[HN Gopher] Die shots of as many CPUs and other interesting chip...
       ___________________________________________________________________
        
       Die shots of as many CPUs and other interesting chips as possible
        
       Author : uticus
       Score  : 198 points
       Date   : 2025-10-17 18:39 UTC (5 days ago)
        
 (HTM) web link (commons.wikimedia.org)
 (TXT) w3m dump (commons.wikimedia.org)
        
       | roflmaostc wrote:
       | Someone should sponsor that guy a gigapixel microscope such as
       | those
       | 
       | https://gallery.ramonaoptics.com/gallery/viewer/42009871001#...
        
       | 7222aafdcf68cfe wrote:
       | If you're into this, and enjoy more details, many more get
       | published weekly on Mastodon under the #nakeddiefriday tag -
       | https://infosec.exchange/tags/nakeddiefriday
        
       | rft wrote:
       | Another great resource is this site:
       | https://misdake.github.io/ChipAnnotationViewer/?map=Zeppelin...
       | It has a Google Maps like interface for exploring die shots and
       | even some annotated versions of chips.
        
       | 7373737373 wrote:
       | See also: https://siliconpr0n.org/
        
       | ttoinou wrote:
       | Reminds me of Koooooooyanisqatsi
        
       | mrcsharp wrote:
       | I'm always fascinated by how brilliant us humans can be. So much
       | so that we can put billions of transistors in very small spaces
       | and in complex structures while also mass producing it.
       | 
       | I highly recommend watching this video about lithography and the
       | machine that makes it all possible [1].
       | 
       | [1] https://www.youtube.com/watch?v=B2482h_TNwg
        
         | solarwindy wrote:
         | Outstanding video, thank you. No wonder this took months' worth
         | of research and animation to make.
        
         | Velocifyer wrote:
         | Branch Education makes good videos.
        
         | atonse wrote:
         | Never heard of this channel. I was mesmerized by this video,
         | thanks. But I have to get back to Adulting for now.
        
         | chrisgarand wrote:
         | On the less serious side I like EvilmonkeyzDesignz youtube
         | shorts.
         | 
         | They generally looks for the doodles on older chips, but also
         | looks at MEMs devices as well:
         | https://www.youtube.com/@EvilmonkeyzDesignz
         | 
         | They would definitely be able to add to the list (they have a
         | video of 100 chips for their 100k subscriber mark).
        
       | guerrilla wrote:
       | Which one of the CPUs do you think looks the cleanest,
       | aesthetically? The first Alpha looks rather chaotic, while the
       | Samsung Alpha looks very uniform. That TI PA-7000 FPC looks like
       | chaos. I think the two PowerPCs look the best, which is what I'd
       | expect too. Well, actually I'd expect some of the other RISC to
       | look simpler too...
        
         | gblargg wrote:
         | I was also fixed on PowerPC after noticing how regular the dies
         | were. Cache on the right, then what looks like bank after bank
         | of fairly uniform control logic. It's almost like they decided
         | on a general structure and fit everything into this, rather
         | than letting things organically form as needed.
        
         | leptons wrote:
         | >That TI PA-7000 FPC looks like chaos
         | 
         | I think it looks very well structured compared to the Soviet
         | KM1801VM2
         | 
         | https://commons.wikimedia.org/wiki/User:Birdman86#/media/Fil...
        
       | potato-peeler wrote:
       | Are there any shots of audio amplifiers?
        
         | phkahler wrote:
         | >> Are there any shots of audio amplifiers?
         | 
         | https://www.righto.com/2020/06/reverse-engineering-and-compa...
         | 
         | https://www.righto.com/2019/02/op-amp-on-moon-reverse-engine...
         | 
         | https://www.righto.com/2018/06/silicon-die-analysis-op-amp-w...
        
       | potato-peeler wrote:
       | Realistically, are these enough to replicate the chips?
        
         | dfox wrote:
         | Mostly no. You do not see the lower layers and for anything sub
         | 1um or so the resolution is too poor anyway.
        
         | pbw wrote:
         | To capture the individual transistors on a modern CPU, you'd
         | need an image tens of terabytes in size, and it'd have to be
         | captured by an electron microscope, not an optical image. And
         | even that wouldn't let you see all the layers. Some of the very
         | old CPUs, I'm not sure what resolution would be required.
        
         | dan_hawkins wrote:
         | Absolutely not. It's like opening a hood of your car, taking
         | picture of what you see and then try to build replica of the
         | engine based on that.
        
