Fantastic video Glenn. I fear a reality where I "inspire" anyone to do anything, never mind a PCB and related video lol. One small correction. I wasn't using the yoke for Flight sims, although back in the day, I did one session with it in F18 before realizing that you don't fly a jet fighter with a yoke lol. Love your work mate. Awesome.
Ahh yeah that makes sense, jet fighters don't have yokes. Regardless though it was the push I needed to start some pcb work and I'm so glad I did because it's addictive! Already got my next one designed. Hope the adapter works for you.
Congratulations Glen, nicely done. You've opened the hardware development floodgates now :) I loved (and still love) flight sims, and played a lot of them and F1GP back in the day with an analogue stick, probably using the same schematic. I think I can offer some further insight into how these games read the controller and so on. Many games back in the day (including the flight sims you tried) didn't offer any sort of calibration options, and essentially made assumptions on the centre and the limit values of the stick. It meant that you needed to use the trim controls (those small screw adjustments) to adjust the stick to suit a game instead, and if your stick gave significantly wider ranges of values, it would be basically impossible to control. They assumed the values would be quite low because PC-type sticks were far more common than Amiga-type. This means that although they are analogue readings, the resolution is still quite low considering the maximum reading possible was 255, with only maybe 20 steps in each direction available. Sticks made for the PC typically used 100K pots, which gave quite low values compared to sticks designed specifically for the Amiga (which were based on Atari paddle controllers and thus had higher resistance pots, typically 470K), meaning values were 4-5 times lower with a PC stick than an Amiga one. The optional capacitors in that circuit are intended to compensate for that, however, because games made the assumption for PC sticks, they were redundant in many cases as you saw. A few games do allow calibration, like Tornado and F1GP, and these games should make full use of the smoother input using the capacitors or larger value pots, and will still work with lower value setups. It's just a shame that so many games didn't include a calibration routine.
Congratulations on your first PCB! Even though there were some issues and you needed another revision, the final result was a success! Keep the PCBs of the non-working version - they are a prefect to use as a worktop for tinning copper wires or burning away enamel coating :)
Great work Glen. And i learned a few things watching this. Nice editing at 23:17 btw, slick👍. Just wanted to mention, I bought f15 strike eagle 2 for my Amiga back in the day. It's a cracking game and a natural successor to fa18 interceptor in my opinion. Looking forward to your next vid.
Thanks, F15 does seem like a really good game. I need to spend some more time with it now that I've got the adapter. As for the editing, I like to try a bit of flare and especially when I'd already covered the build before.
Good job with your first foray into PCB design! The first project is always a challenge but super exciting when it actually works. My first ever was designing a memory module for a Compaq SLT/286. Definitely bit off more than I could chew but got it to work in the end. Starting smaller and working your way up is definitely the (less frustrating) way to go :)
Haven't played F1GP since the early nineties but immediately after you started that lap, I remembered every corner like some kind of sleeper agent, activated from deep slumber.
I've never dabbled in PCB creation/design, but it does seem prudent to start out with a breadboard for testing the schematic first. That way you're not waiting 3 weeks only to find out the capacitors were unnecessary.
@@CRG If you do buy a breadboard, I recommend not cheaping out as there's a distinct difference between the cheap ones and expensive ones with how well they work.
First batch of PCBs is perfectly fine, just not populate caps and they are good to go. But I understand the urge for perfection for only $5+shipping :)
KiCad is a big win! I learned a bit on PCB design with Eagle and hadn't touched in years. Then I found KiCad and ran into a project that almost mandated a custom board (really weird shape, nasty to hand wire, etc). Other than a bit of fighting to get foot prints correct for some of the more esoteric components I've needed, it's been a very capable tool for me. I highly recommend it if you need a board design tool for anything DIY.
