@@Fazzwrld That's great to hear: I know the topic is interesting (at least to us, hah) but am still working on how to best share it, and a comment like this helps a lot (both in morale and direction)
@@PsychogenicTechnologies I totally agree with the other person. I’m an electrical engineer (in the making) for microwave technology, but I have never dabbled into the world of ASIC design. Very interesting and good presentation halfway into the video.
@@CSTEnjoyer Hello, and welcome both to the channel and to the ee world! Thanks a lot. Microwave--nice. I've always had a pull towards RF, and have worked on some things but never quite enough. The great news for both of us is that even if the GPIO on the TT chips is relatively slow, you can get really fast on the inside of these things--even with the older sky130 tech (low) GHz is supposed to be doable. I'm currently playing with my ADC and working on some power stuff, but with that under my belt I'm going to start playing with the high freqs. Thanks again, cheers!
Gosh, there's so much knowledge in your videos. It clarified so many things I never really understood of ASIC design. Seriously, thank you for doing this.
That's great to hear, many thanks Bob 😀I am, at this very minute, playing with stuff I'll be doing demos of in the next video (I'm so lucky to find everything I work on so much fun I want to keep doing and sharing it all the time). A bientot!
@@PsychogenicTechnologies Right on! I learned a TON from your video by the way, I don't know how the algo. works but it picked me up from a long way away, I certainly wasn't looking to build an ASIC haha, never used KiCAD or SPICE etc, but just watching you use the software is very reassuring somehow, it's normally so intimidating!
Wow, thanks! Seems we share an appreciation for language or effective transmission of material, anyway. I'm often trying to get rather involved concepts across and, when I'm consuming them, I value precision first and information density a close second. I try and put some thought in how I'll present what I'm sharing, so your comment means a lot to me. Thanks again, cheers.
Wow this is fantastic! I really appreciate how clearly you explained what the process is like, and where to expect frustrations. Now I have to give this a shot!
Awesome! That is exactly what I hope to hear when I started making this video 😀 Thanks a lot for your comment and do give it a go--it's just so easy to get started now. And then gets just as hard as you want it to, which I also find great. Have fun, cheers!
Thanks for demystifying ASIC design! This truly is an amazing time to be alive. I'm blown away that this stuff is now accessible to hobyists and enthusiasts. Thanks for you work here!
Blake: I totally agree. It is bonkers, what we can do. Won't be long until we really are in Stephenson's "diamond age" but this whole accelerating returns thing means that the possibilities just keep growing: here I am, on my computer, designing functional things down in the nanometer range and that wasn't a even a possibility for me like 5 years ago. Loving it, and loving the opportunity to get the word out and maybe help. Thanks for the feedback and encouragement! Cheers
Valeu! That's wonderful to hear, I do hope to make these things accessible or at least expose new possibilities. Thanks a lot for taking the time to comment, cheers.
What I liked best about about your video? Dunno - everything? Maybe the positive DIY vibe, the "hey, it's chip design but you can do it" spirit and the great overview over the design process. I've been working with chips for more than 30 years now, designing analog and mixed-signal ASICs and teaching stuff, and for me it was a joy to watch the video. I just started dealing with tinytapeout, can't wait for your Magic and xschem videos!
Many thanks, Christian! I'm always happy to hear someone found the video useful or fun, and from someone with a solid background it means a lot. I got sidetracked on an FPGA project I'm prepping to talk about next, but now that summer play is winding down, I want to get going on other ASIC vids and get them out there. Thanks again, cheers :)
Man, this is absolutely amazing. We could have more access to free tools like this and develop single ICs with lower prices using projects like this TinyTapout. Waiting for incoming videos. Thanks
Obrigado! TinyTapeout is a great place to try stuff out, and the tools are there and working on real projects big and small, but it does have some limitations on space and speed of I/O and such. However, some people in the TT discord are getting together to do a chipignite directly with efabless--costs a bigger chunk of money, of course, but there you really do control everything. Not saying 10k is cheap but it is imaginable--not some crazy quarter million dollars. And the more stuff we do, the more of us playing, the more we can start getting clout and reducing costs further. I'd say it's pretty good already and even better things are coming. cheers!
@@PsychogenicTechnologies as I’m from Brazil we face a lot of difficulties here in terms of import taxes and so on, but I see this as a great opportunity for a Master’s or doctors degree in an University for example. I had worked on front end design 15 years ago, but give up and now I’m working on others stuff. Are they using 130nm technology, aren’t they?Thanks
It would be a dream to have a community to develop chips here in Brazil. Unfortunately our closed border policies regarding electronics make things way more difficult.
Awesome video again. I've been following the open source ASIC/Tiny Tapeout space for a few months now and working on a digital design for TT07. But very curious about mixed signal design. This was a great overview for me a software developer turned circuit designer. Now I know what I need to learn more about to go deeper. Really cool stuff!!
Thanks Ben! It's super you'll be on the next run and if being set on the digital side means you'll only have one learning curve to confront for mixed signal next time 'round. We're on similar paths: professionally, I also went from software to hardware to wind up playing with chips. Means there's plenty of mystery left to explore in the whole analog design side of things, which is pretty exciting. I went rather blind, first time around, "thinking analog" but ignoring (and often being completely ignorant of) all the zillions of little effects that impact how real devices function down there. There are a lot, but if you're curious about how the universe works, it's fascinating to figure out. Good luck on TT07 (I still don't even know if I'll have time to submit my own, with all that's been going on!)
I just found your channel and I want to encourage you to continue making this type of content. Your presentation is great, and there is not much content I can find on such topics. I would love to see you do a few series say on KiCAD, PCB Design, High Speed Digital design, and low level analog design, and all the related tools. Keep up the GREAT work!
Hi! Welcome! This counts a lot, for me: it's great to hear and does encourage me to put in the time to make more. I'm already planning some more ASIC stuff, as well as KiCAD, that much is certain and the others are all actually topics I'm into, so I think it'd be fun to do and useful to share. Thanks again and keep an eye out, I hope to increase the pace of video production a little bit over the summer.
Jason, that's music to my ears! Thanks so much, really glad you're going to play... warning: can be addictive (it's 3am and I'm still fighting with a bug, but if it works gonna a be pretty coool! maybe some shuteye will help me solve it tho). Thanks again for the comment, have fun :)
This was a great intro to the analog work flow! I've found it really difficult to find good resources for learning this side of the open source tools so I'm looking forward to watching this video through again with the tools up and ready to follow along. Thanks for sharing this stuff! Looking forward to your future videos that get deeper into the weeds!
Thanks Calvin--I agree: what little I found was either too much of a basic intro and focus on installing stuff, or sometimes just too deep, and I was hoping to provide something that gives a overview while being useful. I live in the weeds and def want to share some more focused content, with a mix of the how and the why of bits I find particularly useful or interesting--a comment like this encourages me to actually get it done :)
I actually found the whole thing rather fascinating. I'm pretty good when it comes to discrete components or actual tubes/valves but this is a whole extra level and something I've always been curious about. I'll definitely be going through the rest of your videos in the near future.
Yes! The e-nerds are who I live for: it's my tribe, hah! Thanks a lot for your comment, I appreciate it, and am glad you liked the vid. I'll be putting out more on this front in the not-too-distant future, though I'm still not certain of the ordering (I want to keep the kicad crew happy, too!)
Thanks William! I'm working out exactly what I'll put in there, but I want to publish at least more detailed spice, xschem, and magic videos. Though magic will probably be the main one for talking about parasitics, all three feel like good spots to get into a mix of practical and more specific talk about the low-level components, how they're made, tradeoffs, matching all that good stuff. Personally, I'm better at doing anything when I actually understand *why* so I try to reflect that in the content and want to get more on all that in the next ASIC-related vids. Your comment encourages me to let myself loose in that department :-D
Thanks! A lot :) I'm planning to, have a few in the pipe. Going to do an FPGA thing next, and long-promised pick and place thing too, but they're coming.
That was great, both highly educational and very entertaining. There was obviously a lot of work behind this and a lot of work going into the video. I really appreciate the end-to-end approach as so much contents jumps right into the middle and spend two hours on one detail. I'll now go hunt for more of your videos. PS: Though I do not have any KiCad experience (and next to no EE background), I still think I could follow all of this. 👍
Hi Tommy! Thanks a lot: it is kindof a load of work to make these, and with the TT06 stuff and some summer play, I took a little break but am now working on the next ones--going to have some FPGA stuff, some KiCad and definitely some deeper dives on specific bits of ASIC land. Thanks for the great feedback, cheers!
Thanks a lot! The blessing and curse is that there is *so much* to learn, but there's never been a better time to get into it. Many thanks for the feedback, I hope to put out a lot more ASIC specifics in the near future.
Victor, I find that a fitting, funny and oddly poetic description: love it, I think I'll use it a my linkedin bio, hah! Thanks for the comment, welcome and I hope to keep putting out stuff that you appreciate. Cheers.
