DDR3-1333 Speed and Latency Shootout

By Thomas Soderstrom
published

Conclusion

We wanted to add a value element to our DDR3-1333 comparison Compare Prices on DDR3-1333 Memory and waited months for enough modules to reach the market so that we could get a complete collection of prices. Unfortunately, several of the modules we tested still do not appear on any of our favorite price- search engines, but we can at least consider the prices of those modules we did find. Yet, before we consider pricing, let's take a look at the performances the devices were able to provide.

We've said it before and we have no problem repeating that the best reason to buy DDR3 now is to remove any CPU overclocking limitations imposed by slower RAM. Anyone considering insanely-priced DDR2-1200 or faster modules will find better value in several DDR3-1333 devices.

With overclocking our primary consideration, OCZ's Platinum Edition PC3-10666 wins hands down by beating its own elaborately-cooled ReaperX modules as well as the rest of the field. While many OCZ fans will simply accept this win as par for the course, it's the first time OCZ has actually won any overclocking competition that this author has been involved with. It's great to see the company finally earning a bit of the good reputation it had formerly purchased through clever marketing.

Buyers who are shooting for an overclock to around a 1600 MHz data rate might instead prefer Super Talent's PC3-10600, and buyers looking at stealth might go for the mundane-appearing Kingston ValueRAM.

This was the first time we've included Aeneon modules in a broad comparison, and we're pleased to see the former Infineon memory division alive and well. These might not be winning any awards just yet, but the company might still compete against mid-performance models on price.

Wintec AMPX PC3-10600 took second place in overclocking, and though we've yet to find any available in the market, we know the company plans to undercut OCZ on price. Once again, buyers should weigh any loss in overclocking capability against a potentially lower price.

And now for a brief "Bang for the Buck" segment for the prices we did find: OCZ's PC3-10666 Platinum Edition has a starting price of around $300, which easily beats out the super-high-speed overclocked DDR2 it replaces. On the other hand, Super Talent's PC3-10600 kit clocks almost as high, and it retails for around $250. Edging out Super Talent on the pricing front is G.Skill, but its $240 module set finished at the bottom of both our overclocking and our under-latency tests.

PDP Patriot's 4 GB kit provides twice the capacity of the other modules we tested and, at $580, costs around twice as much. Though it finished around the middle-of-the-pack in overclocking and under-latency testing, high speeds and tight timings are especially difficult to achieve with high-capacity modules. We're sure anyone who would like to use 4 GB or more in their systems will appreciate the fact that these haven't suffered the degraded timings or inflated prices normally associated with high-capacity RAM.

In fact, the only egregious pricing we found was for Kingston's HyperX PC3-1100 kit. Starting at around $480 for only 2 GB of total capacity, buyers could instead choose even faster DDR3-1600 from various brands and still save money. The fact that these "performance" modules were beaten by the brand's own "value" line in both overclocking and under-latency testing bodes poorly for its pricing scheme.

Join our discussion on this article!

Thomas Soderstrom
Thomas Soderstrom
Thomas Soderstrom is a Senior Staff Editor at Tom's Hardware US. He tests and reviews cases, cooling, memory and motherboards.
6 Comments Comment from the forums
  • dv8silencer
    I have a question: on your page 3 where you discuss the memory myth you do some calculations:


    "Because cycle time is the inverse of clock speed (1/2 of DDR data rates), the DDR-333 reference clock cycled every six nanoseconds, DDR2-667 every three nanoseconds and DDR3-1333 every 1.5 nanoseconds. Latency is measured in clock cycles, and two 6ns cycles occur in the same time as four 3ns cycles or eight 1.5ns cycles. If you still have your doubts, do the math!"

    Based off of the cycle-based latencies of the DDR-333 (CAS 2), DDR2-667 (CAS 4), and DDR3-1333 (CAS8), and their frequences, you come to the conclusion that each of the memory types will retrieve memory in the same amount of time. The higher CAS's are offset by the frequences of the higher technologies so that even though the DDR2 and DDR3 take more cycles, they also go through more cycles per unit time than DDR. How is it then, that DDR2 and DDR3 technologies are "better" and provide more bandwidth if they provide data in the same amount of time? I do not know much about the technical details of how RAM works, and I have always had this question in mind.
    Thanks
    Reply
  • Latency = How fast you can get to the "goodies"
    Bandwidth = Rate at which you can get the "goodies"
    Reply
  • So, I have OCZ memory I can run stable at
    7-7-6-24-2t at 1333Mhz or
    9-9-9-24-2t at 1600Mhz
    This is FSB at 1600Mhz unlinked. Is there a method to calculate the best setting without running hours of benchmarks?
    Reply
  • Sorry dude but you are underestimating the ReapearX modules,
    however hard I want to see what temperatures were other modules at
    a voltage of ~ 2.1v, does not mean that the platinum series is not performant but I saw a ReapearX which tended easy to 1.9v(EVP)940Mhz, that means nearly a DDR 1900, which is something, but in chapter of stability/temperature in hours of functioning, ReapearX beats them all.
    Reply
  • All SDRAM (including DDR variants) works more or less the same, they are divided in banks, banks are divided in rows, and rows contain the data (as columns).
    First you issue a command to open a row (this is your latency), then in a row you can access any data you want at the rate of 1 datum per cycle with latency depending on pipelining.

    So for instance if you want to read 1 datum at address 0 it will take your CAS lat + 1 cycle.

    So for instance if you want to read 8 datums at address 0 it will take your CAS lat + 8 cycle.

    Since CPUs like to fill their cache lines with the next data that will probably be accessed they always read more than what you wanted anyway, so the extra throughput provided by higher clock speed helps.

    But if the CPU stalls waiting for RAM it is the latency that matters.
    Reply
  • what is on pc3-10600s "s" ?
    Reply