Author: Terren Tong
Editor: Howard Ha
Publish Date: Tuesday, February 22nd, 2005
Originally Published on Neoseeker (http://www.neoseeker.com)
Article Link: http://www.neoseeker.com/Articles/Hardware/Reviews/dfinf4ultralp/
Copyright Neo Era Media, Inc. - please do not redistribute or use for commercial purposes.
The transition between AGP and PCIe has also brought out a crop of new chipsets from everybody including NVIDIA's nForce 4 platform which brings arguably the richest feature set out of the current generation of K8 boards. The current chipsets that have recently hit the market are also of much interest to consumers as it finally brings PCIe and socket 939 into one package both of which are seen as longer term technologies compared to socket 754 and AGP, something that a lot of enthusiasts have been holding out for.
While there is never really any shortage of motherboard manufacturers, the same cannot be said about enthusiast level boards. While most manufacturers do realize that overclocking is not merely just a fringe hobbyist activity, and do include some tweaking options, there are few that go all out and really let loose the options in the BIOS to allow the end user to really break something. One of the fast rising stars in the enthusiast department is DFI. After setting their mark with their LanParty nForce 2 Ultra board, as the enthusiast board to get for the Socket A platform, DFI has further extended their LanParty lineup to both the Socket 754 and Socket-T line ups all of which have been well received.
Today we take a look at one of DFI's entries into the Socket 939 and PCIe market in the LanParty UT nF4 Ultra-D. DFI combines the rich feature set of the nForce 4 Ultra chipset with their rich overclocking palette along with a few extras like the additional x16 PCIe slot. DFI looks like they may have the combination for success and we'll take a closer look at the nF4 Ultra-D right now.
|
CPU
|
|
Chipset
|
|
Front Side Bus
|
|
Memory
|
|
BIOS
|
|
Power Management
|
|
Hardware Monitor
|
|
Audio
|
|
LAN
|
|
IDE
|
|
Serial ATA with RAID
|
|
IEEE 1394
|
|
Rear Panel I/O Ports
|
|
I/O Connectors
|
|
Expansion Slots
|
|
PCB
|
DFI and the nForce 4 chipset have most the basics covered here including the standard four SATA ports along with two IDE channels. One of the nice things about the nForce 3 and 4 chipsets is the ability to span RAID partitions across both PATA and SATA. DFI has implemented a dual Gigabit ethernet connection, one provided through the nForce 4 MCP and the second with the ubiquitous Marvell chip. The nForce 4 LAN connection is more feature complete as it includes NVIDIA's ActiveArmor and second generation firewall. Things get interesting quickly however as we pore through the rest of the feature set.
Audio is provided through what DFI calls the Karajan audio module - when the motherboard is unwrapped the back panel looks a bit bare, there are the standard PS/2 ports, SPDIF in and out, dual Gbe, a single 6 pin firewire connection, and 6 USB2 ports. Conspicuously absent are the audio jacks. The Karajan audio module provides 8 channel audio courtesy of the ALC850 codec through six audio jacks. Instead of having the ALC850 chip on the actual motherboard PCB, it is moved onto the Karajan module, presumably to reduce the effects of noise from other motherboard components.
The Karajan audio module powered by an ALC850 chip
The second big surprise is the presence of the second PCIe x16 slot - remember this is an Ultra chipset that we are looking at today, not a SLI. DFI has something that they are touting called Dual Xpress Graphics (DXG), which simply allows the addition of a second video card for an additional two displays for desktop output. Unfortunately, the official line is that it will not support SLI as there is no SLI bridge piece included with the board. By default, the lane configuration is set to x16/x2 for x16 slots but the infamous pencil trick will change it to a x8/x8 configuration. There is talk that NVIDIA will be cracking down on modded Ultra chips so buying an Ultra board in hopes of modding it to a SLI board are hampered by what NVIDIA can do on the driver side and the ability of end users to get a hold of the SLI bridge piece.
DFI is not done with their bag of PCIe tricks yet however. Astute readers would have noticed that there is a x4 and a x1 slot in addition to the two x16 slots. The nForce 4 supports a maximum of 20 PCIe lanes so the math does not quite work out. This will get confusing fast, so bear with me a minute. As mentioned previously, one of the x16 slots actually has only x2 lanes as far as bandwidth goes. The x4 slot is allocated x1 worth of bandwidth and the x1, x1. This adds up to exactly twenty, the maximum supported by the nForce 4. However, through a set of jumpers this can be reconfigured. We noted previously the mod trick will change the lane configuration of the two x16 slots to x8 each. The jumper can also change give the x4 slot full x4 bandwidth. A table is conveniently silkscreened onto the motherboard. Officially only the nForce 4 SLI board supports reconfigurable PCIe lanes but the sneaky engineers over at DFI seem to have put in some OT to give us end users something that is pretty neat. Big thumbs up from us for further future proofing technology.