Love to hear it Daniel, thanks! If I can open the door a bit and get others joining in on the fun, my work is done--so thanks again for the encouraging comment, am working on more goodies on this front. Cheers!
It will be super educational to create a basic inverter , with the capability of the experimental characterization in mind from 0 to tapeout to actual measurements, everything recorded in videos. Also, one can add different inverters (same design) in different places in the wafer to showcase the process variability. Its a lot effort to do the whole thing, so , thank you for your efforts, much appreciated !
hm, that's a good idea. I did an inverter "live" with Robert Feranec (on his channel a few weeks back) but it was just the design parts--doing it in a way to allow some deep probing and comparisons would be really interesting... maybe I can reserve a section of my TT08 submission for this. Thanks!
@@PsychogenicTechnologies thanks for the quick reply! The thing that's missing from academia are all the practical considerations from design, cost of EDA tools, physical limitations, experiment methodologies, optimization flows...etc. these series of videos from you and the whole community will be a first pass for beginners. And the more we tap out the more confident we get to try even more complex projects. And who knows, maybe 10 years down the line we will get an open source Cerebras 🙄😁.
I really enjoyed this, and I made it all the way through to the end. To that end, here's *why* I enjoyed it. Back in my '86-87 ish 2nd - 3rd year university classes, we did some units on VLSI design, standard text being as you might expect. We learnt about design rules, components and had an assignment to put together a small hand-checked library, and be able to draw these ito a layout. It was cool, but seeing this actually become an accessible to Jo Everyhobbiest thing is incredible, thought yes, it's kinda of been 'accessible' at the university level for a while. Thanks for your videos. Any 'surprising' applications of these techniques? What applications of analog ASIC design are most 'happening' or most under the radar right-now?
Thanks! That's pretty awesome: I didn't get any opportunity like that at uni (wrong program, but not even sure they offered anyone) but the fact that it's now possible both let me play and is one of the reasons I make this sort of content, just so more people will know about it, see that it's doable and join the fun. With loads of people working in parallel and cross-pollinating crazy cool stuff can happen. As for what's going on, there's more than I can manage to keep track of. Of course AI related things but application specific compute is one way to squeeze out more juice as we hit the bottom of possible granularity and the end of Moore's law, so as it looks to be exploding everywhere. One thing that got that piqued my curiosity with all this was a contract I did with a company doing analog neural nets (in essence shoving charge into the gate insulator, to change the Vth, to basically weight the neuron): makes the computation extremely low power, so you wind up being able to stuff ML into the *components* on the edge. I thought that was pretty cool.
Super cool! I did my thesis designing a phase detector for a digital phase-locked loop in Cadence; however, it never moved beyond simulation. This video inspires me to try one day bringing it into reality 🤩.
Yeah! You get far with sim, but there's something I find magical about getting these things you've designed in your hands. About 2 months until the TT08 deadline--maybe a project for rainy summer days :) Thanks for your comment, Jeremiasz. Cheers!
Love this video! Looking to build ASIC skills and trying TinyTapeout projects soon, so your videos are a goldmine. Any advice for an undergrad trying to learn RF/Mixed Signal ASIC design?
Thanks Rohit! Basic advice is "play" -- get xschem, start doing experiments in simulation. But then, yeah, this goes so (so) deep, you could write a thesis on track width and tradeoffs, and when you get into the FETs well, feels like the level of interesting goes on forever. Also, some of these are beginner or intermediate projects, but TT has a number of RF things in the last two tapeouts--oscillators, a gilbert cell based double balanced mixer, lots of things. So, if you do want to take a look, you can just clone those repos and see how they're built, tweak things, see the impact, improve the projects themselves or use them as building block (say you take the mixer as a module and dev an env detector or fm demod, that should be fun).
this is awesome. going into grad school for mixed signal design so I appreciate that there are many more intelligent people making videos like this for me to consume lol
This reminds me of 6 years ago when i was trying to figure out how to even open magic, it was not that easy. After a while i figured it out. Cool stuff.
Ah, well so you've been through it as well: good job, oof :-D What I really needed was a) a little perseverance and b) to actually go through the tutorials... it was just a bit too alien to "pick up" like I usually do with most software (since, it seems, we've come to some sort of agreements in terms of UI that I'm leaning on, hah)
This is just great!!! You have explained the workflow way better than many others. Looking forward to see more content from you. Maybe a more intertwined AMS? 😅
@@PsychogenicTechnologies Sorry, my bad 😅 Actually, one of the most confusing parts of these workflows are the Three Letter Abbreviations 🤣 AMS stands for Analog and Mixed Signals, basically your whole project including the digital part of it 😉👍
@@gabrielbarrientos468 Hahah, yes, IDK, TBH IRL I use them a whole lot, LOL. Ah, yeah, I want to talk a good deal more about mixed signal--to me, that's where all the power of this stuff is, leveraging the goodness of both sides!
Thanks! There are a few ways to do this, but really virtualbox and the VM on github.com/TinyTapeout/analog-virtualbox-vm-sky130a are what I've used for this, so far--all the analog tools are in there. I'll talk about the tweaks I did in xschem and magic videos, but that's all optional.
Thanks Abhishek, I really appreciate the comment and it's awesome that getting to talk about this on Robert's channel has expanded the reach. More on the way, on many topics, but some cool ASIC stuff for sure. Cheers
Thanks a lot, Gary! I took some time off this summer, but am getting back into it and hope what I'm working on now is well received and reaches lots of interested/interesting people. Thanks for the feedback! Cheers
Thanks! That has been the plan all along, but I kept getting sidetracked. So, right now I have one in the works on FPGA, and two related to ASICs (though further down the line). And I'm releasing an unrelated quickie video this morning, just 'cause it wasn't too hard to get done while I was doing the boards related to that. So, I'll be on videos a lot more fulltime as soon as I get back from the next little trip. Thanks again, cheers.
@@PsychogenicTechnologies No hurry mate, I'll be looking forward to these projects, and yeah the new video is great. Being a beginner all these things are kinda overwhelming but I'm slowly understanding stuffs . The enthusiasm you bring in your videos is outstanding, wish I discovered your channel earlier.
Thanks! Yes, I put some time into getting it to work in a way I find usable both with key mappings (no, just no: ctrl-s really does need to be "save", lol) and scripts, and I just need to document and put online--will def be doing along with a magic-specific video
This is very interresting. Digital is very useful analog is beautiful. I guess I stick with tht though since analog doesn't require so many components most often and many things just need special components.
That's a nice way of putting it! I'm still working on getting my analog creations to be streamlined to the point of beautiful, but it sure has been lots of fun.
Thanks for the overview, this is very helpful. A bit annoying to get everything setup, either you run a virtual amchine or try to get those old tools to play nice with your distro, still very far away from the state of where kicad is. Hopefully more people join the space and we can get some standardization going on
Thanks! Yeah, I used to insist on getting things working on the bare metal of my machine, but with everything interacting with the PDK and between themselves, and everything in dev at the same time, it can get pretty awful. So the VM was the easiest way for me, and it worked well. Something I haven't tried yet but that looks pretty good is another packaged solution, but using docker (which might be lighter than running in a VM? I dunno). This is another Harald Pretl thing, and it's well maintained: it's the iic-osic-tools, hub.docker.com/r/hpretl/iic-osic-tools Might be worth checking out.
I know! I think, with this, we're finally getting pretty close to the bottom. The good news is: there's *tons* to explore and play with down there, so lots of fun to come :-D
Thanks! Took a quick look at your mag trap filtration system... that's pretty cool! Have always been curious about the wet sciences: a highlight experience at uni was a molecular bio class where I saw (stained) DNA with my own eyes... and the smaller things get, the less you can differentiate between wet and dry, anyway. Thanks again, cheers!
@@PsychogenicTechnologies Cool, thanks for checking it out. Wet stuff does have some interesting properties. I think of DNA, and its compiler the ribosome, as a programming language for atoms. Too bad its way too tricky and powerful. I also think the sciences are converging and its probably not so productive anymore to say where one begins and the other ends.
Super cool, I'm thinking about whether we could do something like that with our apprentices (electornics and application & coding). I think we could do some thing like a spi port expander. Is it possible to do bga packages?
Hi Lukas, I think the SPI port expander project is a really cool idea to get started with. For packaging: with tinytapeout the answer is no. At this time we're on QFNs and that's that. If you're going direct through efabless and doing a chipignite on your own, there are more options and you can even get bare dies. I'm curious why you'd want BGA: what's the advantage for you, density, low inductance pads, something else?
Thanks! Is it normal to view the output as 2 dimensional? I believe the incorrect bits you saw after a certain point are indicating your signals might be inverting. If you link them via imaginary axis, you might see when and where the signal flips and why. That stems from using sin and cos for wave generation. It embeds the pi geometry in the signal, and after a certain point, you're seeing the drift in the bits.
