EEVblog
Vložit
- čas přidán 24. 10. 2018
- Getting the 3 cent Padauk microcontroller to blink a LED.
Or at least the in-circuit emulator blinking a LED with C.
3 cent micro & LCSC: • EEVblog #1132 - The 3 ...
UPDATE: LCSC have just said that they will offer a free programming service for a limited time to anyone. Just email them when you place your order.
www.padauk.com.tw
Forum: www.eevblog.com/forum/blog/ee...
EEVblog Main Web Site: www.eevblog.com
The 2nd EEVblog Channel: / eevblog2
Support the EEVblog through Patreon!
/ eevblog
AliExpress Affiliate: s.click.aliexpress.com/e/c2LRpe8g
Buy anything through that link and Dave gets a commission at no cost to you.
Stuff I recommend:
kit.com/EEVblog/
Donate With Bitcoin & Other Crypto Currencies!
www.eevblog.com/crypto-currency/
T-Shirts: teespring.com/stores/eevblog
Likecoin - Coins for Likes: likecoin.pro/@eevblog/dil9/hcq3 - Věda a technologie
Those chips are total candy. I've never felt so excited about a microcontroller since the PICs and their development system and programmers first became affordable to minions like us. (PIC16C54 and EPROM development version era) It would be great if they had a flash (Or EPROM?) variant in the family and the programming algorithm got added to the TL866 style universal programmers. The English manual for the chips is very good, with a very readable guide to the assembly code instructions. As you were looking through the data I particularly liked the reference to the ready-baked 1Hz interrupt on a 32.768 kHz crystal.
Yeah I noticed that 1Hz thing, that's groovy
Have you played with it, Clive?
@@NicksStuff I've not. It requires a specialist programmer. It would have been nice if it had used a common multi-chip programmer.
UPDATE: LCSC have just said that they will offer the free programming service for a limited time to anyone. Just email them when you place your order.
errr, you're not planning on using this in uSupply are you ?
why not?
3 cent? Does that include shipping?
One of the authorized vendor ( skywiner.taobao.com/ ) can do the programming at 0.02 yuan /pcs if the volume is above 10K.
MowSong Ng - Wow, 0.3 cents per chip programmed, practically free. 10k qty at 2.5 cents is only $250 worth!
6:22 Say what you will, but the start-up time of that IDE is exactly what the start-up time of an IDE should be on a 2018 computer.
Truth. No software companies seem to care about speed anymore, it's all about how much bullshit bloated bells and whistles they can cram in.
Performance is a feature.
Why would they care when they have i7's and 8GB ram to play with? Same with games. It's got to the point now where a family member upgraded to a 1080 because a new game is early access and is choppy on a GTX980. This game will never leave early access. Devs are spoiled for resources and aren't responsible with them.
Somewhat lacking in features though.
Hell yeah, that's why I switched to visual studio code from the community version, fuck that bloated piece of shit.
How many time you start up your IDE? Maybe once a day, or maybe you leave it always open. Is not that big deal. I prefer waste a couple of seconds the start-up but have a rich IDE like the ones from JetBrains that will make me save hours of writing code and debugging.
You've possibly found out what the unmarked 8 pin micro controller is that Big Clive always seems to come across. Something like this must be in literally millions (maybe billions) of Chinese made items, which would explain the low price. Maybe yourself and Big Clive could team up with the 'electronupdate' guy who likes decapping IC's?
At these prices I doubt there is any issue with intellectual property. They would probably sell the bare die or even give you a copy of the documentation if asked. I don't think there is any possibility of someone trying to undercut the design. This is already as low as it could possibly go. They would probably find it amusing if someone were dumb enough to try. The more attention that is brought to their products the more they benefit from market exposure. It is in their best interest to share anything they can possibly share right now in order to increase this current surge in popularity.
Nothing to do with 'stealing' a design, just solving Big Clives mystery.
czcams.com/video/5jw5D0F008c/video.html he decapped it
You’re like a boy in a candy store; so very fun to share your excitement.
