> are you sure you did not get an extra zero in there?
Yes, that is right. But I tend to make things just a tad better than needed.
There is no such engine to my knowledge. I have to admit, the 60000 RPM are for four strokes, 2 strokes could make 30000 RPM. The biggest problem are the ignition coils. Mine do have a charge time of 500 ns @22 Amps. 200ns longer, and they go up in smoke. They need a recovery time of about the charge time. So the limit is 1µs. I have wound my own coils. They are much smaller than conventional ones.
That work isn't completed, I stalled with the engine I built the system for. But not because of the RPM.
But it has been running on a different engine. One of the features is sparc sequences at lower RPM. This improves starting them a lot.
Nick
▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
Never use force, just go for a bigger hammer!
The DIY Digital-Readout for mills, lathes etc.: YADRO
> Let's get this straight. How many R.P.M. exactly ?
No THIS engine will certainly never ever make 30000 RPM! It would fall apart at 3000 RPM.
And I do not intend to build a 24 cyl. engine that does 60000 RPM. I'm not that stupid.
This V-twin is only for show and tell and for sound. RPM at about 1000 to 1500. The open crankcase and the small valves won't allow any serious power.
The design is taken from "Model Engine Builder". I redraw it in CAD in metric and made some changes to look a bit better.
Nick
▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
Never use force, just go for a bigger hammer!
The DIY Digital-Readout for mills, lathes etc.: YADRO
heater said...
Actually I'm totally fascinated by the idea of a little PCB carrying an ARM and a gob of memory and a Propeller or two.
I just did that with an AVR and a prop.
The project has 3 boards, one was made by the EE types here and sniffs
data from a clamp that goes onto a cable (I have no idea how it works).
This data stream then goes to board2 where an AVR looks over the data and makes
decisions based on what it finds....this board also has a prop on board to handle some timing and
VGA and kbrd duties if ever needed. Board3 is in a separate case and has a prop and can be used at
a remote location to connect to the main package via a 2 wire cable...it can power
the remote package and communicate over the 2 wires... it's a remote terminal
that uses a prop...it replaced a laptop computer. The external 12v power brick for the VGA monitor
also can power all 3 boards from the remote terminal's location. When this is done then no power
supply is needed at the cable location which is really nice since it's not always easy to find
power. I'm still experimenting to see how far apart the 2 packages can be before loss in the cable makes it unworkable.
The boards don't use lots of power so the one 12v brick seems to be enough. When I check the total power
used from the power brick it is below it's stated rating when powering the boards and the monitor.
Originally the 2nd board was to have an ARM but I managed to squeeze enough
juice out of a 20mips Atmega644 to do the job.
▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
- Some mornings I wake up cranky.....but usually I just let him sleep -
Nice work Nick. Veering even further off topic [noparse]:)[/noparse], but have you ever tried to make a Stirling engine?
heater and Holly, I'm also interested in such a board. Andre's board is moderately interesting, but I would want something with closer to 0.5MB code space (LPC2048). What I miss when I work with the LPC2048 is convenient JTAG ... are they all printer port dongles? ... ugh! Is there anything USB2.0 like for JTAG? If so, please recommend it!
"Yes, but they don't stink, don't make smoke, make no noise. They are LAME!"
Nonsense [noparse]:)[/noparse]
A handful of us built Sterling engines in metal working class at school when I was 13/14.
Covered in oil and heated from a blow lamp all that burning oil makes a fine stink and plenty of smoke.
Lots of clunky noise due our fairly lax metal working tolerances.
No power to speak of but a good turn of speed.
▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
For me, the past is not over yet.
> Lots of clunky noise due our fairly lax metal working tolerances.
Naaa! You have no idea how an IC-engine has to sound! First of all, Stirlings are are no ICs, so they fail completely. Second, the sound is coming from the intake and the exhaust. The rest is noise. And I don't want noise, I want SOUND!
I have built this engine from a set of castings:
<http://www.motor-manufaktur.de/ellwe/pics/ellwe_2fb_m10.JPG>
The seller of the castings was astonished when he first heard my engine. His sounded like a bunch of cans torn behind a car ("Just married"), mine sounds like this:
<
No, in the beginning, that's one of my cats!