Hello! My guess is that you've started magic "bare"... like with just the "magic" command. I have a shortcut, "magicsky" and what it does is this: alias magicsky='magic -rcfile /home/ttuser/pdk/sky130A/libs.tech/magic/sky130A.magicrc It is loading the magic rc from the SKY130 PDK, and that is in turn importing a bunch of stuff (like sky130.tcl etc) that has the import spice stuff added to the menu.
hah, not exactly what I was expecting/hoping to inspire, but hey, I'll take it too :-D Welcome, enjoy the vid, and should have more related stuff on the way soon.
Hello Khang, maybe but not that I know of. The VM and the IIC-JKU docker are the two easy ways, and Matt's analog course does have a detailed section on doing it "by hand" on your local system (bit.ly/analog-waitlist ) but, because there are bundled versions, I just haven't focused on manual style installs.
@@PsychogenicTechnologies oh thanks for your reply! So, how could you set it up on your machine though? I wanna be able to get access to the tools (software) and do it on my laptop so that I can do these amazing things that you showed in the video!
In short, I've been getting one chip per tinytapeout run where I have a project. The cost depends on the size you want (a single "tile", enough for say 1k gates or so if doing digital) is $50, and the physical chip + the carrier board it's on and the demoboard is an extra 100... so if it's for a class, or a group, you could get many tiles and only a few chips--or do it the other way around if you wanted. See app.tinytapeout.com/ for current pricing.
@@nurahmedomar awesome. I juuuust missed the TT08 deadline (left things to last minute, crunched like an idiot, and then failed to get the project in before deadline), but am certain the next run will be announced soon and fully intend to be on that one. Hope to see you there! Cheers
Been looking for a decent software for chip designing, potentially with the ability to design transistor library for chips smaller than 65 nanometers (or even 2 nanometers node), as I figured why not put together a general purpose RISC-V VLIW processor. Hopefully this IC CAD app will do.
Thanks! I hope to put out some more, including some challenge-based learning experiences you may particularly enjoy ;-) I appreciate the comment, cheers
@@PsychogenicTechnologies Yes I spend too much time trying to push open source into our universities here in Denmark. From now I just need to point them to your videos.
No but fr, I just somehow feel like we got the same curios type of mentality about hardware. Is like , I be thinking if it could be possible to go further with HDL design but using only open source tools, and bam!! There you go. Please keep up the good work of spreading knowledge of free.
Awesome! Am glad to hear it answered some of your Qs and that you actually happened upon it :-D And yeah, I'm curious on a number of fronts and tend to hop around a lot... funny thing is that, by putting out these videos, I'm learning that there aren't tons of us but I'm certainly not alone in my eclectic hardware fascinations--so cool!
I have not. However, from what I gather, IHP's newly open process is BiCMOS--not sure that TT will support it, at least right away, but it's happening right now and pretty interesting. They have settled on klayout, I think, in terms of doing layout, rather than magic but in any case the PDK is open and they seem very motivated to get us involved. I was at latch-up a month or two ago, and Frank Vater gave a talk you might be interested in: ua-cam.com/video/abAmUjBE60M/v-deo.html (a lot of the talks on the FOSSi channel from the conference are pretty interesting).
To load a .tcl file like the parax.tcl in magic, use the "source" command. e.g., "% source parax.tcl". It took me a while to find this, so I hope it helps others.
Hi! We've had workshops with people going from no knowledge to tapeout in 3 hours, no joke. Curiosity is all it takes--if you really do know nothing about it, checkout the design guide on tinytapeout.com/digital_design/wokwi/ It uses wokwi, a browser based playground, so 0 install and you can just try stuff out, and the guide walks you through the basics of digital design. If you're tempted, can be a fun evening. If you really get into it, you can actually tapeout but there's nothing forcing you to go that far.
Yeah, one of the first things I did was create a quick way to import all the standard cells into kicad by generating a library for those schematic parts, and that worked well, but magic really kills as it's so specialized--would be tough to get something like that with pcbnew, and I've come to think that it's a question of the right tool for the job. I *love* kicad, but after getting over the bump am finding that xschem and magic really are pretty good when you're down at the level of MOSFETs.
Hello Deng! Well, if I've understood your question correctly, the limit is basically the pad frame/packaging, if you're doing fully custom ASICs. At some point you need to get them wirebonded and in a package, so you need to come to some sort of agreement with whoever is doing that. TinyTapeout goes through efabless and used to use the Caravel harness to stick everything in a project area, and now uses openframe instead which gives more space and flexibility with the I/O (there used to be an MCU in there and other stuff, now instead we've dedicated pads to analog, have 24 digital GPIO as well as a few pins dedicated to internal management. There's a webinar on efabless and openframe at ua-cam.com/video/eflKFBCsor4/v-deo.html if you're curious about internals on that front.
Hey, it's amzing. Could you please tell , in terms of pdk version compatibility how do you combine your analog layout with the gds of digital design from openlane?
Hi! What I did for this was look at which version, specifically, had been setup for the analog tools in the VM (easy to tell, since all the $PDK_ROOT etc env variables point to it). Then, I had a system with openlane (github.com/efabless/openlane) already installed, I switched to that and just sync'ed the PDKs using `volare enable --pdk `. I'm not *certain* this was required, but it sure can't hurt. Finally, did face some issues in getting the GDS created by the digital hardening into the mixed signal... put my hints on that here: github.com/psychogenic/tt06-analog-wowa/tree/main/openlane Good luck!
I haven't: I actually started the whole ASIC journey when I found out there were open source alternatives to Cadence (which I didn't use but heard about pricing and endless complaining about from the teams using it). Had I known about Glade, might have poked around sooner... now: am liking the OSS stuff, and there's something about having a tool that powerful and free but not perfect that *I* can actually contribute to and make better that really hooks me. I'll tell you up front that the learning curve, to me, was steeper than normal. I was coming in from PCB land, so that accounts for some of it, but magic has a particular way of doing things that I find philosophically pleasing but seems different than "standard". I cannot tell, but am told, that klayout functions more like other things, so you might have a peek at that if you find magic too alien.
There are some hard to get really great sounding dac chips like the tda1541 and the ad1862 which are long since disconinued, but are still well loved and sought after. Would it be possible to remake them using this sort of method?
Without a doubt! Rej did a z80 CPU and a AY-891x (programmable sound generator). Check that out for an idea: github.com/rejunity/tt07-analog-dac-ay8913 really cool stuff
@@PsychogenicTechnologies Wow that's awsome Imagine the best sounding dacs of the future are open source and put into collaborative projects like tiny tape out. Would be so cool
How are you moving components around at the 27ish min mark? Whenever I import a spice netlist into magic no matter what tool im using it just starts painting in new materials(?)
Hi! First off, sorry it took me a while to see and answer this. I use a funny keyboard and am a little obsessive about comfortable key mappings, so I don't even remember what the original shortcuts are. However, if you ignore my specific keys (or set to something more useful to you), you can see part of my magicrc file here: --- set HOTKEY_WIRING XK_apostrophe set HOTKEY_ROTATE XK_grave set HOTKEY_COPY XK_comma set HOTKEY_MOVE XK_1 set HOTKEY_TEXT XK_t set HOTKEY_SELECT XK_y set HOTKEY_SELECT_MORE XK_Y # Actual Macro setup # same key goes to and out of wiring mode macro $HOTKEY_ROTATE {rotate -90} macro $HOTKEY_COPY {copy} macro $HOTKEY_MOVE {move} # select what's in the box macro $HOTKEY_SELECT select macro $HOTKEY_SELECT_MORE {select more} --- So, in essence I'm doing select on the cell, then using "move" (which is key 1 for me, from this setup). It'll jump whatever is currently selected to the position of the cursor. I'm not too clear on the other part/question. Are the spice imports working? Are you using the TT VM and/or sky130 tech?
@PsychogenicTechnologies Hey! Thank you for providing the template on how to modify key binds, I appreciate it. I did end up watching a few other magic tutorials on UA-cam and (somewhat) figured out how to use the commands. I was using the TT VM, but got stuck on an entirely different design/testbenching my design step further down the road. Been trying to figure out Prof. Carsten Wulff's aicex repo for now, but as an EEE Undergrad, the level of programming/software literacy(?) needed makes the learning curve steeper. Looking forward to any future videos you might have in the pipeline for Analog Design using the open source tools :)) Really enjoyed this one.
I watched for awhile then you mentioned 'magic'. Magic? We were using magic when I got my PhD in analog circuit design in 1990 time frame. After all these 30+ years, what you showed here doesn't look that much different. I checked out TTO, wow, a tile is only $300 for a packaged bonded out part. That's amazing! Back then, we'd get back a wafer, blah blah, not going to bore you reminiscing. Hey thanks for the video and the time travel back to my youth. Good luck, have fun and may your ASIC function as designed. Here's good reference for you "Analog Circuit Design: Art, Science and Personalities" by Jim Williams. PDF is easy to find.