Hey Glenn, Is that a Commodore PC motherboard infrastructure of your Amiga 2000? I would too like to master Kicad. Do you know of any good training resources for it? Regarding four fire buttons; the CD32 controller has these natively via the 9 pin Atari controller ports. Competition Pro pads are what you want. Cannot think of any good beat 'em ups on Amiga; I think Elfmania had best visual effects. Street Fighter 2 really was a poor port (very slow when compared to a Megadrive).
It is indeed a Commodore PC. Video for that is coming soon. For KiCAD I followed a few starter tutorials here on UA-cam but I also work with AutoCAD in my day job so getting to grips with the interface wasn't that much of a jump for me. There are a few decent beat em ups on the Amiga, Fightin Spirit is probably my favourite although thats a late game, didn't arrive until 1996 I think. I was just rambling at that point and yes I know the CD32 controller has 7 buttons (4 face, 2 shoulder and start). I'm actually toying with a follow up PCB idea relating to that.
"It maybe is a shame that not more games on the Amiga support this sort of stick" Support for just a two button stick (let alone three) is up there with Unicorns being ridden by Bigfoot over the rainbow, even though the Amigas port was ready to go. If you read posts about some Amiga games on YT there are still people that believe the Amiga was a one button joystick computer.
I wonder if you were getting such different values because it looked like the calibration program was reading a byte (0-255) for each axis, whereas F1 appeared to be displaying a short (0-65535) based on the number of 0's
Once I got used to the interface it's easy enough. I work with AutoCAD in my day job though so I think that helps a little as they do have similarities.
@@CRG ahh, I learned a little AutoCAD some 25 years ago.... I have a feeling KiCad (and others) might have evolved a wee bit since those Pentium 166 machines with NT4 on..... Last electronic thing I used was called Electronic Workbench and ran under Win 3.1.....
Glen, you've done 2 things in this video that bug me You apologized for your circuit diagram, yeah it could be neatened up but it was functional, and you apologized for working with Kicad as a beginner, we all start somewhere and you made an amazing product A good bit of advice for Kicad is to download the digikey libraries, lots of useful components with premade footprints that are pretty good Practice altering footprints, I use Kicad for work for making student board products, but some of the premade footprints are a bit tight, for beginners especially, but sometimes solves issues just moving the centre leg of a transistor forward to avoid bridging Great work again mate
Yeah making PCBs is pretty amazing... something that was impossible 30 years ago now costs just a few dollars. I'm at the other end of the scale... After my Commodore 128D 1571 disk controller repro PCB I decapped the read/write hybrids in all Commodore floppy drives and fully reversed them all (there are 3 different versions with minor changes) using Eagle PCB. After a few weeks of work they were done. It's basically all common smd parts (caps/resistors/transistors) so should only cost a few dollars to build.
It was very much possible 30 years ago, just slow and annoying :) DOS Protel design, printing on transparency, cutting pcb to size, spraying photo resist on pcb, exposing with UV lamp, developing photo resist, fixing errors with special pen, etching with ferric chloride, drilling. In late nineties you could buy PCBs with photo resist pre applied, pull off protective foil and its ready for exposing. Later in 2000s someone figured out Laser print can be thermally transferred directly to PCB with clothes iron :o Steps went down to print on normal paper->clothes iron->etch, made turnaround much faster. Finally in 2010 OSH Park brought cheap fast PCBs to order a reality.
@@rasz Heh! Years ago I did one photo etch PCB and it was a bit rough but worked. I still have one photo-etch pcb in a new packet unopened but it will sit there for some years then eventually get thrown in the bin hehe! Nowadays JLC is the cheapest pcb company and pcbway is 2nd. Osh is just way too expensive but people in US like to get their boards from there to keep everything in the good old U.S. of A hehe!
The caps should be there but do not connect them together at the other end, just use a double-pole switch to disconnect them or 2 separate jumpers. With the caps not connected to gnd (with switch/jumper open) the caps should be completely open at the end. You could try it on your older design, add the parts and cut the trace on the caps so they are not joined. As the readme says, some games require the caps and some don't.... that's why in your testing some games worked and other didn't. Back to the old drawing board for version 3 ;-)
That would require actual knowledge of how USB works and probably a PIC microcontroller and some code. This is not a thing you can just knock up with a few loose parts in 5 minutes. Anyway, there is already a 9-pin to USB joystick adapter out there which should work fine on the Amiga.