Who is going to be the first person to program it as a 555 replacement?
I though you were on to something but the 555 is a relatively precision timing chip, so to match it you would likely have to add an external crystal to the micro
@@DavidScheiber Not everything using a 555 needs to have precision timing, especially when timing at the second or minute level.
I already did that with the attiny4. iirc i've put it on github
EDIT: there it is gist.github.com/5N44P/3b7f3be883065c6469940819f6ff9fe0
@@DavidScheiber how the 555 can be a precision timer if the time constant of the system is given by an RC circuit?
David Scheiber - 555 is not even close the relatively precision timing. The internal RC oscillators on micros are typically much better controlled than a 555 (typical >=5% cap value)
This is really great, I'd love to see you do more projects with these little micro's :) The fact that they wrote the development software nice and compact with all of that code-help is just downright respectable. I'm impressed!
Not sure if you missed it, but on the lcsc site there is a link to the manual/data sheet for each micro, which includes description of the instruction set, registers, etc. It's in Chinese and due to its size you can't just go to Google Translate and select the file, however you can import it into Google Docs which has a translate option under to the Tools menu. You can then download it (From the File menu) using quite a few different formats including PDF, docx, etc. The translation isn't bad at all and is very usable, not sure if this is how they generate the English versions of their other tools since it seems pretty similar.
It's not like Google Translate _ever_ works for Chinese...
I've had success with some datasheets, but I agree, there was WTF moments for sure, like pins "hanging" themselves... (that's a pullup btw hehehe)
www.onlinedoctranslator.com/translationprocess-pdf
There is an english version with instructions on the Padauk site.
You just need to look for PMS150 series instead of C model.
www.padauk.com.tw/upload/doc/PMC150,%20PMS150%20datasheet%20V007_EN_20170613.pdf
@@robimarko 404 error now.
This is exactly where Moore's law still is valid. A speck of smart silicon dust that we can sprinkle over almost anything at almost zero price.
That's not Moore's law. It's half of it. Trace width is down to 5 nm I think now...this uses 19nm process(which I am not saying is "old") but Moore's law is that you get 2x the performance for 1/2 the price... 8Mhz with 1K ram is pretty fast and large!!! For 1974. What this does is use some cool tricks in the architecture to use minimal die size, thus making it higher yield per silicon wafer run and cheaper per IC. Smaller size die = more chips per run(fewer fails during process) & more chips altogether squeezed on one wafer. IE These are 3 cents because they are so simple and small. Brilliant idea! The idea of producing stripped down stuff hasn't been popular in mainstream companies for eons now...if it ever was. This kind of stuff is more reflective of 1950s/1960s USA computer tech where every transistor was scrutinized.
I once bought ~2300 AtTiny2313 and programmed them myself. Had a little robot + script that did most of the "work". you needed to drop a stack of them in a cardboard "feeder", it would align, program and "test" them (mostly if they booted at all). Finished in ~90h, only 2 of which required actual human attention if you don't count refilling the stack and giving it a light tap on a semi-regular basis.
You could probably automate the light tap with some simple mechanism, like with a slow motor or something
Dude, definitely buy another small stack and make a video of that
That sounds very intriguing. Any chance of a CZcams video to explain I.
More than the chip I'm impressed with the IDE. 3mb download and does everything it should with nice help. Bugfixes and better translation and it would really be awesome
They seem to work on that. Lot of chinese crap just dont give a shit about manuals and documentation but here it seems promising. Lets hope those chips will appear on aliexpress soon ;) to get them to CZ.
the IDE is no longer aviable, where i can find them?
@LazicStefan Got to store the ads somewhere
@@robertohurtado6458 I mean there is probably copy of that on eevblog forum
@@Shonicheck padauk respond me email and fix the official link
My parents went to see Coldplay, and they were given these light-up wrist band with LEDs and a radio receiver inside, which were remote controlled to flash in different patterns based on the music. This is the sort of throw-away application that a super-cheap micro would be useful for. Something simple, maybe one-time use, that you might need in huge quantities.