Hi Craig! Yeah, magic--and I don't think its changed all that much, on the surface anyway. And we're on sky130, which was cutting edge... a good while back. What has changed is just how easy it is to play along, now. I didn't get the chance to do this in skewl, so I'm loving the opportunity to try something different every couple of months--nuts. It's great to be living in the future. Looks like you've been concentrating on macro scale tools (another something I still don't know but am curious about). If you want to detour down to the atoms, well you've got a head start with all the familiar tooling. Thanks for the comment and mini-reminiscing, cheers!
@@PsychogenicTechnologies @RevolutionEDA I went the analog route because I was music/stereo guy. At GT 88-93, I specialized in device modeling and circuit simulation (PE Allen advisor, CMOS red book). After graduation, I didn't want my passion to be my career. Did a UT aerospace post-doc in CFD supercomputing then started my career. Never had a full-time job, only consulting/contracting at start-ups or small companies, sometimes embedded hardware design. I wrote the first Embedded Linux book in 2022. I've had a wonderful career and like to keep current. Then your video popped up, I checked out TTO and was amazed. Things sure have changed in the past 30 yrs. Anyway, thanks again. I'm easy to find if you have other questions.
Would it be possible.. to take the spice output and convert it to verilog/vhdl for fpga physical testing?! Or even turn from xschem output to Amaranth?
I don't have any experience with this, and have wondered myself, but your question pushed me over the edge and I finally did a little search. We are not the first to think of this, it seems. Lots of the results are, however, really old. But I think some digging might reveal interesting things... This one github.com/Eyantra698Sumanto/Spice-to-Verilog-Converter looks really dumb and basic, and I can't say I've tried it or how it would react to raw mosfets... may look it to this further (but the stack is so deep, hah)
@@PsychogenicTechnologies i had a suspicion that this should have been done before. But might be problematic to go from analog circuit descriptions to Verilog, but i'm thinking a digital concept should translate bit easier. A good starting point prob. would be to make a simple circuit for a digital comparator and then make the same in Amaranth and try making a ... a parser/translator that route.. (oh my brain now are hurting and speeding at the same time).. The idea of taking your asic design and somewhat able to "verify it" in a fpga kinda makes sense to me for some reason.
Suggestion, use white (light) background, with black text/wires. They always work better and easier to watch. This is basic UI, and there are a reason ink are black and paper white. What you use on your own doesn't matter, use as you want. If you want people to read the contents on the screen, light background and dark/black forground is just better.
I'd like to say I left all the interfaces as you'll find them by default, to make things realistic, but in reality I didn't even think of it. It's a good idea and I'm pretty certain there's an easy way to get xschem out of "darkmode", so I'll try to do that next time. As for magic, I never did find an easy way to play with the palette on that one but at least it's more in line with what you're saying. Thanks for the suggestion! Cheers
The reason ink is black is because that was what was figured out. Paper is white to accommodated the ink. A black screen can show more colors distinctively than a white screen. It really becomes individual preference to choose one vs the other.
@@LeviYourchuck no. Well, you can use a sub optimal theme if you want to. But objectively light themes are better then datk themes. And black are the worst. Read text/lines on a black background is plain evil when it comes to User Interfaces and eye ergonomics. That is not an option, it is how it works. You might call it science if you want. Just try to read small red text on black background, and then red on white background. So if you are concern with eye ergonomic, you use a white or light background and everything written on it is much easier to read and detect, compared to a black background. And if it still is to bright for your eyes, there are control that could be used to turn down the intensity on the screen. Or use some other bright theme, not not a dark theme.
@@PsychogenicTechnologies actually it is bright themes that have better eye ergonomics, and dark themes that are hard on eyes. Just try to read red text on black background, and then compare to a white background.
i have seen the video, and i think the production of an asic is very very very complicated. for this devices, i have learned, the physics is the most important aspect to consider, therefor the eda software is secondary. as of my understanding, if you understand the physics of such devices everything else just consequently follows. but i respect your effort.
I think you're correct. The perfect tool would be absolutely transparent, meaning it would let you get that vision into silicon without being in the way at all. To me, that transparency goes with both how well you can work with the tool, but also how much you can count on it and how many obstacles are in your way. High cost and NDAs are big pains in the butt, at least for me, and an obstacle to learning and just experimenting, which is why I really love the open source stuff and want to get the word out about it.
Reading this makes my day, thank you! I really love sharing the cool stuff I find, and making the concepts accessible without diluting/deforming them is a main goal--so I'll try to keep it up and improve. Cheers!
Dude your content is amazing
Wow: thanks!! :D
@@PsychogenicTechnologies Yeah it's a total joy to listen to your stuff!
@@Fazzwrld That's great to hear: I know the topic is interesting (at least to us, hah) but am still working on how to best share it, and a comment like this helps a lot (both in morale and direction)
@@PsychogenicTechnologies I totally agree with the other person. I’m an electrical engineer (in the making) for microwave technology, but I have never dabbled into the world of ASIC design. Very interesting and good presentation halfway into the video.
@@CSTEnjoyer Hello, and welcome both to the channel and to the ee world! Thanks a lot. Microwave--nice. I've always had a pull towards RF, and have worked on some things but never quite enough. The great news for both of us is that even if the GPIO on the TT chips is relatively slow, you can get really fast on the inside of these things--even with the older sky130 tech (low) GHz is supposed to be doable. I'm currently playing with my ADC and working on some power stuff, but with that under my belt I'm going to start playing with the high freqs.
Thanks again, cheers!
Gosh, there's so much knowledge in your videos. It clarified so many things I never really understood of ASIC design. Seriously, thank you for doing this.
I simply could not receive more encouraging feedback than this--thank you so much!
Which parts I liked? Everything. From content to presentation and from editing to your energy. Keep it up, amazing work.
That's great to hear, many thanks Bob 😀I am, at this very minute, playing with stuff I'll be doing demos of in the next video (I'm so lucky to find everything I work on so much fun I want to keep doing and sharing it all the time). A bientot!
Please keep making these! This is fantastic and timely!
Thanks, that's great to hear! Putting out more is def the plan :-D See you soon
$5 says this guy knows how to make a guitar peddle and grows his own vegtables.
Hah, I am *such* a stereotype. Right on both fronts, though it's been a while for the guitar peddles.
@@PsychogenicTechnologies Right on! I learned a TON from your video by the way, I don't know how the algo. works but it picked me up from a long way away, I certainly wasn't looking to build an ASIC haha, never used KiCAD or SPICE etc, but just watching you use the software is very reassuring somehow, it's normally so intimidating!
Why did you only bet $5
@@Ariccio123 I could have just been projecting :P
@@PsychogenicTechnologies and smoke weed?
Your word economy is amazing. So concise. You grabbed me so fast too. Nice job.
Wow, thanks! Seems we share an appreciation for language or effective transmission of material, anyway. I'm often trying to get rather involved concepts across and, when I'm consuming them, I value precision first and information density a close second. I try and put some thought in how I'll present what I'm sharing, so your comment means a lot to me. Thanks again, cheers.
Wow this is fantastic! I really appreciate how clearly you explained what the process is like, and where to expect frustrations. Now I have to give this a shot!
Awesome! That is exactly what I hope to hear when I started making this video 😀 Thanks a lot for your comment and do give it a go--it's just so easy to get started now. And then gets just as hard as you want it to, which I also find great. Have fun, cheers!
Thanks for demystifying ASIC design! This truly is an amazing time to be alive. I'm blown away that this stuff is now accessible to hobyists and enthusiasts. Thanks for you work here!
Blake: I totally agree. It is bonkers, what we can do. Won't be long until we really are in Stephenson's "diamond age" but this whole accelerating returns thing means that the possibilities just keep growing: here I am, on my computer, designing functional things down in the nanometer range and that wasn't a even a possibility for me like 5 years ago. Loving it, and loving the opportunity to get the word out and maybe help.
Thanks for the feedback and encouragement! Cheers
Thank you for your content, the academic community is very grateful, you influence new minds. a hug from Brazil
Valeu! That's wonderful to hear, I do hope to make these things accessible or at least expose new possibilities.
Thanks a lot for taking the time to comment, cheers.
What I liked best about about your video? Dunno - everything? Maybe the positive DIY vibe, the "hey, it's chip design but you can do it" spirit and the great overview over the design process. I've been working with chips for more than 30 years now, designing analog and mixed-signal ASICs and teaching stuff, and for me it was a joy to watch the video. I just started dealing with tinytapeout, can't wait for your Magic and xschem videos!
