Table Of Contents

Accumulator (Signed) (Clock-Driven Logic)

Last Modified: September 10, 2016

Accumulates or decrements x. This node supports multi-channel operations and feedback scaling.

This node supports only Boolean arrays and scalar values of integer and fixed-point data types.

connector_pane_image
datatype_icon

x

Data to add to or subtract from the value of sum. If x is a fixed-size Boolean array, the first array element represents the least-significant bit (LSB) and the last element represents the most-significant bit (MSB). In applications with multiple channels, interleave the data and wire one element per clock cycle to x.

datatype_icon

sub?

A Boolean that specifies whether to perform subtraction or addition. Use this input to change the operation programmatically.

TRUE Calculates sum - x.
FALSE Calculates sum + x.

Default: FALSE

datatype_icon

load

A Boolean that specifies whether to start a new accumulation from 0.

TRUE Ignores any previous value of sum and begins a new accumulation by loading the value of x.
FALSE Continues accumulating from previous executions.

Restarting the Accumulation from a Specific Channel

In multi-channel applications, this input affects the channel whose data arrives at x during the clock cycle when load changes. For example, if you accumulate data from two channels for eight clock cycles, x receives data from channel 1 during the first, third, fifth, and seventh clock cycles. To restart accumulation for channel 1, set load to True during any of those clock cycles. The accumulation from channel 2 continues unchanged.

Default: FALSE

datatype_icon

enable

A Boolean that specifies whether to operate on the value of x. Use this input to operate on only valid values of x. For example, you can wire the output valid output of a High Throughput Math node to this input to ensure Accumulator operates only if the upstream node produces a valid result.

TRUE Accumulates the current value of x.
FALSE Ignores x and does not change sum.

Ignoring Values from a Specific Channel

In multi-channel applications, this input affects the channel whose data arrives at x during the clock cycle when enable changes. For example, if you accumulate data from two channels for eight clock cycles, x receives data from channel 1 during the first, third, fifth, and seventh clock cycles. To ignore only the second value from channel 1, set enable to FALSE during the third clock cycle, and then set it back to TRUE during the fifth clock cycle. The accumulation from channel 2 continues unchanged.

Default: TRUE

datatype_icon

carry in/inverted borrow in

A Boolean that specifies whether Accumulator accounts for an extra least-significant bit (LSB) when calculating sum. For example, the LSB could come from the bit pattern of the carry out/inverted borrow out output of an upstream Accumulator node.

TRUE

If sub? is also TRUE—Calculates sum - x.

If sub? is FALSE—Calculates sum + x + 1(LSB).

FALSE

If sub? is TRUE—Calculates sum - x - 1(LSB).

If sub? is also FALSE—Calculatessum + x.

datatype_icon

sum

Result this node calculates.

This result depends on not only the value of x, but also the values of any Boolean inputs on the node, such as load and enable. In multi-channel applications, sum accumulates separately for each channel.

Effect of Overflow on Sums

If overflow returns TRUE, LabVIEW discards significant bits of sum until the value is within the range of the output type.

datatype_icon

overflow

A Boolean that indicates if the theoretical computed value of the sum exceeds the valid range of its data type.

True This node wraps the value of sum.
False This node does not wrap the value of sum.
spd-note-note
Note  

If you cascade nodes, you must use unsigned versions of the nodes in the intermediate operations. Set the encoding of the final node to the encoding you want for the entire cascaded operation.

Resetting the Accumulation

To start a new accumulation, set the load input to TRUE. Then, reset load to FALSE for as long as you want sum to accumulate values.

Multi-Channel Operation

Two factors define how this node operates on data from multiple channels:

  • This node always takes one cycle to compute a value of sum.
  • For each channel, this node maintains separate internal values of sum.

Given an interleaved data set from two channels, this behavior means that during a cycle when x is from channel 1, the node returns a sum value for channel 2. For a detailed example of this behavior, refer to the Examples tab for this node.

Performance Impact on Simulation

This node can increase simulation run time significantly when you use it in conjunction with downloading, stopping, or running the FPGA VI.

The following table shows an example of how this node computes sum + x in an application that accumulates interleaved data from two channels.

Clock cycle x input Explanation of x sum output Explanation of sum
1 6 Channel 1, sample 1 - Accumulator does not return a sum during the first clock cycle because it requires one clock cycle to execute.
2 8 Channel 2, sample 1 6 (the accumulation of channel 1) Although x receives data from channel 2, sum returns the calculation from channel 1 data received during clock cycle 1. During cycle 1, sum was 0 and x was 6, so this node returns 6.
3 44 Channel 1, sample 2 8 (the accumulation of channel 2) Although x receives data from channel 1, sum returns the calculation from channel 2 data received during clock cycle 2. During cycle 2 sum for channel 2 was 0 and x was 8, so this node returns 8.
4 5 Channel 2, sample 2 50 (the accumulation of channel 1) 6 (sum from clock cycle 2) + 44 (x from clock cycle 3)
5 2 Channel 1, sample 3 13 (the accumulation of channel 2) 8 (sum from clock cycle 3 ) + 5 (x from clock cycle 4)
6 3 Channel 2, sample 3 52 (the accumulation of channel 1) 50 (sum from clock cycle 4) + 2 (x from clock cycle 5)
7 - There is no more data from channel 1. 16 (the accumulation of channel 2) 13 (sum from clock cycle 5) + 3 (x from clock cycle 6)

Where This Node Can Run:

Desktop OS: none

FPGA: All devices


Recently Viewed Topics