
    • Low base clock (1395 MHz)
    • High TDP (350 W)
    • Slow VRAM clock (1219 MHz)
    • Small L2 cache (6 MB)
    • Long card length (313 mm)
    • Limited fan cooling (2)
    • High PSU requirement (750 W)
    double-arrow
    • Avg. price: ~£1,710
    • VRAM: 24 GB
    • Memory bus width: 384 bit
    • Thermal Design Power (TDP): 350 W

    The Nvidia GeForce RTX 3090 is a powerhouse built on the Ampere architecture, featuring 10,496 CUDA cores and a massive 24GB of GDDR6X VRAM on a 384-bit bus, delivering approximate...Read more

    Compare

    nvidia-geforce-rtx-3080-ti
    7.7
    Overall score
    7.6
    Technical Score
    9.2
    User score
    7.2
    Ratio quality/price

    Why buy?

    • Many compute units (80)
    • Many ray tracing cores (80)
    • High memory bandwidth (912.4 GB/s)
    • Wide memory bus (384 bit)
    • Many TMUs (320)
    • Many ROPs (112)
    • Many AI cores (320)
    • High texture rate (532.8 GTexel/s)

    Why avoid?

    • Low boost clock (1665 MHz)
    • Low base clock (1365 MHz)
    • High TDP (350 W)
    • Slow VRAM clock (1188 MHz)
    • Small L2 cache (6 MB)
    • Limited fan cooling (2)
    • High PSU requirement (750 W)
    • Old GPU release (2021)
    double-arrow
    • Avg. price: ~£870
    • VRAM: 12 GB
    • Memory bus width: 384 bit
    • Thermal Design Power (TDP): 350 W

    The NVIDIA GeForce RTX 3080 Ti is an enthusiast-class graphics card featuring 10,240 CUDA cores, 80 Ray Tracing cores, and 320 Tensor cores on the 8nm GA102 architecture. It is equ...Read more

    Compare

    nvidia-geforce-rtx-4070-ti
    7.4
    Overall score
    7.3
    Technical Score
    8.5
    User score
    7.1
    Ratio quality/price

    Why buy?

    • High boost clock (2610 MHz)
    • High base clock (2310 MHz)
    • Many compute units (60)
    • High texture rate (626.4 GTexel/s)
    • Many ray tracing cores (60)
    • Many TMUs (240)
    • High pixel rate (208.8 GPixel/s)
    • Advanced process node (5 nm)

    Why avoid?

    • Narrow memory bus (192 bit)
    • Slow VRAM clock (1313 MHz)
    • High TDP (285 W)
    • High PSU requirement (700 W)
    • Low HDMI bandwidth (12 Gbps)
    • High peak power draw (285 W)
    • Weak FP64 ratio (1:64)
    • Few HDMI ports (1)
    double-arrow
    • Avg. price: ~£800
    • VRAM: 12 GB
    • Memory bus width: 192 bit
    • Thermal Design Power (TDP): 285 W

    The NVIDIA GeForce RTX 4070 Ti is a high-end graphics card built on the 5nm Ada Lovelace architecture, featuring 7,680 CUDA cores, 60 third-generation Ray Tracing cores, and 240 fo...Read more

    Compare

    nvidia-geforce-rtx-5070
    7.2
    Overall score
    7.2
    Technical Score
    7.1
    User score
    7.4
    Ratio quality/price

    Why buy?

    • High boost clock (2512 MHz)
    • High memory speed (28000 MHz)
    • High memory bandwidth (672 GB/s)
    • Many compute units (48)
    • High base clock (2160 MHz)
    • High texture rate (481.9 GTexel/s)
    • Advanced process node (5 nm)
    • Modern PCIe support (5.0)

    Why avoid?

    • Few ray tracing cores (48)
    • Narrow memory bus (192 bit)
    • Limited fan cooling (2)
    • High TDP (250 W)
    • Small Infinity Cache (48 MB)
    • No RGB lighting
    • Few NVENC sessions
    • High load noise (37.7 dB)
    double-arrow
    • Avg. price: ~£520
    • VRAM: 12 GB
    • Memory bus width: 192 bit
    • Thermal Design Power (TDP): 250 W

    The Nvidia GeForce RTX 5070, launched in March 2025, is built on the 5nm Blackwell architecture and features the GB205-300 GPU with 6,144 CUDA cores, 48 Ray Tracing cores, and a bo...Read more

    Compare

    nvidia-geforce-rtx-4070-super
    7.0
    Overall score
    7.1
    Technical Score
    6.7
    User score
    7.1
    Ratio quality/price

    Why buy?

    • Many compute units (56)
    • High boost clock (2475 MHz)
    • High texture rate (554.4 GTexel/s)
    • Many ray tracing cores (56)
    • Many TMUs (224)
    • Advanced process node (5 nm)
    • High base clock (1980 MHz)
    • High memory bandwidth (504.2 GB/s)

    Why avoid?

