主板
规格
- EATX - 330mm * 305mm
- ATX - 244mm * 305mm
- MATX - 244mm * 244mm
- Mini-ITX - 170mm * 170mm
- Nano-ITX - 120mm * 120mm
芯片组
Intel
700 Series
13th Gen Core
| Z790 | 
600 Series
12th Gen Core
| Z690 | B660 | H610 | 
500 Series
11th Gen Core
400 Series
10th Gen Core
PCIe
Peripheral Component Interconnect Express
Power
- x1 cards are limited to 0.5 A at +12 V (6 W) and 10 W combined.
- x4 and wider cards are limited to 2.1 A at +12 V (25 W) and 25 W combined.
- A full-sized x1 card may draw up to the 25 W limits after initialization and software configuration as a “high power device”.
- A full-sized x16 graphics card may draw up to 5.5 A at +12 V (66 W) and 75 W combined after initialization and software configuration as a “high power device”.
Optional connectors add 75 W (6-pin) or 150 W (8-pin) of +12 V power for up to 300 W total (2x75 W + 1x150 W).
| x1 | x2 | x4 | x8 | x16 | |
|---|---|---|---|---|---|
| 1.0 | 250 MB/s | 0.50 GB/s | 1.00 GB/s | 2.0 GB/s | 4.0 GB/s | 
| 2.0 | 500 MB/s | 1.00 GB/s | 2.00 GB/s | 4.0 GB/s | 8.0 GB/s | 
| 3.0 | 984.6 MB/s | 1.97 GB/s | 3.94 GB/s | 7.88 GB/s | 15.75 GB/s | 
| 4.0 | 1969 MB/s | 3.94 GB/s | 7.88 GB/s | 15.75 GB/s | 31.51 GB/s | 
| 5.0 | 3938 MB/s | 7.88 GB/s | 15.75 GB/s | 31.51 GB/s | 63.02 GB/s | 
USB
Universal Serial Bus
| Transfer speed | |
|---|---|
| USB 2.0 | 480 Mbit/s (60 MB/s) | 
| USB 3.x Gen 1 | 5Gbit/s (625MB/s) | 
| USB 3.x Gen 2 | 10Gbit/s (1.25GB/s) | 
| USB 3.2 Gen 2 x 2 | 20 Gbit/s (2.5GB/s) | 
USB 3.1, released in July 2013 has two variants. The first one preserves USB 3.0’s SuperSpeed transfer mode and is labeled USB 3.1 Gen 1, and the second version introduces a new SuperSpeed+ transfer mode under the label of USB 3.1 Gen 2. SuperSpeed+ doubles the maximum data signaling rate to 10 Gbit/s.
USB 3.2, released in September 2017,preserves existing USB 3.1 SuperSpeed and SuperSpeed+ data modes but introduces two new SuperSpeed+ transfer modes over the USB-C connector with data rates of 10 and 20 Gbit/s (1.25 and 2.5 GB/s).
- Mini-A
- Mini-B
- Micro-A
- Micro-B
- Type-A
- Type-B
- Type-C
网卡
| 千兆 | 2.5 G | 万兆 | 
|---|---|---|
| 125M/s | 300M/s | 1.25G/s | 
Wifi
| Gen | IEEE | Mbit/s | Date | GHz | 
|---|---|---|---|---|
| Wi-Fi 6 | 802.11ax | 600 ~ 9608 | 2019 | 2.4/5 | 
| Wi-Fi 5 | 802.11ac | 433 ~ 6933 | 2014 | 5 | 
| Wi-Fi 4 | 802.11n | 72 ~ 600 | 2008 | 2.4/5 | 
| Wi-Fi 3 | 802.11g | 6 ~ 54 | 2003 | 2.4 | 
| Wi-FI 2 | 802.11a | 6 ~ 54 | 1999 | 5 | 
| Wi-Fi 1 | 802.11b | 1 ~ 11 | 1999 | 2.4 | 
SSD
M.2
| M.2 2280 | M.2 2260 | M.2 2242 | |
|---|---|---|---|
| Size | 22 x 80 | 22 x 60 | 22 x 42 | 
主控
缓存
闪存
| SLC | MLC | TLC | QLC | 
TBW - max terabytes written - 最大 TB 写入量
GPU
CUDA
CUDA (Compute Unified Device Architecture) is a parallel computing platform and application programming interface * (API)* model created by Nvidia.
Scalable Link Interface
Scalable Link Interface (SLI) is a brand name for a multi-GPU technology developed by Nvidia for linking two or more video cards together to produce a single output.
At a transfer rate of 1 GT/s, the data rate would be 8 × 10^9 B/s, i.e. 8 GB/s, or approximately 7.45 GiB/s. The bit rate for this example is 64 Gbit/s (8 × 8 × 109 bit/s).
内存
时序
延迟
RAID
- RAID 0
    - split data evenly across disks
- no redundancy
- no fault tolerance
- ntimes read and write
 
