Newegg has
ASUS B650E MAX GAMING WIFI W AMD B650 AM5 ATX Motherboard + 32GB (2x16GB) Team T-Force Delta RGB DDR5 6400 CL32 Desktop Memory Kit (FF4D532G6400HC32ADC01) on sale for
$169.99.
Shipping is free.
Thanks to Community Member
BeigeRoad455 for sharing this deal.
Specs:
- ASUS B650E MAX GAMING WIFI W
- AMD AM5 Socket: Ready for AMD Socket AM5 for AMD Ryzen 9000 & 8000 & 7000 Series Desktop Processors
- Enhanced Power Solution: 8+2+1 phase power design, 6-layer PCB, alloy chokes and durable capacitors for stable power delivery
- Ultrafast Connectivity: Wi-Fi 6E, PCIe 5.0 x16 slot, PCIe 5.0 M.2 slot, Realtek 2.5Gb Ethernet, rear USB 10Gbps Type-A & Type-C ports, and front USB 5Gbps Type-C support
- Comprehensive Cooling: VRM and M.2 heatsinks, PCH heatsink, hybrid fan headers and Fan Xpert 2+
- DIY Friendly Design: Pre-mounted I/O shield, BIOS FlashBack, Q-Antenna, Q-LED Core
- Unmatched Personalization: Three Addressable RGB Gen 2 headers, all configurable with ASUS-exclusive Aura Sync RGB lighting
- CEC Tier II ready
- 32GB Team T-Force Delta RGB
- DDR5 6400 (PC5 51200)
- Timing 32-39-39-84
- CAS Latency CL32
- Voltage 1.35V
Top Comments
The only meaningful downside to this board is its 8+2+1 phase vrm, which leans towards the budget side. Even so, it should be more than sufficient for anything below a 16 core cpu (and can probably grudgingly handle an un-overclocked 16-core). The other specs are overall competent for a midrange board. It does support pcie gen5 on both the primary x16 slot and one of the m.2 slots.
The memory included as a "free gift item" is 32gb (2x16gb) ddr5 6400, Timings: 32-39-39-84, 1.35v. Considering the speed and timings, this memory almost certainly uses hynix memory chips. It's important to note that most zen4/5 cpus will NOT be stable at base expo/xmp timings for this memory in 1:1 mode, since overclocking the memory controller to 3200mhz often requires greater than 1.3v soc voltage (which is unsafe) depending on the silicon lottery.
Basic am5 memory overclocking/tuning information: https://www.youtube.com/watch?v=Xcn_nvW
These are the settings I'd personally recommend overwriting on the xmp/docp profile as a starting point (primarily derived from buildzoid ddr5 6000 hynix fallback timings, simply taking the loosest between a or m die since I don't know which this particular memory kit is):
Ddr5 speed set to 6000 (mclk will be 3000, remember "ddr" is "double data rate")
uclk div1 mode set to uclk=memclk (mclk will be 3000, uclk will be 3000)
FCLK left on auto (2000)
Voltages:
VSOC 1.25v (if you disable your igpu in bios (or have decent luck with the silicon lottery) this can likely be set lower)
DRAM VDD 1.35v
DRAM VDDQ Auto (should be 1.35v)
VDDIO Auto (should be 1.35v)
VDDP 0.99v (0.95 should be easily possible, but I've heard of very rare anecdotal cases where stability suffers below 1v even at ddr5 6000)
Timings:
tCL 32 (30 easily possible)
tRCDWR 40 (38 likely possible)
tRCDRD 40 (38 likely possible)
tRP 40 (38 likely possible)
tRAS 126 (or auto; tras doesn't appear to be used on single rank with am5)
tRC 70 (60 likely possible)
tWR 48
tRFC 512 (if your memory ever goes over 60C, you might want to raise to 544)
tRFC2 auto
tRFCSB auto
tREFI 40000 (if you memory stays below 60C raise to 50000, if it goes over 65C lower to 30000)
tRTP 16
tRRDL 8
tRRDS 8
tFAW 32
tWTRL 16
tWTRS 6
tRDRDSCL 4
tRDRDSC 1
tRDRDSD 6
tRDRDDD 6
tWRWRSCL 4
tWRWRSC 1
tWRWRSD 8
tWRWRDD 8
tWRRD 4
tRDWR 16
These are a basic starting point that should be completely stable unless you really lose the silicon lottery, there's tons of headroom for additional overclocking/tuning. Still, even with these very conservative settings, make sure to thoroughly stress test your system. Running memtest86 (free) from a flashdrive is good as an initial test to check for obvious hardware defects in all memory chips, but it's not very good at picking up errors due to overclocking. For free stress testing, at a minimum I'd recommend prime95 large ffts (4+hrs, 8+ preferred) and y-cruncher vt3 (4+hrs, 8+ preferred). For additional validation you could try occt memory and cpu+ram large dataset tests, hci memtest, and tm5 (testmem5) with 1usmus and absolut configs.