Actually, the Coldplay thing is a great idea. Say you wanted it to only switch on the radio receiver during the actual concert, then afterwards it switches to autonomous mode, controlled by the micro. Not having to run a radio receiver would probably use less power, so it'd last longer, and it could have user-replaceable batteries so it's not useless after it runs flat. The micro could be programmed to switch to the autonomous mode if it doesn't hear a signal from the receiver after a short time.
Crowdsync?
but please, can we reduce trash? most of it never gonna be used or recycle again.
@@youkofoxy my parents still have their wrist bands sitting on a shelf with other trinkets from their travels, they just don't do anything. But I agree, they should definitely think about making these sort of single-use gadgets useful after their intended purpose is finished.
13:48 actually the manual explains that you do not need 9v power as the ice is powered directly by usb. The 9v input is only used if there is something power hungry plugined to the ice.
Instruction set: PMS150C datasheet, pages 50-61. Summary on 62.
Awesome video. Learned a lot here and looking forward to follow ups
Great video! Looking forward to the next part :)
Up/Down buttons with PWM and passive filter as a replacement for a potentiometer.
PID controller
Simple single push button interface to differentiate single press/double press/hold with separate output functions.
Incredible, I remember when I started with microcontrollers you had to get special (expensive) windowed devices and each build run cycle required erasing the devices in a UV eraser and reprogramming them so I tended to write much larger blocks before testing whereas today it is easier to almost write line by line and test each small block as you go. The first ICE I purchased made life so much easier as I could develop in 'real time' but it cost £68,000.00 and you needed a separate (£2800.00) module for each device that you wanted to use so you had to be really serious about product development before starting. It is good that today the same can be had for just a few $'s.
I'll love to think this will led us to a new kind of industrial revolution.
Yes that would be good but I suspect it will lead to things such as pencils with flashing LED's. What we really need to complement these cheap devices is a cheaper power source. My company develops embedded systems and we are now in a position that in the devices using very low cost microcontrollers the battery is now the dominant cost so saving a few pence on the micro would not really make much difference. In fact the microcontrollers have generally been the lower cost components for a while. It will be interesting to see what develops.
Came to learn something new, stayed for the boundless enthousiasm :)
Not sure if you missed it, but on the lcsc site there is a link to the manual/data sheet for each micro, which includes description of the instruction set, registers, etc.
1:53 D-AvE's inner Canadian coming out.
These are one step in the evolution of electronics toward being software-defined. Gone are the times when you had to fiddle with caps, resistors, etc. when you wanted to have something like a "reset" button you have to press for 3 seconds. How you can just add a 3-cent micro and code that condition. Want a flashing LED? Add another one for that. Want your monitor to fail after 2 years but you cannot get those "bad" caps tuned perfectly? ...
I've used PICs to flash a single channel of LEDs in the past. It was for a TV prop and I had an existing module with a MOSFET output drive on it, so it was faster and easier just to whack a loop in assembler with two delays. It flashed the indicators (custom LED panels) on a small car driven by puppets.
Easy peasy lemon squeezy
They need a transistor-lookalike version so you won't even notice it on the board when checking for bad caps.
With built-in clock, higher voltage to program and single output it is theoretically possible.
Now I realize that 1-wire was a Chinese plot to infiltrate the world. /s
The reason you can't find the explanation of why runs at 2Mhz is because it's clock is "1/2n" but you are missing "n" at 15:19.
Some chap already has some kit using this to provide full instrumentation on a Honda lawnmower engine. Waltzing Mathilda!
ohhhh, been waiting for this!
They custom wrote a compiler for their CPU to maximize ram/speed on their extremely small size mcu.
I doubt gcc would work, or if it did it would be slow or use up most of the ram
That's what I was thinking, too. There's not a lot of memory on there and there are a lot of opportunities for optimization the manufacturer would be in a better position to take advantage of.
