The Ice Lake Benchmark Preview: Inside Intel's 10nm
by Dr. Ian Cutress on August 1, 2019 9:00 AM EST- Posted in
- CPUs
- Intel
- GPUs
- 10nm
- Core
- Ice Lake
- Cannon Lake
- Sunny Cove
- 10th Gen Core
Gaming Results (15W and 25W)
One of the biggest changes to the Ice Lake design is in the integrated graphics – Intel is now giving more focus and more die area to graphics, something it has arguably been neglecting for several years now. With Ice Lake, we move to a Gen11 graphics architecture, which is almost like the previous Gen9.5 but now with added support for variable rate shading (VRS), moving from 24 EUs to 64 EUs, and memory support up from LPDDR3-2133 to LPDDR4X-3733.
World of Tanks is a very CPU driven benchmark, and having the extra frequency of the 25W processor does help here. We're getting a sizeable uplift from Whiskey Lake, due to the extra EUs and memory frequency.
Our Final Fantasy test seemed to regress in 25W mode, although still within the noise. This test is still GPU bound, so adding the extra TDP to the CPU didn't actually help much. However, comparing to the Whiskey Lake integrated graphics, we've got over a 2x speedup.
Similarly with Civilization, with what is normally our 'IGP' settings, we are still GPU limited here.
One of Intel's newest features is Variable Rate Shading.
If developers add the option, soon to be an easy checkbox in Unity and Unreal, the game can decide to control the rate at which it shades pixels, from calculating every pixel down using one result across a 4x4 grid, to save compute power. Currently the only way to test this is with the 3DMark functional demo.
The new VRS test in 3DMark is designed as a feature test to show the potential uplift effect from enabling variable rate shading within a game. In both 15W and 25W modes, the data saw a good uplift, and we seemed to get more out of the 25W mode than the 15W mode.
261 Comments
View All Comments
Rookierookie - Thursday, August 1, 2019 - link
Other reports have indicated that Ice Lake graphics actually beats Vega, so...psychobriggsy - Thursday, August 1, 2019 - link
Except you've been able to get Zen APUs for a long time, and this isn't out yet. For all we know, Zen 2 APU will arrive in a similar timeframe. But it is good that it is competitive.Phynaz - Thursday, August 1, 2019 - link
AMD laptop parts have always been garbageFulljack - Thursday, August 1, 2019 - link
they're now in the same ballpark with integrated vega graphics.still, looking at oem, they'll probably add mx250 or gtx 1650 max-q.
StormyParis - Thursday, August 1, 2019 - link
They'll be, in 6 months, assuming this demo unit can be produced at scale, priced competitively, and AMD doesn' t improve in the mean time.I understand why Anand did this review, and I enjoyed it, but it is really about Intel making vague promises in the hope we don't all buy AMD stuff for our 2-3-4-5 yearly upgrade in the next 6 months. And then we'll do it anyway ;-p
Fulljack - Sunday, August 4, 2019 - link
sadly, oem doesn't push amd based laptop as much as intel based laptop.just recently we've got ryzen 3000 (zen+) laptop, despite it's released early this year. it only improve performance over previous generation by ~10%, but it greatly improves battery life though.
GNUminex_l_cowsay - Thursday, August 1, 2019 - link
For once the Civ VI graphics benchmark actually tells us something useful; but would it kill you guys give turn time numbers?Ian Cutress - Thursday, August 1, 2019 - link
Didn't have time, we're we're running to the edge with the game numbers.zealvix - Thursday, August 1, 2019 - link
Any news/dates for the desktop releases?And graphics seem only abit better than the iris plus 655 in my NUC, but a large part of that is from the increased bandwidth as well.
Any news of a 2 tflop or at least 1.7+ tflop variant of the graphics to match the best from ryzen mobile?
psychobriggsy - Thursday, August 1, 2019 - link
That would require Intel to design a SKU with 96EUs at least.That doesn't exist, and likely won't on 14/10nm.
AMD is likely months away from 7nm Zen 2 APU with over 2 TFLOPS, assuming memory bandwidth to feed it (I truly hope they have LPDDR4X support for their mobile APUs).