I'm reading bascom manual right now. The situation is a bit worse here, since for PIC I already have proto boards, bag of different MCUs, ICDs and so on. For 8051 and AVR I have none.
Somewhat OT. Earlier we were talking about how cheap and easy-to-find '51 tools are. I found this nifty little programmer on ebay, which will program (socket and/or ISP) 8051s from Atmel, Winbond, SST, STC, SYNCMOS, NUVOTON,... and it does PICs and AVRs too! It just might be the best $35 I've spent in a while. (Software for the programmer came on a CD along with the programmer, but I think you can download it here. It has a demo mode, so no programmer required to poke around.)
..
PicAXE uses same approach as bs & prop - small vendor specific pre-programmed IC required.
Regarding the ASM, I was writing quite effective code for Z80 in ASM, but that was almost 2 decades ago, and I won't recall these days as pleasant and sunny ones
The situation is a bit worse here, since for PIC I already have proto boards, bag of different MCUs, ICDs and so on. For 8051 and AVR I have none.
If you want to avoid vendor specific pre-programmed IC, but also want minimal kit, then I'd suggest the AT89LP51RB2 (or ED2 is you need a lot of frequent EE writes).
We have input, logcal level. Where, the specific sequence of pulses should be isolated and processed (there may be tons of different pulses, but we should care only for specific sequence): first pulse duration 1ms, then pause 3ms, then 2nd pulse 5ms. And after 2nd pulse finishes, you have to output pulse of specific length (1-8ms), length of which is defined by user. User sets the length of pulse by pressing the button repeatedly. Selected value is shown on 7 segment led. User defined value is memorized and being restored and shown on power restore.
You do not mention precision, but the upper end 51's have a pre-scaled PCA, which can capture on multiple edges, and create pulses on the same timebase (with some adding), and that would give a good HW base for your Pulse R/W Manager.
If you want to avoid vendor specific pre-programmed IC, but also want minimal kit, then I'd suggest the AT89LP51RB2 (or ED2 is you need a lot of frequent EE writes).
Don't know if it fits the OP's wants/needs-- an HLL software dev environment well removed from any nasty hardware business--but +1 for Turbo51, one of the best free development tools to be had.
But then, I'm kind of partial to Pascal anyway. In fact, I like its mix of readability, reasonably strong type checking, and low-level control.
The Proton BASIC compiler can be had for free if you are willing to be limited to the PIC18F25K20 processor. The usb development/programmer board has a 28 pin dip socket and chips are in the $2-$2.50 range. It is hardware compatible with Arduino shields.
The circuit I mentioned above, with pulse capture and etc. Is an intermediate solution to allow wirelessly use old generation flashes with modern cameras, while having ability to control the pulse power.
I'd like to thank everyone once again. I asked this question at several other places too, but this forum seems to be most responsive and helpful !
Frankly, I think you might find the 8051 quite disappointing and overly complex. At best, you might use a 8052, but even that has a lot of the same quirks of the 8051.
The AVRs do have a C cross-compiler that is free, and that makes them interesting. But there is a lot to learn.
The disadvantage with trying these microprocessors is that the currently active support communities are smaller. The 8051 has had a lot written about it, but it was really intended to have an add on ROM and an add on RAM chip... so it becomes a big project to get a complete system built. The AVRs have it all in one package.
The 8051 has had a lot written about it, but it was really intended to have an add on ROM and an add on RAM chip... so it becomes a big project to get a complete system built. The AVRs have it all in one package.
Wow, it is along time since you looked at 8051....
True, way back ~ 1980, the 8051 often had external RAM, and FLASH was not invented yet... the first ones were even NMOS, and needed 12 clocks per opcode...
However, in 2013, the modern Flash, Wide Vcc, 1T 89C51 variants are certainly 'all in one package'
Here is an example :
So far, I disliked PICBASIC PRO quite much. With these guys, you have to know and remember all registers of specific MCU. ... and have to memorize thousands of register names and variables for various MCU.
