SiS 756: PCI Express for AMD Socket 939
by Wesley Fink on September 13, 2005 12:05 AM EST- Posted in
- CPUs
TRas and Memory Stress Testing
Memory tRAS Recommendations
In past reviews, memory bandwidth tests established that a tRAS setting of 11 or 12 was generally best for nForce2, a tRAS of 10 was optimal for the nForce3 chipset, a tRAS of 7 was optimal for the nForce4 chipset, the ULi 1695 was best at tRAS of 10, and ATI is best at 7 to 8.
Since this is our first review of the SiS 756 chipset, 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 13 to determine the best setting.
Two things stood out in our tRAS tests. First, the SiS 756 exhibits the best bandwidth at a tRAS setting of 8 or 9. We decided to use 8 for all our testing. Second, the bandwidth numbers on the SiS 756 were lower across the tRAS tests than what we have seen with the NVIDIA nForce4 and the ATI Rx480. This points to SiS having a poorer front end to the AMD on-chip memory controller than either nForce4 or ATI Rx480 - at least at this stage of development.
Memory Stress Test
Our memory stress test measures the ability of the SiS 756 to operate at its officially supported memory frequency (400MHz DDR), at the lowest memory timings that OCZ PC3200 Platinum Rev. 2 modules will support. All DIMMs used for stress testing were 512MB double-sided (or double-bank) memory. To make sure that memory performed properly in Dual-Channel mode, memory was only tested using either one dual-channel (2 DIMMs) or 2 dual-channels (4 DIMMs).
Using two DIMMs in Dual-Channel 128-bit mode, the memory performed in all benchmarks at the fastest 2-2-2-8 timings at default voltage, which was the only memory voltage available.
Tests with all four DIMM slots populated on the SiS 756 required a 2T Command Rate with 4 DIMMs in two dual channels. This is the pattern seen on other top-performing Socket 939 boards. Since the Athlon 64 memory controller is on the processor, there were no real surprises in the memory stress tests.
Memory tRAS Recommendations
In past reviews, memory bandwidth tests established that a tRAS setting of 11 or 12 was generally best for nForce2, a tRAS of 10 was optimal for the nForce3 chipset, a tRAS of 7 was optimal for the nForce4 chipset, the ULi 1695 was best at tRAS of 10, and ATI is best at 7 to 8.
Since this is our first review of the SiS 756 chipset, 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 13 to determine the best setting.
| Memtest86 Bandwidth SiS 756 Reference Board with Athlon 64 4000+ |
|
| 5 tRAS | 1954 |
| 6 tRAS | 1954 |
| 7 tRAS | 1995 |
| 8 tRAS | 2039 |
| 9 tRAS | 2039 |
| 10 tRAS | 1995 |
| 11 tRAS | 1995 |
| 12 tRAS | 1875 |
| 13 tRAS | 1875 |
Two things stood out in our tRAS tests. First, the SiS 756 exhibits the best bandwidth at a tRAS setting of 8 or 9. We decided to use 8 for all our testing. Second, the bandwidth numbers on the SiS 756 were lower across the tRAS tests than what we have seen with the NVIDIA nForce4 and the ATI Rx480. This points to SiS having a poorer front end to the AMD on-chip memory controller than either nForce4 or ATI Rx480 - at least at this stage of development.
Memory Stress Test
Our memory stress test measures the ability of the SiS 756 to operate at its officially supported memory frequency (400MHz DDR), at the lowest memory timings that OCZ PC3200 Platinum Rev. 2 modules will support. All DIMMs used for stress testing were 512MB double-sided (or double-bank) memory. To make sure that memory performed properly in Dual-Channel mode, memory was only tested using either one dual-channel (2 DIMMs) or 2 dual-channels (4 DIMMs).
| Stable DDR400 Timings - One Dual-Channel (2/4 DIMMs populated) |
|
| Clock Speed: | 200MHz |
| CAS Latency: | 2 |
| RAS to CAS Delay: | 2T |
| RAS Precharge: | 8T |
| Precharge Delay: | 2T |
| Command Rate: | 1T |
Using two DIMMs in Dual-Channel 128-bit mode, the memory performed in all benchmarks at the fastest 2-2-2-8 timings at default voltage, which was the only memory voltage available.
| Stable DDR400 Timings - 4 DIMMs (4/4 DIMMs populated) |
|
| Clock Speed: | 200MHz |
| CAS Latency: | 2.0 |
| RAS to CAS Delay: | 2T |
| RAS Precharge: | 8T |
| Precharge Delay: | 2T |
| Command Rate: | 2T |
Tests with all four DIMM slots populated on the SiS 756 required a 2T Command Rate with 4 DIMMs in two dual channels. This is the pattern seen on other top-performing Socket 939 boards. Since the Athlon 64 memory controller is on the processor, there were no real surprises in the memory stress tests.
Facebook
Twitter
RSS
31 Comments
View All Comments
tanekaha - Tuesday, September 13, 2005 - link
I dunno if it`s cos i`m using the latest firefox beta butthe ethernet page only loads down to this line
NTttcpr - m 4 ,0,
to get to final words i have to go back a page and select it from the
index.
Otherwise thx for the review.
It`s good 2 see anandtech review budget boards :)
a lot of us outside the US use them.
