Searching for the Memory Holy Grail: Part 1
by Wesley Fink on July 27, 2003 11:13 PM EST- Posted in
- Memory
Intel White Papers Confirm Results
We have seen that SiSoft Sandra UNBuffered Memory benchmarks show four DIMMs to perform better than two in all configurations at DDR400 or higher base speeds on the 875/865 chipsets. However, MemTest86 Bandwidth and Sandra Standard Memory benchmark do not show improvement in going from two to four double-sided DIMMs. So why do we believe Sandra UNBuffered Memory benchmarks?
The answer, surprisingly, is in Intel 875P Chipset Memory Configuration Guide White Paper and the Intel 865P Chipset Memory Configuration Guide White Paper. The tables below are taken from Page 13 of the 875P White Paper, but the tables are identical in both the 865 and 875 documents, except that the first and second positions are reversed in the DDR400 chart for the 865. The Intel performance rankings exactly match the rankings we have measured using SiSoft Sandra UNBuffered Memory Test. It is interesting that UNBuffered Sandra was able to distinguish the performance differences between #1 and #2 –- four double-sided vs. two double-sided. This performance distinction was not apparent in either MemTest86 Bandwidth or SiSoft Sandra Standard Memory Test.
| DDR400 Performance Configurations on Intel 875/865 from Intel White Papers |
|||||
| DDR Speed | Number of DIMMS | Sides per DIMM | Mode | SC or DC | Performance |
| 400 MHz | 4 | 2 | Dynamic | Dual Channel | 1 (2 for 865) |
| 400 MHz | 2 | 2 | Dynamic | Dual Channel | 2 (1 for 865) |
| 400 MHz | 4 | 1 | Dynamic | Dual Channel | 2 |
| 400 MHz | 2 | 1 | Dynamic | Dual Channel | 3 |
| 400 MHz | 4 | any | Normal | Dual Channel | 4 |
| 400 MHz | any | 2 | Dynamic | Single Channel | 5 |
| 400 MHz | any | 1 | Dynamic | Single Channel | 6 |
| 400 MHz | any | any | Normal | Single Channel | 7 |
| DDR266/333 Performance Configurations on Intel 875/865 from Intel White Papers |
|||||
| DDR Speed | Number of DIMMS | Sides per DIMM | Mode | SC or DC | Performance |
| 233/333 MHz | 2 | 2 | Dynamic | Dual Channel | 1 |
| 233/333 MHz | 4 | 1 | Dynamic | Dual Channel | 1 |
| 233/333 MHz | 2 | 1 | Dynamic | Dual Channel | 2 |
| 233/333 MHz | 4 | 2 | Dynamic | Dual Channel | 3 |
| 233/333 MHz | any | any | Normal | Dual Channel | 4 |
| 233/333 MHz | 2 (1 in each channel) |
any | Dynamic | Single Channel | 5 |
| 233/333 MHz | 1 | 2 | Dynamic | Single Channel | 5 |
| 233/333 MHz | any | 1 | Dynamic | Single Channel | 6 |
| 233/333 MHz | any | any | Normal | Single Channel | 7 |
While we didn’t test Asynchronous 5:4 or 3:2 performance, or Memory Performance when using a 533FSB CPU, the DDR266/333 table should be useful for those situations. When running DDR333/266 as a base speed, two double-sided DIMMS or four single-sided DIMMS perform the fastest. Two single-sided DIMMs are second in speed performance, and four double-sided DIMMs – the fastest performer at DDR400 and higher – drops to third place.
Facebook
Twitter
RSS
42 Comments
View All Comments
Anonymous User - Tuesday, July 29, 2003 - link
Wesley Fink - Monday, July 28, 2003 - link
Yes, you read it correctly. the mixed dimms can vary from very large drops - 20 to 27% or so - to very little if any drop for closely matched pairs. Performance seems to fare best when mixed pairs are the same capacities and the same "bank" configuration.ViRGE - Monday, July 28, 2003 - link
Very good article, Wesley, but I'm a bit confused on your mixed vs. matched DIMM comparison. On the 1st configuration, 2x256DS + 2x512DS, the timings look to be the same as in the 1st table on page 4, the DDR 400 numbers. Am I reading this correctly? It seems odd(although believable) that memory bandwidth dropped 25%, even though it stayed as 4 DS DIMMs at the same timings.PrometheusN - Monday, July 28, 2003 - link
Thanks to a reader for pointing out one correction to this article. The Intel White Paper from the 865 chart reverses 1 and 2 memory positions compared to the 875 white paper. While I did not test 865 performance in the review, I did make reference to the White Paper Chart.We can also wonder, with Intel not sanctioning PAT on the 865, exactly how the 865 boards WITH PAT will behave, and if memory on an 865 with PAT changes back to the same as 875.
Anonymous User - Monday, July 28, 2003 - link
You did not close your table on page eight.3 4 DS/td> Dual Channel
Add a < to the "/td>" ;)
Otherwise... Great article, very informing, thanks for taking the time to write this up.
PrometheusN - Monday, July 28, 2003 - link
Single-Sided or Double-Sided IS a functional description - but it normally is also a visual description. BEWARE dimms that skip every other chips on both sides. These have 4 chips on each side, but are functionally Single-Sided.PrometheusN - Monday, July 28, 2003 - link
According to CPU-Z 1.18C, 4 Sticks of ram - single or double - do NOT disable "PAT" on the DFI 875PRO or the Asus P4C800-E. It depends on how the motherboard BIOS handles the 4 Dimms.Wesley Fink
Shalmanese - Monday, July 28, 2003 - link
Nice article but what are the real world consequences of these numbers? When DDR was 1st introduced, we only saw a ~10% peroformance increase from a 100% increase in memory bandwidth so I am thinking that the difference between 2 and 4 banks would be trivial.ghg - Monday, July 28, 2003 - link
Hi PromNice review, as we expected from you 8-).
Using 4 sticks of doublesided OCZ 3700 Gold disables PAT according to CPU-Z 1.18c.
Same behavior when taking 4 sticks of singlesided OCZ 4000 ?
Ciao
Gary
ghg - Monday, July 28, 2003 - link