@@g4z-kb7ct Sadly, I don't know of any USB adaptors that present analogue control to the Amiga in this way. Any I've seen that read analogue sticks convert them to digital control signals. But yes, that's a far more involved project, though with plenty of open-source, Arduino-based projects out there that do things along these lines, it's not as "black magic" as it used to be.
Haha, this takes me back to learning AutoCAD in school :) I only made one single-sided PCB though, and we etched it on site! (Well... gave to the teacher and watched him dunk 'em. The school's insurance didnae cover us working with the acid oursels!) I absolutely second the suggestion to get a breadboard - and make sure it's a good one! Cheap ones will do you no favours, with extra capacitance and dodgy connections. In fact, given the issues the capacitors gave you here, this circuit might never have worked quite right on a cheap'yun. But... aye. In the aforementioned AutoCAD adventure, the teacher made us all test our circuit on a breadboard first. Before we wasted any precious, UV-activated-whatever, blank PCB material on a useless circuit! Those transparencies for the printer didnae come cheap either ;) I suppose with how cheap professional PCBs are nowadays, that's all a wee bit less of a concern.
Fantastic video Glenn. I fear a reality where I "inspire" anyone to do anything, never mind a PCB and related video lol. One small correction. I wasn't using the yoke for Flight sims, although back in the day, I did one session with it in F18 before realizing that you don't fly a jet fighter with a yoke lol. Love your work mate. Awesome.
Ahh yeah that makes sense, jet fighters don't have yokes. Regardless though it was the push I needed to start some pcb work and I'm so glad I did because it's addictive! Already got my next one designed.
Hope the adapter works for you.
Congratulations Glen, nicely done. You've opened the hardware development floodgates now :)
I loved (and still love) flight sims, and played a lot of them and F1GP back in the day with an analogue stick, probably using the same schematic. I think I can offer some further insight into how these games read the controller and so on. Many games back in the day (including the flight sims you tried) didn't offer any sort of calibration options, and essentially made assumptions on the centre and the limit values of the stick. It meant that you needed to use the trim controls (those small screw adjustments) to adjust the stick to suit a game instead, and if your stick gave significantly wider ranges of values, it would be basically impossible to control. They assumed the values would be quite low because PC-type sticks were far more common than Amiga-type. This means that although they are analogue readings, the resolution is still quite low considering the maximum reading possible was 255, with only maybe 20 steps in each direction available.
Sticks made for the PC typically used 100K pots, which gave quite low values compared to sticks designed specifically for the Amiga (which were based on Atari paddle controllers and thus had higher resistance pots, typically 470K), meaning values were 4-5 times lower with a PC stick than an Amiga one. The optional capacitors in that circuit are intended to compensate for that, however, because games made the assumption for PC sticks, they were redundant in many cases as you saw.
A few games do allow calibration, like Tornado and F1GP, and these games should make full use of the smoother input using the capacitors or larger value pots, and will still work with lower value setups. It's just a shame that so many games didn't include a calibration routine.
I'll have to do a 3rd board to add the caps back in and another DPDT switch to enable/disable them properly.
Congratulations on your first PCB! Even though there were some issues and you needed another revision, the final result was a success! Keep the PCBs of the non-working version - they are a prefect to use as a worktop for tinning copper wires or burning away enamel coating :)
great work mate! the magic of having your own PCBs made never gets old :)
Thanks. It is very satisfying even for something as simple as this. Addictive too as I'm already working on the next one.
Great work Glen. And i learned a few things watching this. Nice editing at 23:17 btw, slick👍. Just wanted to mention, I bought f15 strike eagle 2 for my Amiga back in the day. It's a cracking game and a natural successor to fa18 interceptor in my opinion. Looking forward to your next vid.
