nForce4 SLI Roundup: Painful and Rewarding
by Wesley Fink on February 28, 2005 7:00 AM EST- Posted in
- Motherboards
The New Motherboard Test Suite
One of the ongoing concerns at AnandTech has been the tight clustering of performance results in our recent motherboard tests. In general, it is rare to see really wide variations in stock performance with motherboards these days. This has been made even clearer by the AMD Athlon 64 CPU, which has the memory controller on the CPU itself, removing another variable from the chipset equation. This is not bad news for buyers, since more consistent performance at stock speeds makes choosing a motherboard an easier task. Readers have pointed out that we need to do more tests, which would really differentiate boards, and we have been working on updates to our tests.First and foremost, we have been including overclocking tests and memory stress testing for some time - simply because motherboards can vary a great deal in these capabilities. This tells you which motherboards overclock well and which ones are poor, and even if you don't ever plan to overclock, the ability of a motherboard to run at much higher than stock speeds tells you something about the quality of components used in a motherboard. Good overclockers generally use better components and regulate power on the board better, so the good overclocking boards often make sense to buy even if you will never overclock. You can reasonably expect better stability and a longer service life.
Features are increasingly important in motherboards these days as well. With USB, Firewire, IDE, SATA controllers, RAID, LAN, and audio commonly found on top-line motherboards, you are buying much more than sockets for a processor and memory. There are potentially great variations in performance of these features, which could be very important for certain uses of the board. AnandTech has done a good job of detailing these features in past motherboard reviews, but we confess that we have not done very well in actually testing and comparing performance of these features. Our new motherboard tests are designed to correct that.
Starting with this roundup, we are adding iPeak storage tests first used by Anand in his storage reviews. iPeak will be used to establish baseline performance for nForce4 on-chip IDE and SATA performance; iPeak will also be used to test the throughput of the various additional SATA controllers on these SLI motherboards. USB 2.0 and Firewire 400/800 throughput will be measured with a new test developed for motherboard testing. Basically, we create a RAM disk in Windows XP, write a standard test file to the RAM disk, and then copy the file from the RAM disk to a USB 2.0, Firewire 400 or Firewire 800 connected hard drive. We record the time to copy from RAM disk to the connected drive with a timer program developed by our IT Manager.
Ethernet testing uses the Windows 2000 DDK to connect two computers with a CAT 6 crossover cable. We then use a standard host computer as the server and measure the transmission rate and CPU overhead at the client side, which is our test motherboard. We have talked about the advantages of PCIe over PCI gigabit Ethernet in the past, and in this roundup, you will be able to see the actual difference in the performance of Gigabyte LAN over these two busses.
Audio is an area that is still under development and we will be adding tests of audio quality, as well as do subjective listening in future testing. For this roundup, we have included results from the industry standard RightMark benchmark suite for CPU utilization or overhead. There is only room for so much in a roundup, but in the future, we will definitely be including additional audio benchmarks to our motherboard tests.
Last, we have added some new benchmarks, like the popular video synthetic benchmarks from FutureMark - 3DMark 2005 and 3DMark 2003. These tests are particularly useful for testing SLI, since current nVidia drivers support SLI mode in both benchmarks. We are continuing Winstones 2004 for Business and Multimedia, PCMark04, and AutoGK for media encoding. Games are now more heavily weighted toward the most current games with Half Life 2, Far Cry, Doom 3, and Unreal Tournament 2004. Aquamark 3, which is better known as a benchmark than the game on which it is based, is also continued. Return to Castle Wolfenstein-Enemy Territory and Quake 3 have been retained primarily because of their sensitivity to memory performance. It is also a useful reference to include Open GL-based games with so many new game offerings based on Direct X or sporting DX9 front ends.
We are still experimenting with methods of presenting this new data to you, so please let us know if you have suggestions for the future.
Changes in Memory tRAS Recommendations
In past reviews, memory bandwidth tests established that a tRAS of 10 was optimal for the nForce3 chipset and a tRAS setting of 11 or 12 was generally best for nForce2. In the first memory stress test of a production nForce4 board, tRAS timings were first tested with memtest86, a free diagnostic program with its own boot OS that will boot from either a floppy disk or optical disk. Bandwidth of OCZ PC3200 Platinum Rev. 2, based on Samsung TCCD chips, was measured from tRas 5 to tRAS 11 to determine the best setting.| Memtest86 Bandwidth DFI nForce4 with Athlon 64 4000+ |
|
| 5 tRAS | 2191 |
| 6 tRAS | 2242 |
| 7 tRAS | 2242 |
| 8 tRAS | 2242 |
| 9 tRAS | 2141 |
| 10 tRAS | 2141 |
| 11 tRAS | 2092 |
The best bandwidth was achieved with this combination of nForce4/4000+/TCCD in the 6 to 8 range, so a mid-value tRAS of 7 was chosen for all tests. It appears that optimal tRAS timings may also be memory dependent on the nForce4, so we recommend a quick series of memtest86 to establish the optimum tRAS timings for other memories.
