Hi, yes, I called it the St Fanboys have come out of the woodwork and they said in my previous video Part two of my new Uh Project series that I'm doing here is that I chose the wrong micro controller because I I stopped when I You know, Look, there are dozens of different types of microcontrollers that can do this application. I'm sure of it. and what are you going to go through every single one? Sometimes you've just got to stop pick one and run with it because it suits your application. But it's interesting to compare microcontrollers.

You can do it until the cows come home. So let's actually do it. Yes, the Stmr Fanboys they're as rabid as the Apple uh Fanboys they've all come out of the woodwork and they've said no I chose the wrong microcontroller. We should be using the Um and the Stm32l Zero series because they're lower power, didn't you know? Okay, yes, I'm pretty sure they are, but that's not the only part of the equation.

I'm actually very well aware of the Stmr 32 2L series because I actually specified the Stm32l1 series, not The l0 the L1 series into the 121 GW multimeter. Pretty powerful. It was relatively uh, cheap at the time and and available at the time. We'll get into that um and it had the you know, the LCD driver signal was it was pretty low power and you know all that sort of stuff.

So yeah, I'm very familiar with uh, that. so let's do a quick comparison here. Between the Stm32l0 and the Pic24 Fj128, it's not going to be hugely in depth because if you want to go down the rabbit hole, you might not even be able to get the power consumption figures from the data sheet because it depends on all the peripherals that you're using. you're driving the LCD What's your type of LCD Like, how are you driving it all that sort of stuff.

So you might find yeah, that uh, the on paper one micro might be lower power than the other. but when you actually use it in your application, the only way to tell is to actually build it up and measure it. So let's have a squares here at the 32l0 ultra low power Mcus. Now they're going to be lower power because they're like using a better process node than the older Pic24f Series So you know you can normally get a lower power it.

They claim to be the world's lowest power consumption at 125 degrees C and that's another trap for young players. Is that on on the top level better spec for the power consumption. You know the uh micro amps per megahertz figure might be lower. but then you might use your application in a you know, it's working in a 50 degree C environment or something like that or even 125 degrees for example.

And yeah, it's not the same power consumption. It can radically go up with, uh, temperature. So yeah, you've got to be careful with that. um that.

once again, that's something that. you may not actually get from the data sheets. And you may actually, if you're really critical about your power consumption, you may actually have to build it up and test it. Current Consumption Reference values: Dynamic Run mode down to 49 microamps per megahertz hurts and right off the bat.

I Know that that's actually lower power than the 24 F series. We're gonna have a look at that in a minute, but that's with an external uh DC to DC uh converter. Whereas at the actual figure, if you use the internal Uh audio because I believe this has an internal Ldo in it, it's a 76 microamps uh per megahertz. As a comparison, the peak 24 Fga family here.

Um, once again, this is not in the data sheet. This is just like a top level family. Uh, spec. Uh, run currents down to 150 microamps per megahertz.

So we're basically talking double the power consumption and that sounds pretty horrible, right? You go. Oh well, of course that's the wrong choice, right? The the 24f series. Why would you use that when this a bad boy here is is half the power consumption in run mode, in at one at a one megahertz reference value which might be a frequency that I'd be using. uh, my particular project at.

well, it's Horses for courses. Now there's a couple of things you need to consider when you're talking about and analyzing. uh, battery power consumption for your product. uh, design like this.

Why the first thing is is how often does it run in that, say, for example, that one megahertz mode? my particular product, it's probably going to be spend most of its time sleeping and just updating a clock display. Most of the time, you could say 99 of the time. For example, it's not actually operating at that one megahertz figure. The fact that the peak 24f was double the power consumption of the Um STM 32 one Yeah, really doesn't make a difference.

And the next thing to consider is what type of batteries are using my product might use uh, for example I haven't finalized it yet, but might use a couple of double A batteries for example. Let's run the numbers. Let's get the confuser out here. Even at 25 milliamp discharge, you would still get like three thousand.

You know, or approaching three thousand. You can see that on the graph. here. This is is actually a good graph to look.

It's got the milliamp hour capacity here versus the drain current. And of course, we're operating down because we're using a segmented LCD Um, and you know it's during that four right sniff of an oily rag stuff. We're almost approaching 3000. but let's call it 2500 milliamp hours.

So even if we were running this thing continuously 24, 7, 365 days a year at that full operating frequency. which we're not. But let's just take that as a worst case. Uh, well.

2.5 amp hours Okay, divided by 150 microamps. That gives us 16 666 hours divided by 24 hours in a day. Divided by 365 days in a year is 1.9 years. Okay, so we're approaching two-year battery life in our product from a couple of Double A's even when it's running at slam full one megahertz, right? Processing 24 7, 360 five, which is not.