Thanks, F15 does seem like a really good game. I need to spend some more time with it now that I've got the adapter.
As for the editing, I like to try a bit of flare and especially when I'd already covered the build before.
Congrats Glen. The first of many to come !!
Nice work Glen. F1 I used to play full races back in the 90s. Happy days man.
Good job with your first foray into PCB design! The first project is always a challenge but super exciting when it actually works. My first ever was designing a memory module for a Compaq SLT/286. Definitely bit off more than I could chew but got it to work in the end. Starting smaller and working your way up is definitely the (less frustrating) way to go :)
Excellent little project there Glenn :)
Thanks, was fun to pull together.
Great job Glenn =D
Haven't played F1GP since the early nineties but immediately after you started that lap, I remembered every corner like some kind of sleeper agent, activated from deep slumber.
Great Project! Your Videos are Awesome👍🏻
I've never dabbled in PCB creation/design, but it does seem prudent to start out with a breadboard for testing the schematic first. That way you're not waiting 3 weeks only to find out the capacitors were unnecessary.
Ahh but that would be the sensible approach 😅
Will have to get a breadboard for the next project!
@@CRG If you do buy a breadboard, I recommend not cheaping out as there's a distinct difference between the cheap ones and expensive ones with how well they work.
First batch of PCBs is perfectly fine, just not populate caps and they are good to go. But I understand the urge for perfection for only $5+shipping :)
KiCad is a big win! I learned a bit on PCB design with Eagle and hadn't touched in years. Then I found KiCad and ran into a project that almost mandated a custom board (really weird shape, nasty to hand wire, etc). Other than a bit of fighting to get foot prints correct for some of the more esoteric components I've needed, it's been a very capable tool for me. I highly recommend it if you need a board design tool for anything DIY.
Nicely done!
Well done learning a new skill.
Good job Sir .
Hey Glenn,
Is that a Commodore PC motherboard infrastructure of your Amiga 2000?
I would too like to master Kicad. Do you know of any good training resources for it?
Regarding four fire buttons; the CD32 controller has these natively via the 9 pin Atari controller ports. Competition Pro pads are what you want. Cannot think of any good beat 'em ups on Amiga; I think Elfmania had best visual effects. Street Fighter 2 really was a poor port (very slow when compared to a Megadrive).
It is indeed a Commodore PC. Video for that is coming soon.
For KiCAD I followed a few starter tutorials here on UA-cam but I also work with AutoCAD in my day job so getting to grips with the interface wasn't that much of a jump for me.
There are a few decent beat em ups on the Amiga, Fightin Spirit is probably my favourite although thats a late game, didn't arrive until 1996 I think. I was just rambling at that point and yes I know the CD32 controller has 7 buttons (4 face, 2 shoulder and start). I'm actually toying with a follow up PCB idea relating to that.
"It maybe is a shame that not more games on the Amiga support this sort of stick"
Support for just a two button stick (let alone three) is up there with Unicorns being ridden by Bigfoot over the rainbow, even though the Amigas port was ready to go. If you read posts about some Amiga games on YT there are still people that believe the Amiga was a one button joystick computer.
Good work. Exercise those geek muscles!
I wonder if you were getting such different values because it looked like the calibration program was reading a byte (0-255) for each axis, whereas F1 appeared to be displaying a short (0-65535) based on the number of 0's
Very cool. I have a few ideas myself for different projects. How did you find KiCAD to use?
Once I got used to the interface it's easy enough. I work with AutoCAD in my day job though so I think that helps a little as they do have similarities.
@CRG kool will have a play at some point
@@CRG ahh, I learned a little AutoCAD some 25 years ago.... I have a feeling KiCad (and others) might have evolved a wee bit since those Pentium 166 machines with NT4 on.....
Last electronic thing I used was called Electronic Workbench and ran under Win 3.1.....