- RAID 1
    - copy data across disks
 
- RAID 2
- RAID 3
    - rarely used in practice
 
| Level | Desc | Min | Space | Tolerance | Read | Write | 
|---|---|---|---|---|---|---|
| RAID 0 | Block-level striping | 2 | 1 | None | n | n | 
| RAID 1 | Mirror | 2 | 1/n | n - 1 | n | 1 | 
| RAID 4 | Block-level striping + dedicated parity | 3 | 1 - 1/n | 1 | n - 1 | n - 1 | 
| RAID 5 | Block-level striping + distributed parity | 3 | 1 - 1/n | 1 | n | |
| RAID 6 | Block-level strubg + 2x distributed parity | 4 | 1 - 2/n | 2 | n | 
Nested RAID
- RAID 10
Intel
Core
酷睿
| Suffix | Meaning | 
|---|---|
| X/XE | Unlocked, High End | 
| K | Unlocked | 
| F | Requires discrete Graphics | 
| T | Power-optimized | 
| Y | Moblie extremely low power | 
| E | Embedded | 
| H | High performance optmized for mobile | 
| U | Mobile power efficient | 
Series
- 12/- 11/- 10/- 9/- 8gen- i9
- 12/- 11/- 10/- 9/- 8/- 7/- 6/- 5/- 4gen- i7/- i5/- i3
- 8/- 7/- 6gen- m(mobile)
12th
| i9 | i7 | i5 | i3 | |
|---|---|---|---|---|
| Cores | 14 ~ 16 | 10 ~ 16 | 6 ~ 12 | 4 ~ 10 | 
| Turbo (GHz) | 4.8 ~ 5.5 | 4.6 ~ 5.0 | 4.2 ~ 4.9 | 4.4 | 
10th
| i9 | i7 | i5 | i3 | |
|---|---|---|---|---|
| Thermal Velocity Boost | Y | N | N | N | 
| Turbo Boost | Y | Y | N | N | 
| Single Core Turbo | 5.1 | 5.0 | 4.8 | 4.6 | 
| Max Cores | 10 | 8 | 6 | 4 | 
| Smart Cache (MB) | 20 | 16 | 12 | 8 | 
| Memory | 2933 | 2933 | 2666 | 2666 | 
Pentium
奔腾
- Silver
    - (?<prefix>[A-Z])(?<sku>[0-9]{4})
- optimized for cost
 
- Gold
    - (?<sku>[0-9]{4})
- optimized for performance
 
A Higher number indicates improved features and benchmarks.
| Series | Cores | Date | Platforms | 
|---|---|---|---|
| Gold | 2 ~ 5 | Q1’19 ~ Q1’22 | D / E / M | 
| Silver | 4 | Q4’17 ~ Q1’21 | M | 
| D | 2 ~ 4 | Q4’15 ~ Q2’16 | E / S | 
| G | 2 | Q1’13 ~ Q1’17 | D / E | 
| J | 4 | Q3’13 ~ Q1’21 | D / E | 
| N | 4 | Q3’13 ~ Q1’21 | E / M | 
| 6800 | 2 | Q4’20 | M | 
| 4000 | 2 | Q3’15 | M | 
| 3000 | 2 | Q3’13 ~ Q1’15 | M | 
| 2000 | 2 | Q3’12 ~ Q3’15 | M | 
| 1000 | 2 ~ 4 | Q2’12 ~ Q1’16 | E / M / S | 
D - Desktop
E - Embedded
M - Mobile
S - Server
Xeon
至强
- Scalable Processor
- D Processor
    - Workload optimized
 
- E Processor
    - Essential
 
- W Process
    - Workstation
- creative processionals (VFX, 3D rendering, 3D CAD)
 
Scalable Processor
| First Digit | Meaning | 
|---|---|
| 9/8 | Platinum | 
| 6/5 | Gold | 
| 4 | Silver | 
| 3 | Bronze | 
| Second Digit | Meaning | 
|---|---|
| 1 | 1st gen | 
| 2 | 2nd gen | 
| 3 | 3rd gen | 
Third and fourth digits indicate the SKU number
| Suffix | Meaning | 
|---|---|
| N | Network | 
| … | … | 
Celeron
赛扬
- (?<sku>[0-9]{4})
- (?<prefix>[A-Z])(?<sku>[0-9]{4})
A Higher number indicates improved features and benchmarks.
Series
| Series | Cores | Date | Platforms | 
|---|---|---|---|
| G | 2 | Q1’13 ~ Q1’22 | D / M | 
| J | 2 ~ 4 | Q3’13 ~ Q1’21 | D / E / M | 
| N | 2 ~ 4 | Q3’13 ~ Q1’22 | E / M | 
| 7000 | 5 | Q1’22 ~ Q1’22 | E / M | 
| 6000 | 2 | Q4’20 ~ Q3’21 | E / M | 
| 5000 | 2 | Q4’19 ~ Q2’20 | M | 
| 4000 | 2 | Q1’19 | E / M | 
| 3000 | 2 | Q1’15 ~ Q1’19 | E / M | 
| 2000 | 2 | Q3’13 ~ Q2’14 | E / M | 
| 1000 | 2 | Q1’13 ~ Q3’13 | E / M | 
- J4125
- N5105
Code Name
| CodeName | Node | Year | 
|---|---|---|
| Sandy Bridge | 32mm | 2010 | 
| Ivy Bridge | 22nm | 2011 | 
| Haswell | 22nm | 2013 | 
| Broadwell | 14nm | 2014 | 
| Skylake | 14nm | 2015 | 
| Coffee Lake | 14nm | 2017 | 
| Cascade | 14nm | 2019 | 
| Comet Lake | 14nm | 2019 | 
| Cooper Lake | 14nm | 2020 | 
| Ice Lake | 10nm | 2021 | 
| Rocket Lake | 14nm | 2021 | 
| Tiger Lake | 10nm | 2021 | 
| Alder Lake | 10nm | 2021 | 
超频
AMD
PBO
锁频降压
| 3900x | 1.1v | 3.95 Ghz | 45 ~ 66 °C | 
| 3900x | 1.1v | 4.00 Ghz | 48 ~ 67 °C | 
网线
| CAT-5 | 100 Mbps | 
| CAT-5e | 1000 Mbps | 
| CAT-6 | 1000 Mbps | 
| CAT-7 | 10 Gbps |