With the Nvidia 4000 series set to release later this year, gamers are anxiously waiting to see what the next-gen GPUs have to offer. Before we start upgrading our PCs though, let’s take a step back and look at what the current generation graphics cards have to offer.
NVIDIA RTX 3090 vs AMD RX 6900 XT
RTX 3090
Clock Speeds
| Base Clock | 1395 MHz |
| Boost Clock | 1695 MHz |
| Memory Clock | 1219 MHz |
Render Config
| Shading Units | 10496 |
| TMUs | 328 |
| ROPs | 112 |
| SM Count | 82 |
| Tensor Cores | 328 |
| RT Cores | 82 |
| L1 Cache | 128 KB (per SM) |
| L2 Cache | 6 MB |
Memory
| Memory Size | 24 GB |
| Memory Type | GDDR6X |
| Memory Bus | 384 bit |
| Bandwidth | 936.2 GB/s |
Graphics
| DirectX | 12 Ultimate |
| OpenGL | 4.6 |
| OpenCL | 3.0 |
| Vulkan | 1.3 |
| CUDA | 8.6 |
| Shader Model | 6.6 |
RX 6900 XT
Clock Speeds
| Base Clock | 1825 MHz |
| Game Clock | 2015 MHz |
| Boost Clock | 2250 MHz |
| Memory Clock | 2000 MHz |
Render Config
| Shading Units | 5120 |
| TMUs | 320 |
| ROPs | 128 |
| Compute Units | 80 |
| RT Cores | 80 |
| L0 Cache | 32 KB per WGP |
| L1 Cache | 138 KB per Array |
| L2 Cache | 4 MB |
| L3 Cache | 128 MB |
Memory
| Memory Size | 16 GB |
| Memory Type | GDDR6 |
| Memory Bus | 256 bit |
| Bandwidth | 512.0 GB/s |
Graphics
| DirectX | 12 Ultimate |
| OpenGL | 4.6 |
| OpenCL | 2.1 |
| Vulkan | 1.3 |
| Shader Model | 6.5 |
While both graphics cards boast high performance, it’s no surprise that the RTX 3090 comes out on top – it’s why the majority of gamers prefer Nvidia over AMD. Compared to the RX 6900, the RTX 3090 has better memory, made faster with its wider memory bus. The RTX 3090 excels when it comes to ray tracing but shows similar performance to the RX 6900 when ray tracing isn’t used. And though the RX 6900 has less than half the stream processors as the RTX 3090 has CUDA cores, the RX 6900 doesn’t necessarily work half as fast. Given all the specs, the RX 6900 is more equivalent to the RTX 3080.
NVIDIA RTX 3080 vs AMD RX 6800 XT
RTX 3080
Clock Speeds
| Base Clock | 1440 MHz |
| Boost Clock | 1710 MHz |
| Memory Clock | 1188 MHz |
Render Config
| Shading Units | 8704 |
| TMUs | 272 |
| ROPs | 96 |
| SM Count | 68 |
| Tensor Cores | 272 |
| RT Cores | 68 |
| L1 Cache | 128 KB (per SM) |
| L2 Cache | 5 MB |
Memory
| Memory Size | 10 GB |
| Memory Type | GDDR6X |
| Memory Bus | 320 bit |
| Bandwidth | 760.3 GB/s |
Graphics
| DirectX | 12 Ultimate |
| OpenGL | 4.6 |
| OpenCL | 3.0 |
| Vulkan | 1.3 |
| CUDA | 8.6 |
| Shader Model | 6.6 |
RX 6800 XT
Clock Speeds
| Base Clock | 1700 MHz |
| Game Clock | 18 MHz |
| Boost Clock | 2105 MHz |
| Memory Clock | 2000 MHz |
Render Config
| Shading Units | 3840 |
| TMUs | 240 |
| ROPs | 96 |
| Compute Units | 60 |
| RT Cores | 60 |
| L0 Cache | 32 KB per WGP |
| L1 Cache | 128 KB per Array |
| L2 Cache | 4 MB |
| L3 Cache | 128 MB |
Memory
| Memory Size | 16 GB |
| Memory Type | GDDR6 |
| Memory Bus | 256 bit |
| Bandwidth | 512.0 GB/s |
Graphics
| DirectX | 12 Ultimate |
| OpenGL | 4.6 |
| OpenCL | 2.1 |
| Vulkan | 1.3 |
| Shader Model | 6.6 |
Unlike the RTX 3090 and RX 6900, the RTX 3080 and RX 6800 are more evenly matched. Despite the RX 6800’s greater memory, the RTX 3080’s upgraded GDDR6X plus its wider memory width gives it a greater standard memory bandwidth. The RX 6800 however, uses 10% less memory and generally there’s very little FPS difference between them. Like all Nvidia cards, the RTX 3080 does better with ray tracing, but both cards show similar performance when ray tracing isn’t in use.
