Is there any known bug persist ? I want to consider GCBASIC as my prime development tool to teach some of students in robotic application using PIC micro.
How it is comparable with MElab's PIC Basic Pro compiler and Parallex Basic Stamp ?
Regards,
Deepak.
If you would like to refer to this comment somewhere else in this project, copy and paste the following link:
I am hesistant to respond to this as I am not the developer, and do not have an intimate knowledge of the compiler. But as an observer of the forums, there are few things that have cropped up. And I would not characterize these items as deal breakers, because with knowledge, and foresight these can be worked out.
1) The wait/delay routines, at this point in time, are under review. Improved routines have been started, and suspect that an update will be released as time permits. Using a 20Mhz osc will give you the most flexible, and perhaps the most stable wait states at the moment. At 20Mhz the wait ms, wait 10ms, and wait s routines look very solid. In a pinch, there is a Delay Code Generator on the net for assembly instructions.
2) Serial comms like the RS232 SerSend, SerReceive commands. The reason for this is the wait 10us routine, and how different chipMhz settings are affected by it. So someone may find that these commands work fine with a 20Mhz crystal osc, but not so at 4Mhz internal osc. Using a Pic device that has a hardware Usart and MSSP, along with the code from the contributors section, will solve most current software serial problems.
3) Various items like a bootloader (don't ask me about them), and going over the 2k page boundary code limit for the 16f devices, have caused concern in the past.
The nice thing about GCBasic is the direct access to the PIC internal registers/hardware, enabling you to do as you please, or to work around problems if necessary.
If larger pin device (28-40 pin) is in the budget, then an 18F device with a hardware USART, and MSSP (hardware I2C and SSP) would be recommended.
Here is an example program that exercises a MKIII sumo bot servo motors, Sharp IR sensors, and reflectance (i.e. line follower) sensors. https://sourceforge.net/forum/message.php?msg_id=5520186
The servo code is just inline and not interrupt driven as some would want. Have used a 16f 684 for a RC controlled tank too. So all in all, I think Hugh's GCBasic is very capable.
On a side note, Hugh has developed a Graphical Great Cow Basic that he has targeted for the educational market.
The question on the Basic Stamp, PicBasic Pro, no experience with it. But everytime mentioned, I think of the small pile of bare chips you can get, for just one of their modules.
If you would like to refer to this comment somewhere else in this project, copy and paste the following link:
I'm not very experienced with GCBasic, just starting with it, but now it's my compiler.
There are somethings that make it very transparent and very flexible... i think GCBasic is made with the idea of keeping things simple , and you can stand in the easy side, but when you want to come in details or inprovements it's very easy to do.
>The nice thing about GCBasic is the direct access to the PIC internal registers/hardware, >enabling you to do as you please, or to work around problems if necessary.
I agree... and as an open-source software you can access to the source and solve problems or create a custom feature.
In my short experience i have seen a good and robust compiler, it just need some debugging and perhaps optimizing in some lybraries, but the compiler's core looks to be very solid.
For me GCBasic is a powerful and flexible tool. I like the asm it generates, very clear and useful to work in it.. indeed GCBasic can be used to generate a base asm code to work in.... Other compilers generate a difficult to understand and impossible to re-use asm code... this is what i call a powerful tool for experienced PIC users.... but if you just want the easy side, you have it.. that's what i call a flexible tool.
Then i think GCBasic in hands of an experienced PIC programmer can be a very good tool to develop educational issues.
If you would like to refer to this comment somewhere else in this project, copy and paste the following link:
Hello,
Is there any known bug persist ? I want to consider GCBASIC as my prime development tool to teach some of students in robotic application using PIC micro.
How it is comparable with MElab's PIC Basic Pro compiler and Parallex Basic Stamp ?
Regards,
Deepak.
I am hesistant to respond to this as I am not the developer, and do not have an intimate knowledge of the compiler. But as an observer of the forums, there are few things that have cropped up. And I would not characterize these items as deal breakers, because with knowledge, and foresight these can be worked out.
1) The wait/delay routines, at this point in time, are under review. Improved routines have been started, and suspect that an update will be released as time permits. Using a 20Mhz osc will give you the most flexible, and perhaps the most stable wait states at the moment. At 20Mhz the wait ms, wait 10ms, and wait s routines look very solid. In a pinch, there is a Delay Code Generator on the net for assembly instructions.
2) Serial comms like the RS232 SerSend, SerReceive commands. The reason for this is the wait 10us routine, and how different chipMhz settings are affected by it. So someone may find that these commands work fine with a 20Mhz crystal osc, but not so at 4Mhz internal osc. Using a Pic device that has a hardware Usart and MSSP, along with the code from the contributors section, will solve most current software serial problems.
3) Various items like a bootloader (don't ask me about them), and going over the 2k page boundary code limit for the 16f devices, have caused concern in the past.
The nice thing about GCBasic is the direct access to the PIC internal registers/hardware, enabling you to do as you please, or to work around problems if necessary.
If larger pin device (28-40 pin) is in the budget, then an 18F device with a hardware USART, and MSSP (hardware I2C and SSP) would be recommended.
Here is an example program that exercises a MKIII sumo bot servo motors, Sharp IR sensors, and reflectance (i.e. line follower) sensors.
https://sourceforge.net/forum/message.php?msg_id=5520186
The servo code is just inline and not interrupt driven as some would want. Have used a 16f 684 for a RC controlled tank too. So all in all, I think Hugh's GCBasic is very capable.
On a side note, Hugh has developed a Graphical Great Cow Basic that he has targeted for the educational market.
The question on the Basic Stamp, PicBasic Pro, no experience with it. But everytime mentioned, I think of the small pile of bare chips you can get, for just one of their modules.
I'm not very experienced with GCBasic, just starting with it, but now it's my compiler.
There are somethings that make it very transparent and very flexible... i think GCBasic is made with the idea of keeping things simple , and you can stand in the easy side, but when you want to come in details or inprovements it's very easy to do.
>The nice thing about GCBasic is the direct access to the PIC internal registers/hardware, >enabling you to do as you please, or to work around problems if necessary.
I agree... and as an open-source software you can access to the source and solve problems or create a custom feature.
In my short experience i have seen a good and robust compiler, it just need some debugging and perhaps optimizing in some lybraries, but the compiler's core looks to be very solid.
For me GCBasic is a powerful and flexible tool. I like the asm it generates, very clear and useful to work in it.. indeed GCBasic can be used to generate a base asm code to work in.... Other compilers generate a difficult to understand and impossible to re-use asm code... this is what i call a powerful tool for experienced PIC users.... but if you just want the easy side, you have it.. that's what i call a flexible tool.
Then i think GCBasic in hands of an experienced PIC programmer can be a very good tool to develop educational issues.