There are a few tricks how to build a muffler that sounds good.
Sorry for OT!
Nick
▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
Never use force, just go for a bigger hammer!
The DIY Digital-Readout for mills, lathes etc.: YADRO
jazzed said...
Nice work Nick. Veering even further off topic [noparse]:)[/noparse], but have you ever tried to make a Stirling engine?
heater and Holly, I'm also interested in such a board. Andre's board is moderately interesting, but I would want something with closer to 0.5MB code space (LPC2048). What I miss when I work with the LPC2048 is convenient JTAG ... are they all printer port dongles? ... ugh! Is there anything USB2.0 like for JTAG? If so, please recommend it!
I use a Rowley CrossConnect USB JTAG interface with their CrossWorks tools for ARM development.
Leon
▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
Amateur radio callsign: G1HSM
Suzuki SV1000S motorcycle
I don't think benchmarks like Dhrystone and Whetstone would mean much. Heck I don't even know if they could run on a Prop for that matter.
Real world benchmarks say like controlling a engine from a BMW 700 series OTOH would be more interesting. Though that would comparing the Prop against Freescale's micros which do the processing in that case.
As you have seen above, each cog is totally independent, each with it's own registers, flexible counters and 496 longs (32bits) of instruction/data. In addition to this, it has access to shared hub memory which is 32KB or RAM and 32KB of ROM.
Each cog instruction takes 6 clocks (a few are different) but due to single pipelining, they result in 4 clocks. Each cog runs it's own independent program. Each cog can run the Spin Interpreter, which is itself a PASM program. So this means with the standard 80MHz clock, each instruction takes 4x12.5nS=50nS.
Now, if you want to see an example of interleaving 4 cogs to do sampling of the pins at 12.5nS see my Data Logger object. BTW I am not the only one to have done this. see the "Prop Tools under Development or Completed" link in my signature below and look for the Data Logger post for the link to the thread and code.
▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔ Links to other interesting threads:
I had only read the datasheet. It's actually not very good. It only says that the Cogs can execute some limited commands on their own without the hub. On reading a different doc I see that thats all wrong.
What people are saying about real time deterministic behavior. Thats all doable with regular micros. I've rarely met a real time situation that was so hard that I had to synchronize behavior using hardware exact to the picosecond or whatever. I'm not saying they don't exist. Just for the types of apps I typically write they aren't that hard. You don't need to mess with an RTOS, which is nice and you don't need to do context switching, which is also nice
I need to think about this and whether the performance is enough, with enough memory to do what I want for the price. This is an important issue. You also get a USB interface (if I've read the specs correctly) with almost no extra parts needed.
$8 and $1.50 for the EEPROM.
Too bad it doesn't have an internal flash instead of requiring the external EEPROM, but it's not a huge deal. One application I'm really interested in pursuing lately is using small microprocessors instead of FPGAs. FPGAs are about $13 for a spartan 3. The propeller is an easier chip to add to a circuit. I like the idea of stamping a preprogrammed chip like it's an ASIC. Check out, for example, the SoundGin processor using a PIC.
Oh, BTW, can you protect your EEPROM data so it's only readable by the propeller so someone doesn't steal your software?
I'm sold on the propeller now for what it can do. I need to resolve the security issue above and then decide on a few issues. Long term I definitely want to get one of these to play with It does look like fun to play with. I like pulling that extra 30% out of chips and making them doing things that make people scratch their heads on how I got it to do that.
Thanks for all the good suggestions....not sure how we got onto an app about engines
_D
There's really no practical way to protect your EEPROM data from reading by someone else. There have been many long thread discussions on this topic and you're welcome to search for and peruse them for details.
Nick, That's really nice work on the V twin. I didn't think ic engines had got to 60,000 RPM, but things change so fast you never know. I used to race 2 stroke sprint carts in my late teens and early twenties, and thought the 10~12K RPM of those engines was pretty impressive. Flat out they sounded like a swarm of really angry hornets. Nowadays the formula 1 engines do somewhat better than that.