Facebook
Twitter
RSS
108 Comments
View All Comments
Rike - Monday, February 28, 2005 - link
And then he double posts and can't spell "graphs." (not "graphes!") *bangs head on wall* Oh well. To err is human. Happy Monday! :)Wesley Fink - Monday, February 28, 2005 - link
#33 - That is correct, but to implement PCIe Ethernet the mfg must use a PHY gigabit ethernet controller. In fact, as I state in the review, all 4 SLI boards implement PCIe on Gigabit #1, but all 4 boards have dual gigabit ethernet. Most implement PCI on Gigabit #2 with the results you can see in the ethernet performance charts.#30 - It IS a significant point and I thought we were clear that the 3132 is PCIe. I will add that to the chart to removee any confusion. However, there is another side to MSI using PCIe on all the on-board features. With both PCIe LAN's and PCIe SATA2 add-on there are no channels left from the 20 lanes available for PCIe slots.
Rike - Monday, February 28, 2005 - link
Rike - Monday, February 28, 2005 - link
Minor typo: HL2 resolutions on the graphes are listed as 16,000 x 1200 instead of 1600 x 1200. Either that or you're using a seriously wide screen! ;)mechBgon - Monday, February 28, 2005 - link
Wes, an academic point: unless something's changed with nVidia southbridges, the nForce3/4 gigabit Ethernet controller isn't a PCI-based device, it's native to the southbridge and rides the Hypertransport bus. If you're getting >900Mbit/sec in your test, it's pretty obvious it's not on a 32-bit PCI bus ;)AlanStephens - Monday, February 28, 2005 - link
#28 - I know for a fact that Creative doesn't support Dolby Digital Encoding. I wish they did though. Here is a quote from Creative's Knowledge Basse on this:"Computer games written with support for 3D audio do not require a Dolby Digital Interactive Content Encoder (DICE) to output multichannel sound, with no exceptions. Sound devices that support the real time encoder technology from Dolby will simply receive the multichannel wave file output and encode it in real time to a somewhat modified Dolby Digital bitstream. Creative does not support the Dolby Digital Interactive Content Encoder on any of its sound cards.
The only difference between a Sound Blaster card and an audio card that has a real time encoder, is that you can make a one-wire, digital connection from your audio card to your home theater receiver and enjoy discrete multichannel sound from the game. However there will be a continuous, slight delay, known as "latency", as the encoder is creating and transmitting the bitstream, and of course the compression scheme being used is "lossy" (i.e. not bit-accurate).
If you want to enjoy 3D audio in 3D enabled PC games in multichannel surround sound with a Sound Blaster card, it is recommended that you connect the analog outputs of the sound card directly to the analog inputs of the receiver."
EODetroit - Monday, February 28, 2005 - link
I wanna know the answer to #28's question. I've been looking for the next Soundstorm... ie I want to output a dolby digital 5.1 (or better) signal through a spdif connection from a non-pre-encoded source, like only Soundstorm can.Can the MSI do this now too??
RyanVM - Monday, February 28, 2005 - link
Why didn't you guys bother to note that the Si3132 SATA controller is PCIe? I think that's a fairly significant point in comparison to the PCI Si3114 controller and it likely explains why the Si3132 was faster.Lakku - Monday, February 28, 2005 - link
I wish you would have discussed 6600GTs in more detail. I am perturbed at a number of sites saying the 6600GT is not worth it for SLI. Specifically from X-Bit (though many have echoed it) "We guess it is the 6600GT SLI configuration that's not very appropriate". The only fact for this statement is the lack of a 256MB 6600GT card. This means its high resolution on some games and FSAA capabilites are limited. But so what? Noone has discussed REAL WORLD prices of 6800GT and Ultra cards. They range from 430 to 800 dollars, for ONE card. Yet sites claim it's better to just get a 6800GT rather then two 6600GTs. I picked up my 6600GT for 170 bucks, brand new retail. It overclocked to 550/1100 easily and I kept it at that, even though it went quite a bit higher with air cooling. I put that in because it gave me another 5%-10% or so increase over stock speed. I could get two for just over 350, almost $100 cheaper, and in a majority of tests, it equals or BEATS a 6800GT. It only suffers when you get to 1600x1200 with FSAA, as its smaller frame buffer and 128-bit interface is a hinderance. But for such a HUGE price difference, is it not worth it? If I had bought two outright (my plan was to get a start on a PCIe based system to bring me up to date for the future, and then buy the next generation cards for SLI), I could use that extra 100 for the GF7600GT (I am assuming this name, and the fact that if the next generations are 24 pipes, then the mainstream will be 12) that when SLId, will be up there in terms of todays speed on 700 to 1000 dollar video combinations, all for under 400. The point is that I wish someone would actually take an IT type approach to this situation instead of badmouthing SLI or the price for running an SLI setup. You get SLI now, be cost effective and get 6600GTs, and you have a great platform for the next generation (if it still exists, but nVidia seems to have put its eggs in this basket so I assume it will be around for awhile) of cards that you can SLI and get great performance. Is SLI worth it? I say the answer SHOULD be a resounding yes if the same performance holds true for the future. It means you can always get two mainstream cards to equal ONE high end card, for a much cheaper price.chup - Monday, February 28, 2005 - link
Is the SB Live! 24 really capable of encoding audio stream into Dolby Digital stream?