DDR2 SDRAM
 |
In electronic engineering, double-data-rate two, synchronous dynamic random access memory (DDR2 SDRAM) is a random access memory technology used for high speed storage of the working data of a computer or other digital electronic device.
It is a part of the SDRAM (synchronous dynamic random access memory) family of technologies, which is one of many DRAM (dynamic random access memory) implementations, and is an evolutionary improvement over its predecessor, DDR SDRAM (double-data-rate synchronous dynamic random access memory).
Its primary benefit is the ability to run its bus at twice the speed of the memory cells it contains, thus enabling faster bus speeds and higher peak throughputs than earlier technologies. This is achieved at the cost of higher latency.
Overview
Like all SDRAM implementations, DDR2 stores memory in memory cells that are activated with the use of a clock signal to synchronize their operation with an external data bus. Like DDR before it, DDR2 cells transfer data both on the rising and falling edge of the clock (a technique called double pumping). The key difference between DDR and DDR2 is that in DDR2 the bus is clocked at twice the speed of the memory cells, allowing transfers from two different cells to occur in the same memory cell cycle. Thus, without speeding up the memory cells themselves, DDR2 can effectively operate at twice the bus speed of DDR.
DDR2's bus frequency is boosted by electrical interface improvements, on-die termination, prefetch buffers and off-chip drivers. However, latency is greatly increased as a trade-off. The DDR2 prefetch buffer is 4 bits wide, whereas it is 2 bits wide for DDR and 8 bits wide for DDR3.
The cost of these optimizations is increased latency, as the cells take twice as long (in terms of bus cycles) to produce a result, and additional buffering adds yet more delay. While DDR SDRAM has typical read latencies of between 2 and 3 bus cycles, DDR2 may have read latencies between 3 and 9 cycles. Because of this higher latency, DDR SDRAM running at the same bus speed as DDR2 is generally considered superior; DDR2 is, however, able to run at substantially higher bus speeds.
Although the effective clock speeds of DDR2 are higher than for DDR, the total performance was no greater in the early implementations, primarily due to the high latencies of the first DDR2 modules. DDR2 started to be effective by the end of 2004, as modules with lower latencies became available.[1]
Power savings are achieved primarily due to an improved manufacturing process, resulting in a drop in operating voltage (1.8 V compared to DDR's 2.5 V). The lower memory clock frequency may also enable power reductions in applications that do not require the highest available speed; DDR2 can use a real clock frequency 1/2 that of DDR SDRAM whilst maintaining the same throughput.
Specification standards
Chips
| Standard name | Memory clock | Bus clock | Data transfers per second |
| DDR2-400 | 100 MHz | 200 MHz | 400 Million |
| DDR2-533 | 133 MHz | 266 MHz | 533 Million |
| DDR2-667 | 166 MHz | 333 MHz | 667 Million |
| DDR2-800 | 200 MHz | 400 MHz | 800 Million |
Modules
For use in PCs, DDR2 SDRAM is supplied in DIMMs with 240 pins and a single locating notch. DIMMs are identified by their peak transfer capacity (often called bandwidth).
| Module name | Bus clock | Chip type | Peak transfer rate |
| PC2-3200 | 200 MHz | DDR2-400 | 3.200 GB/s |
| PC2-4200 | 266 MHz | DDR2-533 | 4.267 GB/s |
| PC2-5300 | 333 MHz | DDR2-667 | 5.333 GB/s1 |
| PC2-6400 | 400 MHz | DDR2-800 | 6.400 GB/s |
Note: DDR2-xxx (or DDR-xxx) denotes effective clockspeed, whereas PC2-xxxx (or PC-xxxx) denotes theoretical bandwidth (though it is often rounded up or down). Bandwidth is calculated by taking transfers per second and multiplying by eight. This is because DDR2 memory modules transfer data on a bus that is 64 data bits wide, and since a byte is comprised of 8 bits, this equates to 8 bytes of data per transfer.
1 Some manufacturers label their DDR2-667 sticks as PC2-5400 instead of PC2-5300. At least one manufacturer has reported this reflects successful testing at a faster-than standard speed. [2]
Backwards compatibility
DDR2 DIMMs are not backwards compatible with DDR DIMMs. The notch on DDR2 DIMMs is in a different position than DDR DIMMs, and the pin density is slightly higher than DDR DIMMs. DDR2 is a 240-pin module, DDR is a 184-pin module.
