Value DDR2-800 from Super Talent: Where Can You Go?
by Wesley Fink on November 29, 2006 12:05 AM EST- Posted in
- Memory
Stock Memory Performance
Most Intel Socket T (Socket 775) motherboards provide a wide range of memory ratios that match available DDR2 memory. End-users can select the memory ratio that matches their DDR2 memory speed. Our memory testing begins with the same approach. We first test all of the stock ratios at the fastest stable timings we can achieve at the given ratio. With these ratios, CPU speed remains the same at 2.93GHz in our memory test bed, and memory speed is varied by selecting different ratios.
DDR2 memory is then pushed from the highest stock ratio that could be achieved in testing - in this case 889 (3:5) - to the highest FSB speed at the stock multiplier. We could not reach DDR2-1067 with this value Super Talent DDR2-800, so we settled for the intermediate ratio of 889. The system was then pushed to DDR-1000, which was the highest stable speed reached that ran our complete test suite. We were able to boot at speeds much higher than 1000, and we could actually navigate the boot screen up to DDR2-1050, but the system was not stable in Windows and could not complete our test suite, even at the highest 2.4V supported by the ASUS board. Other systems may achieve higher test results with more voltage.
At DDR2-800 the memory completed our benchmarks at 4-4-4-11 timings at just 1.95V. This is significantly faster than the rated 5-5-4-12 timings. Most boards can easily supply 1.95V, but if your board has a fixed memory voltage you may need to use slower timings for complete stability with this memory. DDR2-667 similarly required 2.10V to achieve 3-3-3-9 timings. If you can't supply 2.1V you will likely need 4-4-3-9 or slower timings at stock 1.8V.
Many buyers of value DDR2-800 memory will not have a motherboard that can supply voltages like the highest 2.30V required in our testing at DDR2-1000. If your top available voltage is less than 2.3V you will not be able to achieve DDR2-1000 performance with this memory. Top results at stock voltage will likely be in the DDR2-900 range at the rated 5-5-4-12 timings
The performance results found in our testing give most every buyer some choices when using this value Super Talent DDR2-800 in their system. If their board is voltage limited, the memory is still usable at slightly slower timings. If the board supplies a very wide voltage range, then you can likely squeeze better timings from this memory with higher voltage. With air cooling, however, the best you can expect to achieve is around DDR2-1000.
Most Intel Socket T (Socket 775) motherboards provide a wide range of memory ratios that match available DDR2 memory. End-users can select the memory ratio that matches their DDR2 memory speed. Our memory testing begins with the same approach. We first test all of the stock ratios at the fastest stable timings we can achieve at the given ratio. With these ratios, CPU speed remains the same at 2.93GHz in our memory test bed, and memory speed is varied by selecting different ratios.
DDR2 memory is then pushed from the highest stock ratio that could be achieved in testing - in this case 889 (3:5) - to the highest FSB speed at the stock multiplier. We could not reach DDR2-1067 with this value Super Talent DDR2-800, so we settled for the intermediate ratio of 889. The system was then pushed to DDR-1000, which was the highest stable speed reached that ran our complete test suite. We were able to boot at speeds much higher than 1000, and we could actually navigate the boot screen up to DDR2-1050, but the system was not stable in Windows and could not complete our test suite, even at the highest 2.4V supported by the ASUS board. Other systems may achieve higher test results with more voltage.
Click to enlarge.
At DDR2-800 the memory completed our benchmarks at 4-4-4-11 timings at just 1.95V. This is significantly faster than the rated 5-5-4-12 timings. Most boards can easily supply 1.95V, but if your board has a fixed memory voltage you may need to use slower timings for complete stability with this memory. DDR2-667 similarly required 2.10V to achieve 3-3-3-9 timings. If you can't supply 2.1V you will likely need 4-4-3-9 or slower timings at stock 1.8V.
Many buyers of value DDR2-800 memory will not have a motherboard that can supply voltages like the highest 2.30V required in our testing at DDR2-1000. If your top available voltage is less than 2.3V you will not be able to achieve DDR2-1000 performance with this memory. Top results at stock voltage will likely be in the DDR2-900 range at the rated 5-5-4-12 timings
The performance results found in our testing give most every buyer some choices when using this value Super Talent DDR2-800 in their system. If their board is voltage limited, the memory is still usable at slightly slower timings. If the board supplies a very wide voltage range, then you can likely squeeze better timings from this memory with higher voltage. With air cooling, however, the best you can expect to achieve is around DDR2-1000.
Facebook
Twitter
RSS
28 Comments
View All Comments
slayerized - Wednesday, November 29, 2006 - link
The last week has seen a sudden drop in memory prices. There are quite a few DDR2 800 modules in the 200 price (-1/+30-40) range on newegg with better timings.G.Skill F2-6400CL4D-2GBPK is one of the products out there ($204) with rated timings of 4-4-4-12. I realize anandtech posted another g.skill ddr2 800 review a few days back, but i am pretty sure this is different. I think things are looking better in the memory side of things and prices should stabilize in the affordable range even for enthusiast memories soon (hopefully). Right now the trend is ridiculous.
Turin39789 - Thursday, November 30, 2006 - link
It would have been nice for them to look at several memories, I'm not even sure why they tested this module except the manufacturer is launching the product and needs press. When you are looking for valueram and there are products that are more than 15% cheaper, that's what the budget minded are looking at. I want to know how the $200 sticks work and I want to see them compared with each other, similar to the treatment the high end stuff gets.Turin39789 - Thursday, November 30, 2006 - link
It would have been nice for them to look at several memories, I'm not even sure why they tested this module except the manufacturer is launching the product and needs press. When you are looking for valueram and there are products that are more than 15% cheaper, that's what the budget minded are looking at. I want to know how the $200 sticks work and I want to see them compared with each other, similar to the treatment the high end stuff gets.Spacecomber - Wednesday, November 29, 2006 - link
Are those the proper results with a mis-labeled heading, or did the wrong results find their way into the article? The results are titled G.Skill F2-6400PHU2-2GBHZ.Wesley Fink - Wednesday, November 29, 2006 - link
Web Editing used an older template, updated the results, but not the title. The results were for the Super Talent DDR2-800. The table is now corrected.Wesley Fink - Wednesday, November 29, 2006 - link
Web Editing used an older template, updated the results, but not the title. The results were for the Super Talent DDR2-800. The table is now corrected.lopri - Wednesday, November 29, 2006 - link
Probably mis-labeled. I'm seeing G-skill, too.ashegam - Wednesday, November 29, 2006 - link
is it me, or do these memory modules look like Walmart branded memory, at first glance? :)n7 - Wednesday, November 29, 2006 - link
There's almost no way this is D9GMH/D9GKX.This is likely very good Promos or possibly Elpida.
Definitely not too bad results considering it's using a crap IC.
OrSin - Wednesday, November 29, 2006 - link
"Since when is pretty good? The article is ok, but this line made me laugh. Reminds me of dells new pretty cases and the crap inside. Not all dells are crap but please dont buy a gaming rig from them and expect a true gaming computer.