But even if it was right, you're still talking two years battery life and that might be fine. So you can see why for a particular application. In this particular one, it doesn't matter that this thing is like half the power consumption. Okay, you might get four years at full.

Okay, and it might be better. All right, and that might make a difference. Okay, if you're powering it from like a CR2032 coin sell or something, you know. Yeah, Okay, you need to start.

you know, considering this sort of stuff. But really, it's you know. the fact that's half the power consumption sounds fantastic. But really, it's not a reason that you go.

Oh, I must use the Stm32 over the pick because it's half the power consumption It? it just doesn't really matter in this particular case. So here's where we can start going down the rabbit hole. And trust me, this video would be hours and hours long if I actually went down the full rabbit hole on uh, you know power consumption comparison between these two two devices and you can spend a week Actually, you know, really comparing uh devices like this. If you're you know, doing like a watch, uh, product or something like that really ultra low, uh Power type stuff, you know you can.

You really have to go down the rabbit hole here and look, uh, let's go into the uh pick uh 24 FJ Let's actually have a look at uh, the LCD for example because not only are there many different modes, both microcontrollers have many different modes. They have sleep and deep sleep and you know, wake up and then full run mode and partial run modes and all sorts of interrupt modes. All sorts, of, uh, different modes, right? ones where you're running your ADC one when you're not and you know all sorts of things. Uh, like that.

But in this particular case, really, all our micro is doing is, uh, when it's running okay, it's running at, say, the one megahertz, uh, frequency range. so we've got the known current for that. and then you've got the current for the LCD So incremental current driving a segmented LCD And you've got different modes here. You've got low power resistor ladder mode, your medium power resistor ladder power high power resistor ladder mode.

you know, depending on contrast and stuff that you actually want. So this is the incremental current above. so the Delta current. That's why it's got Delta LCD So the additional current required above the operating current of the microcontroller.

And then we've got typical and maximum figures here. and this is from 2 volts to 3.6 volts operation for example. And yes, the Uh STM arm one I think goes down to 1.65 volts. So if you really needed like you know to go down to a lower voltage, that might be a better part.

That might be a deal breaker for you in this particular case. uh, two volts minimum operation is just fine. Typical figure for an LCD here though it might be you know you might have the high power resistor ladder mode. you know that could be 64 microamps or it could be worst case if you're doing worst case uh design uh figures which you know you probably should be right, it's a 140 micrograms once again that'll be at like the higher temperatures.

And they've got different figures for the 2 volts and 3.3 volts here so it's going to dynamically, uh, change as your battery uh, drops in voltage. For example, there's an extra 100 microamps right there. Anyway, if you're talking specific devices here, we really have to use. you can see down here we have to use the Stm32l0x3 because the X3 is the only one that has a segmented LCD driver down here.

We don't need much memory, for example, right? and they're even 32k And AK of SRAM is overkill, right? So we're only talking about because you want the cheapest device you can. We're only talking about these two parts down here: the L05 3c6 or the L053 R6 So here's our data sheet. Okay, yeah, as I said, 1.65 to 3.6 volt operation. Uh, point, you know, 270 Nano Amper standby mode 400 Nano amp stop mode and then you've got the RTC as well.

Well, we haven't even gone into the RTC which I'm going to have and that 88 microamps per megahertz in run mode? half the value of the nominal Peak. But let's see if we can find uh, some LCD consumption data, shall we? optional LCD power supply scheme So yeah, it's got internal thing for a step up our converter and stuff like that. that's all. cool and groovy.

Supply Current characteristics We're getting there. These things are often buried in the parameters of the data sheet. You've really got to have a look. Well, hello, right off the bat.

What's going on here Is that Banner spec Some BS in the band? The banner spec. Maybe check this out right under these conditions range: 3 V core is 1.2 volts. so that's the core voltage that it's running at. As I said, it's got an internal um low Dropout regulator in there at one megahertz 165 Typical microamps.

What happened to the 88 or whatever we were running at Supply current in run mode code executed from flash. Once again, you've got to go into all the different clocking modes and the clocking sources. All sorts of jazz like that. But I've got a figure here of 165 microamps that's basically the same as the as the Pick and this one up here was Flash code executed from Flash.

Well, wouldn't you execute your code from flash? Isn't that a given? Okay, and it starts to lower with the flash. Switched Off 135 Microamps Okay, we're still not like at the 87.88 or whatever. it was okay. Supply Current in sleep mode.

