Intel 11th Generation Core Tiger Lake-H Performance Review: Fast and Power Hungry
by Brett Howse & Andrei Frumusanu on May 17, 2021 9:00 AM EST- Posted in
- CPUs
- Intel
- 10nm
- Willow Cove
- SuperFin
- 11th Gen
- Tiger Lake-H
SPEC CPU - Multi-Threaded Performance
Moving onto multi-threaded SPEC CPU 2017 results, these are the same workloads as on the single-threaded test (we purposefully avoid Speed variants of the workloads in ST tests). The key to performance here is not only microarchitecture or core count, but the overall power efficiency of the system and the levels of performance we can fit into the thermal envelope of the device we’re testing.
It’s to be noted that among the four chips I put into the graph, the i9-11980HK is the only one at a 45W TDP, while the AMD competition lands in at 35W, and the i7-1185G7 comes at a lower 28W. The test takes several hours of runtime (6 hours for this TGL-H SKU) and is under constant full load, so lower duration boost mechanisms don’t come into play here.
Generally as expected, the 8-core TGL-H chip leaves the 4-core TGL-U sibling in the dust, in many cases showcasing well over double the performance. The i9-11980HK also fares extremely well against the AMD competition in workloads which are more DRAM or cache heavy, however falls behind in other workloads which are more core-local and execution throughput bound. Generally that’d be a fair even battle argument between the designs, if it weren’t for the fact that the AMD systems are running at 23% lower TDPs.
In the floating-point multi-threaded suite, we again see a similar competitive scenario where the TGL-H system battles against the best Cezanne and Renoir chips.
What’s rather odd here in the results is 503.bwaves_r and 549.fotonik_r which perform far below the numbers which we were able to measure on the TGL-U system. I think what’s happening here is that we’re hitting DRAM memory-level parallelism limits, with the smaller TGL-U system and its 8x16b LPDDR4 channel memory configuration allowing for more parallel transactions as the 2x64b DDR4 channels on the TGL-H system.
In terms of the overall performance, the 45W 11980HK actually ends up losing to AMD’s Ryzen 5980HS even with 10W more TDP headroom, at least in the integer suite.
We also had initially run the suite in 65W mode, the results here aren’t very good at all, especially when comparing it to the 45W results. For +40-44% TDP, the i9-11980HK in Intel’s reference laptop only performs +9.4% better. It’s likely here that this is due to the aforementioned heavy thermal throttling the system has to fall to, with long periods of time at 35W state, which pulls down the performance well below the expected figures. I have to be explicit here that these 65W results are not representative of the full real 65W performance capabilities of the 11980HK – just that of this particular thermal solution within this Intel reference design.
Facebook
Twitter
RSS
229 Comments
View All Comments
eastcoast_pete - Monday, May 17, 2021 - link
A key question is now if Intel can further improve its 10 nm superfin process in time for Alder Lake. This TL is at least getting somewhat close to AMD's monolithic Zen3, but Rembrandt is around the corner, and that'll be an even more capable competitor. Right now, Intel retains its market share mostly because AMD can't deliver as many CPUs as people would buy.hfm - Monday, May 17, 2021 - link
I'm looking forward to an i5 or i7 design with no dGPU and that sweet embedded TB4. Probably going to be a small subset of the offerings like the lower tier XPS 15/17. Interested to see what comes about later in summer.evilspoons - Monday, May 17, 2021 - link
It'd be helpful to see power-normalized graphs, some of these results are a bit of a double-edged sword.outsideloop - Monday, May 17, 2021 - link
I want to see a comparison of the 11980HK with the 5800X clocked down to 65W, apples to apples. Let's compare the thermals of the AMD 8-core desktop to this "desktop replacement" Intel 8-core.Hifihedgehog - Monday, May 17, 2021 - link
Core i9-11980HK:https://images.anandtech.com/doci/16680/P95-45W_57...
Ryzen 9 5980HK:
https://images.anandtech.com/doci/16446/Power-P95-...
Just to add some further insight on these two charts, here is some breakdown. What ultimately happens is under sustained heavy loads, Ryzen 5000 series can maintain 8 cores at higher sustained clock speeds (~3.7 GHz) with lower power draw (35W). Bear in mind also that Zen 3 has a slight IPC advantage over Tiger Lake, meaning at a lower clock speed, it performs the same or does the same amount of work as Tiger Lake at a higher clock speed. Yet here we are where Ryzen 5000 can clock higher, do even MORE work, and draw less power. Meanwhile, Intel is struggling to reach similar sustained clock speeds (~3.2 GHz) at 45W sustained and it has slightly lower IPC. That is, Ryzen 5000 is ~10% faster clock-for-clock (or in IPC) than Intel's Willow Cove (Tiger Lake 11th Gen), Cypress Cove (Rocket Lake 11th Gen) or Sunny Cove (Ice Lake 10th Gen) which all three share roughly the same IPC amongst the whole trio (link: https://www.guru3d.com/articles_pages/intel_core_i... ). Ryzen 5000's synergy (higher clocks, lower power, higher IPC) sounds like a recipe for disaster for Intel, no? Alder Lake needs to come sooner rather than later if you ask me. Let's just hope Intel is throwing the full weight of their massive workforce of software developers at the problem of Alder Lake's heterogenous architecture because they will certainly need it working smoothly for it to shine.
lmcd - Monday, May 17, 2021 - link
Can we pause to note that MSI delivering bad thermal performance is characteristic of every single one of their laptops, bar none?lmcd - Monday, May 17, 2021 - link
Like, obviously that doesn't solve all the problems here, but how on earth with all the delays TGL-H(45+) got is this the OEM Intel picked?Spunjji - Tuesday, May 18, 2021 - link
It's a very odd choice. But then, didn't they use MSI for the original TGL reference platform too? That one actually outperformed most shipping devices, although IIRC it did so by running the fans at 100% 🤷♂️lmcd - Tuesday, May 18, 2021 - link
I put it on twitter also but my opinion is that the Tiger Lake MSI design is actually about the cooling and size that MSI would use for an H series processor normally, such as the MSI Creator.This implies that the MSI's thermal overhead is pure luck on Intel's part in that MSI didn't have time to reduce the thermal performance accordingly.
Spunjji - Thursday, May 20, 2021 - link
That would make sense!