Why you can trust Tom's Hardware Our expert reviewers spend hours testing and comparing products and services so you can choose the best for you. Find out more about how we test.
Power, Heat, Efficiency And Overclocking
Does a network controller pull more energy than a USB 3.0 controller? These are the major differences between ASRock’s and EVGA’s hardware configurations, and both ASRock boards require slightly more power.
A look at the voltage regulator temperatures might be revealing, as the more power-hungry ASRock X99M-series runs a little warmer than the X99 Micro. Whatever the cause for added hunger, the X99M Extreme4 and Fatal1ty X99M Killer end up around 2% below average in efficiency.
Note that all averages are for all X99 motherboards reviewed to-date. It’s a running tally that will cause class-average power consumption to go up as we address boards with better feature sets.
Overclocking
| BIOS Frequency and Voltage settings (for overclocking) | |||
|---|---|---|---|
| Row 0 - Cell 0 | ASRock X99M Extreme4 | ASRock Fatal1ty X99M Killer | EVGA X99 Micro (131-HE-E995-KR) |
| BIOS Version | P1.50 (09/04/2014) | P1.60 (09/04/2014) | 5.6.5 (09/19/2014) |
| Base Clock | 90-300MHz (0.1MHz) | 90-300MHz (0.1MHz) | 80-250MHz (0.05MHz) |
| CPU Multiplier | 12-120x (1x) | 12-120x (1x) | 12-80x (1x) |
| DRAM Data Rates | 800-2666 (200/266.6MHz) | 800-2666 (200/266.6MHz) | 1200-2666 (200/266.6MHz) |
| CPU Vcore | 0.80-2.00V (1mV) | 0.80-2.00V (1mV) | 0.80-2.00V (5mV) |
| VCCIN | 1.20-2.30V (10mV) | 1.20-2.30V (10mV) | 1.50-3.05V (1mV) |
| PCH Voltage | 0.90-1.50V (5mV) | 0.90-1.50V (5mV) | 0.90-1.60V (1mV) |
| DRAM Voltage | 1.00-1.80V (10mV) | 1.00-1.80V (10mV) | 0.80-2.00V (1mV) |
| CAS Latency | 4-31 Cycles | 4-31 Cycles | 5-31 Cycles |
| tRCD | 5-31 Cycles | 5-31 Cycles | 5-31 Cycles |
| tRP | 5-31 Cycles | 5-31 Cycles | 5-31 Cycles |
| tRAS | 10-63 Cycles | 10-63 Cycles | 10-63 Cycles |
Based on the same circuit board, the X99M Extreme4 and Fatal1ty X99M Killer have the same overclocking options and ranges in firmware. Like the EVGA X99 Micro, these boards are able to hit 4.4GHz at 44 x 101MHz BCLK and 1.28V core, but aren't stable at 45 x 100MHz BCLK even with a 1.30V core. Because of this, I’ve settled with a 1.28V core as the new reference setting.
EVGA stood out as the best memory overclocker in its review, and retains that position here. The Fatal1ty X99M Killer surprises with a slightly higher BCLK at the processor’s 1.25x internal core/uncore strap ratio, and the biggest cause for surprise is that it outpaces its nearly-identical sibling.
The ASRock boards have better memory bandwidth at DDR4-2666, just as they had better bandwidth at default SPD setting (DDR4-2133). EVGA might be getting its DRAM overclocking advantage through looser timings.
Alarming amount of Texas Instrument chips found in Russian-based weapons in Ukraine — Russian military using third parties to purchase U.S-made chips
Aoostar AG02 eGPU dock with OCuLink support and 500W PSU announced for $219
China levels anti-trust charges against Nvidia – country accuses chipmaker of violating anti-monopoly regulations, examining the $7 billion Mellanox acquisition