PCIe Lane settings are silkscreened on the board
While the full LanParty comes with the kitchen sink, the UT series scales back and is more barebones as far as the packaged accessories go. Included are a pair of SATA cables, a pair of rounded IDE cables, a floppy cable, a SATA driver floppy, a molex to dual SATA power cable, a jumper removal tool, warning sheet, the manual and two baggies of extra jumpers. End users are left to their own devices as far as more firewire and USB ports are concerned though that may not be a big issue as a good number of cases include built in USB if not firewire.
A nice touch on the driver CD are recent drivers for both NVIDIA and ATI cards, .NET 1.1 for Catalyst Control Center users and DX9c. Sure beats redownloading this stuff everytime a format comes around.
The nF4 board looks great. The board itself is decidedly enthusiast oriented with the black, yellow and orange color scheme, it does not feel over the top. The color scheme feels more lively than the Kawasaki green used on the nF2 Ultra II board. Though color preferences are highly subjective, I think that most enthusiasts will like what DFI has done with the nF4. It is not as gaudy as Gigabyte boards and I prefer it over Soltek's purple. DFI has kept up with the enthusiast / modder philosophy as all the plastics and the cabling are UV reflective.
DFI does a good job on the board layout and it is obvious they put a lot of thought into it but it is not perfect and we do have some gripes. The DIMM slots are at the top of the board, like the MSI K8N Neo board we took a look at last year. The both the ATX and the 12v power connection sit beside each other on the right edge of the board which should make things easier for those who are sticklers are cable cleanliness. The ATX connector is actually a BTX style connector with the 4 extra pins. The nF4 Ultra-D require an additional molex connector to be plugged in for even more juice and is located further down the board, above the fan that covers the MCP. Though this is a bit out of the way it is not horribly places as it resides between the PATA and SATA ports so the molex can be bundled with some of those cables.
The expansion header placement could use a bit of work. The firewire headers are located just below the back panel near the edge of the board. The placement there is not the most ideal as it may be competing for space with the PCIe slots on the back of the case. The headers for the USB ports are beside the first classic PCI slot. It would be better if they were closer to the edge of the board so cabling can be a bit cleaner.
The single x1 PCIe slot is sandwiched between the two x16 slots which is similar to what the Gigabyte nForce 4 SLI also does. There is very little clearance with a fairly low profile single slot card like the 6600GT and the expansion slot is a write off with any sort of taller card. It is probably not too feasible from the economics side to lengthen the board, but with the extra x16 port, there is definitely some compromise for the x1 slot placement.
Because enthusiasts like their fans, DFI has included not one but five fan headers scattered around the board. LED diagnostic readouts are available in the corner of the board. While they do their job adequately, I've really grown to like the numbered readout found on the Soltek FGR and EpoX boards. One awesome feature is the built in power button and reset switches located by the front panel connectors. For the tinkerer it sure beats shorting out the front panel leads. The DR versions of the nF4 boards also come with an extra four SATA ports courtesy of a Sil3114 controller. The markings and chip position are outlined on the board but none of the connectors are present on the Ultra-D revision.
The MCP fan unit is also of note. DFI specifically mentioned this in their PR material. The fan is neither a sleeve or ball bearing fan like most but is a mag lev fan. DFI makes several claims about this fan
The longevity issue is something of note. Too often do manufacturers put junk fans on motherboards that break within months and make a huge racket. It would be interesting to visit this issue again a few months down the road and see how well it has held up.
Another core strength of DFI's engineering prowess is the BIOS and the options available on the nF4 Ultra-D should make any enthusiast salivate. Those that like to tweak their memory in particular will have a field day. The number of memory adjustment options available is almost overwhelming.
All of the overclocking settings are stored under the Genie BIOS menu. There are a lot of things that stand out but several in particular will be highlighted. The front side bus is adjustable from 200-465. While a lot of manufacturers have absurdly high FSB values, there are few that can claim that the majority of the FSB range is usable. DFI will be one of those that can make such a claim as will be seen later on.