You may have to do some work to find the correct settings but to say that there are thousands of things to memorize is not at all accurate - if it were, I would not be able to use it.
They do not offer "plug & code" capability, as parallax does. And that is not due to huge number of MCUs or any other technical reason. This is just because most IDEs being written using left ankle of the rear feet.
Wow, it is along time since you looked at 8051.... in 2013, the modern Flash, Wide Vcc, 1T 89C51 variants are certainly 'all in one package'
Precisely.
Look, the 8051 may not be "glamorous" but in many many instances it will get the job done just fine, maybe even better than other solutions. The parts have been modernized so that now you can get 8051s with a variety of peripherals, memory sizes, supply voltages, and clock speeds (up to 100 MIPS, I believe). If unit cost truly is a critical factor in a (re)design--and sometimes this can be the case even at several hundred units per year or less--then it would be silly not to at least consider the many 8051 options out there. In terms of bang-for-buck, 8051s are very hard to beat and will likely remain so for years to come.
I want ease of coding of Basic Stamp, but with price at least 3 times less than basic stamp module.
Features of basic stamp are enough, and I don't need fancy multi-core, onboard bluetooth, wifi and other bells and whistles.
We got ya, Curious. These threads invariably (it seems) go off on tangents.
I still think, again going only by what you've said, that BASCOM plus AVR or 8051 should get your serious consideration. Although, a quick search turned this up. Might even be the best option yet: looks affordable, and IIRC you've already invested in PIC hardware (hopefully PIC18).
What actually do you want to do? I mean what are you going to make? How many of them? Commercially or just for hobby fun. What processing do you need to do? What I/O do you need? etc etc.
We cannot advise on selection of any components without knowing these basics.
As it is I'm confused because on the one hand you want easy to program and cheap. On the other hand you have blown 400 dollars on whatever dev kit whilst at the same time saying the Propeller is too expensive despite the fact that the chip is 10 dollars and the dev kit is free. Not to mention that the Prop is about one of the easiest devices you will ever find to program.
So, please say what it is you are aiming at, then we can think about it rationally.
Ditto what Heater said. You're not going to find quite the same ease of programming of a Basic Stamp with anything else. You'll find pretty good development systems, some free and some quite costly. Any time you have a bunch of hardware peripheral units on a chip, you're going to have some mess in setting them up and making them work. Some compilers are better than others at hiding the flaws and rough spots in whatever architecture you're using. That's what the cost vs. convenience balance is all about. You're stuck with making tradeoffs and there's no way around that. I've used PICBasicPro and it's pretty good, but it's not a Basic Stamp. You've got to spend the time to learn about the PIC peripherals and those vary from chip to chip and some are buggy or odd. On the other hand, PICs can be very cheap. The same sort of thing can be said for many other microcontrollers. If you're making a lot of whatever it is, you can afford the time to learn an architecture and development platform. For making a few of something, it's hard to beat the Propeller with its free development tools and straightforward architecture. At about $10 for the processor, xtal, and EEPROM, it's pretty cheap once you factor in the development tools.
If you're launching a new product and you want to do everything yourself, then it might not hurt to start out with something like the Propeller which gives you plenty of wiggle room to make changes later on. Then, after you see how your product is doing in the marketplace, and you're sure you've got everything finely tuned and all customer requirements solidly defined, you can hire some code monkeys software experts to take your microprocessor requirements and pare them down to the bare essentials and cram everything into super-cheap PICs. If the Propeller seems like overkill, then consider something like PicAxe. As Mike and others have noted, your trade-off here is ease-of-use vs. expense per chip. Maybe your best approach is to launch your product in phases, so if Phase 1 flops, you haven't sunk too much money and time into the inner workings of PICs, etc. Perhaps your best approach would be to just get something up and out the door, learn what you can from its performance in the marketplace, then make the next one better and cheaper by taking advantage of the economies of scale. If what you've got is that good and you still don't want to learn the inner workings yourself, then you'll be able to afford hiring people to do this sort of thing for you.