We have all the programming procedures in the data sheet. So we should be able to rig up an ICP header on your boards. It might be a viable option for lowish runs.
I don't see the programming info in the datasheet?
Excellent video Dave. Hopefully your interest in their parts prompts the manufacturer to further improve the tools and documentation for english speaking engineers.
Kind of interesting you didnt mention that timing glitch with the Logic Analyser is called aliasing - most people know the term but wouldnt connect that this is a form of it
It could have been as well an ICE glitch. I actually suspect that ICE is a dummy USB-to-parallel adapter and all the emulation happens on the PC side. No timing guaranteed.
When he zooms out it can also be viewed as a beat frequency, since the two portions of different frequency had a “duty cycle” that looked equal to the duty cycle of the toggling pin. It would be nice if those logic analysers could utilise the oversampling he’s getting, or to have an external trigger and frequency multiplier of some sort. Even just error bars. But it is a fairly archetypal description of time-domain aliasing.
error bars would actually be freaking awesome
Not quite. Aliasing happens when you try to sample a frequency that is higher than the half the sample rate - the apparent sampled frequency is reflected around f/2. See Nyquist Theorem.
So sampling @16kHz can only resolve up to 8kHz. If you try to sample 9kHz with it, you get a 7kHz signal. Importantly, you get the same 7kHz *aliased* frequency when trying to sample 23, 25, 39, 41 kHz etc.
Obviously this sounds horrible as all high harmonic content gets mangled and reflected into hearing range. This is why sharp low pass filters are used that start their cut-off at 2/5 of the sample frequency below the Nyquist frequency, f/2.
The sampling distortion with the logic analyser and the pic signal is more as a result of phase shifting that causing Moiré fringes, much like looking through two combs placed on top of each other or the railings on motorway bridges.
Phase locking or synchronizing both clocks would give more accurate results.
@@frigzy3748 The timing couldn't be happening on the PC at 2 Mhz, USB is 1khz max. I believe the ICE is totally emulating it.
I'm not gonna lie, the IDE's functionality looks pretty decent for my standards. China you scary.
Also I kind of chuckled over "Somebody bought 10.000 of these", because at first I was really impressed, then I realized that's 300$.
Why scary ? Pretty much everything is made there nowadays and they're even growing more and more into the open source mindset be it hardware and software.
The in stock quantity has gone down from 1600 to 500 just since you filmed the video. Seems you are good advertising for them.
they have earned $33 for that lol
Plus those people might have thrown other things into their cart, and perhaps some repeat business in the future. Not bad for a couple of hours.
How are we gonna program these things? are we all waiting for a opensause and openhardware programmer?
@@gn_ghost4757 i think they get subsidies from the Gouvernement when they sell stuff outside CN.
maybe not subsidies but export tax reduction.
Fabulous.
Truly we live in the future.
Superb video again, how about an idea for a future one about using a logic analyser?
I think that to run it in full speed you need to disable divide clock bit or something like this, example as in atmega micros.
Hi Dave.
Can you hook up analyzer when programing and post results?
Whole flash program if possible.
Post binary source also.
Thank you.
The 2MHz sample isn't taking into account the loop instruction? So it should be 4MHz assuming it's a single JMP and flip.
if it has a single cycle instruction like that. And it might not.
On stm32 to write the shortest blinky possible I used a rotate instruction to shift 1 between upper and lower 16 bits on one of the GPIO registers, so I didn't have to use two delays and GPIO writes in the loop.
3:24 sailing create brilliant. best slogan ever 😂
Thank you!
Dave, can you make a self contained u current. With adc , outputing readings via Usb using this 3 cent microcontroller?
no
For 3c, you could replace a whole load of SSI logic that does not run at any speed with it, or even just put a port expander on it to make a low cost random number display using a cheap dot matrix LED. You could even do that to make a custom font 7 segment display, one IC for clock input plus another 4 driving the LED display (after all 3 of them are cheaper than a shift register) and upgrade the old clock circuit with things like slow fade from digit to digit, or even a roll to emulate a flip display.
