golang 单线程原子操作性能怎么这么差？

package main
import (
    "sync/atomic"
    "fmt"
    "time"
)


func main() {

    var t1 uint64 = 0
    var t2 uint64 = 0

    endChan := make(chan int)
    for i := 0; i < 1000; i++ {
        go func() {
            for i := 0; i < 10000; i++ {
                atomic.AddUint64(&t1, 1)
                t2 += 1
            }
            endChan <- 1
        }()
    }

    for i := 0; i < 1000; i++ {
        <-endChan
    }

    // 测试非原子操作造成的值不正确
    // t1= 10000000
    // t2= 8513393
    fmt.Println("t1=", t1)
    fmt.Println("t2=", t2)


    // 性能测试
    func() {
        var t1 uint64 = 0

        startTime := time.Now()
        for i := 0; i < 1000000000; i++ {
            t1 += 1
        }
        endTime := time.Now()
        fmt.Println("非原子操作耗时：", endTime.Sub(startTime))
        // 非原子操作耗时： 535.0303ms

    }()

    func() {
        var t1 uint64 = 0

        startTime := time.Now()
        for i := 0; i < 1000000000; i++ {
            atomic.AddUint64(&t1, 1)
        }
        endTime := time.Now()
        fmt.Println("原子操作耗时：", endTime.Sub(startTime))
        //原子操作耗时： 14.7758413s
    }()
}