Intel’s flagship ARC Alchemist GPU-powered graphics card has been noticed throughout the Geekbench database as soon as once more. This time, the chip appears to be an early pattern because it runs at decrease clocks than the utmost frequency which is listed throughout the similar benchmark.
Intel’s Flagship ARC Alchemist GPU Powered Graphics Card Noticed in Geekbench Benchmark
The Intel ARC Alchemist graphics card can simply be recognized with its 512 Execution models (4096 ALUs). The precise variant is talked about to function a most enhance clock of two.1 GHz however might be seen working at a a lot decrease 1.33 GHz throughout the JSON Information listed at Geekbench. As such, the GPU is working at 63% of its most efficiency potential, says Videocardz. It’s possible that the variant is an early pattern and nonetheless must be sorted out earlier than we are able to see its full efficiency.
This specific ARC Alchemist graphics card pattern was additionally examined on an older Core i5-9600K take a look at platform. The efficiency itself is not one thing to be enthusiastic about because the OpenCL rating ranged between 65000-69000 factors which is sort of GeForce RTX 2060 efficiency. As soon as once more, the Xe HPG powered graphics card is not prepared and never in its last state so the efficiency numbers listed below are just about ineffective for comparability towards any modern-day graphics card.
Intel ARC Alchemist 32 Xe-Core Graphics Card Geekbench Benchmarks (Picture Credit: TUM_APISAK / Benchleaks)
Intel Xe-HPG 512 EU ARC Alchemist Graphics Card
The highest Alchemist 512 EU variant has only one configuration listed thus far and that makes use of the total die with 4096 cores, 256-bit bus interface, and as much as 16 GB GDDR6 reminiscence that includes a 16 Gbps clock although 18 Gbps can’t be dominated out as per the rumor.
The Alchemist 512 EU chip is anticipated to measure at round 396mm2 which makes it greater than the AMD RDNA 2 and NVIDIA Ampere choices. The Alchemist -512 GPU will come within the BGA-2660 package deal which measures 37.5mm x 43mm. NVIDIA’s Ampere GA104 measures 392mm2 which implies that the flagship Alchemist chip is comparable in dimension whereas the Navi 22 GPU measures 336mm2 or round 60mm2 much less. This is not the ultimate die dimension of the chip nevertheless it needs to be very shut.
NVIDIA packs in tensor cores and far greater RT/FP32 cores in its chips whereas AMD RDNA 2 chips pack a single ray accelerator unit per CU and Infinity Cache. Intel can even have devoted {hardware} onboard its Alchemist GPUs for Raytracing & AI-assisted super-sampling tech.
The Xe-HPG Alchemist 512 EU chip is usually recommended to function clocks of round 2.2 – 2.5 GHz although we do not know if these are the common clocks or the utmost enhance clocks. Let’s assume that it is the max clock pace and in that case, the cardboard would ship as much as 18.5 TFLOPs FP32 compute which is 40% greater than the RX 6700 XT however 9% decrease than the NVIDIA RTX 3070.
Additionally, it’s said that Intel’s preliminary TDP goal was 225-250W however that is been upped to round 275W now. We are able to count on a 300W variant with twin 8-pin connectors too if Intel needs to push its clocks even additional. In both case, we are able to count on the ultimate mannequin to rock an 8+6 pin connector config, The reference mannequin can also be going to look very very like the drone advertising and marketing shot Intel put out through the ARC branding reveal. That reference design was leaked some time again by MLID too. There’re additionally talks a few customized lineup being labored upon by Intel’s AIB companions.
Intel ARC Alchemist vs NVIDIA GA104 & AMD Navi 22 GPUs
GPU Title | Alchemist DG-512 | NVIDIA GA104 | AMD Navi 22 |
---|---|---|---|
Structure | Xe-HPG | Ampere | RDNA 2 |
Course of 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 Items | 40 Compute Items |
FP32 Items | 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 Items | 48 RT Cores (V2) | 40 RA Items |
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 |
Extra Cache | 16 MB Good Cache? | N/A | 96 MB Infinity Cache |
Reminiscence Bus | 256-bit | 256-bit | 192-bit |
Reminiscence Capability | 16 GB GDDR6 | 8 GB GDDR6X | 16 GB GDDR6 |
Launch | Q1 2023 | Q2 2023 | Q1 2023 |
The Intel ARC Alchemist GPUs are anticipated to launch in Q1 2023 with the desktop variants now rumored for a Q2 2023 launch. Anticipate Intel to share extra particulars on ARC Alchemist GPUs & their next-gen Alder Lake-P/M laptop computer CPU lineup at CES 2023.
Information Sources: Benchleaks , TUM_APISAK , Videocardz