Ale, Not really sure the "We want to know what kind of motors are you talking about !?" question was aimed at me, but just in case it was, I was talking about the turbo molecular turbine pump motors used for high vacuum work. They are brushless DC motors similar to what is used in the hard drive spindle motors. See http://en.wikipedia.org/wiki/Turbomolecular_pump
"What people are saying about real time deterministic behavior. Thats all doable with regular micros."
No it is not. A hyper simple example.
0) Wait for an input clock edge on a pin.
1) Read two bits from other 2 pins.
2) Form some logical result from those two bits, oh something simple like AND.
3) Output that result on a fourth pin after exactly some multiple of 50ns. +/- 25ns.
Easy perhaps any suitably fast micro can do that.
BUT now do that or something similar with 3 or 4 sets of similar pins each with their own unrelated clock inputs at the same time.
OK this is a very artificial example to make the point. But that kind of timing requirement does exist in many applications. Without the Prop one ends up having to implement the solution in hardware logic or an FPGA.
▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
For me, the past is not over yet.
kwinn: Sorry, it was not aimed at you. I know what a turbomolecular pump is as I work in mass spectrometry and our machine has 3 of them. They spin up to 60 krpm in a couple of minutes: the one on the picture spins up way faster www.chemie.uni-freiburg.de/orgbio/w3platt
I think Nick answered about the engines but I seem to have missed the bit where he said which engine went up to 60 krpm.
With apologies to continuing to pull this thread off topic, no discussion of minature engines should be complete without seeing this 1" x .9" stroke V8, complete with supercharger
The next thing you know, you want a 1/4 scale V8 to put into your 1/4 scale racer ;-)
Heater, what you described is a gate. Like you said, a 5 cent chip can do that.
I'm talking real time apps. An RTOS can do what I (note I'm not saying you) need to do typically. Anything that requires immediate behavior....well, lets just say, there are situations that require hardware. The question is whether it's so complicated you need a propeller to do it cheaper than (like you suggest) an FPGA or CPLD.
Like I said, I'm sold on the propeller for what it can do.
I have some solutions for security on the EEPROM that I posted on a separate discussion I created....basically involves some clever decoding and the use of a 41 cent locked PIC on the I2C bus to hold part of the decode key.
I've gotten my answer from this forum, but I still haven't seen any comparison other than the math, which looks good.
This chip is capable of 160 MIPS for $8 which is pretty good. I have a few questions about the chip, but I'm going to put those on a different forum. This one is almost unreadable because of the engine chat
Thanks to all who replied. I got the answers I needed and those that are available.
-D
I am also talking about "real-time" apps. As I understand them from working on machine controls, avionics etc.
Yes indeed my hyper simple example is a gate. That was the whole point. Such a simple thing but as the problem is put most micro-controller would stumble on it.
I claim that the example is a very good example of a real-time application. "But you can do it with a 5c logic chip" you quite rightly say. Well it is only an example to demonstrate the sharp end of the real-timeness. What if I extend the specification with one more requirement.
"The logic functions to be performed on the two input bits is to be configurable at run time by a human operator with either a graphical or command line interface to be either AND, OR, XOR, or NOT. The output can also be set to 1, 0, input A or input B my the same interface."
Now we have a full up real-time app. The 5c chip won't do it. An FPGA could. Most mico-controllers won't without hardware assistance. Whilst one is designing/implementing that the Prop user has gone home early.
I think the main design idea of the Prop is nothing to do with compute performance. It's just the simple concept that - instead of bolting a bunch of peripheral functions to a CPU like UARTS, SPI, I2C, USB etc etc. Lets have some little CPUs that can be programmed to do many of those things. Then there need only be one chip instead of the huge array different PIC, AVR etc variants. There is a lot more flexibility.
As a bonus side effect that design decision results in something that is dead easy to program because all the component parts are isolated.
▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
For me, the past is not over yet.
No, there is no such Internal Combustion engine! Even the small competionion 2 strokes (for R/C models) stop at about 30000. But they don't have spark plugs.
Nick
▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
Never use force, just go for a bigger hammer!
The DIY Digital-Readout for mills, lathes etc.: YADRO
Comments
Yes, that is right. But I tend to make things just a tad better than needed.
