OCZ 3700 GOLD: Building High-Performance Memory
by Wesley Fink on August 7, 2003 8:34 PM EST- Posted in
- Memory
Retail Performance
Both the Samsung stock DIMMs and the Lasered DIMMs were assembled from unbinned chips – chips pulled prior to sorting by performance. The final step is sorting the lasered chips and selecting the best-performing chips for the best-performing memory. Chips that do not meet the highest performance levels are used in lower-speed products.The chips are also only one component in the final performance of the product. The PCB on which the chips are mounted is very important to performance. Six-layer boards are generally lower in noise and the quality of the PCB can significantly impact the performance of the memory DIMM. The programming of the SPD is also very important in determining the final performance of the memory module.
Using the best lasered chips, we finally get to the performance of the completed Retail OCZ 3700 GOLD. In this case, we looked at Maximum Speed at SPD timings, Maximum Speed at CAS2, which is faster than rated SPD CAS of 2.5, and performance at Specifications of DDR466, 2.5-7-3-3.
| OCZ3700 GOLD DS 256Mb Performance Intel 875 Chipset, Dual-Channel, Maximum Overclock |
|||
| DDR Memory Speed | Memory Timings | Memory Voltage (vDIMM) |
UNBuffered Sandra 2003 Memory Test (MB/Second) |
| 466 | Specification 2.5-7-3-3 |
2.65V | 3079 INT 3146 FLT |
| 480 | 2.0-7-3-3 | 2.7V | 3175 INT 3216 FLT |
| 510 | Maximum SPD 2.5-7-3-3 |
2.8V | 3403 INT 3471 FLT |
Speed-sorting of chips yields a retail product that now reaches to a speed of DDR510 at stock SPD timings with a voltage increase to 2.8V. We also see that speed-binning is producing even faster CAS 2.0 settings, up to DDR480. Probably the most significant performance improvement from speed-binning is the memory-bandwidth, as measured by the UNBuffered Sandra 2003 Memory Test.
DDR510 is the highest stable setting that we could achieve at SPD timings, but the memory will reach even higher speeds at more relaxed timings. OCZ 3700 GOLD is one of the many included in our evaluation of high-speed DDR500/DDR466 memory modules. Look for the results in 'Searching for the Memory Holy Grail — Part 2'. Our performance comparison of the fastest memory from Corsair, Kingston, Geil, Adata, and OCZ will be published soon.
Facebook
Twitter
RSS
68 Comments
View All Comments
Anonymous User - Saturday, August 9, 2003 - link
Which board did you use #17? Early 865PE and 875P boards had BIOS issues with lots of memory modules (Corsair, Kingston, etc.), not just OCZ.Anonymous User - Saturday, August 9, 2003 - link
Companies get turned around all of the time - for better or for worse.If I had read this before my June upgrade, things may have gone differently.
In my case, I got OCZ for the first time (2 x 256 3200) for my 865PE board. Despite Anandtech's report on compatibility, I have to run in "slow" timing mode to make sure my system won't crash. Never have I had such poor performance in memory. The full featured board (I skipped a few options) and the 875 version cost about the same as good memory replacements. Oh well...
Anonymous User - Friday, August 8, 2003 - link
Apparently the 'lasering' process is performed after binning, so why were unbinned DIMMs used? Unless the sample size was significant (I would not be confident with under 30 DIMMs of each type, personally) it would make a lot more sense to have the modules binned before modification so that at least some baseline is established. Without this then all you have is a statistical correlation (by the way, what was the nature of this beyond it being positive?) hence the higher number of modules required. That is not to say of course that even if they are binned then you can get away with a very small population, but at least reduce it to perhaps 10 to 15 of each type.I would be very interested to learn what n and r were in each case. Also, what statistical method was used to determine correlation? Spearman's rank method? Was this tested at 10%? 5%? 1%?
On a different note, I've become aware that some of OCZ's banner adverts suffer unfortunate spelling and punctuation errors...
Anonymous User - Friday, August 8, 2003 - link
I've read alot of things about OCZ a lot of them bad....but that was probably a year or two back. I've also read the reviews for this ram on Hexus and they seem to reach similar conclusions as the article here. It makes me wonder is everyone just shilling for this product or has OCZ turned their image around and are they now producing good cutting edge products....being in the market for some ram for my IS7 I'm really tempeted to try these.Anonymous User - Friday, August 8, 2003 - link
Of course there's no word about OCZ's shady practices. OCZ bought favor over on ABX, and now they've infested AT with their garbage. Hell, why not just print an OCZ press release instead of an article?Watching what's happening to AT is like watching Tom's HW after Tom decided he was too important to write. When you pass off all of your work onto unqualified lackeys, articles like this and the power supply one are the result.
Anonymous User - Friday, August 8, 2003 - link
Can someone please explain memory timings to me?What's the differance from 2-7-3-3 to 3-8-4-4? Is it more than the amount of data written per CPU cycle? Or does it have nothing to do with that?
Thank-you!
Anonymous User - Friday, August 8, 2003 - link
Im not a huge OCZ fan, but it does piss me off that so many people continue to believe that they are this fraudulent enterprise that doesnt exist. I suppose you think Mushkin doesnt buy reviews? Maybe you have the notion that Kingston doesnt relable memory? Ive seen Corsair shut sites down for comments they didnt agree with.Anonymous User - Friday, August 8, 2003 - link
Kishkumen,Everyone at one point or another has a bad piece of hardware, it happens. You just need to RMA it. You think no one else has RMAs? LOL check out Corsair's forums and their RMA rate, then get back to me.
Kishkumen - Friday, August 8, 2003 - link
OCZ makes me extremely nervous. I've had 2X256MB of PC-2700 stuff from them that has never worked at 333MHz from day one on both modules. So I relegated them to a KT266A motherboard running 266MHz with crappy timeing and wrote them off as a bad decision. Well, all of a sudden this Gold stuff suddenly appears and since I've recently upgraded to a Barton core and don't have memory that with work at 333MHz, I ask myself why I'm living with memory that is bordering on fraud so I decide to give tech support a ring and was surprised to find them somewhat helpful. I'm sending back the modules and if they send me back two modules that actually work as advertised, perhaps they'll warrant a second chance. In the meantime, however, I'm going to stay the hell away from OCZ. There are no shortcuts in this business and it seems OCZ has tried them all and failed 90% of the time while pawning off those failures to their customers.Radelon - Friday, August 8, 2003 - link
Great article, for me personally this article is showing the average experience that people will receive from their OCZ 3700 or higher ram. I have 2x256 OCZ PC3700 Gold that will reach DDR540 @ 2-6-3-3 timings. I have 2x256 OCZ PC3700 Premier that will reach DDR530 @ 2.5-6-3-3. That's way beyond spec and I'm very pleased with them. You just can't go wrong using OCZ these days.For comparison, I have 2x256 Corsair 3700xms that will only do DDR490 3-8-4-4, and they aren't 100% stable there. From personal experience I will never buy Corsair again, not when I can get OCZ which will run way over spec.
Again great article, it seems things are always gettin' better for OCZ.