Flash off All right. Hats off to the St data sheet here they're They're giving you like really comprehensive figures here. But but this is in sleep mode I Can go into the different modes and everything right? But but we're still talking 50 Seven microamps, right? 43 Microamps This is not nano amps, right? this is microamps in sleep mode. Supply Current in low power run mode.

This is where it's going to work. Oh no. Okay, so that's only low power clocks. Okay, because it has an internal RC 32 kilohertz uh clock.

Okay, so that's that's not our one megahertz mode anymore, which is great. If you're in clock mode or something, you know your products shut down and you just updating. You know, a clock display or something like that, then that's uh, that's that's fine and groovy. That's why you can get you know 4.7 microamps and stuff like that.

Yeah, no, where's this? 88 87 microamps? Where are we see in that in the data sheet, right? Maybe I can search for 88 and can I do a control F 88 Uh no no no no no no no no. So if we go back to the pick over here, uh and then if you want to learn once again, go down the rabbit hole. There's a little three next to that right if you go down here and have a look at three base idle current is measured with all of these different modes, right? So this is why I'm saying if you're really serious about this sort of stuff and you're really comparing devices like this: If I was designing a professional product at a you know, at a professional company and we haven't had all the money and all the time and all the resources to actually you know choose a proper micro controller uh, like this to choose like the optimum absolute Optimum part for the application, then I would be building this up. I'd be building up two different but you know I'd narrow it.

used data sheets to narrow it down to, you know, two, maybe three. uh, you know competing micros that I'm uh, trying to use and then I would build up. Uh, just demo boards for them and uh, then just use that and then design onto those boards like proper Little Current Consumption shunts and like you know, jumpers and all sorts of things that I can easily you know measure the in-circuit uh, current consumption I'd run code on both of them, right? But this takes time and effort, right? So you've got to have the time and the resources to actually do this. Doesn't cost much to spin up a cheap board and buy a couple of one-off parts, and you know, hand assemble them right.

But you know it's it's time and effort. but you would do that if you were genuinely trying to, uh, do a shootout between, you know, a couple of different competitors. And there's just so many variables in here. It's just.

it's just absolutely nuts, right? Especially when you're driving LCDs and you're using timers. You might be using Adcs, you might be using comparators and other peripherals coming out the Wazoo right? And and in different modes and things like that. And then, yeah, you'd have to write. You know, test, um, software, because you wouldn't write the full application.

Generally, just write some. You know, test operations that drives the LCD. You know it does your ADC thing. Whatever you want to do, does the time of things it does, You know you, comparative stuff and things like that, and the things that, um, you know that you're operationally draw in power with.

But yeah, anyway, but first thing on the table here for the pick: I'm seeing operating current Idd uh, right, we're talking like Max 350 typical 208 for example. So you know that's actually operating at one megahertz. Okay, so that's at half half a million mips as million instructions, uh, per second. And if you're really comparing, it's not just microamps per megahertz, it's you know how many instructions per uh, micro amp? It's great to run to one megahertz, but if you can only process half as much as compared to the other uh, device or whatever, then that could make a difference.

It's how many instructions? um, per micro amps really. But once again, that's only on the processor side of things. Right When you're talking about all your sleep modes and power down modes or low power modes and stuff like wow, like I honestly could not cover it, even for a relatively simple application like this, which is just driving an LCD Really? To do a shootout between two different devices is like hours. many hours worth of videos and then building up like comparison boards to actually seeing that they're You know that they're not BS in you in the specs and they do actually, uh, perform as claimed or you simply cannot get all the information in there.

But anyway, we've got like operating current okay, which is Idd. Then we've got idle currents so it's sitting there idle. It's not in some power down mode, but it's it. But no, it says it's doing one mips.

So once again, you'd have to go into the differences between idle mode and one mips at two megahertz because this is lower consumption. Okay, you'd have to go down here and check what things are all turned on and stuff like that. This one's actually lower power consumption at twice the mips as this one here. Like what the? and this this is half the Mips, half the frequency and it's more because it's probably got more stuff on.

Anyway, back to the SST I'm still trying to find some um LCD power consumption I Haven't found yet where it's getting this 88 microamp per megahertz figure from. Okay, now they do have some figures for the LCD controller. They're saying the Ilcd so is there a voltage? Uh, for the LCD pin and that's only three typically three microamps low. Drive resistor Network Right, there's a high drive resistor Network That's what microchip we're specifying.

but that's it, right? they're not. They're not telling you. So going back to the microchip over here. Actually, the microchip is lower.

typical, right? One microamp for the low power resistor ladder 1.3 microamps typical. It could be a maximum uh, as it could be as high as 12. Over here we're talking about three microns, but you know one again. this is the Delta current on top of the operational.

so we we're still not sure what the operational current of the STM is over here. look. I'm sure a whole bunch of people oh yeah, I've measured that. No, it's in here this page of the data sheet, but I'm like not readily finding stuff and it's just Ilcd.

