Professional Documents
Culture Documents
Rev. 1.1
LATTICE AT A GLANCE
Fully
Production-
Ready HD
Video
Camera
Major
with HDR Application:
Security &
Surveillance
Other Cameras
Applications:
Traffic,
Automotive,
Video Full HD
Conferencing 1080p60
Capable HDR
Image
Signal
Processing Industry
(ISP) Pipeline Leading Auto
Exposure,
High Dynamic
Range &
Auto White
Balance
Price $399
Industry-leading features:
- Fast auto exposure
- 120dB+ High Dynamic Range
- Support for up to 16MP sensors
• XP2-5 and XP2-8 devices support the required Orion Nanovesta Board
price-points and code footprints
Transmitter Receiver
• 4 Evaluation Boards
– ECP3 Video Protocol Evaluation Board (Board #1)
– 7:1 LVDS to DVI Converter Board (Board #2)
– DVI to 7:1 LVDS Converter Board (Board #3)
– HDMI Mezzanine Card
– All Necessary Cables and Power Supplies
• Documentation
– User’s Guides for Reference Designs, IP Core, Eval
Boards, & Demos
– Instructions to Setup and Operate the Demos
– Lattice Sales Tools on eLearn
• Customer Presentation, Lobby Pitch, Marketing
Bulletin, etc.
*
These are “Loaner Kits” Only, for Lattice FAEs in Various Regions
They are not meant for Customer Purchase
Please Contact Your Local HDBU Marketing Rep for Obtaining the Kit
LED
HDMI
OUT
Lvl
Shft
HDMI
Enc
Tx
IN2
Dec
Rx
HDMI
IN1
EQ
Key Features:
• DSP Processors need to operate ~ 100x the Sample Rate of a System. FPGAs
can be used in High Sample Rate Applications where Parallelism is present.
• FPGAs use smaller independent memory elements, removing data bottlenecks and
faster access.
• FPGAs can exceed the computing power of DSPs with their inherent parallelism,
breaking the paradigm of sequential execution and accomplishing more per clock
cycle.
XP2/ECP2/M: 18
ECP3:
1. Has Dual Slice and Cascade
capabilities
2. Internal accumulate or cascade to next
slice for wider tree 18
3. New Modes (MMAC, Adder Tree, Wide
18 18
Mux, Barrel Shifter)
4. 54-bit Cascadable ALU supports
rounding and truncation. M M M
M
5. Neighboring ALU output chainable as 54
third input for ternary adders A 54
A
6. Total DSP capability of up to 128,000
Million Multiply accumulates per second
(MMACs) SLICE 0 SLICE 1
Data_A
z-1 z-1 z-1 z-1 z-1 z-1 z-1 z-1
+ + + +
slice ‘3’ slice ‘2’ slice ‘1’ slice ‘0’
ECP3: 370 MHz for sysDSP
MULTADDSUB
ECP3: ~260 MHz for Fabric
+ +
ECP2: ~260 MHz for Fabric
Lower Latency
+
ADDER TREE
Data_A
z-1 z-1 z-1 z-1 z-1 z-1 z-1 z-1 z-1 z-1 z-1 z-1
MULTADDSUBSUM
ECP3:400 MHz, Higher Latency