Intel ARC Alchemist graphics card efficiency and value positioning slide has leaked out displaying the NVIDIA and AMD GPUs the lineup will probably be competing towards.
Intel ARC Alchemist All Set To Deal with NVIDIA GeForce & AMD Radeon GPUs In The $100 To $500 US Market Vary
The leaked slide appeared on Baidu Forums (through Videocardz) and has been shared with retail companions by Intel. The slide lists down the proposed ARC Alchemist ‘DG2’ lineup for 2022 and respectively opponents they are going to be positioned towards. In response to the slide, the lineup will embrace at the least 4 SKUs however there could also be extra. Additionally, the slide seems to be a bit previous because it does not point out the ARC branding that Intel only recently launched for its graphics lineup so inner plans may need modified since then.
Intel’s Excessive-Finish ARC Gaming Graphics Card Lineup ($300 – $500 US)
With that stated, let’s discuss concerning the higher echelon of the lineup which is able to function Fanatic ($400 – $499 US) and Efficiency+ ($300 – $399 US) graphics playing cards. These Intel ARC Alchemist GPUs will function the ‘SOC1’ which may very well be the interior codename for the Alchemist 512 EU chip. This chip has a TDP vary of 175W to 225W. The Fanatic half will sort out the NVIDIA GeForce RTX 3070 & the AMD Radeon RX 6700 XT.
Do observe that Intel does not plan to launch an ‘Fanatic+’ contender in its first-gen ARC lineup as that will probably be reserved for the next-gen. Which means the highest graphics playing cards from NVIDIA & AMD such because the GeForce RTX 3080 and the Radeon RX 6800 XT will stay uncontested.
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 Sensible 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 2022 | Q2 2022 | Q1 2022 |
Intel’s Entry-Stage ARC Gaming Graphics Card Lineup ($100 – $200 US)
Transferring on to the lower-end section, that is one place the place Intel could hit AMD and NVIDIA exhausting contemplating each GPU producers have not launched a single sub-$300 US desktop graphics card available in the market but. The mainstream section will function the ‘SOC2’ which is the Alchemist 128 EU GPU and can function a TDP of 75W and up. ARC graphics playing cards primarily based on this chip will probably be featured within the ‘Mainstream+’ ($150 – $199 US) and ‘Mainstream’ ($100 – $149 US) section.
Each AMD and NVIDIA do have the chips to launch within the entry-level section however they’re being pushed to notebooks and laptops first. If launched previous to Q1 2022, then entry-level ARC GPUs will sort out AMD’s Navi 24 (RX 6500 Collection) and NVIDIA’s GA107 GPUs (RTX 3050 Collection) although these could find yourself over $200 US pricing contemplating the place the mainstream playing cards are at the moment featured.
A serious lead that Intel might have over the competitors, particularly NVIDIA since AMD lacks on this division, is AI help in supersampling applied sciences. Intel has already showcased a formidable demo of its XeSS know-how and primarily based on the anticipated numbers, Intel GPUs might outperform NVIDIA’s Tensor Core implementation (DLSS) with its XMX structure. Intel can be anticipated to function a small however helpful sport cache on its GPUs and will probably be geared up with larger VRAM capacities of as much as 16 GB (GDDR6) throughout a 256-bit bus interface. This could be twice as a lot reminiscence as NVIDIA’s RTX 3070 and RTX 3070 Ti so they could have to arrange a refresh to counter it.
We now have seen the Intel Xe-HPG Alchemist GPU-based discrete graphics card engineering pattern leak out a couple of months again together with some rumored efficiency and pricing figures, you may learn extra on that right here. Based mostly on the timeline, the Xe-HPG Alchemist lineup will compete towards NVIDIA’s Ampere & AMD RDNA 2 GPUs since each corporations aren’t anticipated to launch their next-gen elements by the very finish of 2022. The Xe-HPG ARC GPUs will probably be coming to the mobility platform too and will probably be featured in Alder Lake-P notebooks.
Intel Xe-HPG DG2 GPU Based mostly Discrete Gaming Graphics Card Specs:
GPU Variant | GPU SKU | Execution Items | Shading Items (Cores) | Reminiscence Capability | Reminiscence Bus | TGP |
---|---|---|---|---|---|---|
Xe-HPG 512EU | Alchemist-512EU | 512 EUs | 4096 | 16/8 GB GDDR6 | 256-bit | ~275W |
Xe-HPG 384EU | Alchemist-384EU | 384 EUs | 3072 | 12/6 GB GDDR6 | 192-bit | TBC |
Xe-HPG 256EU | Alchemist-384EU | 256 EUs | 2048 | 8/4 GB GDDR6 | 128-bit | TBC |
Xe-HPG 192EU | Alchemist-384EU | 192 EUs | 1536 | 4 GB GDDR6 | 128-bit | TBC |
Xe-HPG 128EU | Alchemist-128EU | 128 EUs | 1024 | 4 GB GDDR6 | 64-bit | TBC |
Xe-HPG 96EU | Alchemist-128EU | 86 EUs | 768 | 4 GB GDDR6 | 64-bit | ~120W |