But anyway, the resistor divider is like across the rail. So let's say you're operating at uh, even even the two volts, right? You know your batteries right down. you're operating at, you know, the minimum. uh, battery, right? So two volts divided by uh, 240k? We're talking like eight.

uh, an additional eight micro amps there. But yeah, I don't know. The microchip over here is LCD Plus you know. charge pump a low power resistor ladder.

Okay, 10 microamps over here and then LCD high power resistor level. It's like we have to go down the whole Rabbit Hole of what resistive ladder we're using. What? uh, chart? you know, charge pumps we're using for the LCD what type of LCD we've got and you know the contrast that we want and all sorts of things. This is why from the data sheets you would if you wanted to do a real shootout, you'd as I said, you've probably got to build these things up and measure it.

So yeah, I'm not sold on this. 88 microamps per megahertz in run mode. It might be taking that from a a like a higher frequency figure and divide in it or or something like that because let you go in here: Supply Current in row mode. even when you're executing the code from Ram I assume how easy is it to copy the road from flash to Ram I Don't know, it might be trivial, but you know, um, if you actually even when you're executing it from Ram I'm getting like 135 microamps at one megahertz.

wouldn't be the first time that a manufacturer is, you know, rub some snake oil on the top level Banner spec. Leave it in the comments down below. Am I missing it? Is it somewhere else in here in run mode code with data processing running from RAM and if it's running from flash, it's like it's on par with the with the pick or even potentially worse. So anyway, as I said, huge, Absolute Huge.

Rabbit Hole many many hours of videos just to do that. um, and many, many weeks of work if you wanted to actually build up to and compare it. But of course, one of the things with the STM that they're Infamous for the chip again the component Supply crisis. Everyone's project stopped because they couldn't get STM arm micros right.

They built it. It was the flavor of the month. Everyone was going giddy over the STM micros. And then the component Supply crisis happened and you couldn't get them right.

So then, whereas you could still get a lot of the microchip Parts they weren't hit nearly as hard as the STM ones I were. I've heard of people who simply folded their business because they built their little hobby business around the product. They were using STM micro in it and they couldn't get it. and they went.

This is ridiculous. It's Year's lead time right? I did just they. They just shut up shop right? because they couldn't get these parts. So let's get these two different parts.

the C6 R6 here. Okay. quantity. Uh, their prices are pretty much on par between uh, the pick and the STM here.

So really, yeah, it's not. There's real, no, no difference in the price. They're two. They're two bucks something.

Um, in uh, quantity? volume? But look. Stock: Zero? Zero Zero. Oh 180. they've Then they've got 180 in the tray.

Twenty Six thousand, Four hundred. There you go. So future Electronics have 26 000 but in trays once again, like you might prefer them in reels. um, right and 2.96 So you're actually paying a premium for that, but it looks at that's 160.

Quantity: Looks like you have to pay that minimum. Looks like you made it I Don't know. You might get a little bit of a discount for higher quantity on that. Zero Zero Zero 1900, right? It's not looking good here.

A Ming's holding so that's some weirdo distributor com Sit. Never heard of them right? So there's 29 000 in some where in? Europe maybe. But yeah, nah, right, it isn't a good vibe. Let's let's try this other part.

Stock Zero, right? So we're talking Arrow there one practically one of the biggest uh providers. Digikey. Okay, they have 2600 in stock, but you might get a oh there we go. if you've got to buy 2500.

So if you bought all of them, you know you're paying almost three bucks a a part there. Um, but zero. Zero Stock? Zero? Okay, 1400, right? So okay. you might be able to get your first run: buy the parts before you design the PCB.

Just buy the parts, Have them in stock. you know? Take that Financial Risk Buy them now so that you're guaranteed to have them. But yeah, not that terrific. Is it? So we can order direct here? Okay, and we can go over to here.

Here's the part. All they tell you is that it's in stock. There's no quantity, right? Yeah, the goddamn stock. Trust us trying to order ten thousand of them.

