Intel has officially unveiled its first two Arc Alchemist A-Series GPUs, the ACM-G10 with up to 32 and the ACM-G11 with up to 8 Xe Cores.
Intel's First Discrete Gaming GPUs Unveiled: Arc A-Series In ACM-G10 32 Xe & ACM-G11 8 Xe Flavors
The Intel ARC Alchemist GPUs are designed primarily for desktop and notebook platforms. From the information that we have currently gathered, the lineup will consist of two GPUs and each of those will have various SKUs, and each of those SKUs will be featured in a range of solutions for desktop graphics cards and mobility GPUs.
All Intel ARC Alchemist GPUs will utilize the TSMC 6nm process node and from what we already know, the Intel Xe-HPG Alchemist GPU features a Xe-Core which is the fundamental DNA of the 1st Gen ARC lineup. The Xe-Core is a compute block that is composed of 16 Vector Engines (256-bit per engine) and 16 Matrix Engines (1024-bit per engine).
Each Vector Engine is composed of 8 ALUs so, in total, we are looking at 128 ALUs per Xe-Core. The Xe-Core further features its own dedicated 192 KB L1 cache so that is a total of 6 MB of L1 cache on the complete chip.
Intel fuses four Xe-Cores together to form a Render Slice which is composed of 4 Ray Tracing Units, four Sampler Units, Geometry/Rasterize/HiZ engines, and two Pixel Backend blocks with 8 units on each. These Render Slices are put together to form the main GPUs. The flagship is composed of an 8 Render Slice configuration which features 32 Xe-Cores, 512 Vector Engines, and 4096 ALUs. There will be different configurations with 2, 4, 6 Render Slices but we are focusing on the flagship part in this report.
Intel’s Xe HPG architecture will be able to achieve 1.5x higher clock rates than Xe LP and also deliver 1.5x higher performance per watt. This means we are looking at clocks in the 2.1 GHz range considering the Xe LP discrete GPUs were clocked at 1.4 GHz. It also means that Intel will be able to squeeze additional power out of the architecture, should they want to, without increasing the power draw (or reducing the power draw while keeping performance constant.
Out of the two GPUs, the first one will be known as ACM-G10. This is the top SKU and will be featured in mainstream and high-end designs. The second GPU is known as ACM-G11. This is an entry-level SKU and as such, will be featured in entry-level and mainstream designs. Each GPU has its own SKUs with different specifications.
Each Matrix Engine block is also referred to as an XMX block which will handle tensor operations in both FP16 (128 FP16 ops/clock), INT8 (256 INT8 ops/clock), & INT4 (512 INT4/INT2 ops/clocks) mode. The XMX engines allow for up to 16 times the compute capabilities for INT8 inferencing. This helps boost the performance through the dedicated XeSS (Xe Super Sampling) technology.
Coming to the media engine, the Intel Arc Alchemist discrete GPUs are equipped with the latest tech which includes up to 8K60 12-bit HDR decode, up to 8k 10-bit HDR encode, VP9, AVC, HEVC, and AV1. Intel is partnering with top brands to leverage the AV1 capabilities of their Arc GPUs through the Alliance for Open Media. The new AV1 engine will provide 50x faster GPU encoding capabilities versus software mode. The tech will be integrated into several creation tools such as FFMPEG, Handbrake, Adobe Premiere Pro, DaVinci Resolve, and XSplit on launch.
In addition to that, the Intel Arc Xe Display Engine will support the latest display standards such as HDMI 2.0b, DisplayPort 1.4a, and 2.0 10G. Support for a variety of high resolution and high refresh rate modes is also going to be available which include:
- 2x 8k60 HDR
- 4x 4k120 HDR
As for adaptive syncing technology, Intel's discrete Arc Alchemist GPUs will support both Adaptive-Sync and Speed Sync technology. The former will sync the display's refresh rate to provide a smooth and tear-free experience while the latter will speed up the latest frame, delivering low latency, no tearing, and no cap. There's also a third mode known as Smooth Sync which will blur distracting screen tears with a dithering filter.
Intel Xe-HPG Based Discrete Alchemist GPU Configurations:
|GPU Variant||Graphics Card Variant||GPU Die||Execution Units||Shading Units (Cores)||Memory Capacity||Memory Speed||Memory Bus||TGP|
|Xe-HPG 512EU||ARC A780?||Arc ACM-G10||512 EUs||4096||Up To 32/16 GB GDDR6||18 / 16 / 14 Gbps||256-bit||~225W (Desktops)
|Xe-HPG 384EU||ARC A580?||Arc ACM-G10||384 EUs||3072||Up To 12 GB GDDR6||16 / 14 Gbps||192-bit||150-200W (Desktops)
|Xe-HPG 256EU||ARC A550?||Arc ACM-G10||256 EUs||2048||Up To 8 GB GDDR6||16 / 14 Gbps||128-bit||60-80W (Laptops)|
|Xe-HPG 128EU||ARC A380?||Arc ACM-G11||128 EUs||1024||Up To 6 GB GDDR6||16 / 14 Gbps||96-bit||~75W (Desktops)|
|Xe-HPG 128EU||ARC A350?||Arc ACM-G11||128 EUs||1024||Up To 4 GB GDDR6||16 / 14 Gbps||64-bit||35-50W (Laptops)|
|Xe-HPG 96EU||ARC A330?||Arc ACM-G11||86 EUs||768||Up To 4 GB GDDR6||16 / 14 Gbps||64-bit||~35W (Laptops)|
Intel Arc Alchemist 'ACM-G10' GPU Specifications - The Top GPU
The top Arc Alchemist GPU is going to be the ACM-G10. It is said expected to measure 407mm2 which makes it larger than both NVIDIA's GA014 and AMD's Navi 22 chips. The ACM-G10 powered GPUs are going to compete against NVIDIA's GeForce RTX 3070(Ti) and AMD Radeon RX 6700 XT.
