The following illustrates the identification method of memory capacity by taking the memory granule coding rules of several large factories as examples.
Samsung memory granules
At present, many manufacturers use Samsung's memory particles to produce memory chips, which has a high market share. Due to the huge product line, the naming rules of Samsung memory particles are very complicated. The model of Samsung memory granule is named by 16 digit code. Among them, users are more concerned about the identification of memory capacity and working rate, so we focus on the meaning of these two parts.
Coding rule: K4XXXXXXXXX-XXXXX
Main meaning:
Bit 1- chip function k stands for memory chip.
Bit 2-Chip Type 4 stands for DRAM.
The third bit-further description of the chip type, S stands for SDRAM, H stands for DDR, and G stands for SGRAM.
4th and 5th digits-capacity and refresh rate. Memory with the same capacity will use different refresh rates and different numbers. 64, 62, 63, 65, 66, 67 and 6A represent the capacity of 64Mbit; 28, 27 and 2A represent the capacity of 128Mbit; 56, 55, 57 and 5A represent a capacity of 256Mbit; 5 1 represents the capacity of 5 12Mbit.
6th and 7th bits —— the number of data line pins, and 08 represents 8-bit data; 16 stands for 16 bit data; 32 represents 32-bit data; 64 denotes 64-bit data.
Bit 1 1- connects "-".
Bits 14 and15-the speed of the chip, for example, 60 means 6 ns70 or 7 ns7B means 7.5 ns (Cl = 3); 7C is 7.5ns (cl = 2); 80 is 8 ns 10 or 10ns (66MHz).
Knowing the meaning of the main numbers of the memory particle code, it is very easy to calculate the capacity of the memory stick after getting it. For example, a Samsung DDR memory encapsulates 18 pieces of Samsung K4H280838B-TCB0 particles. The 4th and 5th digits "28" of the block number indicate that the block is 128Mbits, and the 6th and 7th digits "08" indicate that the block has an 8-bit data bandwidth, so we can calculate that the capacity of the memory bank is 128Mbits (megabytes) ×16chips/8bit.
Note: "bit" is "digit" and "b" is byte“byte ".If a byte is 8 bits, divide it by 8. Regarding the calculation of memory capacity, there are two cases in the example given in this paper: one is non-ECC memory, and every eight particles with 8-bit data width can form a memory; In another ECC, an 8-bit ECC check code is added after every 64 bits of data. Through the check code, two errors in memory data can be detected and one error can be corrected. Therefore, in the process of calculating the actual capacity, the parity bit is not calculated, and the actual capacity of the memory bank with ECC function of 18 particles is multiplied by 16. You can also judge that 18 or 9 memory bank is ECC memory when purchasing.
Micron memory particles
The capacity identification of micron memory particles is much simpler than that of Samsung. We use the number MT48LC 16M8A2TG-75 to illustrate the coding rules of Meguiar's memory.
Meaning:
Mt- Meguiar's manufacturer name.
Type 48 memory. 48 stands for SDRAM；; 46 stands for DDR.
LC- supply voltage. LC stands for 3v; C stands for 5v; V stands for 2.5V
16M8—— The granularity memory capacity is 128Mbits, and the calculation method is: 16M (address) ×8-bit data width.
A2- Memory kernel version number.
TG packaging method, TG is TSOP packaging.
-75- the working speed of the memory, that is,133mhz; ; -65 means 150MHz.
Example: A micron DDR memory stick consists of 18 particles numbered MT46V32M4-75. Memory supports ECC function. So each bank is an odd number of storage particles.
Its capacity calculation is: capacity 32M ×4bit × 16 chip /8=256MB (megabytes).
Siemens storage particle
At present, Infineon, a subsidiary of Siemens, only produces two kinds of memory particles in the domestic market: 128Mbits and 256Mbits. The number details the capacity and data width of its memory. Infineon's memory queue organization and management mode is that each particle consists of four banks. So its memory particle model is relatively small, and it is also the easiest to distinguish.
HYB39S 128400 means 128MB/ 4bits, and "128" means the particle capacity, and the last three digits mean the data width of the memory. The same is true of others, such as: HYB39S 128800, that is,128mb/8bits; HYB39S 128 160 is128mb/16bits; HYB39S256800 is 256 MB/8 bits.
The working rate of Infineon memory particles is expressed by adding a short line at the end of its model and then marking the working rate.
-7.5- indicates that the working frequency of the memory is133mhz; ;
-8- indicates that the working frequency of the memory is 100MHz.
For example:
1 Kingston memory stick is produced by 16 Infineon HYB39S 128400-7.5 memory granules. Its capacity is calculated as: 128Mbits × 16 chips /8=256MB (megabits).
1 Ramaxel memory chip is produced by 8 Infineon HYB39S 128800-7.5 memory particles. Its capacity is calculated as: 128 megabits× 8 chips /8 = 128 megabits.
Kingmax memory granules
Kingmax memories are all encapsulated in TinyBGA ball grid array. Moreover, the packaging method is a patented product, so we see that the memory chips of Kingmax particles are all produced by the factory itself. Kingmax memory granules have two capacities: 64Mbits and128 bits. You can list the memory granularity model of each capacity series here.
Capacity remarks:
Ksva 44t 4a0a-64mbits, 16M address space × 4-bit data width;
Ksv884t 4a0a-64mbits, 8M address space× 8-bit data width;
Ksv244t4xxx- 128mbits, 32M address space × 4-bit data width;
Ksv684t4xxx- 128mbits, 16M address space× 8-bit data width;
Ksv864t4xxx- 128Mbits, 8m address space × 16 bit data width.
The working rate of Kingmax memory has four states, which are separated by a short line symbol after the model to identify the working rate of the memory:
-7A——PC 133/CL = 2；
-7——PC 133/CL = 3；
-8A——PC 100/CL = 2；
-8——PC 100 /CL=3 .
For example, a Kingmax memory stick consists of 16 KSV884T4A0A-7A, and its capacity is calculated as: 64Mbits × 16 /8= 128MB (megabytes).