I have an asrock 7s41gx that I got for my son and it`s bin rock solid for a year not one hassle from day 1
I know that a lotta ppl rubbish them but no worries here.
tanekaha
Cygni - Tuesday, September 13, 2005 - link
When this chipset and its accompanying reference board circulated the review sites a few months ago, i made the prediction that while clearly faster than most of the competition (at the tiime it was), it would never see the widespread use of the SiS735... and it wouldnt crack the boardmakers open enough to get sales.Unfortunatly, it looks like i was right. This is a never ending battle for SiS. They produce a chipset that is as good or better than the rest of the competition... and then it just sits there on the shelf. Or gets used on 1 or 2 low end boards only sold in China. Its really an awful problem. If this chipset had found a partner back when the first review samples hit the net, it would have gained a fair share of marketplace... but now it is months behind, facing different opposition like the excellent ULi chipset, much updated and tweaked Nforce4 Ultra boards, and SLI boards like the K8N Neo4 and AN8-SLI on the $100 border.
I hope SiS can fix this problem in the future... it really sucks to see excellent chipsets ignored.
Wesley Fink - Tuesday, September 13, 2005 - link
I have also been very frustrated to review an excellent SiS chipset and then never see anything but bottom-dwelling boards appear. It was particularly embarassing to me to give the SiS chipset an Editor's Choice in the past and then have no major manufacturer produce a board with that chipset.However, I have learned that SiS is not blameless in this deilemma. About a year ago SiS lost their foundry, so they have to contract chip production these days. This makes reliable production schedules extremely difficult for SiS, which scares many potential manufacturers. Several mfgs also told me they loved the SiS chipset, but that they could not get consistent, reliable product from SiS either before or after the foundry went. I was told failure rates on the SiS chipset were much higher than the competition. One also confided that it cost the company MUCH more to use a SiS chipset, even though it was cheap, because they ended up doing the chipset QA while they were building boards - that features were not reliably tested by SiS before shipment.
All of this is hearsay, of course, but I've heard it from so many different manufacturers that I have to believe that many of these complaints have a basis in fact.
postler - Tuesday, September 13, 2005 - link
I've been using SIS based mainbords since release of the 735 chipset, so I waited to get my hand on the first SIS756 based chipset.I got a µATX Asrock 939S56-M last week. The board is attractively priced, well laid out, easy to install, very stable and fast...
Since I don't very much care about overclocking, the choice for me was obvious, though there are a few overclocking options in the bios (multiplier setup, modest over voltage settings for cpu and ram etc.) that look promising to get a few extra mhz. Another big advantage is that the board doesn't require a northbridge fan.
The board looks much more solid impression than the Nforce4 based Foxconn NF4K8MC-RS I owned and had a lot of problems with before (the board only worked for about a month and then blew up).
Tujan - Tuesday, September 13, 2005 - link
Im guessing that the premise is if the board runs well on a CPU at any speed and stumps them,then it should run a higher grade CPU and stump them." hmmm..Not finding that the X2s are actually stated as being able to be used on these newer AMD chispet MBs. While the reviews are full of information. A major question of the MB is will it run the AMD X2s from a default manufacturers configured curcuit board.Major Major question...left off.[ ]
What happen to the Saphire board in the gaming benchmarks here [ ] . Was there a barn burner or something that wouldn't sync with the results . ? Or is that the 'secret,the secret of holding to another review/article sometime. Incidently I haven't seen a Sapphire in the retail anywhere from a web search or otherwise for a retailer. Hope this isn't one of those Asus situation where a board is only winning 'over-there (Hi ya Europe). Everything Ive seen on the web is somewhere in a landing in the far east with the boards,and central europe.
.........See ya Mr.Fink.// dont know if I'll be able to reply to this cause don't really understand the layout of this new forum interface that well.
Wesley Fink - Tuesday, September 13, 2005 - link
We have tested and reported Athlon64 X2 compatability in EVERY recent motherboard review, including this one. From p.6, Test Setup:"SiS claims full Athlon64 X2 compatibility on their website. We were able to confirm that claim in brief operation tests with an A64 4800+. Benchmarks were run with the 4000+ 130nm A64 for benchmark comparison with other Athlon64 board reviews."
Tujan - Tuesday, September 13, 2005 - link
Sorry , I must've gleemed over that. I haven't seen a lot of reference to the X2s in practically 'all of the retail packaging stats. Of course it is only a recent announcement I know. Any reference to the X2 has only been that of noticing a comment from someone. Not from actual stats seen for/from them by me or from their retail statements (their boxes). Since 'how could they show being so ? My personal dilema I guess.Here we're doing a MB,so I'll leave it at that.
Thanks for reply.
Tanclearas - Tuesday, September 13, 2005 - link
I notice you didn't even mention ActiveArmor/NAM in "Final Words" when discussing feature comparisons. This topic comes up from time-to-time in the forums (like a discussion going on over the last couple of days). Have you ever spoken to Nvidia about ActiveArmor/NAM being broken? Far too many people are experiencing problems with this hardware/software for it to be coincidence. Despite the large number of people having problems, there is little help coming from Nvidia, and definitely no acknowledgement that there are serious problems.nserra - Tuesday, September 13, 2005 - link
I always had problems with my nforce2 drivers (sound, ide....) i thought i was unlucky.I have also had problem with VIA drivers. Intel some times...
With SIS and ULI never had problems.
Spacecomber - Tuesday, September 13, 2005 - link
Just to add a quick footnote to this, in the recent review of the Sun X2100 server, it stood out in my reading of the article that although Sun was using a nforce4 ultra chipset for the motherboard, active armor had been disabled.I had thought that perhaps the issues with active armor were those that eventually could be addressed with a driver update, but 6 months later, this is looking less likely.
Space