BasicMicroUK - Forums

In case anyone is interested,
I found a way to easily speed up the programming of the Atom. I left a message on the wish list bulletin board. Maybe everyone knows this trick, but I had never seen it posted. I went from a 3 minute program/verify mode down to 20 seconds, for short programs. I use a 233 laptop so your mileage may vary.
-Kerry

Programming the Atom is probably one of the largest handycaps for begginers. Making the upgrade from the Parallax stamp, it is undesirable to wait 3 minutes to program the Atom when on the stamp a program would be loaded in 20 seconds. As a begginer much time is spent simply making trial and error changes. I have noticed that none of the verify routines are needed after the initial program write function. I simply select Cancel. This saves me at least 1 full minute for program changes. In addition, the entire program write function is not needed either.
For short programs, Programming can be canceled at 20% or less depending on program size and yet the Atom program will work fine.
It would be very easy to modify the compiler to have a software switch set to disable the Verify routine. It also would easily be possible to check the program size and only program a percentage of the Atom that is needed, not the entire 100% everytime.
I realize that for production work, critical work etc, the full verify and complete program function would be wanted, but for simple changes, you could allow the programmer the option of a speedy program mode.
You may wish to consider this in your next upgrade.
_Kerry

Your talking two entire different systems. The stamp software is programming an I2C which is fast. The Atom you are programming a Microcontroller. Microchip is the one who set the timings for each byte. If any body could make it faster it would be them. We thought about a faster system but in every aspect the Atom is faster expect when uploading the intial code. Nothing is perfect.

Dale K.

But you missed my point by a country mile. Using the Atom IDE to program the atom, your software routine goes through the complete Program mode, then through various verify modes, all taking considerable time to program. Verify modes are not needed as I have proven now, and the entire program sequence is not needed either. I have been stopping program mode at 20% for my short programs and all work fine. I gave up on the verify mode long ago. It was unnecessary as well. Sure Verify is a necessary item for mission critical programs, but for simple learning and short non critical programs it is not always necessary.
-Kerry

All of this is possible. It just hasn't been a high priority. The current program/verify system was designed for acurate programming not speed. I'll put your suggestion in our todo list. It shouldn't be difficult to add this option.

Nathan

[This message has been edited by Acidtech (edited 02-09-2003).]

Great product BasicMicro has with the NANO line ...

I am connected using the 'Solderless Breadboard Programmer Circuit' diagram
illustrated on page 6 of the Basic Atom Nano 28 Data Sheet.

The programming procedure described in Basic Micro Studio works.

After programming the chip, the program runs as expected on the Nano 28.

However, when initially programming the Nano 28, and/or when making program
changes, the programming procedure takes several minutes to complete.

In your Basic Language Manual, on page 27 is says ...
quote "The Download Begins
After the microcontroller is detected and your program compiles without errors, downloading will
begin and you should see the progress dialog. Downloading the program can take up to 30 seconds
on some modules. If downloading is slower than this you may not have your USB drivers installed
correctly." end quote.

I believe my USB drivers for the FTDI UM232R chip (my RS232 interface method) are installed correctly,
otherwise the Basic Micro Studio IDE would probably not be able to communicate with the chip at all.

My thought was perhaps there is some additional handshaking that could be used between Basic Micro Studio
and the COM port ( FTDI UM232R assigns COM4 in my case ).

Although I do have RES from the Nano is connected to RTS of the RS232 circuit,
I notice we are not using CTS, DSR, DTR signals.

Could you suggest how to speed up the Nano 28 programming process? The 30 seconds described in the
manual is certainly reasonable. How Can I get from several minutes, down to 30 seconds ?

There are several significantly newer posts than this thread that explain the problem. The FTDI chips default to a 17ms latency. You need to go into the Advanced serial port settings and change this to a 1ms latency and your programming speed will be about 30 seconds.

Thanks so much, You are so right.
Changed the serial port advanced settings to 1 ms.
Worked great.
Now why didn't I think of that.
Best Regards,
VideoBeam