Hey all, newbie here.

What im looking for is a quick overview of what each of the tests in the Byte benchmark do. I ran it, and understand some of the tests, but it would be more informative to know what they all do!!

Dhrystone 2 without register variables 459874.0 lps (10 secs, 6 samples)
Dhrystone 2 using register variables 459863.7 lps (10 secs, 6 samples)
Arithmetic Test (type = arithoh) 13255212.7 lps (10 secs, 6 samples)
Arithmetic Test (type = register) 60463.0 lps (10 secs, 6 samples)
Arithmetic Test (type = short) 58394.6 lps (10 secs, 6 samples)
Arithmetic Test (type = int) 60462.6 lps (10 secs, 6 samples)
Arithmetic Test (type = long) 60465.0 lps (10 secs, 6 samples)
Arithmetic Test (type = float) 39231.2 lps (10 secs, 6 samples)
Arithmetic Test (type = double) 39232.1 lps (10 secs, 6 samples)
System Call Overhead Test 125965.0 lps (10 secs, 6 samples)
Pipe Throughput Test 137501.0 lps (10 secs, 6 samples)
Pipe-based Context Switching Test 34583.0 lps (10 secs, 6 samples)
Process Creation Test 1498.7 lps (10 secs, 6 samples)
Execl Throughput Test 250.9 lps (9 secs, 6 samples)
File Read (10 seconds) 412606.0 KBps (10 secs, 6 samples)
File Write (10 seconds) 42753.0 KBps (10 secs, 6 samples)
File Copy (10 seconds) 5325.0 KBps (10 secs, 6 samples)
File Read (30 seconds) 414747.0 KBps (30 secs, 6 samples)
File Write (30 seconds) 44699.0 KBps (30 secs, 6 samples)
File Copy (30 seconds) 4893.0 KBps (30 secs, 6 samples)
C Compiler Test 303.7 lpm (60 secs, 3 samples)
Shell scripts (1 concurrent) 564.4 lpm (60 secs, 3 samples)
Shell scripts (2 concurrent) 294.3 lpm (60 secs, 3 samples)
Shell scripts (4 concurrent) 149.7 lpm (60 secs, 3 samples)
Shell scripts (8 concurrent) 75.0 lpm (60 secs, 3 samples)
Dc: sqrt(2) to 99 decimal places 49948.6 lpm (60 secs, 6 samples)
Recursion Test--Tower of Hanoi

lets see drystone is an old 1980's program to test like a typical application
(typical for 1980 that is)
register variables are variables actually stored in the cpu.
type = arithoh
arithmatic overhead -- this is where most of the math time is taken up so is the most important
System Call Overhead
this is mainly switching context from user to kernel space to make underlying system calls
this is a huge slowdown especially for multiple processors and constitutes a major cause of network latency.
Pipe Throughput
this is the cpu pipes i think and is about memory access speed
Pipe-based Context Switching
this one i'm not sure -- must be about the cost to the cpu of handing ram data from userspace
to kernel space ????
Execl is a family of functions that allow one proccess to branch or fork to another process
recursion is when a function calls itself from within itself -- an condition that can get quite out of controll
because when it calls itself within itself it calls itself again and again