|
|
Leadtek K7NCR18D-PRO Motherboard 17th February 2003 Today we are reviewing Leadtek's flagship Athlon motherboard. This board is based on Nvidias NForce2 chipset which is generally acknowledged to be better performing than it KT400 rival. Let's take a look at what extras come with the board.
While many manufacturers leave out additional USB connectors as "optional extras" we find here that everything is provided including digital outputs for sound and firewire connectors and a Serial ATA cable. The Dolby Digital output module is particularly welcome as it allows the decoding to be done by an external source such as an A/V Amp which tend to have several digital inputs but usually only one set of discrete phono input (the six phono inputs you need for DTS). Having the IEEE1394 connectors on a separate card will please those that don't need them as they then needn't be installed, saving a PCI slot. There is a counter argument for having them on the back panel but there would really only be room for one without sacrificing USB 2.0 capability and this isn't really enough for people who make use of these connectors (digital video editing, external devices etc).
The board
itself looks quite spacious and should be able to accommodate the largest of heat sinks.
We prefer the use of passive cooling for chipsets as it's often the small fast
fans that create the most annoying high pitched noise (this is inconsequential
if you're going to use a screaming Delta on the CPU). Here are the
manufacturer's specifications: Form
factor
System
BIOS
One of the notable omissions is Serial-ATA which has become common-place these days but is not really going to be missed until S-ATA drives start appearing in volume which is unlikely to happen before the end of the year. Another feature that is common on other boards is the inclusion of some form of RAID controller - particularly on high-end boards. We have found built-in RAID to be quite useful although many RAID manufacturers (Promise being the main one) have started imposing severe restrictions on these built-in controllers, we have noticed that Albatron boards are now crippled by having only one RAID channel thanks to Promise. At this point in time our advice with RAID is to buy a separate controller when it is needed - you may not need it until S-ATA RAID controllers become more common thereby solving both these problems. The manual is concise and well written without inundating the reader with large amounts of spurious information as is the case with AOpen manuals. Beginners should have no difficulty in starting installation within minutes of reading the manual. The ability to lock AGP and PCI dividers in BIOS is particularly welcome by overclockers as are the generous voltage ranges for CPU and Memory. Bundled software is good for playing around with firewire capture and basic editing of home movies but not what is required by professionals - they would have their own software in any case, so the basic packages provided may serve as a useful introduction into media creation for novice users. Looking at the I/O panel on the board we see the usual connectors.
Combined with the extra back panel connectors (included) there are plenty of connectors to suit every requirement.
Test Set Up To test this board we used an Athlon XP3000+ which has the new Barton core as well as an Athlon XP2700+ with 2 sticks of Crucial PC2700 memory and a Creative Ti4400 graphics card. We were able to successfully over-clock the board to board to 183MHz giving us a speed of almost 2.4GHz and everything was very stable at that speed. This is largely down to the ability to fix the AGP and PCI bus speeds in the BIOS and is sure to please over-clockers in particular. All the available pre-sets worked fine and tests were conducted at CAS2. It should be noted that we also performed tests with TwinMOS PC3200 (DDR400) memory but actually got worse performance due to the asynchronous FSB and memory speeds so all our tests are conducted with PC2700 RAM.. Boot time was very good once XP was installed and we ran the usual SiSoft and Mad Onion benchmarks along with Divx tests. The latter may sound like a CPU test but is actually more a function of bandwidth and latency efficiency. Here's what we found.
SiSoft Sandra We ran CPU Arithmetic, CPU Multimedia, Cache Memory and Memory Bandwidth benchmarks. The board performed as expected without problems. Again we see another solid performance as expected from an NForce2 based board. Another decent result. Results are as expected for this type of board and it is only beaten by the much higher bandwidth of the i850 chipset. Overall the synthetic results tell us that the board performs as well as any other board of this type to within about 1% of other similar boards.
3DMark and PCMark These should be more interesting. This is a phenominal score for a Ti4400 card and shows the efficiency of the NForce2 design and the dual memory channels. Although AGP 8x is supported by the board we have yet to see it make a significant difference during testing. The motherboard has a warning light if the wrong type of AGP card is inserted and a retainer clip to hold the card securely in place. Let's see how it performs with PCMark2002. We see very good results in this benchmark as well since any deficiency in the board would impact one of these three areas.
Unreal Tournament The performance of the board with gaming benchmarks needs to be determined so we ran UT2003 which will have to do until Doom3 becomes available. The K7NCR18D-PRO results are the ones listed as NForce2 since the official title is too long to use as a label.