Voltage adjustments for the CPU is pretty unique. Besides the standard 0.0125 voltage stepping, DFI also has something an additional fine tuner if you will - a multiplier can be applied to the base voltage to fine tune the actual voltage fed to the CPU in even smaller increments. The ranges are listed in the table and with the highest multiplier of 1.36, a whopping 2.108v can be applied to the processor. Voltages are shown on the Genie BIOS page which is good but the functionality could be improved - DFI shows the current voltage setting that was saved. It might be more useful to show the default voltage instead since the set voltage is already shown through the voltage selection menu.
 | 
|
Memory voltage can be cranked up to 4.0v. Because the potential for damage is there, DFI limits the range by default from 2.5v to 3.2v. A jumper on the motherboard has to be set in order to unlock the 3.2-4.0v range. DFI allows for what seems to be every single adjustable option with regards to memory.
 |  |
The large set of memory dividers is also very welcome as this will allow again for better fine tuning of the memory when the FSB starts hitting the stratosphere. To help test for memory stability, DFI also has MemTest86+ built in as a BIOS option. With this enabled, after the post, MemTest86+ will launch. No CD required, all of this is stored on one of the onboard flash memory chips. How cool is that?
In addition to the very impressive voltage range and options, there is also the DFI exclusive, CMOS reloaded. After the initial post screen and boot detection, settings that are bootable are saved to the SEEPROM. The user then has the option of naming and saving this, or any other configuration with the CMOS reloaded configuration. Four save banks are available and at the whim of the user, each of these sets can be loaded.
Here's a not so quick summary of the BIOS settings and values available-
| FSB Bus Speed | 200-456 Mhz in 1 Mhz increments |
| HyperTransport Ratio | 1-3 in 0.5 steppings 3-5 in integer steppings |
| HyperTransport Bandwith | 8/8, 16/8, 8/16, 16/16 |
| CPU Multiplier Ratios | 4 to CPU Max in 0.5 steppings |
| PCI Express Frequency | 100-145Mhz in 1 Mhz steppings |
| CPU Voltage | 0.800-1.550 in 0.0125v steppings; voltage multiplier settings 1.00, 1.04, 1.10, 1.13, 1.23, 1.26, 1.33, 1.36 |
| HyperTransport Voltage | 1.20v to 1.50v in 0.1v steppings |
| Chipset Voltage | 1.5v to 1.8v in 0.1v steppings |
| DRAM Voltage | 2.5v to 4.9v in 0.1v steppings |
| Memory Divider | 1:2, 3:5, 2:3, 7:10, 3:4, 5:6, 9:10, 1:1 |
| Command Per Clock | On/Off |
| CAS Latency Control (TCL) | 1.0-4.5 in 0.5 steppings |
| RAS Latency Control (TCL) | 0-7 Bus Clocks |
| Min RAS active time (Tras) | 0-15 Bus Clocks |
| Row Precharge Time (Trp) | 0-7 Bus Clocks |
| le time (Trc) | 7-22 Bus Clocks |
| Row Refresh Cycle Time (Trfc) | 9-24 Bus Clocks |
| Row to Row Delay (Trrd) | 0-7 Bus Clocks |
| Write Recover Time (Twr) | 2-3 Bus Clocks |
| Write to Read Delay (Twtr) | 1-2 Bus Clocks |
| Read to Write Delay (Trwt) | 1-8 Bus Clocks |
| Refresh Period (Tref) | 16-4708 Cycles; varied stepping size |
| Write CAS Latency (Twcl) | 1-8 Bus Clocks |
| DRAM Bank Interleave | On/Off |
| DQS Skew Control | Increase/Decrease Skew |
| DQS Skew Value | 0-255 |
| DRAM Drive Strength | Level 2-8 |
| DRAM Data Drive Strength | Level 1-4 |
| Max Async Latency | 1-15ns |
| Read Preamble Time | 2-9.5ns in 0.5 ns steppings |
| Idle Cycle Limit | 0-256 Cycles; stepping increments are double the previous value |
| Dynamic Counter | Enable/Disable |
| R/W Queue Bypass | 0,4,8,16 |
| Bypass Max | 0-7 |
| 32 Byte Granularity | 4 or 8 Bursts |
Hardware & Benchmark Setup
Benchmark Setup
Our benchmark system consisted of the following:
NVIDIA's Forceware 66.93 was used for the 6600GT across all platforms. DFI's Board was tested with the ForceWare 6.37 drivers. VIA's Hyperion 4.55vp1 and the driver revisions for the nForce 4 SLI board can be found here.
