Thursday, April 8th 2021
Intel Xe-HPG DG2 GPU Engineering Sample Pictured
We have recently received pictures of any early engineering sample of Intel's upcoming DG2 GPU from YouTuber Moore's Law is Dead. The card features 512 Execution Units and will be the flagship model for Intel's upcoming Xe-HPG lineup reportedly targeting performance between the RTX 3070 and RTX 3080. The final product is rumored to feature a base clock of 2.2 GHz along with 16 GB GDDR6 memory and a 256-bit bus. The sample has a TDP of 275 W with 8 + 6 pin power connectors up from original targets of 225 W - 250 W.
The report also notes that Intel is still deciding between three cooler designs with the finished card potentially featuring a white shroud. Intel also appears to be working on a NVIDIA DLSS/AMD FidelityFX Super Resolution competitor codenamed XeSS which confirms support for hardware-accelerated raytracing and resolution upscaling tech. The card is unlikely to launch until Q4 2021 with wider availability in 2022, lower end 128 EU, and 256 EU cards will follow shortly afterward. The full report can be viewed below.
Full Video
Source:
Moore's Law is Dead
The report also notes that Intel is still deciding between three cooler designs with the finished card potentially featuring a white shroud. Intel also appears to be working on a NVIDIA DLSS/AMD FidelityFX Super Resolution competitor codenamed XeSS which confirms support for hardware-accelerated raytracing and resolution upscaling tech. The card is unlikely to launch until Q4 2021 with wider availability in 2022, lower end 128 EU, and 256 EU cards will follow shortly afterward. The full report can be viewed below.
48 Comments on Intel Xe-HPG DG2 GPU Engineering Sample Pictured
Back on topic, I find the specs of this card hard to believe. I understand that Intel is a big company and all, but how can they go from lackluster HD Graphics to RTX 2080 Ti performance with essentially no products in between?
im aware its a engineering shroud, just saying
As for shortages, Intel has noted they are not making these GPUs in house, so they'll be fighting with everyone else for limited fab space at TSMC, Samsung, etc.
www.cnbc.com/2021/03/23/intel-is-spending-20-billion-to-build-two-new-chip-plants-in-arizona.html
In the words of Green Day, wake me up when September ends, but only if they actually have a product that is
Competitive
Priced right
Available
Bug free
So far they have shown 0/4 of those.
Moreover, if intel isn't utilizing their own fabs for this release, I can't see availability being all that great. Of course, they'd have to actually have a finished product, first.
And Xesse means "defecate" in Greek. Χέσε μας επιτέλους ένα υπαρκτό προϊόν ρε intel.
It's a lot easier to engineer new fab when you're using the same chip across multiple SKUs and that's going to be a huge disadvantage for NVIDIA in the near future.
Nvidia has published a whitepaper on MCM GPUs in 2017.
Intel's Xe-HPC will be MCM as well.
I really doubt you can patent MCM as a concept, since it's been used for so many different things. AMD can patent their specific implementation, of course, yet it doesn't mean nobody else will do MCM GPUs.