There is no such engine to my knowledge. I have to admit, the 60000 RPM are for four strokes, 2 strokes could make 30000 RPM. The biggest problem are the ignition coils. Mine do have a charge time of 500 ns @22 Amps. 200ns longer, and they go up in smoke. They need a recovery time of about the charge time. So the limit is 1µs. I have wound my own coils. They are much smaller than conventional ones.
I'm nuts, I know!
Here's a shot of the terminal window:
<http://www.cncecke.de/bilder/uploadFiles/16348_111971633898_terminal.PNG>
Here of the coil:
<http://www.cncecke.de/bilder/uploadFiles/16348_111971633898_coil-2.jpg>
And here of the engine (unfinished):
<http://www.cncecke.de/bilder/uploadFiles/16348_111971633898_vt-5.jpg>
Made everything by myself. No parts bought (except screws etc.). Even constructed a mandrel bender for the pipes.
That work isn't completed, I stalled with the engine I built the system for. But not because of the RPM.
But it has been running on a different engine. One of the features is sparc sequences at lower RPM. This improves starting them a lot.
Nick
▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
Never use force, just go for a bigger hammer!
The DIY Digital-Readout for mills, lathes etc.:
YADRO
Let's get this straight. How many R.P.M. exactly ?
Edit: I do see 1500 in your screen shot.
▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
For me, the past is not over yet.
No THIS engine will certainly never ever make 30000 RPM! It would fall apart at 3000 RPM.
And I do not intend to build a 24 cyl. engine that does 60000 RPM. I'm not that stupid.
This V-twin is only for show and tell and for sound. RPM at about 1000 to 1500. The open crankcase and the small valves won't allow any serious power.
The design is taken from "Model Engine Builder". I redraw it in CAD in metric and made some changes to look a bit better.
Nick
▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
Never use force, just go for a bigger hammer!
The DIY Digital-Readout for mills, lathes etc.:
YADRO
I just did that with an AVR and a prop.
The project has 3 boards, one was made by the EE types here and sniffs
data from a clamp that goes onto a cable (I have no idea how it works).
This data stream then goes to board2 where an AVR looks over the data and makes
decisions based on what it finds....this board also has a prop on board to handle some timing and
VGA and kbrd duties if ever needed. Board3 is in a separate case and has a prop and can be used at
a remote location to connect to the main package via a 2 wire cable...it can power
the remote package and communicate over the 2 wires... it's a remote terminal
that uses a prop...it replaced a laptop computer. The external 12v power brick for the VGA monitor
also can power all 3 boards from the remote terminal's location. When this is done then no power
supply is needed at the cable location which is really nice since it's not always easy to find
power. I'm still experimenting to see how far apart the 2 packages can be before loss in the cable makes it unworkable.
The boards don't use lots of power so the one 12v brick seems to be enough. When I check the total power
used from the power brick it is below it's stated rating when powering the boards and the monitor.
Originally the 2nd board was to have an ARM but I managed to squeeze enough
juice out of a 20mips Atmega644 to do the job.
▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
- Some mornings I wake up cranky.....but usually I just let him sleep -
heater and Holly, I'm also interested in such a board. Andre's board is moderately interesting, but I would want something with closer to 0.5MB code space (LPC2048). What I miss when I work with the LPC2048 is convenient JTAG ... are they all printer port dongles? ... ugh! Is there anything USB2.0 like for JTAG? If so, please recommend it!
▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
--Steve
Propeller Tools
Yes, but they don't stink, don't make smoke, make no noise. They are LAME!
Nick
▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
Never use force, just go for a bigger hammer!
The DIY Digital-Readout for mills, lathes etc.:
YADRO
Nonsense [noparse]:)[/noparse]
A handful of us built Sterling engines in metal working class at school when I was 13/14.
Covered in oil and heated from a blow lamp all that burning oil makes a fine stink and plenty of smoke.
Lots of clunky noise due our fairly lax metal working tolerances.
No power to speak of but a good turn of speed.
▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
For me, the past is not over yet.
Naaa! You have no idea how an IC-engine has to sound! First of all, Stirlings are are no ICs, so they fail completely. Second, the sound is coming from the intake and the exhaust. The rest is noise. And I don't want noise, I want SOUND!
