Moore’s Law is Dead has offered us the first-ever renders of the Intel’s ARC Alchemist reference graphics cards and the images show us what Intel’s reference design will look like along with some details regarding the launch of the desktop ARC lineup.
We have seen Intel’s ARC Alchemist graphics card reference design for high-end variants pictured on several occasions before, quite recently we witnessed the prototype leak out by MLID back in April 2021 & Intel itself has confirmed the design in its ARC marketing campaign in August.
However, the render provided at Moore’s Law is Dead is very detailed and sheds light on almost every aspect of the reference graphics card. It gives us a detailed look at what might be Intel’s first dedicated design for its first desktop discrete gaming GPU.
According to the leak, the card is based on a high-end ARC Alchemist GPU, which will probably be the DG2-512 (Xe-HPG) chip that is expected to power the upper echelon of Alchemist graphics cards.
Reportedly the card sports a dual-slot and dual-fan design and each of this fan incorporate a 9-blade fan design and we can also see RGB embedded around the fans which gives a nice touch in blue.
Coming to the cooler shroud we can see a circular patterning around the edges which gives a unique touch to the card. The company has also provided the ‘Intel’ logo on the side accent plate which is reportedly made out of acrylic and also embeds RGB LEDs.
The card receives power through an 8+6 pin connector configuration and the display outputs include a single HDMI and triple DP ports. However, we might have to wait for a PCB render for later on.
Intel ARC Alchemist vs NVIDIA GA104 & AMD Navi 22 GPUs
| GPU Name | Alchemist DG-512 | NVIDIA GA104 | AMD Navi 22 |
| Architecture | Xe-HPG | Ampere | RDNA 2 |
| Process Node | TSMC 6nm | Samsung 8nm | TSMC 7nm |
| Flagship Product | ARC (TBA) | GeForce RTX 3070 Ti | Radeon RX 6700 XT |
| Raster Engine | 8 | 6 | 2 |
| FP32 Cores | 32 Xe Cores | 48 SM Units | 40 Compute Units |
| FP32 Units | 4096 | 6144 | 2560 |
| FP32 Compute | ~16 TFLOPs | 21.7 TFLOPs | 12.4 TFLOPs |
| TMUs | 256 | 192 | 160 |
| ROPs | 128 | 96 | 64 |
| RT Cores | 32 RT Units | 48 RT Cores (V2) | 40 RA Units |
| Tensor Cores | 512 XMX Cores | 192 Tensor Cores (V3) | N/A |
| Tensor Compute | ~131 TFLOPs FP16 ~262 TOPs INT8 | 87 TFLOPs FP16 174 TOPs INT8 | 25 TFLOPs FP16 50 TOPs INT8 |
| L2 Cache | TBA | 4 MB | 3 MB |
| Additional Cache | 16 MB Smart Cache? | N/A | 96 MB Infinity Cache |
| Memory Bus | 256-bit | 256-bit | 192-bit |
| Memory Capacity | 16 GB GDDR6 | 8 GB GDDR6X | 16 GB GDDR6 |
| Launch | Q1 2022 | Q2 2021 | Q1 2021 |