What I'm doing, is making some kind of adapters/converters, which will allow to re-use older on-camera flashes with modern digital cameras. I do not intend to make any commercial output from them - after finished, all schematics and firmware will be posted online. And I want construction to be repeatable by everyone, and should not require large investment for programmer, mcu or other parts.
Another example, which I already done via simple logical circuits. I should done it with MCU (for ease of repeatability and simplicity of circuit), but price tag is prohibitive.
The "device" consists of two buttons (up/down), one 7 segment led and 10 logic outputs. When user press "up", the number on 7 segment led is displayed, and apropriate number of outputs are enabled- (say user pressed button "up" 4 times - the outpus 0,1,2,3 will be high, others low) and vice versa.
All above is being done at spare time, as hobby, so I can't invest neither too much time or money into it.
I'm going to echo Loopy's suggestion: Use the SX. It's overkill for such a simple application as yours, but the chips are selling for $1.15 in single piece quantity. I can't say a thing about SX Basic as I've never used it, but if it's anywhere as good as SX assembly, the SX Key, and the SX chip itself, it would have to be outstanding.
Well, for projects needed for at home, it may be ok. But as I can guess, say I wrote a program in SX, so if anyone other wants to use it, he has to buy board+key, up to $90 value, right?
If they don't want to buy the key, you could sell them a pre-programmed SX along with the board. If you want to make the project open-source, you could still sell the board and the pre-programmed SXs for anyone who didn't want to make the board themselves or who didn't want to invest in the SX key. If you didn't want to do that and your device is useful enough for quite a few others to build, I'm sure someone would step in to make a few dollars selling boards and SXs.
One of my possible projects (which I have done via basic stamp). There are many old flashes for film cameras, that, while having compatible hotshoe with modern digital cameras, will always fire at full power, since they don't understand power adjust commands from the digital camera. My idea is to make an adapter, which will accept signals from digital camera, and convert them to analog camera formats. These old flashes are sold for $10-20. On the other hand, the modern flash with same power but digital camera compatibility, costs around $60. So, no one will be willing to invest more than say 20$ for upgrading the old flash. This is why kit making and so on is not feasible here. Thing to be dirt cheap to be usable.
...These old flashes are sold for $10-20. On the other hand, the modern flash with same power but digital camera compatibility, costs around $60. So, no one will be willing to invest more than say 20$ for upgrading the old flash. ....
From a commercial standpoint, I can't see how that would be a viable option. There isn't enough of a price difference to make it commercially worthwhile especially if you factor in the cost of enclosures, connectors, marketing, etc. Of course, I'm sure there are lots of camera aficionados who would like to tinker with bringing the older equipment up to the present but it's probably impossible to develop a "one-chip-fits-all" approach for those DIYers. The DIY community uses all sorts of different chips: Basic Stamps, Propellers, PICs, PicAxes, Arduinos, etc.... I kinda doubt you could please everyone.
The point is, that diy photographer should be able to reproduce the device with minimal cost & effort.
I understand that. And that was my point about trying to take a "one-chip-fits-all" approach for the DIYers. Unless you're willing to provide different versions for Basic Stamps, Propellers, PICs, PicAxes, Arduinos, etc., you'll always leave somebody struggling with the effort. And whining about it. Personally, I've found this is especially true with the Arduino crowd: they seem to get especially indignant if something isn't streamlined just for them.
after finished, all schematics and firmware will be posted online. And I want construction to be repeatable by everyone, and should not require large investment for programmer, mcu or other parts.
Unless YOU are prepared to supply pre-programmed mcu's I dont see many 'non cpu' people going to the trouble to do it themselves. Even the cost of sourcing, programming and supplying a programmed cpu may make it uneconomical.
If they actually had to invest in a programmer - forget it.
Actually I hadn't thought about programming device costs until this thread - you just pay what it takes because its your hobby. Years ago when using atmel (atmega about £2 each then) and assembler it was oh so cheap and easy. Free development system (avr studio) and the parallel port and three resistors and ribbon cable was all it took. Never thought it could be otherwise. Then started playing with arduino and 'C' (yuk to the C), and since the usb coms device came on the board and a complete setup could be purchased for about £15 and extra cpu's with boot for £3-4, I was happy enough.