I think i have read about mini-c in embedded controller context before. IF i remember correctly it's a predecessor+compiler that is designed to be an expanded macro-assembler to look more like C. It will not create much overhead. By example by avoiding a stack if possible. But i can't remember the processor type it was initially created. I can remember that it was ported to a few different systems. (My memory said that i found it while searching for efficient higher 68000 languages, but i relative sure I'm wrong. It seams unlikely that the project is that old ...)
I did a quick search but did not find what i remember.
Just in case anyone is wondering, the manual says that it is recommended to attach 9 volts if a high current is drawn from the emulator. Auxiliary poweryou could say.
Though, i am not sure if they have any chips that can handle that much current anyway. But maybe some of their larger chips can.
This was fun to watch. Makes MPLab X seem almost pleasant to work with ho ho.
Linear feedback shift register generating bits output to a pin. Flicker an electronic candle, or a noise source.
Brilliant. I love it.
One wonders how difficult it would be to build a config file to allow you to use a TL-866 to program it. Does anyone know how the TL-866 handles it's config files? Is there any published standard?
How big of a program can you fit on there, and does the compiler tell you when you're over the limit?
It says it in the lower left corner; ROM size, program size and how much memory is left.
Yes, I showed that in the video, % of memory left.
Thanks! I see "64 unused memory", but is that in bytes? instructions? Or something else?
Thanks Sir for your video n i'am get study in home... Peace from Indonesia
They even make multicore uCs. That could be awesome. I.e. one micro to do serial async communication with other uCs or PC, and others doing stuff more precisely uninterrupted for example, and polling data from main core, without sacrificing timing accuracy. A lot of funny things could be done with this. The IDE and their mini language is a bit funky, but honestly you do not need much for most projects.
This should be perfect as a HSV/RGB LED controller
Make a Banksy frame controller. Receive IR command sequence, turn on shredder motor for 15 seconds.
dumb question: whats the equivalent price for this level of functionality elsewhere?
Nice Dave.
i love the ide style
If this thing contains an ADC you could potentioally used those as a DC-DC buck controller. This would be would be revolutionary because you can never find them below $0.50
One of the bigger models in this series has an ADC
Well you could also just buy some lcsc.com/product-detail/DC-DC-Converters_M3406_C83224.html they are also around the 0.03 UDS mark
@@TalpaDK Nice one, the only problem is the Max input voltage, but that could come in handy if in need of a 5V -> 3.3V buck
@@RemiBusseuil Well they also have some massively more expensive ones (4x the cost at ~ $0.14) than can handle 20V (I just randomly punched in that number :) lcsc.com/product-detail/DC-DC-Converters_Techcode_TD1482AP_TD1482AP_C177808.html haven't checked if a pair of 20V mosfets can be had for less than $0.11
Also if one did roll a custom regulator one could potentially get some additional features, comms over SPI... maybe be able to change the voltage (atmels often officially only runs a 8MHz @ 3.3V) so reduce the voltage during "sleep" and increase it when processing or reading status info back, say the loading (pwm duty cycle)
I cant believe it really exists
If it blinks, it's real.
@@EEVblog I want that on a t-shirt.
Do these chips got runtime memory, or after you program them they run like clockwork? Like, for example, could you make it so they can accept an input that defines the a blinking pattern and duration of the pattern? Could you use them as the "neurons" on a hardware neural network that can learn different patterns?
So why didn't you use the scope instead???
So the FPPA0 stands for main; however it seems that this identifies the processing core...
I think this means that the FPPA1-7 seen in the PWM program are individual processing strings that can execute different programs simultaneously? ._o
3MB for IDE and compiler? Man these guys really have to catch up with modern software development. My mouse driver is 20 times bigger than that. 😅
I really like that it is so small. Makes developing quick-n-dirty projects for super simple non-critical software a breeze.
More like modern software development has to become less bloated
Totally! I hate it when I want to just quickly mess with something, and I have to install 3GB worth of software I might never use again. Instant motivation killer.