NVIDIA packs in tensor cores and much bigger RT/FP32 cores in its chips while AMD RDNA 2 chips pack a single ray accelerator unit per CU and Infinity Cache. Intel will also have dedicated hardware onboard its Alchemist GPUs for Raytracing & AI-assisted super-sampling tech.
The full die features 32 Xe Cores, 512 EUs, 4096 ALUs, a 256-bit bus interface, and up to 16 GB GDDR6 memory featuring clock speeds between 16-18 Gbps from Samsung. Following are some of the ACM-G10 32 EU variants that we can expect::
- 512 EU (4096 ALUs) / 16 GB @ Up To 18 Gbps / 256-bit
- 384 EU (3072 ALUs) / 12 GB @ Up To 16 Gbps / 192-bit
- 256 EU (2048 ALUs) / 8 GB @ Up To 16 Gbps / 128-bit
The Xe-HPG Alchemist ACM-G10 chip is suggested to feature clocks of around 2.2 - 2.5 GHz though we don't know if these are the average clocks or the maximum boost clocks. We have already spotted a flagship configuration with DG2-512 GPU running at 2.4 GHz, delivering nearly 20 TFLOPs of FP32 horsepower so we might very well end up with 2.5 GHz clocks in the final revision.
The TDP target for these chips seems to start at 120W for laptops and go all the way up to 300W for desktop parts. In either case, we can expect the final model to rock an 8+6 pin connector config, The reference model is also going to look very much like the drone marketing shot Intel put out during the ARC branding reveal.
Intel ARC ACM-G10 vs NVIDIA GA104 & AMD Navi 22 GPUs
|Graphics Card Name||Intel ARC A770||NVIDIA GeForce RTX 3070 Ti||AMD Radeon RX 6700 XT|
|GPU Name||ACM-G10||NVIDIA GA104||AMD Navi 22|
|Process Node||TSMC 6nm||Samsung 8nm||TSMC 7nm|
|Transistor Density||53.4M Transistors/mm2||44.4M Transistors/mm2||51.2M Transistors/mm2|
|FP32 Cores||32 Xe Cores||48 SM Units||40 Compute Units|
|Max Clock||2100 MHz||1770 MHz||2581 MHz|
|FP32 TFLOPs||17.2 TFLOPs||21.75 TFLOPs||13.21 TFLOPs|
|Memory Capacity||16 GB GDDR6||8 GB GDDR6X||12 GB GDDR6|
|Launch||Q3 2022||Q2 2021||Q1 2021|
Intel Arc Alchemist 'ACM-G11' GPU Specifications - The Small GPU
The Intel ACM-G11 is going to be the smaller Arc GPU, aiming at the entry-level and mainstream PC platforms. The GPU is said to measure around 156mm2. This is also much smaller than the 200mm2 TU117 die that the chip will be competing against. The GA107 die size isn't known yet but it is likely to be around 160-180mm2. It is a bigger chip compared to the recently released AMD Navi 24 which measures only 107mm2.
There are two configs that feature the full-fat SKU with 1024 cores, a 96-bit, and a 64-bit variant with 6 GB and 4 GB memory capacity, respectively. The cut-down variant will come with 96 EUs or 768 cores and a 4 GB GDDR6 memory featured across a 64-bit bus interface. The chip is expected to feature a clock speed of around 2.2 - 2.5 GHz and have a sub 75 Watt power consumption which means we will be looking at connector-less graphics cards for the entry-level segment.
Following are all the ARC ACM-G11 variants that we can expect:
- 128 EU (1024 ALUs) / 6 GB @ Up To 16 Gbps / 96-bi
- 128 EU (1024 ALUs) / 4 GB @ Up To 16 Gbps / 64-bit
- 96 EU (768 ALUs) / 4 GB @ Up To 16 Gbps / 64-bit
This GPU will be very similar to the DG1 GPU-based discrete SDV board however Alchemist will have a more improved architecture design and definitely more performance uplift over the first-gen Xe GPU architecture. This lineup is definitely going to be aimed at the entry-level desktop discrete market based on the specifications.
Intel ARC ACM-G11 vs NVIDIA GA106 & AMD Navi 24 GPUs
|Graphics Card Name||NVIDIA GeForce RTX 2050||AMD Radeon RX 6500 XT||Intel ARC A380||NVIDIA GeForce RTX 3050|
|GPU Name||NVIDIA GA107||AMD Navi 24||ACM-G11||NVIDIA GA106|
|Process Node||Samsung 8nm||TSMC 6nm||TSMC 6nm||Samsung 8nm|
|FP32 Cores||16 SM Units||16 Compute Units||8 Xe Cores||24 SM Units|
|Memory Capacity||4 GB GDDR6||4 GB GDDR6||6 GB GDDR6||8 GB GDDR6|
|Launch||Q1 2022||Q2 2022||Q2 2022||Q1 2022|