However since the demise of the parrallel and serial port on the modern PC, all we have is that horrible (to understand) usb; the cheapest isp programmers are about £3 from china or £10 in this country - not the end of the world, but annoying nevertheless.
A high level language like Basic is a luxury I would like, and seeing Bascom is available free for up to 4k of code I think. I will have a look into this more thoroughly when I have time, (blimey 4k was more than enough in assembler days), I see they have code for a whole series of different programmers including my (now useless) parallel port version.
To me this seems the line of least resistence especially if you have a parallel port.
Dave H
The model that I'm thinking of for this scheme is construction articles in QST, the magazine for amateur radio operators from the ARRL. Typically, these include a schematic and parts list and, when a microcontroller is involved, there's a link to the source code and pre-compiled binary code as well as a source for pre-programmed microcontroller chips. There's always a source where you can buy the PCB and often the same source can provide a complete kit. Take a look at some construction articles in past issues at your local library.
If you want to do this project as open source online only with no product shipped by you.
The only option I see is to base it on a msp430G2xxx as they get two of those ICs plus the USB Programmer for $4.30 (shipped world wide for free)
I did side by side comparison with $200+ strobes and found almost no difference in critical strobe parameters, such as recycle time, power accuracy and raw power.
Comments
If you want to avoid vendor specific pre-programmed IC, but also want minimal kit, then I'd suggest the AT89LP51RB2 (or ED2 is you need a lot of frequent EE writes).
All you need for that, is a 4 pins at 3/5V RS232, (which you likely already have) and either the Atmel Flip SW , or a more active, and good Compiler + Simulator + Programmer Download
You do not mention precision, but the upper end 51's have a pre-scaled PCA, which can capture on multiple edges, and create pulses on the same timebase (with some adding), and that would give a good HW base for your Pulse R/W Manager.
But then, I'm kind of partial to Pascal anyway. In fact, I like its mix of readability, reasonably strong type checking, and low-level control.
http://www.crownhill.co.uk/product.php?prod=2203
I'd like to thank everyone once again. I asked this question at several other places too, but this forum seems to be most responsive and helpful !
The AVRs do have a C cross-compiler that is free, and that makes them interesting. But there is a lot to learn.
The disadvantage with trying these microprocessors is that the currently active support communities are smaller. The 8051 has had a lot written about it, but it was really intended to have an add on ROM and an add on RAM chip... so it becomes a big project to get a complete system built. The AVRs have it all in one package.
Wow, it is along time since you looked at 8051....
True, way back ~ 1980, the 8051 often had external RAM, and FLASH was not invented yet... the first ones were even NMOS, and needed 12 clocks per opcode...
However, in 2013, the modern Flash, Wide Vcc, 1T 89C51 variants are certainly 'all in one package'
Here is an example :
Modern 1T Flash8051 Selection Guide
You may have to do some work to find the correct settings but to say that there are thousands of things to memorize is not at all accurate - if it were, I would not be able to use it.
Look, the 8051 may not be "glamorous" but in many many instances it will get the job done just fine, maybe even better than other solutions. The parts have been modernized so that now you can get 8051s with a variety of peripherals, memory sizes, supply voltages, and clock speeds (up to 100 MIPS, I believe). If unit cost truly is a critical factor in a (re)design--and sometimes this can be the case even at several hundred units per year or less--then it would be silly not to at least consider the many 8051 options out there. In terms of bang-for-buck, 8051s are very hard to beat and will likely remain so for years to come.
I want ease of coding of Basic Stamp, but with price at least 3 times less than basic stamp module.
Features of basic stamp are enough, and I don't need fancy multi-core, onboard bluetooth, wifi and other bells and whistles.
I still think, again going only by what you've said, that BASCOM plus AVR or 8051 should get your serious consideration. Although, a quick search turned this up. Might even be the best option yet: looks affordable, and IIRC you've already invested in PIC hardware (hopefully PIC18).
