Hi, The last response here was almost a year ago. How is the project coming? That is, dynamically loading native unmanaged code? Thanks

Clever.

Thank you. The competition wasn't so much who could play the best song. It was more like who could play a recognizable song.

Was it the type that was woven on a loom with little ring magnets at the junction of the wires? We had some of that.

Yes. Beautiful creations made by persons who had previously worked in the silk industry.

Now I'm impressed.

In those days you phoned those guys up, asked nicely and sent them media. Then, you waited until the postman returned with a copy of whatever it was on their development machine that day.

We had to keep ours in a cardboard box. At the bottom of a lake. </python>

Now that's clever!

Year's later I missed the front panel bus lights so much I wrote a front panel emulator for the Amiga. With all applications running in a common address space it was easy to capture the program counter of the application that was running at the time of a low level interrupt wedged into the OS (just peek the right number of bytes backward on the stack). Shove the PC into the front panel emulator's memory where it would be picked up in user mode and displayed in little make believe lights.

Ahhh. The nostalgia!

Who you calling a whippersnapper My first digital music player was an AM radio tuned between stations sitting on a drum which caused specific RFI based on where the heads were. My first mainframe had 4K of memory. I got my first Unix distribution directly from Ken Thompson own hands. My first program was entered with switches. I didn't know if I could ever fill an RL05. And and and ... You're right - the whippersnappers ARE spoiled. Wish I were one of them.

Hi,

MoonDragon, a fellow over on TinyCLR, implemented what seems to be a fairly fine grained evaluation of actual performance on the FEZ Spider, a .net Gadgeteer board.

His findings were devastating for me and I wonder if a similar result will be found when I fire up the Go that arrived today.

He found the following:

Some of the results I got (in microseconds per single op, rounded down):
Simple for loop: 54
Assignment of const ( y = 5 ): 7
Assignment of var ( y = z ): 11
Multiply constants ( 3 * 5 ): 0
Multiply with var ( 3 * z ): 10
Compare two constants ( 3 > 5 ): 11
Compare with var ( y > 5 ): 41
Compare 2 vars ( y > z ): 45

MoonDragon's full thread is here: Original thread.

Is the runtime on the Go going to place as high a burden on the processor?

(what that is saying above is you effectively only about 100,000 instructions per second in the managed environment)


Thanks