Relation to GDDR memory
The first commercial product to claim using the "DDR2" technology was the NVIDIA GeForce FX 5800 graphics card. However, it is important to note that this GDDR-2 memory used on graphics cards such as the GeForce FX 5800 is not DDR2 per se but rather an early midpoint of DDR and DDR2 technologies. Using "DDR2" to refer to GDDR-2 is a colloquial misnomer. In particular, the performance-enhancing doubling of the I/O clock rate is missing. It had severe overheating issues due to the nominal DDR voltages. ATI has since designed the GDDR technology further, into GDDR3, which is more true to the DDR2 specifications, though with several additions suited for graphics cards.
After GDDR2's introduction with the FX 5800 series, the 5900 and 5950 series reverted to DDR, but NVIDIA's old mainstream card, the 5700 Ultra, used GDDR2 clocked at 450 MHz (compared to 400 MHz on the regular 5800 or 500 MHz on the 5800 Ultra).
ATI Technologies's Radeon 9800 Pro with 256 MiB memory (not the 128 MiB version) also used GDDR2, but this was because it required fewer pins than DDR. The Radeon 9800 Pro 256 MiB only runs its memory at 20 MHz faster than the 128 MiB version, and primarily to counter the performance hit caused by higher latency and the increased number of chips. It is speculated that the GDDR2 used on ATI's 9800 Pro 256 MiB was actually supposed to be used on the GeForce FX 5800 series, but ended up unused after NVIDIA decided to halt the 5800 line's production. The 9800XT that followed reverted to DDR, and later on ATI began to use GDDR3 memory on their Radeon X800 line.
GDDR3 is now commonly used in most NVIDIA- or ATI-based video cards. However, further confusion has been added to the mix with the appearance of budget and mid-range graphics cards which claim to use "DDR2". These cards actually use standard DDR2 chips designed for use as main system memory. These chips cannot achieve the speeds that GDDR3 can but are fast and cheap enough to be used as memory on mid-range cards.
Integration
DDR2 was introduced in the second quarter of 2003 at two initial speeds: 200 MHz (referred to as PC2-3200) and 266 MHz (PC2-4200). Both perform worse than their DDR equivalents since heightened latency makes total access times twice as long in the worst case scenario. However, DDR will not officially be introduced at any speeds above 266 MHz (533 MHz effective). DDR-533, and even DDR-1200 SDRAM exists, but JEDEC has stated that they will not be standardized. These modules are mostly manufacturer optimizations of highest-yielding chips, drawing significantly more power than slower-clocked modules, and usually do not offer much, if any, higher real-world performance.
Both Intel [On sockets LGA775, and Socket M] and AMD [On the socket AM2] support DDR2.
DDR2 SDRAM DIMMs have 240 pins (as opposed to 184 pins on DDR DIMMs, and 168 pins on SDR DIMMs).
Alternatives
Generally, DDR2 is expected to have little competition in main computer memory sector. However, there are three alternatives:
The first is Rambus XDR DRAM (eXtreme Data Rate DRAM). This technology can achieve very high clock speeds, but Rambus has been virtually disowned by IBM PC compatible chipset makers, and it is considered more likely that XDR will find use in set-top appliances and the like. Sony has selected XDR for use in PlayStation 3.
Next is Kentron Quad Band Memory (QBM), which uses DDR modules with effectively two channels routed to the module. This was briefly supported by VIA, but they have dropped support for the technology, and there are doubts about Kentron's commercial viability.
The third alternative is Quad Data Rate SDRAM (QDR), which is considered the natural successor to DDR technologies (DDR2 uses some QDR transfer methods, though is still very much based on DDR technology). However, QDR is not currently considered to be even a remotely viable product due to high production costs and poor speeds currently achieved by such modules - most barely achieve 66 MHz (266 MHz effective), and the technology may not be viable until late in the decade.
See also
References
- Razak Mohammed Ali. "DDR2 SDRAM interfaces for next-gen systems" (PDF). Electronic Engineering Times.
- Ilya Gavrichenkov. "DDR2 vs. DDR: Revenge gained". X-bit Laboratories.