Intel Core i7-1065G7 Benchmarked: Ice Lake with Iris Plus Graphics

[Julio Franco]

This has been a long time coming, as we're finally getting around to testing Intel's Ice Lake architecture. This is Intel's first real attempt at a 10nm CPU, and in this review we'll be comparing it against their 14nm offerings to see how it stacks up in terms of performance.

Read the full article here.

https://www.techspot.com/review/1944-intel-core-i7-1065g7/

 

[Irata]

Nice review, thanks.

Quick question: Did you see any throttling ? Notebook check experienced this when testing the new Ice Lake equipped Surface Pro 7, but this may also be due to the form factor.

 

[Vulcanproject]

So as suspected Intel face the same kind of problems or at least limitations ramping the clocks of their 10nm process that AMD face with Zen on TSMC's 7nm. They can add more transistors and raise IPC quite a bit but they can't clock them high enough to gain anything dramatic that you would otherwise get on a new process. At the moment at least.

It explains why Intel haven't really bothered with 10nm desktop parts, because they probably won't be able to just match what their 14nm+++ parts can do let alone beat them. Having an imposed TDP limit on mobile parts hides that fact, but it would be exposed on desktop where power limits are far more generous and thus clocks are far higher.

There are some glimmers of hope though. The iGPU performance is a much needed improvement and at least they are actually building the damn things and shipping them. The problem is how long it is going to take for them to refine 10nm so they have a process that clearly surpasses 14nm+++. On this evidence we will be waiting for 2021 for anything next gen on the desktop front.

 

[Dimitrios]

No gaming benchmarks????

 

[neeyik]

@Dimitrios While I can't speak for Tim, gaming benchmarks in tests for CPUs in mobile/laptop devices can be problematic to analyse. Modern games require a lot of memory read/write cycles per rendered frame, and so the indicated fps of a game can be more indicative of the memory subsystem performance than the integrated GPU's execution side of things. Hence why it's better to test applications that test the compute element of the GPU, as these typically have far fewer memory read/writes per processed 'frame'.

If one is going to look at gaming benchmarks, then it should be as a full system review - I.e. the entire mobile device.

Edit: Ha! Turns out neither of us read the article all that well...

This review will be discussing everything you need to know about productivity and compute performance, we’ll be comparing Ice Lake to previous 14nm parts, but gaming is something we’ll be tackling in a separate article.

 

[Julio Franco]

No gaming benchmarks????

@Dimitrios Ha! Turns out neither of us read the article all that well... This review will be discussing everything you need to know about productivity and compute performance, we’ll be comparing Ice Lake to previous 14nm parts, but gaming is something we’ll be tackling in a separate article.

Correct. The iGPU/gaming tests will come later this week. The current review was plenty lengthy and in-depth already at nearly 4,000 words + all those benchmarks.

 

[yeeeeman]

Try with Dell XPS 13. You will get better results. From what notebookcheck says, Razer Blade is not the best Ice Lake implementation available.

 

[Dimitrios]

Correct. The iGPU/gaming tests will come later this week. The current review was plenty lengthy and in-depth already at nearly 4,000 words + all those benchmarks.

My bad, I just skimmed through it.

 

[Lew Zealand]

If at all possible, include an 8th Gen Iris Plus unit for the gaming comparo. I'm pretty sure you guys don't have anything in the studio with Iris Plus but if you can beg, borrow, that'd be great. I got back into gaming via Intel Iris Plus Graphics and still game in my free time at work on a 7th Gen NUC with Iris Plus.

 

[ZedRM]

With the move to a more compact process(10nm) one would expect the i7 to add 2 more cores. 2019@10nm and we're still at 4 cores? Unimpressive.

 

[shady28]

So you are comparing the 4C/8T Ice Lake to a 6C/12T Comet Lake?

Why didn't you compare the 4C/8T 1065G7 to the 4C/8T 10610U.

Your conclusions about Ice Lake are very misleading because of this lopsided comparison.