    • Narrow memory bus (192 bit)
    • Slow VRAM clock (1313 MHz)
    • Limited fan cooling (2)
    • Older encoder generation
    • No RGB lighting
    • No hybrid cooling (N/A)
    • Single-GPU only (0)
    • Low HDMI bandwidth (12 Gbps)
    double-arrow
    • Avg. price: ~£620
    • VRAM: 12 GB
    • Memory bus width: 192 bit
    • Thermal Design Power (TDP): 220 W

    The Nvidia GeForce RTX 4070 Super is a high-performance mid-to-high-end graphics card built on the 5nm Ada Lovelace architecture, featuring 7,168 CUDA cores, 224 Tensor cores, 56 R...Read more

    Compare

    nvidia-geforce-rtx-4070
    7.0
    Overall score
    6.8
    Technical Score
    8.3
    User score
    7.1
    Ratio quality/price

    Why buy?

    • High boost clock (2475 MHz)
    • Many compute units (46)
    • Advanced process node (5 nm)
    • High memory bandwidth (504.2 GB/s)
    • High memory speed (21000 MHz)
    • High texture rate (455.4 GTexel/s)
    • Supports CUDA
    • Large VRAM capacity (12 GB)

    Why avoid?

    • Few ray tracing cores (46)
    • Few ROPs (64)
    • Narrow memory bus (192 bit)
    • Low pixel rate (158.4 GPixel/s)
    • Slow VRAM clock (1313 MHz)
    • Few AI cores (184)
    • Limited fan cooling (2)
    • No dual BIOS
    double-arrow
    • Avg. price: ~£560
    • VRAM: 12 GB
    • Memory bus width: 192 bit
    • Thermal Design Power (TDP): 200 W

    The Nvidia GeForce RTX 4070 is a high-performance graphics card built on the 5nm Ada Lovelace architecture, specifically designed to excel at 1440p gaming. Key technical specificat...Read more

    Compare

    nvidia-titan-rtx
    7.0
    Overall score
    6.8
    Technical Score
    8.0
    User score
    6.4
    Ratio quality/price

    Why buy?

    • Large VRAM capacity (24 GB)
    • Many compute units (72)
    • Wide memory bus (384 bit)
    • Many ray tracing cores (72)
    • Many AI cores (576)
    • Many TMUs (288)
    • High memory bandwidth (672 GB/s)
    • Many ROPs (96)

    Why avoid?

    • Low boost clock (1770 MHz)
    • Low base clock (1350 MHz)
    • Low memory speed (14000 MHz)
    • Older process node (12 nm)
    • High TDP (280 W)
    • No fan stop (No)
    • Small L2 cache (6 MB)
    • Limited fan cooling (2)
    double-arrow
    • Avg. price: ~£1,050
    • VRAM: 24 GB
    • Memory bus width: 384 bit
    • Thermal Design Power (TDP): 280 W

    The NVIDIA TITAN RTX, launched in December 2018 based on the 12nm Turing architecture, is a powerhouse featuring 4,608 CUDA cores, 576 Tensor cores, and 72 RT cores. It is equipped...Read more

    Compare

    nvidia-rtx-pro-4000-blackwell
    6.9
    Overall score
    6.9
    Technical Score
    6.9
    User score
    5.4
    Ratio quality/price

    Why buy?

    • Large VRAM capacity (24 GB)
    • Many compute units (70)
    • Many ray tracing cores (70)
    • Many TMUs (280)
    • High memory speed (28000 MHz)
    • High memory bandwidth (672 GB/s)
    • High pixel rate (251.2 GPixel/s)
    • Many ROPs (96)

    Why avoid?

    • Low base clock (1230 MHz)
    • Low boost clock (2055 MHz)
    • Limited fan cooling (1)
    • Narrow memory bus (192 bit)
    • Few HDMI ports (0)
    • No XeSS support
    • No dual BIOS
    • No RGB lighting
    double-arrow
    • Avg. price: ~£1,680
    • VRAM: 24 GB
    • Memory bus width: 192 bit
    • Thermal Design Power (TDP): 140 W

    The NVIDIA RTX PRO 4000 Blackwell is a professional-grade workstation GPU featuring 8,960 CUDA cores, 280 fifth-generation Tensor Cores, and 70 fourth-generation RT Cores built on ...Read more

    Compare

    nvidia-geforce-rtx-2080-ti-founders-edition
    6.9
    Overall score
    6.7
    Technical Score
    9.1
    User score
    6.5
    Ratio quality/price

    Why buy?

    • Many compute units (68)
    • Wide memory bus (352 bit)
    • Many AI cores (544)
    • Many ray tracing cores (68)
    • Many TMUs (272)
    • High memory bandwidth (616 GB/s)
    • Many ROPs (88)
    • Supports CUDA

    Why avoid?

    • Low boost clock (1635 MHz)
    • Low base clock (1350 MHz)
    • Small VRAM capacity (11 GB)
    • Low memory speed (14000 MHz)
    • Low FP32 performance (14.23 TFLOPS)
    • Low pixel rate (143.9 GPixel/s)
    • Older process node (12 nm)
    • Small L2 cache (5.5 MB)
    double-arrow
    • Avg. price: ~£950
    • VRAM: 11 GB
    • Memory bus width: 352 bit
    • Thermal Design Power (TDP): 260 W

    The Nvidia GeForce RTX 2080 Ti Founders Edition is a high-performance graphics card built on the Turing TU102 architecture, featuring 4,352 CUDA cores, 11GB of GDDR6 memory on a 35...Read more

    As an Amazon Associate, we earn from qualifying purchases.
    • 1
    • 2
    • ...
    • 10
    • Next page