Glen, you've done 2 things in this video that bug me
You apologized for your circuit diagram, yeah it could be neatened up but it was functional, and you apologized for working with Kicad as a beginner, we all start somewhere and you made an amazing product
A good bit of advice for Kicad is to download the digikey libraries, lots of useful components with premade footprints that are pretty good
Practice altering footprints, I use Kicad for work for making student board products, but some of the premade footprints are a bit tight, for beginners especially, but sometimes solves issues just moving the centre leg of a transistor forward to avoid bridging
Great work again mate
Yeah making PCBs is pretty amazing... something that was impossible 30 years ago now costs just a few dollars. I'm at the other end of the scale... After my Commodore 128D 1571 disk controller repro PCB I decapped the read/write hybrids in all Commodore floppy drives and fully reversed them all (there are 3 different versions with minor changes) using Eagle PCB. After a few weeks of work they were done. It's basically all common smd parts (caps/resistors/transistors) so should only cost a few dollars to build.
Sounds like a cool project but it is great just how accessible PCB manufacturing is these days.
It was very much possible 30 years ago, just slow and annoying :) DOS Protel design, printing on transparency, cutting pcb to size, spraying photo resist on pcb, exposing with UV lamp, developing photo resist, fixing errors with special pen, etching with ferric chloride, drilling. In late nineties you could buy PCBs with photo resist pre applied, pull off protective foil and its ready for exposing. Later in 2000s someone figured out Laser print can be thermally transferred directly to PCB with clothes iron :o Steps went down to print on normal paper->clothes iron->etch, made turnaround much faster. Finally in 2010 OSH Park brought cheap fast PCBs to order a reality.
@@rasz Heh! Years ago I did one photo etch PCB and it was a bit rough but worked. I still have one photo-etch pcb in a new packet unopened but it will sit there for some years then eventually get thrown in the bin hehe! Nowadays JLC is the cheapest pcb company and pcbway is 2nd. Osh is just way too expensive but people in US like to get their boards from there to keep everything in the good old U.S. of A hehe!
I've had a 'PcJoy2AMIGA V1.2' (+ a Tornado Logic 3 stick) for a few years --- but never had much luck 😥
The caps should be there but do not connect them together at the other end, just use a double-pole switch to disconnect them or 2 separate jumpers. With the caps not connected to gnd (with switch/jumper open) the caps should be completely open at the end. You could try it on your older design, add the parts and cut the trace on the caps so they are not joined. As the readme says, some games require the caps and some don't.... that's why in your testing some games worked and other didn't. Back to the old drawing board for version 3 ;-)
Nice work! How about for v3 you consider a USB version for more modern devices, like XBox/Playstation controllers? Perhaps even a wheel? :p
That would require actual knowledge of how USB works and probably a PIC microcontroller and some code. This is not a thing you can just knock up with a few loose parts in 5 minutes. Anyway, there is already a 9-pin to USB joystick adapter out there which should work fine on the Amiga.
@@g4z-kb7ct Sadly, I don't know of any USB adaptors that present analogue control to the Amiga in this way. Any I've seen that read analogue sticks convert them to digital control signals. But yes, that's a far more involved project, though with plenty of open-source, Arduino-based projects out there that do things along these lines, it's not as "black magic" as it used to be.
👍
diy stuff is super
Haha, this takes me back to learning AutoCAD in school :) I only made one single-sided PCB though, and we etched it on site! (Well... gave to the teacher and watched him dunk 'em. The school's insurance didnae cover us working with the acid oursels!)
I absolutely second the suggestion to get a breadboard - and make sure it's a good one! Cheap ones will do you no favours, with extra capacitance and dodgy connections. In fact, given the issues the capacitors gave you here, this circuit might never have worked quite right on a cheap'yun.
But... aye. In the aforementioned AutoCAD adventure, the teacher made us all test our circuit on a breadboard first. Before we wasted any precious, UV-activated-whatever, blank PCB material on a useless circuit! Those transparencies for the printer didnae come cheap either ;) I suppose with how cheap professional PCBs are nowadays, that's all a wee bit less of a concern.