Intel Xe-HPG DG2 GPU Spotted With Performance On Par With A NVIDIA GTX 1660 SUPER

 

A brand new Intel Xe-HPG DG2 GPU entry has made its way in the GFXBench database which shows performance on par with an NVIDIA GeForce GTX 1660 SUPER graphics card.

Intel Xe-HPG DG2 GPU Spotted Within GFXBench Benchmark, Performance On Par With An NVIDIA GTX 1660 SUPER

Discovered by TUM_APISAK, an Intel Xe Graphics entry was made within the GFXBench benchmark. Since the graphics card isn’t marked as an Iris variant, we can come to the conclusion that this is a Xe-HPG DG2 GPU-based variant that will be part of the higher-end graphics lineup for desktops & mobility.

No specifications are mentioned but compared to the Iris Xe MAX which is the top Xe-LP GPU, we are looking at a 2.5x performance increase. The Iris Xe Max is composed of 96 Execution units or 768 cores. It looks like this particular variant could be a higher-end variant with either 256 or 192 EUs. We should also expect higher clock speeds thanks to increased power headroom on the HPG variants.

Compared to existing discrete graphics cards, the Xe-HPG DG2 GPU is comparable to the NVIDIA GeForce GTX 1660 SUPER. However, this is not the final performance as we don’t know what clocks it was running at or whether the drivers are optimized enough for this particular benchmark to say that this is the final performance but we are seeing improvements to performance with each new entry.

There are a lot of unknowns in this benchmark but this is the third entry of Xe-HPG GPUs that we have seen in the last couple of days which suggests that more and more samples are now running in the test environments and a launch could be real close.

Here’s Everything We Know About Intel Xe-HPG DG2 Gaming GPU Lineup

Each Xe-HPG based DG2 GPU SKU will come in various configurations which will range from the full-fat chip to several cut-down variants. This is similar to NVIDIA’s Ampere GA102-400, GA102-200 naming schemes, or AMD’s Navi 21 XTX, Navi 21 XT, Navi 21 XL naming conventions.

Intel Xe-HPG DG2 512 EU Discrete Gaming Graphics Cards

The top DG2 512 EU variant has just one configuration listed so far and that utilizes the full die with 4096 cores, 256-bit bus interface, and up to 16 GB GDDR6 memory (8 GB GDDR6 listed too). Based on demand and yields, Intel could produce more variants of this flagship chip but we can’t say for sure right now.

The DG512 EU chip is expected to measure at around 396mm2 which makes it bigger than the AMD RDNA 2 and NVIDIA Ampere offerings. The DG2-512 GPU will come in the BGA-2660 package which measures 37.5mm x 43mm. NVIDIA’s Ampere GA104 measures 392mm2 which means that the DG2 chip is comparable in size while the Navi 22 GPU measures 336mm2 or around 60mm2 less. This isn’t the final die size of the chip but it should be very close.

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 is expected to have hardware-accelerated ray-tracing capabilities onboard its Xe-HPG GPUs & a recent tweet from Raja also seems to point out hardware DL/ML capabilities that would assist in super sampling tech.

The Xe-HPG DG2 512 EU chip is suggested to feature clocks of up to 2.2 GHz though we don’t know if these are the average clocks or the maximum boost clocks. Also, it is stated that Intel’s initial TDP target was 225-250W but that’s been upped to around 275W now. We can expect a 300W variant with dual 8-pin connectors too if Intel wants to push its clocks even further.

Intel Xe-HPG DG2 384 EU Discrete Gaming Graphics Cards

Moving on, we have the Intel Xe-HPG DG2 384 GPU SKU which is expected to comprise at least three variants. The full fat chip will feature 3072 cores, up to 12 GB GDDR6 memory (6 GB GDDR6 listed too), and a 192-bit bus interface. Then we have two variants, a 256 EU and a 192 EU variant which are comprised of 2048 and 1536 cores. While both variants feature a 128-bit bus interface, the 256 EU SKU will come with up to 8 GB GDDR6 memory (4 GB GDDR6 listed too) while the 192 EU variant will stick with just 4 GB GDDR6 memory. Based on the specifications, these GPUs will be positioned as mainstream parts.

Videocardz had earlier leaked out the die configuration of the Intel Xe-HPG DG2 384 GPU variant which should measure 190mm2. The PCB blueprint shows 6 memory module locations which do confirm a 192-bit bus interface and either 6 or 12 GB GDDR6 memory capacity. The 384 and 256 EU SKUs are expected to feature 16 MB and 8 MB smart cache, respectively. The clock speeds for the 384 EU parts are reported at 600 MHz base and 1800 MHz turbo while the 256 EU part will feature a 450 MHz base and 1400 MHz turbo clock.

Intel Xe-HPG DG2 128  EU Discrete Gaming Graphics Cards

Then lastly, we have the Intel Xe-HPG DG2 128 EU parts. The top config is once again a full-fat SKU with 1024 cores, a 64-bit bus interface, and 4 GB GDDR6 memory. 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. This GPU will be very similar to the DG1 GPU-based discrete SDV board however DG2 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.

We have seen the Intel Xe-HPG DG2 GPU-based discrete graphics card engineering sample leak out last month along with some rumored performance and pricing figures, you can read more on that here. Based on the timeline, the Xe-HPG DG2 lineup will compete against NVIDIA’s Ampere & AMD RDNA 2 GPUs since both companies aren’t expected to launch their next-gen parts by the very end of 2022. The Xe-HPG GPUs will be coming to the mobility platform too and will be featured in Alder Lake-P notebooks.