shiny: Order of reactive v reactiveValues - r

Question

I am struggling to wrap my head around when to use reactiveValues and when I can used reactive. For example, I am wanting to create an app where I have a reactive value that is modified by a slider. Take this app:
library(shiny)
# Define UI
ui <- fluidPage(
sliderInput("multiplier", "Multiplier", min = 0, max = 10, value = 1, step = 0.1),
actionButton("update_button", "Add Rows"),
tableOutput("data_frame")
)
# Define server
server <- function(input, output) {
# Create reactiveValues object to store data
rv <- reactiveValues(data = data.frame(x = c(1, 2, 3), y = c(4, 5, 6)))
# Add rows to data frame when button is clicked
observeEvent(input$update_button, {
# Code to add rows to data frame goes here
new_rows <- data.frame(x = c(7, 8, 9), y = c(10, 11, 12))
rv$data <- rbind(rv$data, new_rows)
})
# Multiply values in data frame by the value of the slider
updated_data <- reactive({
rv$data * input$multiplier
})
# Display data frame
output$data_frame <- renderTable({
updated_data()
})
}
# Run the app
shinyApp(ui, server)
This works well but I want to understand reactiveValues. For example can anyone explain why if I make the reactive object first to do some multiplication I get an error. Can reactiveValues only take existing data?
library(shiny)
# Define UI
ui <- fluidPage(
sliderInput("multiplier", "Multiplier", min = 0, max = 10, value = 1, step = 0.1),
actionButton("update_button", "Add Rows"),
tableOutput("data_frame")
)
# Define server
server <- function(input, output) {
my_data <- reactive({
df <- data.frame(x = c(1, 2, 3), y = c(4, 5, 6))
df$x <- df$x * input$multiplier
df
})
# Create reactiveValues object to store data
val <- reactiveValues(rv = NULL)
val$rv = my_data()
# Add rows to data frame when button is clicked
observeEvent(input$update_button, {
# Code to add rows to data frame goes here
new_rows <- data.frame(x = c(7, 8, 9), y = c(10, 11, 12))
val$rv <- rbind(val$rv, new_rows)
})
# Display data frame
output$data_frame <- renderTable({
val()
})
}
# Run the app
shinyApp(ui, server)