No - modern software development needs to catch up with these guys. The 3MB IDE seems to be doing everything what my 3GB+ Eclipse does and in some aspects it actually does better.
I loved eclipse when I started using it 10+ years ago (wow, I'm getting old), but today it's soo bloated and slow. Forced to stay with it due to customer demands though :'(
The PMS154 looks really interesting for a RGB LED controller after your interview with Mike on the amp hour (since it has the 3 11bit pwm outputs). I'm sure Padauk would be willing to sell them in die form for integration into an LED package. And that's where a 3c micro really makes sense. BTW I saw pictures of the bank of automated programmers on the taobao page of one of the distributors and the programmer help file mentions control for the automated programmer, so not surprisingly that is an option.
Ah, nice. Yeah, and LCSC can order parts pre-programmed for you, although this is not yet offered to the public.
The IDE looks to be better than Aurdino IDE !
I was thinking the same thing.
For Arduino you can use the free version of Visual Studio with the free Visual Micro addon and you get all the nice modern features the Arduio IDE is missing.
You can also program Arduinos with VS Code but I prefer Visual Studio.
Notepad is better than the Arduino IDE.
@@hasanmansour6849 , VIM is perfect for Arduino and any GCC / ASM stuff.
Arducrap is not even worth mentioning. Gawd I hate that "IDE" and their ridiculous HAL. Just learn how to read a damn datasheet. :/
The buttons of the IDE are looking like these from the Keil uVision IDE.
Your enthusiasm is infectious !! But for 3 cents OTP we really need the ICE. As a hobbiest, I would be more interested in the Multi Time Programmable parts. I do like their IDE concept, small and clean. A GCC port would be fat, how about convert a TCC or SDCC version? Can anyone translate the ICE docs ?
Is it possible that these micros are usually programmed at the time of PCB installation by a special pick-and-place machine that also does programming? Even with low labour rates, I can't see individual chips being programmed one at a time by a human operator.
In a factory they would use something like that, most likely. Or they might even have as part of their line something that injects code - who knows. Like you said no way are they just using humans doing one chip at a time. No one does that on an industrial scale.
can you sand off the epoxy and look at the chip itself? I would like to know how big it is.
I don't if you still want it
Here you go czcams.com/video/5jw5D0F008c/video.html
I would love to make a servo motor or stepper motor based robot arm (Very short, like 15cm) with a suction tip that could take these ICs out of their carrier strip, place them into a ZIF socket, then another motor/solenoid would close the ZIF socket, then it would program the IC, then remove the IC and put it back into the carrier strip, or into a different strip. Then repeat, and grab the next IC in the carrier strip. Until it has programmed them all, or stop after programming a user-defined number of ICs. There would also be a dual heated roller mechanism that would reseal the cover strip of clear plastic of the carrier strip.
Optionally the ZIF socket could be connected to a test circuit, not a programmer, to ensure that the IC was programmed correctly. Or this could be part of the programming setup (to test after programming).
This would be great for people who didn't need 10,000 ICs programmed, but maybe 50 - 5000
That's Cheaper then a 555 and the R/C's
Yep, crazy.
Could you use a pin to change the blink rate, such a an adc pin?
Pot one of these into a counterfeit 555 times package and really confuse people when they try to reverse engineer your production project.
Dude the original 555 has the die size where you could probably place the hole waver of those in xD
How do you think the average quality of these very cheap ICs is? I suppose they're already a better deal than blackmarket aliexpress chips. Do you think you can rely on the datasheets? I'm not only curious about µC's but also analog IC's.
This is really interesting, what about making a product with them on the next episodes ?? :D ?
I think we all came here for the large program and that is, for want of better phrasing, what I really want to see... Please do your big project idea... Even if it fails it wouldn't be an expensive failure and I reckon if it works it would be epic!!!
That micro would be perfect for debouncing buttons in everything that needs debouncing
I guess it's called FPPA0 instead of main() to support multi-core models. So you can use FPPA1 to execute on the second core, perhaps?
