 Notice. New forum software under development. It's going to miss a few functions and look a bit ugly for a while, but I'm working on it full time now as the old forum was too unstable. Couple days, all good. If you notice any issues, please contact me. |
Forum Index : Microcontroller and PC projects : [MM] 3 bit colo(u)r MaxiMite
  Page 4 of 4 |
|||||
| Author | Message | ||||
Geoffg Guru  Joined: 06/06/2011 Location: AustraliaPosts: 3194 |
To answer a few questions: The SPI output of the PIC32 is digital, so intensity control would require a fourth SPI channel. But that will not work because of the memory requirement (as pointed out by Mick) and anyway, the 100 pin chip only has four SPI channels and we are already using the fourth for the SD card. So, sorry Nick. Why not use the ARM Cortex-M4? Good question and the short answer is that it does not have framed SPI which is critical in generating the rock steady VGA image of the Maximite. We all have our favourites and so by the same token why not use the i.MX233? It has a Video DAC and TV PAL/SECAM encoder, costs less than $6 and runs at 464Mhz! There are many chips out there but none that I know of (other than the PIC32) that have the combination of memory, framed SPI and a package that can be soldered by a mere mortal. Above all, Colour MMBasic is running on the PIC32 here and now. My focus on the Colour Maximite is to keep it simple and in the same vein as the original Maximite. I did consider an alternate design using a second PIC32 to generate the colour (as suggested here) but that would have only given 16 colours which hardly seemed worth the trouble given the extra complexity. Another design I have is based on the PIC32 + Solomon SSD1963 and that design has an output of 800 x 480 pixels (widescreen) with 64K colours but it will require a high volume manufacturer to make it practical and that road has its own set of problems. It is all a matter of tradeoffs in deciding the optimal mix of features, cost and effort. Who said that life was not meant to be easy? Geoff Geoff Graham - http://geoffg.net |
||||
boss Senior Member  Joined: 19/08/2011 Location: CanadaPosts: 268 |
Hello Geoff, yes your choice is logical, iMX233 is much better for video and sound than CORTEX M4 or LCP4350. Isn't 32kB of RAM limitation to get decent resolution? The CORTEX M4 CPU was sugested as MM(MM2) processor because of functionally similar to PIC32 and has almost double of RAM,ROM and speed. Best regards boss |
||||
Nick Guru Joined: 09/06/2011 Location: AustraliaPosts: 512 |
I'm fine with all that Geoff. 8 colours it is. What about the possibility of video padding to allow the Maximite's 480x432 display to be displayed on any LCD display with the firmware video driver providing padding video to keep the image centred and avoid upscaling degradation. I know I keep banging on with this but the image quality is *SO* much better on an old CRT VGA monitor. Maybe not worthwhile for those than run their maximite with no screen or have it connected to their PC but all the rest a looking at a substandard display and I wonder if including color in the next Maximite will show this distortion as colour fringing making for an even more distorted display. Nick |
||||
MicroBlocks Guru Joined: 12/05/2012 Location: ThailandPosts: 2209 |
As i am considering using this chip in another project could you add some specifics why the requirement for high volume? Is it the TFBGA package? Because it is also available in a LQFP package that is still doable for someone with some soldering skills. Or are the large number of pins the problem? Microblocks. Build with logic. |
||||
| Geoffg Guru Joined: 06/06/2011 Location: AustraliaPosts: 3194 |
The issue with this is that the speed at which the bits are clocked out of the SPI have to approximate the speed at which 640 horizontal bits are clocked out. That is about a 50% speed increase over the current. The padding would also result in a narrow aspect ratio which would look strange. The last time I tried to fix this I had no luck but I have become much better at fiddling with the SPI settings so I will go back and see if it can be forced into something that looks better. The design contained two high density LQFP packages and a lot of SMD passives and a few actives. I figured that this would have crossed the threshold of what most would be happy building at home with a soldering iron. Certainly as far as Silicon Chip is concerned this would be true. For more expert constructors I believe that it would not be a problem. The kit suppliers insist on pre soldering all SMD devices so this design would have essentially turned into an expensive assembled device. In this case it would be best to use a dedicated manufacturer to crank them out in volume but I hit problems in trying to set this up. Geoff Geoff Graham - http://geoffg.net |
||||
| JohnS Guru Joined: 18/11/2011 Location: United KingdomPosts: 3801 |
Just to mention that Olimex have an i.MX233 board about to ship. (Geoff will already know I'm sure.) John |
||||
| Page 4 of 4 |
|||||
 Print this page |