I have built this engine from a set of castings:
<http://www.motor-manufaktur.de/ellwe/pics/ellwe_2fb_m10.JPG>
The seller of the castings was astonished when he first heard my engine. His sounded like a bunch of cans torn behind a car ("Just married"), mine sounds like this:
<
No, in the beginning, that's one of my cats!
There are a few tricks how to build a muffler that sounds good.
Sorry for OT!
Nick
▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
Never use force, just go for a bigger hammer!
The DIY Digital-Readout for mills, lathes etc.:
YADRO
cheers
- Howard
▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
I use a Rowley CrossConnect USB JTAG interface with their CrossWorks tools for ARM development.
Leon
▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
Amateur radio callsign: G1HSM
Suzuki SV1000S motorcycle
Real world benchmarks say like controlling a engine from a BMW 700 series OTOH would be more interesting. Though that would comparing the Prop against Freescale's micros which do the processing in that case.
As you have seen above, each cog is totally independent, each with it's own registers, flexible counters and 496 longs (32bits) of instruction/data. In addition to this, it has access to shared hub memory which is 32KB or RAM and 32KB of ROM.
Each cog instruction takes 6 clocks (a few are different) but due to single pipelining, they result in 4 clocks. Each cog runs it's own independent program. Each cog can run the Spin Interpreter, which is itself a PASM program. So this means with the standard 80MHz clock, each instruction takes 4x12.5nS=50nS.
Now, if you want to see an example of interleaving 4 cogs to do sampling of the pins at 12.5nS see my Data Logger object. BTW I am not the only one to have done this. see the "Prop Tools under Development or Completed" link in my signature below and look for the Data Logger post for the link to the thread and code.
▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
Links to other interesting threads:
· Home of the MultiBladeProps: TriBladeProp, RamBlade, TwinBlade,·SixBlade, website
· Single Board Computer:·3 Propeller ICs·and a·TriBladeProp board (ZiCog Z80 Emulator)
· Prop Tools under Development or Completed (Index)
· Emulators: Micros eg Altair, and Terminals eg VT100 (Index) ZiCog (Z80) , MoCog (6809)
· Search the Propeller forums·(uses advanced Google search)
My cruising website is: ·www.bluemagic.biz·· MultiBladeProp is: www.bluemagic.biz/cluso.htm
What people are saying about real time deterministic behavior. Thats all doable with regular micros. I've rarely met a real time situation that was so hard that I had to synchronize behavior using hardware exact to the picosecond or whatever. I'm not saying they don't exist. Just for the types of apps I typically write they aren't that hard. You don't need to mess with an RTOS, which is nice and you don't need to do context switching, which is also nice
I need to think about this and whether the performance is enough, with enough memory to do what I want for the price. This is an important issue. You also get a USB interface (if I've read the specs correctly) with almost no extra parts needed.
$8 and $1.50 for the EEPROM.
Too bad it doesn't have an internal flash instead of requiring the external EEPROM, but it's not a huge deal. One application I'm really interested in pursuing lately is using small microprocessors instead of FPGAs. FPGAs are about $13 for a spartan 3. The propeller is an easier chip to add to a circuit. I like the idea of stamping a preprogrammed chip like it's an ASIC. Check out, for example, the SoundGin processor using a PIC.
Oh, BTW, can you protect your EEPROM data so it's only readable by the propeller so someone doesn't steal your software?
I'm sold on the propeller now for what it can do. I need to resolve the security issue above and then decide on a few issues. Long term I definitely want to get one of these to play with
Thanks for all the good suggestions....not sure how we got onto an app about engines
_D
Thanks...see that topic for some ideas.
▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
--Steve
Propeller Tools
Ale, Not really sure the "We want to know what kind of motors are you talking about !?" question was aimed at me, but just in case it was, I was talking about the turbo molecular turbine pump motors used for high vacuum work. They are brushless DC motors similar to what is used in the hard drive spindle motors. See http://en.wikipedia.org/wiki/Turbomolecular_pump
Yes, almost everything I do is around that or repairing / improving / retrofitting machine tools.
Nick
▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
Never use force, just go for a bigger hammer!