       | intrasight wrote:
       | Nice
       | 
       | I have on my desk the book "State of the Art" by Stan Augarten.
       | It shows the progression of transistors and integrated circuits
       | from conception through 1983.
       | 
       | The book was one of the inspirations for me to become an
       | electrical engineer. My older brother loaned me a copy of it when
       | it was published in 1983.
        
       | Sweepi wrote:
       | see also: https://www.flickr.com/photos/130561288@N04/ (yes,
       | afaik flickr is still their main/only homepage except for
       | Twitter: https://xcancel.com/fritzchensfritz)
        
       | hyperbrainer wrote:
       | I do not know enough to analyse these chips in any meaninful way,
       | but is there a trend or cool feature to be seen across?
        
       | sllabres wrote:
       | A nice collection of die shots is on Fritzchens Fritz [1] on
       | flickr
       | 
       | [1]
       | https://www.flickr.com/photos/130561288@N04/albums/721576504...
        
       | atannen wrote:
       | see also https://wikichip.org/
        
       | foz wrote:
       | I love these. The https://www.youtube.com/@EvilmonkeyzDesignz
       | channel does some wonderful explorations of chips with high
       | powered microscopes, finding easter eggs and graphics left by
       | designers. So much fun.
        
         | avipars wrote:
         | His videos are really fun!
        
       | mk89 wrote:
       | How can someone explain this to a kid? Is there somewhere an even
       | more simplified version than Arduino or similar to show how all
       | these things actually work? I know arduino is not a cpu, but
       | overall, how these things work together, would be great to
       | see/show.
       | 
       | I don't expect to show how electrons move :) I mean, some model,
       | a toy or so, that shows how these things work. I remember it only
       | from books/specs, but even there, at a certain point there are
       | "limits" :)
        
         | thessalchips wrote:
         | I'm teaching a course on computer architecture at my university
         | and there are these model processors called MUx (MU0-7) that
         | explain how a CPU works from the ground up. I'm not aware of
         | any toys (my students keep asking me about that as well), but I
         | wrote an interactive visualizer that illustrates the simplest
         | processor and how data moves through it:
         | https://pascalbormann.de/mu-vis/ Not mobile friendly
         | unfortunately and maybe a bit too advanced for kids, but it
         | could be a starting point. The code is here if you want to
         | build on it: https://github.com/Mortano/mu0-visualization
        
           | thabit wrote:
           | A breadboard CPU would be a good "toy" no?
        
             | thessalchips wrote:
             | True but to do anything interesting you do need a whole
             | bunch of wires and components, even for an 8-bit CPU. Maybe
             | it is a good start to instead combine gates to do some
             | useful work. People are also building CPUs in Minecraft,
             | that should get kids engaged :)
        
               | jjk7 wrote:
               | That's how I was taught, build up a CPU using TTL logic
               | chips.
               | 
               | Even just starting with the building blocks is useful,
               | like build a flip-flop
        
         | Quizzical4230 wrote:
         | There is a famous video from Hitachi[^1] showing the Youngs
         | double slit experiment[^2]. You may like it!
         | 
         | [1]: https://www.youtube.com/watch?v=PanqoHa_B6c
         | 
         | [2]: https://en.wikipedia.org/wiki/Double-slit_experiment
        
         | pkaye wrote:
         | You can see a simulation of an 6502 CPU running here.
         | 
         | http://www.visual6502.org/JSSim/expert.html
         | 
         | Also this person has some simple tutorial on how a toy CPUs
         | work. He even made a simulator so you can make your own using
         | gate logic.
         | 
         | https://www.youtube.com/watch?v=QZwneRb-zqA&list=PLFt_AvWsXl...
        
         | devilbunny wrote:
         | Some of the later steps may be beyond them, depending on the
         | child's age, but Charles Petzold's wonderful book _Code_
         | explains how you could set up a processor from logic gates with
         | only a little handwaving.
         | 
         | If they are a little older and interested, NAND 2 Tetris is a
         | good self-paced course.
        
       | ge96 wrote:
       | Would be interesting if all of a sudden, a paradigm shift, all he
       | chips are designed as concentric rings for some reason
        
         | taraindara wrote:
         | Imagine someone quietly developing, and then suddenly releasing
         | a modern chip with this kind of design in mind. They'd get a
         | lot of publicity just for being different. And the process
         | would be their own, so if anyone likes it, or there turns out
         | to be any benefits to the design, it would be a while until
         | someone replicates it.
        
         | anyfoo wrote:
         | My feeling is that this would make bin packing on a silicon
         | wafer less optimal?
        
       ___________________________________________________________________
       (page generated 2025-10-22 23:01 UTC)