This board should really please gamers as Intel's flagship is comprehensively beaten at the higher resolutions. This can be largely attributed to the dual channel controller. Although the 166MHz FSB only uses half the available capacity, the other half is available to the AGP bus etc. This contrasts with the carefully matched buses of the P4 which although they operate at full efficiency suffer from having bandwidth drawn away to the AGP card which degrades performance at higher resolutions (the higher the resolution - the greater the required bandwidth by the AGP bus).
The results are even more pronounced here beating the P4s by an even higher margin.
Divx Encoding Let's turn to an area where we know that fast CPUs will make a difference - Audio and Video encoding. This is becoming more and more popular and is very computationally intensive with long processing times (relatively speaking that is, this field is not for those that complain about how long their Outlook Express takes to load). For consistency we will use Jet Li's The One as our test matter. It is not interlaced and contains a mixture of action types and is not too long. There will be three tests all using Divx 5.02. Audio will be encoded separately. I will try and keep my commentary to a minimum as all configuration information is shown in the images below.
Xmpeg 4.5 Firstly we will start with the industry standard Xmpeg with its settings at those recommended by the Divx community and as used by other sites such as Toms Hardware. Here are the CODEC settings: and here are the Xmpeg settings: and here are the results
The results show that bandwidth is king here and video encoders would be better off with a dual CPU rig than a single Athlon. We are hoping to have a review of a Dual Athlon MP2600+ system soon.
AviSynth and VirtualDub No serious Divx encoder uses Xmpeg alone and it's just used by the media for benchmarking purposes so let's get serious. We ripped our source material to hard disk and created a DVD2AVI project file using forced film (it was 99% film). Loading this into Gordian Knot we first saved an .avs file with no changes at all (720x480) and no filters of any sort. This was loaded into VirtualDub with the following CODEC parameters : After encoding we got these results:
Here we see the P4 pull ahead again mainly due to the P4 optimizations in VirtualDub. The XP3000+ is on a par with the Dual Athlon MP1800 rig in this test where cache memory is not important.
This is all good stuff but how about a real-world test? To simulate a realistic test we added a neutral bicubic resize filter in the .avs file and used the following CODEC parameters (including two popular Pro settings) which are designed to total 700MB (when the audio is muxed in): Which resulted in the following.
The addition of computationally heavy filters put less emphasis on the memory bandwidth and more on raw CPU speed. It also shows that the extra cache on the XP3000+ helps. This difference would be greater with more filters used in the encoding.
Audio Encoding What about audio? We took the AC3 track from the above sample material and used HeadAC3he to convert it into Vorbis format so our final muxed file could have Ogg containment. There isn't space here to go into the advantages of Ogg Vorbis over MP3 and AVI so let's just say that Vorbis sounds about the same as MP3 for half the file size or twice as good for the same file size (that is subjective though). These are the settings we used: Since it is more meaningful to show throughput than time taken (which depends on the length of the source) we display the results thus:
For the second time we had to take a step back in surprise. The increased cache of the XP3000+ has resulted in a whopping 33% increase in performance in this computationally intensive benchmark. Even the 3.06GHz P4 (in a test where Hyper-Threading adds a 20% increase) is left trailing in the dust. We also tested the digital audio output of the board by connecting the digital output to our Home Theatre system using The Matrix and Saving Private Ryan as our source material and WinDVD 4.0 as the software DVD player. Overall the results were very impressive with each bullet hitting the floor in the lobby scene being discernable in the surround speakers and every explosion in the SPR Normandy landings scene being accurately reproduced by the subwoofer. This performance is easily on par with a mid-range DVD player.
Conclusion There can be no doubt that NForce2 boards are the best solutions for Athlon processors. The extra benefit of a dual channel memory design shows clear advantages in most tests. Although PC2700 memory performed better than PC3200 memory due to synchronization issues we would still recommend PC3200 memory to anyone considering this type of purchase in conjunction with an NForce2 board as you can just run it at a lower speed while having enormous potential for overclocking (just make sure your FSB and memory bus speeds are the same). Having established the supremacy of the NForce2 design what about the K7NCR18D-Pro? We found it to be an excellent board at a very reasonable price point. While it's true that it doesn't offer extra features such as Serial-ATA or on-board RAID, these features push up the price and are not used by most people. Stability was rock solid and we were able to use memory timings that were extremely aggressive. Overclocking potential was also very good and we were able to push our XP3000+ to 2.4GHz with just air cooling. We would recommend this board to anyone looking for performance and high stability - particularly when pushed to its limits and will be using it as our de facto NForce2 test board for future tests. We would like to thank Leadtek UK for the review sample motherboard.
All trademarks are the property of their respective owners. |
|
||