A list of benchmarks performed follows:
Business Winstone 2004
We have a dead heat in Business Winstone on all three of the PCIe K8 boards we look at today. The nF4 Ultra-D brings up the rear despite being just 0.1 Winstone units away from both the Gigabyte SLI and Soltek K8T890 boards.
Multimedia Winstone 2004
There is a bit more of a difference in the Multimedia Winstone test. The nForce 4 boards remain tied but they are both slightly faster than the K8T890.
SiSoft Sandra Memory Test
Memory bandwidth is very close between the Soltek and the DFI boards. The Gigabyte is quite a bit more sluggish and brings up the rear
Three different drives from three different manufacturers are used in our HD I/O tests so scores are relative between boards, not between drive interfaces.
IDE HDTach
Results are pretty similar. The DFI board seems to take a slightly higher hit on the CPU usage front though when margin of error (+/- 2%) is factored in, all the boards are more or less equal.
SATA HDTach
The nF4 Ultra-D is on the flip side for the SATA test. Burst rates are slightly higher than both the K8T890 and the SLI board and CPU usage is lower as well. Again with the margin of error factored in things are pretty close all around.
USB2 Throughput
USB2 throughput is in line with the Gigabyte SLI board but CPU usage is significantly higher. The nForce 4 takes second behind the K8T890. DFI's BIOS includes USB tweaking options but those were not touched - these results were obtained with the stock settings.
LAN Testing
With the nForce 4 MCP's ethernet connector, CPU usage is slightly lower than that of both the Soltek VIA board as well as the K8NXNP-SLI board. CPU usage was pretty consistently the same with the firewall and TCP off-loading set to on regardless of the firewall setting.
We see some very impressive results on the throughput test. DFI manages to hit the 1Gbps barrier and trounces both the Gigabyte and Soltek boards.
There are still some oddities with the nForce 4 firewall/TCP offloading feature - while the processor hit with the firewall on is approximately the same as what we see with the VIA chipset with no firewall, processor usage jumps up significantly with BOTH TCP offloading and the firewall set to off. In the case of the DFI, processor usage shot up to around 72% compared to the high 30s, low 40s when run with TCP offloading. This is a bit of a concern as we do not see this kind of processor usage on any other platform including the older nForce 3 250GB, VIA boards or even the Intel platforms. We touched upon this issue in our Gigabyte K8NXP-SLI review so this is not just a DFI issue. We have a good idea of why it is doing this and we have been working with NVIDIA to resolve this issue.
The second Marvell based ethernet connector seems to lie on the PCI bus with the throughput hitting 751Mbps only. CPU usage was lower though at roughly 33%.
Sound Testing
The difference between 1st/2nd and 2nd/3rd is within the 1% range but we're at nearly a 3% gap when comparing the nF4 Ultra-D to the SLK890Pro.
The differences in Directsound 3D are smaller with everyone coming in within 1.3% of each other. Though the Karajin sound module should produce sound with less noise than a traditional implementation, it still relies on the ALC850 codec so it is not surprising that results are so similar.
Call of Duty has the nF4 Ultra a touch behind the Soltek board with the SLI board bringing up the rear. With EAX2 on the positions change up a bit with the Soltek retaining the lead while the SLI and Ultra boards change positions.
MP3 Encoding
Using LAME version 3.90.3 to encode a 700+ meg wave file takes roughly 15 minutes for all three of the boards. The SLI board is slightly faster to the tune of 3 seconds
Xvid Encoding
A Xvid encode shows more differences than the MP3 test. Over a course of 21 minutes, the nF4 Ultra is a couple seconds faster than the VIA Soltek board but is nearly 30 seconds faster than the Gigabyte SLI board when it comes to this test.
Comanche 4
Comanche 4 remains a pretty reliable indicator of triangle throughput to the video card. Results are very close but a slight edge goes to the nF4 Ultra.
Halo
There is a pretty big gap in Bungie's PC to Xbox and then back to PC shooter. The Soltek K8T890 takes a big leap in front of the nForce 4 boards but the DFI distances itself firmly in front of the Gigabyte SLI board.
X2:Rolling Demo
Like with Halo the K8T890 takes the front while the DFI and Gigabyte trail.
