Updates to Skylake Discrete Graphics Performance: PCIe Optimizations Incoming

[tsk]

 

In our initial review of the two 6th Generation Intel Skylake-K processors launched on August 5th, the i7-6700K and the i5-6600K, our comparative analysis to the previous generations of Intel processors was for the most part, positive. On the whole, clock-for-clock performance was a marginal increase over previous generations but the cumulative end-to-end effort of several generations of upgrades, plus for those that overclock, gave a substantial reason for those in CPU limited workloads to find an upgrade (along with benefits on the chipset and DRAM side as well). However, one element of the equation was puzzling at the time – the performance of games using discrete graphics cards was marginally lower with the new platform compared to older platforms when looking at average frame rates.

 

During our testing, it is not uncommon to see two platforms that perform similarly to have a reasonable margin of error, often ±1%, due to variations in pre-initialised cache structures, or in the case of games like GRID that rely on a random sequence to provide the end-result numbers. Despite this, we noticed that for Skylake-K we saw consistent drop in our discrete GPU testing, often around the -1% to -3% mark but sometimes as low as -5% or -7% when we compared it to both Intel’s 5th Generation (Broadwell) and 4thGeneration (Haswell). Other websites such as The Tech Report also noted these results, placing Broadwell’s numbers at the top of the stack (if only marginal). Some commentary at the time focused on Broadwell’s use of eDRAM in the desktop components which can aid performance while retaining a frequency deficit, although given our analysis of the eDRAM in Broadwell as a victim cache rather than a transparent DRAM cache it seems less likely that this is the case, plus we also now have new information coming post launch about this issue. But if we remove Broadwell as a special case, it was still concerning that the i7-6700K lagged behind the i7-4770K despite being higher in frequency and clock-for-clock performance.

 

 

 

Another couple of weeks later, we were contacted by ASUS who shed a lot more light on the issue.  The register in question is called the FCLK (or ‘f-clock’), which controls some of the cross-frequency compensation mechanisms between the ring interconnect of the CPU, the System Agent, and the PEG (PCI Express Graphics). Basically this means it is to do with data from the processor to the GPUs. So when data is handed from one end to another, this element of the processor manages the data buffers to allow that cross boundary migration in a lossless way. This is a ratio frequency setting which is tied directly to the base frequency of the processor (the BCLK, typically 100 MHz), and can be set at 4x, 8x or 10x for 400 MHz, 800 MHz or 1000 MHz respectively.

TL:DR, Z170 mobos were running with an FCLK of 800MHz instead of intels recommended 1000MHz.

Anandtech retested and saw an increase in performance of about 1,3% with the 1000MHz setting.

Source

[Swndlr]

1.3%

 

[Ohlyver]

1.3% could still be in the error margin, let's see what driver and uefi/BIOS update bring..

[Nup]

Thanks for the TL;DR, good read though. Didn't know it was a thing.

Also, woot 1.3%. Can it be over locked more?

[MEC-777]

1.3%!!!!!!!!!!!11!!1!!!!!!!

 

[Master Disaster]

I think 1.3% is around the same number as the amount of people who have Skylake builds but don't have a dedicated GPU.

[typographie]

I think 1.3% is around the same number as the amount of people who have Skylake builds but don't have a dedicated GPU.

 

As it says right in the title, they're talking about discrete (or dedicated if you prefer) GPU performance. These optimizations deal with the speed of CPU-PCIe communications.

[Master Disaster]

As it says right in the title, they're talking about discrete (or dedicated if you prefer) GPU performance. These optimizations deal with the speed of CPU-PCIe communications.

Oh sorry. For some reason I totally misunderstood, I guess it's cause I only had about 3 hours of sleep last night.

My bad.

[lightningterror]

Anandtech retested and saw an increase in performance of about 1,3% with the 1000MHz setting.

 

Moar performance!

 

[Stomplar]

Moar performance!

 

that gif is hilarious. need more of these.

[IanCutress]

1.3% could still be in the error margin, let's see what driver and uefi/BIOS update bring..

Normally that'd be the case, but Skylake at launch numbers were down consistently (across 5 games x 5 GPUs) around -1% to -3%, some as low as -7%, compared to Haswell/Broadwell. If they were up and down, I'd agree with you, but the base line was lower than expected. Retesting with the enhanced FCLK saw consistent gains across all of the dGPU tests. 

 

Disclosure: I'm the Senior Editor at AnandTech responsible for the article linked above. Twitter at @borandi, @IanCutress etc.

 

Some people will say the game tests we used are a limited set and all GPU limited (which in this case is fine), or not a large enough sample. But testing 5 games x 5 GPUs x 3/4 tests each game takes the best part of a day, and that's when everything works. Always ready to accept new benchmarks though if they can be done consistently.