NVIDIA RTX 3070 vs AMD RX 6700 XT
RTX 3070
Clock Speeds
| Base Clock | 1500 MHz |
| Boost Clock | 1725 MHz |
| Memory Clock | 1750 MHz |
Render Config
| Shading Units | 5888 |
| TMUs | 184 |
| ROPs | 96 |
| SM Count | 46 |
| Tensor Cores | 184 |
| RT Cores | 46 |
| L1 Cache | 128 KB (per SM) |
| L2 Cache | 4 MB |
Memory
| Memory Size | 8 GB |
| Memory Type | GDDR6 |
| Memory Bus | 256 bit |
| Bandwidth | 448.0 GB/s |
Graphics
| DirectX | 12 Ultimate |
| OpenGL | 4.6 |
| OpenCL | 3.0 |
| Vulkan | 1.3 |
| CUDA | 8.6 |
| Shader Model | 6.6 |
RX 6700 XT
Clock Speeds
| Base Clock | 2321 MHz |
| Game Clock | 2424 MHz |
| Boost Clock | 2581 MHz |
| Memory Clock | 2000 MHz |
Render Config
| Shading Units | 2560 |
| TMUs | 160 |
| ROPs | 64 |
| Compute Units | 40 |
| RT Cores | 40 |
| L0 Cache | 32 KB per WGP |
| L1 Cache | 128 KB per Array |
| L2 Cache | 3 MB |
| L3 Cache | 96 MB |
Memory
| Memory Size | 12 GB |
| Memory Type | GDDR6 |
| Memory Bus | 192 bit |
| Bandwidth | 384.0 GB/s |
Graphics
| DirectX | 12 Ultimate |
| OpenGL | 4.6 |
| OpenCL | 2.1 |
| Vulkan | 1.3 |
| Shader Model | 6.5 |
On average, the RTX 3070 is about 20% faster than the RX 6700. Both are tied at 1080p gaming but the RTX 3070 begins to pick up performance, boasting a 3% lead on 1440p gaming and 10% on 4k gaming. Although the RX 6700 is good for 1440p gaming, it can’t handle ray tracing at 1440p, making the RTX 3070 a clear winner between the two.
NVIDIA RTX 3060 vs AMD RX 6600 XT
RTX 3060
Clock Speeds
| Base Clock | 1320 MHz |
| Boost Clock | 1777 MHz |
| Memory Clock | 1875 MHz |
Render Config
| Shading Units | 3584 |
| TMUs | 112 |
| ROPs | 48 |
| SM Count | 28 |
| Tensor Cores | 112 |
| RT Cores | 28 |
| L1 Cache | 128 KB (per SM) |
| L2 Cache | 3 MB |
Memory
| Memory Size | 12 GB |
| Memory Type | GDDR6 |
| Memory Bus | 192 bit |
| Bandwidth | 360.0 GB/s |
Graphics
| DirectX | 12 Ultimate |
| OpenGL | 4.6 |
| OpenCL | 3.0 |
| Vulkan | 1.3 |
| CUDA | 8.6 |
| Shader Model | 6.6 |
RX 6600 XT
Clock Speeds
| Base Clock | 1626 MHz |
| Game Clock | 2044 MHz |
| Boost Clock | 2491 MHz |
| Memory Clock | 1750 MHz |
Render Config
| Shading Units | 1792 |
| TMUs | 112 |
| ROPs | 64 |
| Compute Units | 28 |
| RT Cores | 28 |
| L0 Cache | 32 KB per WGP |
| L1 Cache | 128 KB per Array |
| L2 Cache | 2 MB |
| L3 Cache | 32 MB |
Memory
| Memory Size | 8 GB |
| Memory Type | GDDR6 |
| Memory Bus | 128 bit |
| Bandwidth | 224.0 GB/s |
Graphics
| DirectX | 12 Ultimate |
| OpenGL | 4.6 |
| OpenCL | 2.1 |
| Vulkan | 1.3 |
| Shader Model | 6.5 |
The RTX 3060 and RX 6600 are very close in performance – the RX 6600 being about 3% faster in 1080p gaming and the RTX being 2% faster at 1440p gaming. The RTX 3060 offers a 50% larger frame buffer which will allow it to handle future games and textures better, making it the better option of the two for prolonged use.
NVIDIA RTX 3050 vs AMD RX 6050 XT
RTX 3050
Clock Speeds
| Base Clock | 1552 MHz |
| Boost Clock | 1777 MHz |
| Memory Clock | 1750 MHz |
Render Config
| Shading Units | 2560 |
| TMUs | 80 |
| ROPs | 32 |
| SM Count | 20 |
| Tensor Cores | 80 |
| RT Cores | 20 |
| L1 Cache | 128 KB (per SM) |
| L2 Cache | 2 MB |
Memory
| Memory Size | 8 GB |
| Memory Type | GDDR6 |
| Memory Bus | 128 bit |
| Bandwidth | 224.0 GB/s |
Graphics
| DirectX | 12 Ultimate |
| OpenGL | 4.6 |
| OpenCL | 3.0 |
| Vulkan | 1.3 |
| CUDA | 8.6 |
| Shader Model | 6.6 |
RX 6050 XT
Clock Speeds
| Base Clock | 2310 MHz |
| Game Clock | 2610 MHz |
| Boost Clock | 2815 MHz |
| Memory Clock | 2248 MHz |
Render Config
| Shading Units | 1024 |
| TMUs | 64 |
| ROPs | 32 |
| Compute Units | 16 |
| RT Cores | 16 |
| L0 Cache | 32 KB per WGP |
| L1 Cache | 128 KB per Array |
| L2 Cache | 1024 MB |
| L3 Cache | 16 MB |
Memory
| Memory Size | 4 GB |
| Memory Type | GDDR6 |
| Memory Bus | 64 bit |
| Bandwidth | 143.9 GB/s |
Graphics
| DirectX | 12 Ultimate |
| OpenGL | 4.6 |
| OpenCL | 2.2 |
| Vulkan | 1.3 |
| Shader Model | 6.6 |
Looking at the specs alone, I’m sure it’s no surprise when I say that the RTX 3050 crushes the RX 6500 when it comes to 1080p gaming. The RX 6500 was originally created for laptops and rather than proving it matches the RTX 3050, it struggles to prove it belongs on the GPU market. And while it does technically have ray tracing, good luck getting any games to run smoothly with it. If you’re looking for the cheapest GPU for 1080p gaming, your best bet is the RTX 3050.
Recent Comments