Call of Duty
Both the nForce 4 boards definitely have the upper hand in the Quake 3 engine based shooter.
UT2k4
The nF4 Ultra-D board takes the lead and has about a 4% lead over the Gigabyte SLI board in a very CPU bound test.
Half-Life 2
Very close score between the nF4 Ultra-D and the K8T890. The SLI board trails behind a bit once againn.
DOOM 3
Like with both UT2k4, and DOOM 3, scores remain very close between the nF4 Ultra-D and the Soltek board but the Gigabyte board is about 4% behind.
We've seen a continual progress in the ability to push the FSB on the different 939 boards that we have taken a look at in the past months. Some hit the 250 mark, the Gigabyte hit the high 270s mark and the Soltek K8T90 even managed to push 280. There were many reports that DFI's nForce 3 250Gb board could routinely hit about 300Mhz on the FSB. All of that is nothing though compared to what the nF4 Ultra-D is capable of - it blitzes past the 300 Mhz mark easily. Prime95 would fail within a few minutes at 369 - at 368 it took about 6 hours before an errors showed up. Playing it a bit conservatively, we ended up dropping the FSB down to 365 and the nF4 ran Prime95 for over 48 hours before we concluded that the board and processor were completely stable. We had access to three different processors, all 90nm Winchesters, one would refuse to work at FSB frequencies above 343 while the other two were stonewalled at 368. We had no luck even when raising the chipset and HTT voltages. We cannot conclusively say whether it was the processor or the board that was responsible for the Prime95 errors. What we can say is that we were still able to boot into Windows with a 375Mhz FSB. There is no doubt that the DFI nF4 Ultra will be one of the best, if not the best, board for the hard core overclocker. We touched on the fact that DFI has an unbelievable number of memory options and a great selection of memory dividers both of which will be important for maximizing memory performance at high bus speeds. Those with watercooled or more exotic cooling solution will like the ability to throw up to 2.108v at the processor.
The DFI LanParty UT nF4-Ultra-D is going to make some major waves. The board design is top notch - putting the memory above the processor socket is a good idea as there is arguably more airflow there than on the far side of the board. DFI makes a concerted effort to get most of the cabling in a single spot. For those who put value into the aesthetics of their case and boards, the nF4 has a slick color theme that is not overly flashy until the blacklight is turned on at which point all the plastics start to glow. We liked the fact that a PCIe 4x slot was included along with two x16 and a single x1 which provides more flexibility when PCIe expansion cards hit the market en masse. Dual Xpress Graphics is a nice option for those that are not looking for SLI. The second x16 slot seems to have put some constraints on the board design however. The x1 slot is extremely close to the x16 slot and taller cards will block off access to it. DFI has a pretty nifty jumper system to reconfigure the PCIe lanes. I would have preferred that the onboard USB headers were a bit closer to the edge of the board to limit the amount of cabling that needs to run across the board but that is one of the few complaints I have about the design. The nF4 Ultra-D can be modded into a SLI board but the SLI bridge piece cannot be purchased separately and is manufacturer dependant so this may be a tricky route for those looking to save a few bucks.
On the performance side, the nF4 proved to be no slouch and was generally faster than the Gigabyte SLI board that we took a look at earlier. All in all though, most of the K8 boards seem to perform pretty similarly at stock speeds.
While the DFI is a fine choice for people looking for stability at stock speeds - it is the overclocking prowess that will send droves of enthusiasts to this board. DFI has a BIOS that is not likely to be rivaled by any other company in the short term. Memory options can be as simple or complicated as the end user wants, voltages that can definitely break things are available for the more adventurous. DFI has gone all out with the BIOS to allow enthusiasts to maximize the performance of their systems. What can end users expect? How about stable FSB speeds in the 360, 370 range? The nF4 Ultra-D had been delayed a couple times so that DFI could make some last minute component changes and if our sample is any indication, this should be one of the best overclocking mainboards available.
Fast, stable, great looks and phenomenal overclocking abilities are highlights of the LanParty UT nF4 Ultra-D. It is hard to highlight all the positives about an excellent product without sounding too gushy but the nF4 Ultra-D is just that good. Definitely a must have for anyone serious about overclocking.
Win the DFI LanParty Ultra-D and a Venice 3800+! Neoseeker & NCIX are teaming up to give away the Venice 3800+ along with a DFI LanParty Ultra-D and Thermalright XP-120!
Please do not redistribute or use this article in whole, or in part, for commercial purposes.