|
It seems that Abit now knows better, too. They've bolstered the Fatal1ty lineup with the Fatal1ty AN8 motherboard for Athlon 64 processors. With a 939-pin socket, nForce4 Ultra chipset, temperature-controlled OTES cooling, and a loaded uGuru-powered BIOS, the Fatal1ty AN8 could be the ultimate Athlon 64 platform for gamers, overclockers, and enthusiasts. Read on to see if it was worth the wait.

The specs
As usual, we'll kick things off with a look at some highlights from the Fatal1ty AN8's spec sheet. For a more in-depth look at the capabilities of the board's chipset, see our review of the nForce4 Ultra.
| CPU support | Socket 939-based Athlon 64 processors |
| Chipset | NVIDIA nForce4 Ultra |
| Interconnect | NA |
| Expansion slots | 1 PCI Express x16 2 PCI Express x1 3 32-bit/33MHz |
| Memory | 4 184-pin DIMM sockets Maximum of 4GB of DDR266/333/400 SDRAM |
| Storage I/O | Floppy disk 2 channels ATA/133 with RAID 0, 1, 0+1 support 4 channels Serial ATA II with RAID 0, 1, 0+1 support |
| Audio | 6-channel audio via nForce4 integrated audio and ALC658 codec |
| Ports | 1 PS/2 keyboard 1 PS/2 mouse 4 USB 2.0 with headers for 6 more 1 Firewire via Texas Instruments TS43BAB22A with headers for 1 more 1 RJ45 10/100/1000 Gigabit Ethernet 1 analog front out 1 analog bass/center out 1 analog rear out 1 analog line in 1 analog mic in 1 digital S/PDIF output (TOS-Link) 1 digital S/PDIF input (TOS-Link) |
| BIOS | Phoenix AwardBIOS |
| Bus speeds | CPU: 200-410MHz in 1MHz increments PCI-E: 100-145MHz in 1MHz increments DRAM: 100, 133, 166, 200MHz |
| Bus dividers | HT:FSB: 1:1, 2:1, 2:1, 3:1, 4:1, 5:1 |
| Voltages | CPU: 1.5-1.85V in 0.025V increments DDR: 2.8-2.8V in 0.05V increments VTT: 1.25-1.4V in 0.05V increments HT: 1.2-1.35V in 0.05V increments Chipset: 1.5-1.8V in 0.05V increments |
| Monitoring | Voltage, fan status, and temperature monitoring |
| Fan speed control | CPU, chipset, SYS, OTES1, OTES2, AUX |
For a high-end nForce4 Ultra, the Fatal1ty AN8's spec sheet is surprisingly pedestriannot that the chipset needs much company. The nForce4 Ultra has 20 lanes of PCI Express, which the Fatal1ty board splits between one PCI-E x16 slot and a pair of x1 slots. This arrangement leaves two of the chipset's PCI-E lanes untapped, but given the lack of PCI Express peripherals on the market, that's not much of a loss.
The nForce4 Ultra also takes care of all the board's storage needs, serving up two ATA/133 channels and a quartet of Serial ATA II ports. Serial ATA II drives have yet to hit the market, but users can take advantage of the Ultra's support for Native Command Queuing and the Western Digital Raptor WD740GD's Tagged Command Queuing right away. Those with multiple hard drives can also bask in the glory of NVRAID, which can span multiple RAID 0, 1, 0+1, and JBOD arrays across both ATA/133 and Serial ATA drives.
In addition to handling storage duties, the nForce4 Ultra chipset also handles the Fatal1ty board's networking capabilities. The chipset has an integrated Gigabit Ethernet controller that can accelerate both TCP/IP packet handling and stateful packet inspections in hardware, ideally lowering the CPU utilization of not only standard networking activities, but also NVIDIA's included firewall software. While the nForce4 Ultra's integrated GigE controller and bundled Firewall are no doubt impressive, it's a little disappointing that the Fatal1ty AN8 doesn't come with a second networking option. Given the board's relatively high price, it would have been nice to see a second Gigabit Ethernet port or even a Wi-Fi chip on-board.
The Fatal1ty AN8 could also use a little help on the audio front, where the nForce4 Ultra's basic AC'97 audio looks particularly feeble next to its cutting-edge networking and storage controllers. Abit could have side-stepped the chipset's audio deficiencies by employing a third-party audio chip like VIA's eight-channel Envy24, which is capable of high definition sampling rates and resolutions, but they've elected to slap on a six-channel Realtek ALC658 codec instead.
| Intel moves to achieve deep-learning Nervana with acquisition | 17 |