Many thanks, Christian! I'm always happy to hear someone found the video useful or fun, and from someone with a solid background it means a lot. I got sidetracked on an FPGA project I'm prepping to talk about next, but now that summer play is winding down, I want to get going on other ASIC vids and get them out there. Thanks again, cheers :)
@@PsychogenicTechnologies Please make some videos about moving the design from FPGA to ASIC. Thanks
@@nurahmedomar oh, that's a great idea, thanks! Have an FPGA one coming up, will build on that
Fabio strikes again! Excellent video, very grateful for your videos!
hah! Grazie!
Wow, I'm going to need to rewatch this to be able to absorb it all! A lot of good stuff!!!
Thanks Alexander :-D
A+ on second watch I was able to absorb 90% without even looking at the screen
That's good demonstration
Man, this is absolutely amazing. We could have more access to free tools like this and develop single ICs with lower prices using projects like this TinyTapout. Waiting for incoming videos. Thanks
Obrigado! TinyTapeout is a great place to try stuff out, and the tools are there and working on real projects big and small, but it does have some limitations on space and speed of I/O and such. However, some people in the TT discord are getting together to do a chipignite directly with efabless--costs a bigger chunk of money, of course, but there you really do control everything. Not saying 10k is cheap but it is imaginable--not some crazy quarter million dollars. And the more stuff we do, the more of us playing, the more we can start getting clout and reducing costs further. I'd say it's pretty good already and even better things are coming. cheers!
@@PsychogenicTechnologies as I’m from Brazil we face a lot of difficulties here in terms of import taxes and so on, but I see this as a great opportunity for a Master’s or doctors degree in an University for example. I had worked on front end design 15 years ago, but give up and now I’m working on others stuff. Are they using 130nm technology, aren’t they?Thanks
It would be a dream to have a community to develop chips here in Brazil.
Unfortunately our closed border policies regarding electronics make things way more difficult.
Awesome video again. I've been following the open source ASIC/Tiny Tapeout space for a few months now and working on a digital design for TT07. But very curious about mixed signal design. This was a great overview for me a software developer turned circuit designer. Now I know what I need to learn more about to go deeper. Really cool stuff!!
Thanks Ben! It's super you'll be on the next run and if being set on the digital side means you'll only have one learning curve to confront for mixed signal next time 'round. We're on similar paths: professionally, I also went from software to hardware to wind up playing with chips. Means there's plenty of mystery left to explore in the whole analog design side of things, which is pretty exciting. I went rather blind, first time around, "thinking analog" but ignoring (and often being completely ignorant of) all the zillions of little effects that impact how real devices function down there. There are a lot, but if you're curious about how the universe works, it's fascinating to figure out. Good luck on TT07 (I still don't even know if I'll have time to submit my own, with all that's been going on!)
I just found your channel and I want to encourage you to continue making this type of content. Your presentation is great, and there is not much content I can find on such topics. I would love to see you do a few series say on KiCAD, PCB Design, High Speed Digital design, and low level analog design, and all the related tools. Keep up the GREAT work!
Hi! Welcome! This counts a lot, for me: it's great to hear and does encourage me to put in the time to make more. I'm already planning some more ASIC stuff, as well as KiCAD, that much is certain and the others are all actually topics I'm into, so I think it'd be fun to do and useful to share. Thanks again and keep an eye out, I hope to increase the pace of video production a little bit over the summer.
Every time you upload is a good time.
You're making my day, Benjamin. Thanks!
Genuinely such an underrated channel, great content!
Thanks a lot, Aarnav!
You, sir, are a total rock star for sharing your experience with this process. I can't wait to design my first block!
Jason, that's music to my ears! Thanks so much, really glad you're going to play... warning: can be addictive (it's 3am and I'm still fighting with a bug, but if it works gonna a be pretty coool! maybe some shuteye will help me solve it tho).
Thanks again for the comment, have fun :)
@@PsychogenicTechnologies Sleep does wonders for bug-hunting! Wish you the best. Regarding the addiction - to that I say, "Bring on ASICs Anonymous!"
This was a great intro to the analog work flow! I've found it really difficult to find good resources for learning this side of the open source tools so I'm looking forward to watching this video through again with the tools up and ready to follow along.
Thanks for sharing this stuff! Looking forward to your future videos that get deeper into the weeds!
Thanks Calvin--I agree: what little I found was either too much of a basic intro and focus on installing stuff, or sometimes just too deep, and I was hoping to provide something that gives a overview while being useful. I live in the weeds and def want to share some more focused content, with a mix of the how and the why of bits I find particularly useful or interesting--a comment like this encourages me to actually get it done :)
I actually found the whole thing rather fascinating. I'm pretty good when it comes to discrete components or actual tubes/valves but this is a whole extra level and something I've always been curious about. I'll definitely be going through the rest of your videos in the near future.
Valves! I've never gotten far, with valves. Well that's awesome--another cool tool for the belt.
Hey, this was amazing, thanks for doing this videos with such a high quality for us, electronic nerds!
Yes! The e-nerds are who I live for: it's my tribe, hah! Thanks a lot for your comment, I appreciate it, and am glad you liked the vid. I'll be putting out more on this front in the not-too-distant future, though I'm still not certain of the ordering (I want to keep the kicad crew happy, too!)
Great work, all the transitions are awesome, don't know much about designing for silicon, so appreciate the primer on resistors/primers/inductors.
Thanks William! I'm working out exactly what I'll put in there, but I want to publish at least more detailed spice, xschem, and magic videos. Though magic will probably be the main one for talking about parasitics, all three feel like good spots to get into a mix of practical and more specific talk about the low-level components, how they're made, tradeoffs, matching all that good stuff.
Personally, I'm better at doing anything when I actually understand *why* so I try to reflect that in the content and want to get more on all that in the next ASIC-related vids. Your comment encourages me to let myself loose in that department :-D
I love it!
I love everything!
Do more videos about ASIC. Everybody will love it!
Thanks! A lot :) I'm planning to, have a few in the pipe. Going to do an FPGA thing next, and long-promised pick and place thing too, but they're coming.
That was great, both highly educational and very entertaining. There was obviously a lot of work behind this and a lot of work going into the video. I really appreciate the end-to-end approach as so much contents jumps right into the middle and spend two hours on one detail. I'll now go hunt for more of your videos.
PS: Though I do not have any KiCad experience (and next to no EE background), I still think I could follow all of this.
👍
Hi Tommy! Thanks a lot: it is kindof a load of work to make these, and with the TT06 stuff and some summer play, I took a little break but am now working on the next ones--going to have some FPGA stuff, some KiCad and definitely some deeper dives on specific bits of ASIC land. Thanks for the great feedback, cheers!
Your content is absolute fire, advanced subjects and very approachable presentation.
wow. I have been thinking about trying to get into ASiC stuff for a while . Really useful and awesome content
Thanks a lot! The blessing and curse is that there is *so much* to learn, but there's never been a better time to get into it. Many thanks for the feedback, I hope to put out a lot more ASIC specifics in the near future.
Mighty healthy dose of enthusiasm sprinkled with a lot of deep-level technical autism. You have earned my sub, my friend. 🤠
Victor, I find that a fitting, funny and oddly poetic description: love it, I think I'll use it a my linkedin bio, hah!
Thanks for the comment, welcome and I hope to keep putting out stuff that you appreciate. Cheers.
Super engaging, loving the content! Thanks for sharing and making ASIC design accessible to beginners ❤
Love to hear it Daniel, thanks! If I can open the door a bit and get others joining in on the fun, my work is done--so thanks again for the encouraging comment, am working on more goodies on this front. Cheers!
Such a great content! Got me really interested in learning more, thanks!
That's fantastic--and thank you! It's a whole world that's just there waiting for us to play!
It will be super educational to create a basic inverter , with the capability of the experimental characterization in mind from 0 to tapeout to actual measurements, everything recorded in videos. Also, one can add different inverters (same design) in different places in the wafer to showcase the process variability. Its a lot effort to do the whole thing, so , thank you for your efforts, much appreciated !
hm, that's a good idea. I did an inverter "live" with Robert Feranec (on his channel a few weeks back) but it was just the design parts--doing it in a way to allow some deep probing and comparisons would be really interesting... maybe I can reserve a section of my TT08 submission for this. Thanks!
@@PsychogenicTechnologies thanks for the quick reply! The thing that's missing from academia are all the practical considerations from design, cost of EDA tools, physical limitations, experiment methodologies, optimization flows...etc. these series of videos from you and the whole community will be a first pass for beginners. And the more we tap out the more confident we get to try even more complex projects. And who knows, maybe 10 years down the line we will get an open source Cerebras 🙄😁.
New subscriber! Best content ever in electronic design! Thank you Sir!
Wow, thanks Xavier! It's my pleasure to do these, especially when I know they're appreciated--so thanks again and I'll be making more!
dude the presentation is great, congratz
Hey, thanks!! :-D
I really enjoyed this, and I made it all the way through to the end.
To that end, here's *why* I enjoyed it. Back in my '86-87 ish 2nd - 3rd year university classes, we did some units on VLSI design, standard text being as you might expect.
