Intel is just a few days away from launching its first Xe GPU based graphics chip with its Tiger Lake CPUs. The Intel Xe GPU will come in various shapes and sizes but the initial ones are aimed at the notebook segment, offering a 2x performance uplift versus previous generation Iris GPUs featured on Ice Lake CPUs.
Intel Iris Xe GPU Spotted Days Before Launch, Features An Impressive 1650 MHz Overclock
The Intel Iris Xe GPU was spotted by TUM_APISAK over at the Sisoftware database. The Iris Xe GPU is slightly different than the DG1 GPU which is a discrete notebook graphics chip. The difference is that the DG1 GPU is a standalone discrete GPU while the Iris Xe GPU is featured as an integrated unit on the Tiger Lake CPUs. Both chips do share some similarities as they both peak out at 96 EUs (Execution units) which translates to 768 cores.
Intel’s Iris Xe GPUs in the fastest stock configuration will feature a clock speed of 1300 MHz and we have seen the discrete variant clocking up to 1550 MHz. The latest entry for the Intel Iris Xe GPU is running at 1650 MHz which is a 27% overclock over the reference clocks. For comparison, the Vega GPU featured on the Renoir APUs clocks up to 2100 MHz (stock frequency for the Ryzen 7 4750G) but can be overclocked up to 2300-2400 MHz which is about a 10-15% jump.
The Vega GPU comes with lower cores than Intel’s Iris Xe GPUs but Intel has showcased that their architecture is slightly more tuned for faster graphics performance than AMD’s enhanced Vega GPUs. The GPU was running on a Tiger Lake 28W CPU. The Tiger Lake G7 variants feature the same core and clock speeds of 768 and 1300 MHz, respectively. It cannot be said which particular CPU the Xe GPU was running on.
Intel Tiger Lake vs AMD Renoir Mobility CPU Comparisons:
| CPU Family Name | Intel Tiger Lake-U | AMD Renoir U-Series |
|---|---|---|
| Family Branding | Intel 11th Gen Core (U-series) | AMD Ryzen 4000 (U-Series) |
| Process Node | 10nm | 7nm |
| CPU Core Architecture | Willow Cove | Zen 2 |
| CPU Cores/Threads (Max) | 4/8 | 8/16 |
| Max CPU Clocks | 4.8 GHz (Core i7-1185G7) | 4.2 GHz (Ryzen 7 4800U) |
| GPU Core Architecture | Xe Graphics Engine | Vega Enhanced 7nm |
| Max GPU Cores | 96 EUs (768 cores) | 8 CUs (512 cores) |
| Max GPU Clocks | 1300 MHz | 1750 MHz |
| TDP (cTDP Down/Up) | 15W (12W-28W) | 15W (10W-25W) |
| Launch | Mid 2020 | March 2020 |
The performance of the GPU will translate to roughly 2.5 TFLOPs that would put it faster than both last-generation consoles (PS4 and Xbox One). The performance will also be faster than the desktop Ryzen 7 4750G which features around 2.15 TFLOPs performance at the stock 2.1 GHz clock speed.
With this level of performance, the Iris Xe GPU can easily outperform a desktop discrete GeForce GTX 1050 Ti or the Radeon RX 5300M mobility graphics card.
A performance comparison for the Intel Xe GPUs in the Geekbench 5 OpenCL benchmark was posted by Harukaze5719 on Twitter and shows that the discrete GPU featured alongside the i7-1185G7 is about as fast as the Radeon Pro 5300M. Intel has showcased its latest branding for its Tiger Lake CPUs which launch on 2nd September and we will keep you updated with more info related to Tiger Lake processors and notebooks featuring them.