The DIY Digital-Readout for mills, lathes etc.:
YADRO
Great engines. Reminds me of the sort of thing My father used to do (usually with just a file and a pencil!)
▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
Style and grace : Nil point
No it is not. A hyper simple example.
0) Wait for an input clock edge on a pin.
1) Read two bits from other 2 pins.
2) Form some logical result from those two bits, oh something simple like AND.
3) Output that result on a fourth pin after exactly some multiple of 50ns. +/- 25ns.
Easy perhaps any suitably fast micro can do that.
BUT now do that or something similar with 3 or 4 sets of similar pins each with their own unrelated clock inputs at the same time.
OK this is a very artificial example to make the point. But that kind of timing requirement does exist in many applications. Without the Prop one ends up having to implement the solution in hardware logic or an FPGA.
▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
For me, the past is not over yet.
I think Nick answered about the engines but I seem to have missed the bit where he said which engine went up to 60 krpm.
▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
Visit some of my articles at Propeller Wiki:
MATH on the propeller propeller.wikispaces.com/MATH
pPropQL: propeller.wikispaces.com/pPropQL
pPropQL020: propeller.wikispaces.com/pPropQL020
OMU for the pPropQL/020 propeller.wikispaces.com/OMU
The next thing you know, you want a 1/4 scale V8 to put into your 1/4 scale racer ;-)
▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
Post Edited (Agent420) : 8/31/2009 1:42:15 PM GMT
▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
Visit some of my articles at Propeller Wiki:
MATH on the propeller propeller.wikispaces.com/MATH
pPropQL: propeller.wikispaces.com/pPropQL
pPropQL020: propeller.wikispaces.com/pPropQL020
OMU for the pPropQL/020 propeller.wikispaces.com/OMU
The little-bitty crank and internal parts shown are awefully cute ;-)
▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
For me, the past is not over yet.
I'm talking real time apps. An RTOS can do what I (note I'm not saying you) need to do typically. Anything that requires immediate behavior....well, lets just say, there are situations that require hardware. The question is whether it's so complicated you need a propeller to do it cheaper than (like you suggest) an FPGA or CPLD.
Like I said, I'm sold on the propeller for what it can do.
I have some solutions for security on the EEPROM that I posted on a separate discussion I created....basically involves some clever decoding and the use of a 41 cent locked PIC on the I2C bus to hold part of the decode key.
I've gotten my answer from this forum, but I still haven't seen any comparison other than the math, which looks good.
This chip is capable of 160 MIPS for $8 which is pretty good. I have a few questions about the chip, but I'm going to put those on a different forum. This one is almost unreadable because of the engine chat
Thanks to all who replied. I got the answers I needed and those that are available.
-D
Yes indeed my hyper simple example is a gate. That was the whole point. Such a simple thing but as the problem is put most micro-controller would stumble on it.
I claim that the example is a very good example of a real-time application. "But you can do it with a 5c logic chip" you quite rightly say. Well it is only an example to demonstrate the sharp end of the real-timeness. What if I extend the specification with one more requirement.
"The logic functions to be performed on the two input bits is to be configurable at run time by a human operator with either a graphical or command line interface to be either AND, OR, XOR, or NOT. The output can also be set to 1, 0, input A or input B my the same interface."
Now we have a full up real-time app. The 5c chip won't do it. An FPGA could. Most mico-controllers won't without hardware assistance. Whilst one is designing/implementing that the Prop user has gone home early.
I think the main design idea of the Prop is nothing to do with compute performance. It's just the simple concept that - instead of bolting a bunch of peripheral functions to a CPU like UARTS, SPI, I2C, USB etc etc. Lets have some little CPUs that can be programmed to do many of those things. Then there need only be one chip instead of the huge array different PIC, AVR etc variants. There is a lot more flexibility.
As a bonus side effect that design decision results in something that is dead easy to program because all the component parts are isolated.
▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
For me, the past is not over yet.
No, there is no such Internal Combustion engine! Even the small competionion 2 strokes (for R/C models) stop at about 30000. But they don't have spark plugs.
Nick
▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
Never use force, just go for a bigger hammer!
The DIY Digital-Readout for mills, lathes etc.:
YADRO