We learnt about design rules, components and had an assignment to put together a small hand-checked library, and be able to draw these ito a layout.
It was cool, but seeing this actually become an accessible to Jo Everyhobbiest thing is incredible, thought yes, it's kinda of been 'accessible' at the university level for a while.
Thanks for your videos. Any 'surprising' applications of these techniques? What applications of analog ASIC design are most 'happening' or most under the radar right-now?
Thanks! That's pretty awesome: I didn't get any opportunity like that at uni (wrong program, but not even sure they offered anyone) but the fact that it's now possible both let me play and is one of the reasons I make this sort of content, just so more people will know about it, see that it's doable and join the fun. With loads of people working in parallel and cross-pollinating crazy cool stuff can happen.
As for what's going on, there's more than I can manage to keep track of. Of course AI related things but application specific compute is one way to squeeze out more juice as we hit the bottom of possible granularity and the end of Moore's law, so as it looks to be exploding everywhere. One thing that got that piqued my curiosity with all this was a contract I did with a company doing analog neural nets (in essence shoving charge into the gate insulator, to change the Vth, to basically weight the neuron): makes the computation extremely low power, so you wind up being able to stuff ML into the *components* on the edge. I thought that was pretty cool.
@@PsychogenicTechnologies Spiked Neural Nets ?
Super cool! I did my thesis designing a phase detector for a digital phase-locked loop in Cadence; however, it never moved beyond simulation. This video inspires me to try one day bringing it into reality 🤩.
Yeah! You get far with sim, but there's something I find magical about getting these things you've designed in your hands. About 2 months until the TT08 deadline--maybe a project for rainy summer days :) Thanks for your comment, Jeremiasz. Cheers!
Love this video! Looking to build ASIC skills and trying TinyTapeout projects soon, so your videos are a goldmine. Any advice for an undergrad trying to learn RF/Mixed Signal ASIC design?
Thanks Rohit! Basic advice is "play" -- get xschem, start doing experiments in simulation. But then, yeah, this goes so (so) deep, you could write a thesis on track width and tradeoffs, and when you get into the FETs well, feels like the level of interesting goes on forever. Also, some of these are beginner or intermediate projects, but TT has a number of RF things in the last two tapeouts--oscillators, a gilbert cell based double balanced mixer, lots of things. So, if you do want to take a look, you can just clone those repos and see how they're built, tweak things, see the impact, improve the projects themselves or use them as building block (say you take the mixer as a module and dev an env detector or fm demod, that should be fun).
this is awesome. going into grad school for mixed signal design so I appreciate that there are many more intelligent people making videos like this for me to consume lol
hah, well this is a really great time to do so--and now, you don't even have to wait, you can start exploring on your own as your learn. Have fun!
@@PsychogenicTechnologies seeing people as competent as you is honestly inspiring. Keep up the good work
More of these type of vids please
Hey Bram, thanks for the comment and *yes* more coming as fast as I can get them done and edited, while still actually keeping clients satisfied :)
Very nice overview of the whole process. Thanks!
Thanks for the feedback, Henner, glad you liked!
damn these softwares are crazy creative
I appreciate your wizardry and explanations, thanks!
That's great to hear, trying to get more of the out soon. Cheers!
you just make so many cool things... and god at explaining stuff.... very well done....
That's awesome to hear--thanks a lot, Finn :-D
This reminds me of 6 years ago when i was trying to figure out how to even open magic, it was not that easy. After a while i figured it out. Cool stuff.
Ah, well so you've been through it as well: good job, oof :-D What I really needed was a) a little perseverance and b) to actually go through the tutorials... it was just a bit too alien to "pick up" like I usually do with most software (since, it seems, we've come to some sort of agreements in terms of UI that I'm leaning on, hah)
Love this channel
Thanks Mordecai/Pinto!! 😀The comment is much appreciated. Cheers
Sir, this is awesome! Thank you!
That is very sweet to hear--thanks!! Cheers
This is just great!!!
You have explained the workflow way better than many others.
Looking forward to see more content from you.
Maybe a more intertwined AMS? 😅
Thanks a million Gabriel--that's super encouraging :) Willing to try for more AMS but... what's AMS? Too many TLAs for me to hold 'em all, hah
@@PsychogenicTechnologies Sorry, my bad 😅
Actually, one of the most confusing parts of these workflows are the Three Letter Abbreviations 🤣
AMS stands for Analog and Mixed Signals, basically your whole project including the digital part of it 😉👍
@@gabrielbarrientos468 Hahah, yes, IDK, TBH IRL I use them a whole lot, LOL.
Ah, yeah, I want to talk a good deal more about mixed signal--to me, that's where all the power of this stuff is, leveraging the goodness of both sides!
@@PsychogenicTechnologies can't agree more
Looking forward to see more content and new adventures from your side
Excellent content. Do you have a beginners guide to installing and setting up the tools etc ?
Thanks! There are a few ways to do this, but really virtualbox and the VM on github.com/TinyTapeout/analog-virtualbox-vm-sky130a are what I've used for this, so far--all the analog tools are in there.
I'll talk about the tweaks I did in xschem and magic videos, but that's all optional.
@@PsychogenicTechnologies Thanks. Managed to get the tools installed from github. Plugging away at it now...
@@GodzillaGoesGaga Ah, that's great to hear! Welcome to the teeeeeny, have fun
I get lost for 1 second and the next second I get smacked in the face with jewels of information. Top Top !!! content
hahaha, that is great to hear and an awesome way to put it: going onto the resume, this one😄 Thanks!
This side of youtube is awesome!!!!
Aid, thanks! This inspires me to make more for the youtube geekdom!
Thank you very much, you have enormous in this field, I cam thru Robert Feranec'c channel. Subscribed!
Thanks Abhishek, I really appreciate the comment and it's awesome that getting to talk about this on Robert's channel has expanded the reach. More on the way, on many topics, but some cool ASIC stuff for sure. Cheers
You're awesome man ❤
Wow, thanks! I love sharing this stuff, and hope it's appreciated--a comment like this really encourages me to put out more. thanks again :)
really cool channel. i hope you grow fast 👍
Thanks a lot, Gary! I took some time off this summer, but am getting back into it and hope what I'm working on now is well received and reaches lots of interested/interesting people. Thanks for the feedback! Cheers
@@PsychogenicTechnologies don't overtake yourself, live comes first. Cheers 😉
Great video! Can u please make detailed videos on xschem and magic?
Thanks! That has been the plan all along, but I kept getting sidetracked. So, right now I have one in the works on FPGA, and two related to ASICs (though further down the line). And I'm releasing an unrelated quickie video this morning, just 'cause it wasn't too hard to get done while I was doing the boards related to that. So, I'll be on videos a lot more fulltime as soon as I get back from the next little trip. Thanks again, cheers.
@@PsychogenicTechnologies No hurry mate, I'll be looking forward to these projects, and yeah the new video is great. Being a beginner all these things are kinda overwhelming but I'm slowly understanding stuffs . The enthusiasm you bring in your videos is outstanding, wish I discovered your channel earlier.
Great video, especially liked magic related scripts to toggle layers and run pex etc. Any chance they can be shared ?
Thanks! Yes, I put some time into getting it to work in a way I find usable both with key mappings (no, just no: ctrl-s really does need to be "save", lol) and scripts, and I just need to document and put online--will def be doing along with a magic-specific video
This was awesome. Thanks!
Thank you!! super encouraging 😀
This is very interresting. Digital is very useful analog is beautiful. I guess I stick with tht though since analog doesn't require so many components most often and many things just need special components.
That's a nice way of putting it! I'm still working on getting my analog creations to be streamlined to the point of beautiful, but it sure has been lots of fun.
This was really cool!
Thanks Nicholas!!
Thanks for the overview, this is very helpful. A bit annoying to get everything setup, either you run a virtual amchine or try to get those old tools to play nice with your distro, still very far away from the state of where kicad is. Hopefully more people join the space and we can get some standardization going on
Thanks! Yeah, I used to insist on getting things working on the bare metal of my machine, but with everything interacting with the PDK and between themselves, and everything in dev at the same time, it can get pretty awful.
So the VM was the easiest way for me, and it worked well. Something I haven't tried yet but that looks pretty good is another packaged solution, but using docker (which might be lighter than running in a VM? I dunno).
This is another Harald Pretl thing, and it's well maintained: it's the iic-osic-tools,
hub.docker.com/r/hpretl/iic-osic-tools
Might be worth checking out.
Wow, awesome Video!
Many thanks, Sönke!
Am I correct to assume that by magic software you mean Magic VLSI? (& not Image Magick nor MagickQ software)
that's so crazy... It's as low level as it gets! You inspire me dear
I know! I think, with this, we're finally getting pretty close to the bottom. The good news is: there's *tons* to explore and play with down there, so lots of fun to come :-D
Well done!