But if you compare that to the microchip part, digikey have 11 000 in stock over here. Okay, so that's that's pretty. Schmick Yeah, 12 000 is a better Vibe And just based on the history, I would trust being able to get microchip Parts better than I Trust trying to get uh, the St parts. Now, as for the other major requirement in my product was the 32-bit uh timer here and this one it does actually have the St has actually nine times.

one 16 bit with up to 4chan channels to 16 bit with up to two channels. So somebody said you can actually uh Cascade them internally and I will believe that. but once again, you'd have to check specifics. But of course, one of the cool things about the microchip is it had, um, a core independent animation of the LCD So you could do autonomous blinking of displays like this.

you could do alternating, complete, alternating displays. Animation happens in low power mode, for example. Can you get the same sort of thing? Once again, you have to build these up and you'd have to actually measure them. Spend a month of design effort just really comparing.

You know, even just two micros, let alone more. So does the STM have that? Um, I Do believe it has like blinking. Yeah, here it is. Here it says it supports blinking mode whatever that means.

But that is that. Core independent. Does it operate in low power modes? Can you? Is it? Is it just a flash a thing off and on? Or can you do like complete LCD swaps like you can on the microchip? one. That could be your decision made right there because you wanted all that animation capability that the microchip offered, for example.

So there you go, up to eight pixels can be programmed to Blink That's it. So it can't do the full swap so it looks like that would be. yeah, blinking would be core independent. So I Guess you could set the blink rate maybe and um, and which and but you can only blink eight, uh, pixels.

eight segments. So yeah, um, the microchip wins there hands down. So anyway, this video has been long enough. That's a a brief comparison between the Stm32l.

This is just top top level. Me off the cuff, like I didn't you know plan? you know, just just press record. Once again, if I wanted to do a huge detail comparison, it would take me days and days of work to actually uh, you know, truly go in there. That's just the data sheets, let alone you know, building things up.

So yeah, even that like Banner spec down here. 88 microamps per megahertz. Okay, Stm's half the power consumption of the if the pick. well is.

it doesn't seem to be that clear-cut so you can't Clearly say you can't confidently say that. Oh yeah, yeah, and this. this part is absolutely the choice over the pick. here.

you just, you can't say it. I would just run with the pick because it's not that hugely critical as you saw with you know, the double. A If you're pounding from double A batteries or something it it doesn't really matter a much of a rat's ass really. but you know you just want something moderately low power.

You want something that handles the LCDs that animation thing that was really cool right? that the Pit can do and it's good enough and there might be some other once again Lower power than the Stm1 again in certain modes and well, you'd have to go through the whole process for different manufacturers then. uh, some people talked about the psoc uh parts for example um which have the Uh which have some like Hardware routing you know, digital logic blocks are built in and stuff like that. that's that's kind of cool, but there's another whole comparison uh video and and you go down the rabbit hole. Hey oh God if you're not used to going down the design rabbit hole on stuff like this in that I cover in like a 20 30 minute video.

Here it can takes days, weeks, even months to do a real detailed comparison of Uh parts for depends on your application. Some applications doesn't matter, so you can just use the same micro again and again and again. and there's a lot to be said for that, right? I'm familiar with the Pigs I like the picks they've been available for me, they've done me well over decades right? So you know I'm naturally um, you know, when for microchip but then the 121 GW actually I'll link in my uh design uh video of well no no, it's like a historical video of the of the development of this thing this actually did use originally used a pick um a micro in it and I specified a pic micro. But then we decided a decided data to actually switch to an Arm Micro because the company who designed it they were you know wanted to sort of like switch to Arm for other you know operating company reasons and stuff like that.

So I specified in that Stml one series into there. they were going to use another part that didn't have building LCD but I I specifically found an and a suitable L1 uh part and that's been really good. 32 and I didn't actually program that low level uh myself so you know. whereas the microchips I have so I'm familiar with them I'm confident with them.

So yeah, so sorry are you? STM um Fanboys um I have not uh, changed my mind I think I'll still run with the pick 24? Um, oh well. a variant I think this is the variant I'll probably run with though. but I'll do a double check. but yeah, nah I it has not changed my mind.

The fact that it might be lower power. Whoop-dee-doo Um, it's the proof's in the pudding when you actually build it up. So yeah, and yeah, a lot of people have been bitten and refused to use St arm again because they've been bitten by the supply. uh crisis.

Anyway, that's enough. Waffle Found the video interesting, give it a big a thumbs up. And if you like this uh Design series even though you don't really know what I'm designing, just I just randomly press let me know. leave it in the comments please because I know these videos don't get a lot of views.

Um, so you could argue that you know they're not good for the channel. It drags the channel down a bit when I do videos like this. But if you do like it, please engage thumbs up all that sort of stuff and leave it in the comments to encourage me to do future design videos like this. Catch you next time.

Foreign.