Yes, I think that's the reason.
This is all seriously impressive. I bet eventually someone will figure out a way to program these with the Arduino ecosystem... If it hasn't been done already
Also funny how the software looks arguably better than some FPGA suites and such lol. Hats off to them
Internal interrupts might change that timing.
hmmmm i could make a software emulator for this if it doesn't already exist. i wonder if anyone would find that useful?
I'm sure someone would!
build a gcc backend instead it's much more useful!
I agree a gcc backend would be more useful, as we have the fairly cheap ICE for real circuit emulation.
A reverse engineered bitbanging flasher would also be valuable....
How much support do we need to make this project a thing?
SDCC is easier to retarget than GCC, clang I think would also be easier but nobody seems to re target clang?
I wish most of the IDEs I use were that straightforward.
Very useful 👍👍👍👍
1600 units in stock. Dave estimates LCSC is making 1c per unit, That's a good $16 worth of profit there!
Gcc needs a lot of resources, like registers. Try SDCC first.
This uC makes me want to attempt some stupid crazy project, like a high-res LED matrix w/o special drivers, just pure i/o lines.. or a LED cube :D
Analyser timing: A video on why nyquest doesn't apply, in that twice the signal frequency is enough to accurately reproduce the input might be worthwhile. It took me a bit of thinking to realise why.
Even though the cycle rate is only 2MHz the fact its a square wave means you can't band limit.
I'm almost tempted to buy an entire reel of these things because, well, why not?
I was looking for this kind of stuff(Cheap Chinese micros)so long... and here it is... COOOL...
G'day Dave :)
G'day
Awesome!
Why Up and DOWN cycle not the same? (on first example)
while loop take time?
maybe put
LED ON
LED OFF
LED ON
LED OFF
to see a difference
Because of aliasing from the logic analyzer not sampling fast enough. Dave mentions it in the video.
If you look at the example code, between the line that sets the output to 0 and the line that sets it to 1 there is not code, but, in order to set it back to 0, the program have to jump back to the previous line (that is what the while loop does) and that takes a bit of extra time. So the output will be always a bit longer on high state than on low state.
On the other hand, you also have the aliasing issue with the logic analyser, as @drelephanttube already pointed out.
en.wikipedia.org/wiki/Aliasing
@@drelephanttube Sampling is fast enough to show the unequal markspace.
It does not show aliasing at all - it just shows sample error because of phase shift.
Aliasing happens when trying to sample a frequency greater than half the sample rate, which is not the case in this example!
The instruction set is listed in the datasheet starting at page 50.
www.padauk.com.tw/upload/doc/PMS150C%20datasheet%20V004_EN_20180124.pdf
It looks like a pretty decent instruction set for such a cheap microcontroller. It seems pretty simple to work with and there are only like 19 configuration registers. If the ICE wasn't so hard to get, it would be fun to play around with.
Good vid
that softwares got some good features
MiniC is a subset of C that is far simpler and easier to write a compiler for. It's not a standard by any means so this IDE will be modifying it slightly.
This feels like the ICs from the game Shenzhen I/O.
3mb IDE and compiler - I wish Atmel and TI would do the same instead of the bloat they push.
You 'Got to Blinky'! Have you informed Chris Gammell yet?
Fast or slow?
No idea.
I cracked up lol.
cool. It takes me using a STM32 uController to blink an LED.
Wow, you do that without IFTTT? Impressive.
Idea for a project: Take I/Q outputs from a shaft encoder and make "up" and "down" pulses come out.
Another idea: Take I/Q outputs from a mouse (4 inputs) and make standard mouse serial output. This is probably a use case for a small processor. It should easily fit in an 8 pin micro. Good luck.
I was thinking exactly the same thing. Let's go do that!
Sold out MCUs, free programmers... That company loves you, Dave.
I want to see a decapping of one of these microcontrollers.
Now WHY did I know that you couldn’t resist, hee hee … :-)