Thanks! Took a quick look at your mag trap filtration system... that's pretty cool! Have always been curious about the wet sciences: a highlight experience at uni was a molecular bio class where I saw (stained) DNA with my own eyes... and the smaller things get, the less you can differentiate between wet and dry, anyway.
Thanks again, cheers!
@@PsychogenicTechnologies Cool, thanks for checking it out. Wet stuff does have some interesting properties. I think of DNA, and its compiler the ribosome, as a programming language for atoms. Too bad its way too tricky and powerful. I also think the sciences are converging and its probably not so productive anymore to say where one begins and the other ends.
I love your channel! Keep going! 😎🤖
Thanks so much: I put in some effort making these things, so knowing they're appreciated is very encouraging. Cheers!
Just WOW!
Thanks Zeta!!
Super cool, I'm thinking about whether we could do something like that with our apprentices (electornics and application & coding). I think we could do some thing like a spi port expander. Is it possible to do bga packages?
Hi Lukas, I think the SPI port expander project is a really cool idea to get started with. For packaging: with tinytapeout the answer is no. At this time we're on QFNs and that's that. If you're going direct through efabless and doing a chipignite on your own, there are more options and you can even get bare dies.
I'm curious why you'd want BGA: what's the advantage for you, density, low inductance pads, something else?
Thanks! Is it normal to view the output as 2 dimensional? I believe the incorrect bits you saw after a certain point are indicating your signals might be inverting. If you link them via imaginary axis, you might see when and where the signal flips and why. That stems from using sin and cos for wave generation. It embeds the pi geometry in the signal, and after a certain point, you're seeing the drift in the bits.
what a fantastic vedio! But I wonder why there is no "Import SPICE" option in the File tab in my Magic(version 8.3).
Hello! My guess is that you've started magic "bare"... like with just the "magic" command.
I have a shortcut, "magicsky" and what it does is this:
alias magicsky='magic -rcfile /home/ttuser/pdk/sky130A/libs.tech/magic/sky130A.magicrc
It is loading the magic rc from the SKY130 PDK, and that is in turn importing a bunch of stuff (like sky130.tcl etc) that has the import spice stuff added to the menu.
@@PsychogenicTechnologies It really works! Thank you so much!
God, I am missing my long mane. Just subbed your channel and gonna watch this.
hah, not exactly what I was expecting/hoping to inspire, but hey, I'll take it too :-D Welcome, enjoy the vid, and should have more related stuff on the way soon.
a fan from Indonesia, great videos!!
Indonesia, awesome! I'm still amazed that we get to collaborate and interact like this :) Thanks for the comment!
awesome video
vielen dank! :-D
Is there a video on how to set up the environment from scratch?
Hello Khang, maybe but not that I know of. The VM and the IIC-JKU docker are the two easy ways, and Matt's analog course does have a detailed section on doing it "by hand" on your local system (bit.ly/analog-waitlist ) but, because there are bundled versions, I just haven't focused on manual style installs.
@@PsychogenicTechnologies oh thanks for your reply! So, how could you set it up on your machine though? I wanna be able to get access to the tools (software) and do it on my laptop so that I can do these amazing things that you showed in the video!
Great content. Can you please share how many samples did you order, and how much does it cost? Waiting for your chip measurement videos in the future!
In short, I've been getting one chip per tinytapeout run where I have a project. The cost depends on the size you want (a single "tile", enough for say 1k gates or so if doing digital) is $50, and the physical chip + the carrier board it's on and the demoboard is an extra 100... so if it's for a class, or a group, you could get many tiles and only a few chips--or do it the other way around if you wanted. See app.tinytapeout.com/ for current pricing.
@@PsychogenicTechnologies Thanks, it's way cheaper than what I thought. Definitely will follow your design steps and try to tape out one!
@@nurahmedomar awesome. I juuuust missed the TT08 deadline (left things to last minute, crunched like an idiot, and then failed to get the project in before deadline), but am certain the next run will be announced soon and fully intend to be on that one. Hope to see you there! Cheers
Been looking for a decent software for chip designing, potentially with the ability to design transistor library for chips smaller than 65 nanometers (or even 2 nanometers node), as I figured why not put together a general purpose RISC-V VLIW processor. Hopefully this IC CAD app will do.
Inspiring
Thanks! I hope to put out some more, including some challenge-based learning experiences you may particularly enjoy ;-)
I appreciate the comment, cheers
Cool work - You have put me out of work 🙂
hah, since you're saying it with a smile, I'm hoping that means you can focus on other cool bits!
Thanks :-D
@@PsychogenicTechnologies Yes I spend too much time trying to push open source into our universities here in Denmark. From now I just need to point them to your videos.
No but fr, I just somehow feel like we got the same curios type of mentality about hardware. Is like , I be thinking if it could be possible to go further with HDL design but using only open source tools, and bam!! There you go. Please keep up the good work of spreading knowledge of free.
Awesome! Am glad to hear it answered some of your Qs and that you actually happened upon it :-D And yeah, I'm curious on a number of fronts and tend to hop around a lot... funny thing is that, by putting out these videos, I'm learning that there aren't tons of us but I'm certainly not alone in my eclectic hardware fascinations--so cool!
Have you ever played with BiCMOS logic?
I’m interested in it because of [relatively] recent developments in heterojunction bipolar transistors.
I have not. However, from what I gather, IHP's newly open process is BiCMOS--not sure that TT will support it, at least right away, but it's happening right now and pretty interesting. They have settled on klayout, I think, in terms of doing layout, rather than magic but in any case the PDK is open and they seem very motivated to get us involved. I was at latch-up a month or two ago, and Frank Vater gave a talk you might be interested in: ua-cam.com/video/abAmUjBE60M/v-deo.html
(a lot of the talks on the FOSSi channel from the conference are pretty interesting).
To load a .tcl file like the parax.tcl in magic, use the "source" command. e.g., "% source parax.tcl".
It took me a while to find this, so I hope it helps others.
Yes! Thank you Jeremy!
I wish I could dip my toes in a lot of this stuff but I have no EE knowledge. Still great video and loved it through the end.
Hi! We've had workshops with people going from no knowledge to tapeout in 3 hours, no joke. Curiosity is all it takes--if you really do know nothing about it, checkout the design guide on tinytapeout.com/digital_design/wokwi/
It uses wokwi, a browser based playground, so 0 install and you can just try stuff out, and the guide walks you through the basics of digital design. If you're tempted, can be a fun evening. If you really get into it, you can actually tapeout but there's nothing forcing you to go that far.
Waiting for the ngspice video ;)
Thank you!!!
Glad you like: I hope it's useful and that, if you haven't done this flow yet, it encourages you to go for it and helps you get started 😀
If we can use and link KiCAD schematics with a microelectronic layout tools will be an amazing using a plugin or others things, I think!
Yeah, one of the first things I did was create a quick way to import all the standard cells into kicad by generating a library for those schematic parts, and that worked well, but magic really kills as it's so specialized--would be tough to get something like that with pcbnew, and I've come to think that it's a question of the right tool for the job. I *love* kicad, but after getting over the bump am finding that xschem and magic really are pretty good when you're down at the level of MOSFETs.
How many input channels could you pack into an asic, assuming normal processes?
Hello Deng! Well, if I've understood your question correctly, the limit is basically the pad frame/packaging, if you're doing fully custom ASICs. At some point you need to get them wirebonded and in a package, so you need to come to some sort of agreement with whoever is doing that. TinyTapeout goes through efabless and used to use the Caravel harness to stick everything in a project area, and now uses openframe instead which gives more space and flexibility with the I/O (there used to be an MCU in there and other stuff, now instead we've dedicated pads to analog, have 24 digital GPIO as well as a few pins dedicated to internal management. There's a webinar on efabless and openframe at ua-cam.com/video/eflKFBCsor4/v-deo.html if you're curious about internals on that front.
Hey, it's amzing. Could you please tell , in terms of pdk version compatibility how do you combine your analog layout with the gds of digital design from openlane?
Hi! What I did for this was look at which version, specifically, had been setup for the analog tools in the VM (easy to tell, since all the $PDK_ROOT etc env variables point to it). Then, I had a system with openlane (github.com/efabless/openlane) already installed, I switched to that and just sync'ed the PDKs using `volare enable --pdk `. I'm not *certain* this was required, but it sure can't hurt. Finally, did face some issues in getting the GDS created by the digital hardening into the mixed signal... put my hints on that here: github.com/psychogenic/tt06-analog-wowa/tree/main/openlane
Good luck!
Design an opamp next please
Yeah, that does fit with what I've been thinking of--thanks for the suggestion :)
Ever used Glade (from peardrop design)? I'm having some trouble in the layout part and might switch to your setup
I haven't: I actually started the whole ASIC journey when I found out there were open source alternatives to Cadence (which I didn't use but heard about pricing and endless complaining about from the teams using it). Had I known about Glade, might have poked around sooner... now: am liking the OSS stuff, and there's something about having a tool that powerful and free but not perfect that *I* can actually contribute to and make better that really hooks me.