5 Comments
Sign up for a Slickdeals account to remove this ad.
Our community has rated this post as helpful. If you agree, why not thank BeigeRoad455
The only meaningful downside to this board is its 8+2+1 phase vrm, which leans towards the budget side. Even so, it should be more than sufficient for anything below a 16 core cpu (and can probably grudgingly handle an un-overclocked 16-core). The other specs are overall competent for a midrange board. It does support pcie gen5 on both the primary x16 slot and one of the m.2 slots.
The memory included as a "free gift item" is 32gb (2x16gb) ddr5 6400, Timings: 32-39-39-84, 1.35v. Considering the speed and timings, this memory almost certainly uses hynix memory chips. It's important to note that most zen4/5 cpus will NOT be stable at base expo/xmp timings for this memory in 1:1 mode, since overclocking the memory controller to 3200mhz often requires greater than 1.3v soc voltage (which is unsafe) depending on the silicon lottery.
Basic am5 memory overclocking/tuning information: https://www.youtube.com/watch?v=Xcn_nvW
These are the settings I'd personally recommend overwriting on the xmp/docp profile as a starting point (primarily derived from buildzoid ddr5 6000 hynix fallback timings, simply taking the loosest between a or m die since I don't know which this particular memory kit is):
Ddr5 speed set to 6000 (mclk will be 3000, remember "ddr" is "double data rate")
uclk div1 mode set to uclk=memclk (mclk will be 3000, uclk will be 3000)
FCLK left on auto (2000)
Voltages:
VSOC 1.25v (if you disable your igpu in bios (or have decent luck with the silicon lottery) this can likely be set lower)
DRAM VDD 1.35v
DRAM VDDQ Auto (should be 1.35v)
VDDIO Auto (should be 1.35v)
VDDP 0.99v (0.95 should be easily possible, but I've heard of very rare anecdotal cases where stability suffers below 1v even at ddr5 6000)
Timings:
tCL 32 (30 easily possible)
tRCDWR 40 (38 likely possible)
tRCDRD 40 (38 likely possible)
tRP 40 (38 likely possible)
tRAS 126 (or auto; tras doesn't appear to be used on single rank with am5)
tRC 70 (60 likely possible)
tWR 48
tRFC 512 (if your memory ever goes over 60C, you might want to raise to 544)
tRFC2 auto
tRFCSB auto
tREFI 40000 (if you memory stays below 60C raise to 50000, if it goes over 65C lower to 30000)
tRTP 16
tRRDL 8
tRRDS 8
tFAW 32
tWTRL 16
tWTRS 6
tRDRDSCL 4
tRDRDSC 1
tRDRDSD 6
tRDRDDD 6
tWRWRSCL 4
tWRWRSC 1
tWRWRSD 8
tWRWRDD 8
tWRRD 4
tRDWR 16
These are a basic starting point that should be completely stable unless you really lose the silicon lottery, there's tons of headroom for additional overclocking/tuning. Still, even with these very conservative settings, make sure to thoroughly stress test your system. Running memtest86 (free) from a flashdrive is good as an initial test to check for obvious hardware defects in all memory chips, but it's not very good at picking up errors due to overclocking. For free stress testing, at a minimum I'd recommend prime95 large ffts (4+hrs, 8+ preferred) and y-cruncher vt3 (4+hrs, 8+ preferred). For additional validation you could try occt memory and cpu+ram large dataset tests, hci memtest, and tm5 (testmem5) with 1usmus and absolut configs.