Should I use a higher BCLK, what about FCLK?
I decided to use a 100MHz BCLK with common multipliers and then 125MHz with reduced multipliers to match up to the 100MHz test multiplier. This effectively sets the same frequencies on all domains using two different BCLK settings so I can see if there is a benefit to running higher BCLK. I also decided to use this chance to test out the effect of FCLK clock since everything else was equal at a higher BCLK and since FCLK only has a few multipliers (4x, 8x, and 10x).
Take Away: Using higher BCLK at the same frequencies doesn't provide any noticeable gains, but also doesn't show any deficiencies. Running a higher FCLK does seem to boost 3D performance a little bit since it helps alleviate bottlenecks between the PCI-E controller (in the CPU's System Agent/Uncore) and the rest of the CPU.
What effect does cache frequency have on Performance?
CPU Performance as a result of scaling Cache (same as Uncore or Ring on some boards) frequency.
Memory Bandwidth and Latency Performance as a result of scaling Cache frequency.
In these tests, memory frequency and CPU frequency are kept the same and then the cache is changed; this is done at many relevant CPU and memory frequency ranges. Everything has been normalized so the findings can be distilled into something easily understood. While the cache frequency has negligible effects (< 1%) on CPU and memory, there does seem to be a notable increase of memory latency performance (increasing cache decreases memory latency by a small amount). This effect is increased when the memory frequency is increased (2133MHz saw ~1% increase, 3200MHz saw ~2% increase, and 4000MHz saw ~3% increase), so the importance of maintaining a higher cache frequency is more important at higher memory speeds, and negligible at lower memory frequencies.
Take Away: Cache frequency has very little impact on CPU performance, but can have a noticeable effect at higher memory frequencies by improving latency results. I would not expend too many resources (voltage/temperature) on maximizing cache over CPU or memory frequency.
What effect does scaling Memory frequency have on Performance?
CPU Performance as a result of scaling Memory frequency.
Memory Bandwidth and Latency Performance as a result of scaling Memory frequency.
In these tests, CPU frequency and Cache/Uncore/Ring are kept the same and then the memory frequency is changed, this is done at many relevant CPU and cache frequency bins. While the memory has negligible effects (< 1%) on CPU performance at higher speeds, it does seem to have an effect at lower CPU frequencies (~5% by increasing memory frequency to 3200MHz and negligible effects going from 3200MHz to 4000MHz). Scaling the memory frequency obviously increases memory bandwidth performance, and you can see that the increased primary, secondary, and tertiary timings from increasing frequency to 4GHz do negatively affect the memory bandwidth performance (I used XMP, but these timings can easily be tightened for better performance), but memory latency is improved by increasing the memory frequency (even with higher timings).
Take Away: Memory frequency has an obvious impact on the performance of memory bandwidth and latency. CPU performance benefits a lot at lower CPU frequency ranges from increasing memory frequency, and this effect is reduced as CPU frequency increases perhaps because 3.2GHz memory frequency is enough to remove any bottlenecks from stock memory frequency (2133MHz).
PRICING: You can find products similar to this one for sale below.
United States: Find other tech and computer products like this over at Amazon's website.
United Kingdom: Find other tech and computer products like this over at Amazon UK's website.
Canada: Find other tech and computer products like this over at Amazon Canada's website.
- Page 1 [Introduction to Skylake Overclocking]
- Page 2 [Disclaimer and Before You Begin Overclocking]
- Page 3 [Overclocking Flow Chart, Non-K SKUs, BCLK, and Multipliers]
- Page 4 [BCLK, Core, Cache, Memory, and FCLK Scaling]
- Page 5 [Skylake Overclocking Voltages]
- Page 6 [Power Savings and Voltage/Power Analysis]
- Page 7 [The Durability of Intel's 14nm Node]
- Page 8 [Skylake Memory Overclocking: Corsair and Frequency Scaling]
- Page 9 [Skylake Memory Overclocking: G.Skill and Memory Timings]
- Page 10 [ASRock Z170 Overclocking: Z170 OC Formula]
- Page 11 [ASUS Z170 Overclocking: Maximus VIII Extreme]
- Page 12 [EVGA Z170 Overclocking: Z170 Classified 4-Way]
- Page 13 [GIGABYTE Z170 Overclocking: Z170X-SOC Force]
- Page 14 [MSI Z170 Overclocking: Z170A XPOWER GAMING TITANIUM EDITION]
- Page 15 [Stability Testing, Delidding, Crashing, and Throttling]
- We at TweakTown openly invite the companies who provide us with review samples / who are mentioned or discussed to express their opinion of our content. If any company representative wishes to respond, we will publish the response here.
Latest News Posts
- EVGA teases its GTX 1080 Ti FTW3 Hybrid graphics card
- ARM announces new CPUs and GPU at Computex 2017
- ASUS leaks its own Strix X299-E motherboard
- You've less than 7 days to enter our 'Rings' giveaway!
- NVIDIA rumored to unveil Volta-based GeForce tomorrow
- Asrock X99 Extreme11 reports only 48GB instead of 64 GB
- H270m Pro4 Can't boot with my SSD, stuck on splash
- Gigabyte z170 UD5: Question about Voltage Spikes/Offset
- ASRock AliveXFire-esata2 compatibility with Phenom II X4 980
- m4a88td-m/usb3 & Windows 10
- Qualcomm fuels IoT growth by currently delivering more than 1 million chips a day into a wide range of connected applications
- Team Group announces theme for COMPUTEX 2017 showcase: go beyond the limit and reach for the top
- SAPPHIRE announces PULSE Radeon RX 560 graphics card
- ELITEGROUP computer to stand out at Computex for its smart campus deployment, robotic technology, education laptops, tablets, mini PC, and motherboards
- ADATA Shares a Symphony of Technology at Computex 2017