I'll tell you up front that the learning curve, to me, was steeper than normal. I was coming in from PCB land, so that accounts for some of it, but magic has a particular way of doing things that I find philosophically pleasing but seems different than "standard". I cannot tell, but am told, that klayout functions more like other things, so you might have a peek at that if you find magic too alien.
There are some hard to get really great sounding dac chips like the tda1541 and the ad1862 which are long since disconinued, but are still well loved and sought after. Would it be possible to remake them using this sort of method?
Without a doubt! Rej did a z80 CPU and a AY-891x (programmable sound generator). Check that out for an idea: github.com/rejunity/tt07-analog-dac-ay8913 really cool stuff
@@PsychogenicTechnologies Wow that's awsome
Imagine the best sounding dacs of the future are open source and put into collaborative projects like tiny tape out. Would be so cool
How are you moving components around at the 27ish min mark? Whenever I import a spice netlist into magic no matter what tool im using it just starts painting in new materials(?)
Hi! First off, sorry it took me a while to see and answer this.
I use a funny keyboard and am a little obsessive about comfortable key mappings, so I don't even remember what the original shortcuts are. However, if you ignore my specific keys (or set to something more useful to you), you can see part of my magicrc file here:
---
set HOTKEY_WIRING XK_apostrophe
set HOTKEY_ROTATE XK_grave
set HOTKEY_COPY XK_comma
set HOTKEY_MOVE XK_1
set HOTKEY_TEXT XK_t
set HOTKEY_SELECT XK_y
set HOTKEY_SELECT_MORE XK_Y
# Actual Macro setup
# same key goes to and out of wiring mode
macro $HOTKEY_ROTATE {rotate -90}
macro $HOTKEY_COPY {copy}
macro $HOTKEY_MOVE {move}
# select what's in the box
macro $HOTKEY_SELECT select
macro $HOTKEY_SELECT_MORE {select more}
---
So, in essence I'm doing select on the cell, then using "move" (which is key 1 for me, from this setup). It'll jump whatever is currently selected to the position of the cursor.
I'm not too clear on the other part/question. Are the spice imports working? Are you using the TT VM and/or sky130 tech?
@PsychogenicTechnologies Hey! Thank you for providing the template on how to modify key binds, I appreciate it.
I did end up watching a few other magic tutorials on UA-cam and (somewhat) figured out how to use the commands. I was using the TT VM, but got stuck on an entirely different design/testbenching my design step further down the road. Been trying to figure out Prof. Carsten Wulff's aicex repo for now, but as an EEE Undergrad, the level of programming/software literacy(?) needed makes the learning curve steeper.
Looking forward to any future videos you might have in the pipeline for Analog Design using the open source tools :)) Really enjoyed this one.
Evildragon, where did you accent go?
After all this time, you've tracked me down! I was advised to appeal to a wider audience, a creamy canadian-neutral would do the trick. heh
I watched for awhile then you mentioned 'magic'. Magic? We were using magic when I got my PhD in analog circuit design in 1990 time frame. After all these 30+ years, what you showed here doesn't look that much different. I checked out TTO, wow, a tile is only $300 for a packaged bonded out part. That's amazing! Back then, we'd get back a wafer, blah blah, not going to bore you reminiscing. Hey thanks for the video and the time travel back to my youth. Good luck, have fun and may your ASIC function as designed.
Here's good reference for you "Analog Circuit Design: Art, Science and Personalities" by Jim Williams. PDF is easy to find.
That was a very good book. I used to occasionally pass by Bob Pease's cubicle when I used to work at National Semiconductor.
@@RevolutionEDA That's cool, you must be old school like me. I was going to mention Pease's "What all this *** stuff , anyhow?" series.
Hi Craig! Yeah, magic--and I don't think its changed all that much, on the surface anyway. And we're on sky130, which was cutting edge... a good while back. What has changed is just how easy it is to play along, now. I didn't get the chance to do this in skewl, so I'm loving the opportunity to try something different every couple of months--nuts. It's great to be living in the future.
Looks like you've been concentrating on macro scale tools (another something I still don't know but am curious about). If you want to detour down to the atoms, well you've got a head start with all the familiar tooling.
Thanks for the comment and mini-reminiscing, cheers!
@@PsychogenicTechnologies @RevolutionEDA I went the analog route because I was music/stereo guy. At GT 88-93, I specialized in device modeling and circuit simulation (PE Allen advisor, CMOS red book). After graduation, I didn't want my passion to be my career. Did a UT aerospace post-doc in CFD supercomputing then started my career. Never had a full-time job, only consulting/contracting at start-ups or small companies, sometimes embedded hardware design. I wrote the first Embedded Linux book in 2022. I've had a wonderful career and like to keep current. Then your video popped up, I checked out TTO and was amazed. Things sure have changed in the past 30 yrs. Anyway, thanks again. I'm easy to find if you have other questions.
Would it be possible.. to take the spice output and convert it to verilog/vhdl for fpga physical testing?! Or even turn from xschem output to Amaranth?
I don't have any experience with this, and have wondered myself, but your question pushed me over the edge and I finally did a little search. We are not the first to think of this, it seems. Lots of the results are, however, really old. But I think some digging might reveal interesting things... This one
github.com/Eyantra698Sumanto/Spice-to-Verilog-Converter
looks really dumb and basic, and I can't say I've tried it or how it would react to raw mosfets... may look it to this further (but the stack is so deep, hah)
@@PsychogenicTechnologies i had a suspicion that this should have been done before. But might be problematic to go from analog circuit descriptions to Verilog, but i'm thinking a digital concept should translate bit easier. A good starting point prob. would be to make a simple circuit for a digital comparator and then make the same in Amaranth and try making a ... a parser/translator that route.. (oh my brain now are hurting and speeding at the same time).. The idea of taking your asic design and somewhat able to "verify it" in a fpga kinda makes sense to me for some reason.
I’m going to try to build an asic for my custom ai architecture, thanks
Cool, that should be a really interesting project
Suggestion, use white (light) background, with black text/wires. They always work better and easier to watch. This is basic UI, and there are a reason ink are black and paper white.
What you use on your own doesn't matter, use as you want. If you want people to read the contents on the screen, light background and dark/black forground is just better.
I'd like to say I left all the interfaces as you'll find them by default, to make things realistic, but in reality I didn't even think of it. It's a good idea and I'm pretty certain there's an easy way to get xschem out of "darkmode", so I'll try to do that next time. As for magic, I never did find an easy way to play with the palette on that one but at least it's more in line with what you're saying.
Thanks for the suggestion! Cheers
The reason ink is black is because that was what was figured out. Paper is white to accommodated the ink. A black screen can show more colors distinctively than a white screen. It really becomes individual preference to choose one vs the other.
@@LeviYourchuck no. Well, you can use a sub optimal theme if you want to. But objectively light themes are better then datk themes. And black are the worst.
Read text/lines on a black background is plain evil when it comes to User Interfaces and eye ergonomics. That is not an option, it is how it works. You might call it science if you want.
Just try to read small red text on black background, and then red on white background.
So if you are concern with eye ergonomic, you use a white or light background and everything written on it is much easier to read and detect, compared to a black background.
And if it still is to bright for your eyes, there are control that could be used to turn down the intensity on the screen. Or use some other bright theme, not not a dark theme.
@@PsychogenicTechnologies actually it is bright themes that have better eye ergonomics, and dark themes that are hard on eyes.
Just try to read red text on black background, and then compare to a white background.
Nice 1
Application Specific Integrated Circuit.
i have seen the video, and i think the production of an asic is very very very complicated. for this devices, i have learned, the physics is the most important aspect to consider, therefor the eda software is secondary.
as of my understanding, if you understand the physics of such devices everything else just consequently follows.
but i respect your effort.
I think you're correct. The perfect tool would be absolutely transparent, meaning it would let you get that vision into silicon without being in the way at all. To me, that transparency goes with both how well you can work with the tool, but also how much you can count on it and how many obstacles are in your way. High cost and NDAs are big pains in the butt, at least for me, and an obstacle to learning and just experimenting, which is why I really love the open source stuff and want to get the word out about it.
Please maintain your presentation style. It's uniqueness allows you to condense huge amount of information in an easy way to follow and understand
Reading this makes my day, thank you! I really love sharing the cool stuff I find, and making the concepts accessible without diluting/deforming them is a main goal--so I'll try to keep it up and improve. Cheers!
very great content! ngspice video would be fine
Oh no, you again.
** had to watch **
Nice... am working on showing up here more consistently, but main focus is to try and make it worth your time. Cheers :)
What kind of sorcery is this?
The best kind: the kind you can try at home! (btw, it's pronounce "levio-SAH") 😀
👏👏👏👏👏