What actually do you want to do? I mean what are you going to make? How many of them? Commercially or just for hobby fun. What processing do you need to do? What I/O do you need? etc etc.
We cannot advise on selection of any components without knowing these basics.
As it is I'm confused because on the one hand you want easy to program and cheap. On the other hand you have blown 400 dollars on whatever dev kit whilst at the same time saying the Propeller is too expensive despite the fact that the chip is 10 dollars and the dev kit is free. Not to mention that the Prop is about one of the easiest devices you will ever find to program.
So, please say what it is you are aiming at, then we can think about it rationally.
Another example, which I already done via simple logical circuits. I should done it with MCU (for ease of repeatability and simplicity of circuit), but price tag is prohibitive.
The "device" consists of two buttons (up/down), one 7 segment led and 10 logic outputs. When user press "up", the number on 7 segment led is displayed, and apropriate number of outputs are enabled- (say user pressed button "up" 4 times - the outpus 0,1,2,3 will be high, others low) and vice versa.
All above is being done at spare time, as hobby, so I can't invest neither too much time or money into it.
From a commercial standpoint, I can't see how that would be a viable option. There isn't enough of a price difference to make it commercially worthwhile especially if you factor in the cost of enclosures, connectors, marketing, etc. Of course, I'm sure there are lots of camera aficionados who would like to tinker with bringing the older equipment up to the present but it's probably impossible to develop a "one-chip-fits-all" approach for those DIYers. The DIY community uses all sorts of different chips: Basic Stamps, Propellers, PICs, PicAxes, Arduinos, etc.... I kinda doubt you could please everyone.
The point is, that diy photographer should be able to reproduce the device with minimal cost & effort.
I understand that. And that was my point about trying to take a "one-chip-fits-all" approach for the DIYers. Unless you're willing to provide different versions for Basic Stamps, Propellers, PICs, PicAxes, Arduinos, etc., you'll always leave somebody struggling with the effort. And whining about it. Personally, I've found this is especially true with the Arduino crowd: they seem to get especially indignant if something isn't streamlined just for them.
If they actually had to invest in a programmer - forget it.
Actually I hadn't thought about programming device costs until this thread - you just pay what it takes because its your hobby. Years ago when using atmel (atmega about £2 each then) and assembler it was oh so cheap and easy. Free development system (avr studio) and the parallel port and three resistors and ribbon cable was all it took. Never thought it could be otherwise. Then started playing with arduino and 'C' (yuk to the C), and since the usb coms device came on the board and a complete setup could be purchased for about £15 and extra cpu's with boot for £3-4, I was happy enough.
However since the demise of the parrallel and serial port on the modern PC, all we have is that horrible (to understand) usb; the cheapest isp programmers are about £3 from china or £10 in this country - not the end of the world, but annoying nevertheless.
A high level language like Basic is a luxury I would like, and seeing Bascom is available free for up to 4k of code I think. I will have a look into this more thoroughly when I have time, (blimey 4k was more than enough in assembler days), I see they have code for a whole series of different programmers including my (now useless) parallel port version.
To me this seems the line of least resistence especially if you have a parallel port.
Dave H
Granted the last Time I bought a new modern Strobe I paid well over 200 for it.
I dont get How you can find a E-TTL modern Strobe for 60 bucks ,.
The only option I see is to base it on a msp430G2xxx as they get two of those ICs plus the USB Programmer for $4.30 (shipped world wide for free)
https://estore.ti.com/MSP-EXP430G2-MSP430-LaunchPad-Value-Line-Development-kit-P2031.aspx
http://www.mouser.com/ProductDetail/Texas-Instruments/MSP-EXP430G2/?qs=sGAEpiMZZMv1ORdfpzTN%252bMwZ3%252b5KGk2B
http://www.amazon.com/s/ref=nb_sb_noss_1?url=search-alias%3Daps&field-keywords=yongnuo+460+ii
I did side by side comparison with $200+ strobes and found almost no difference in critical strobe parameters, such as recycle time, power accuracy and raw power.