Join Hackaday Editor-in-Chief Elliot Williams and Staff Writer Dan Maloney as they review the top hacks for the week. It was a real retro-fest this time, with a C64 built from (mostly) new parts, an Altoids Altair, and learning FPGAs via classic video games. We also looked at LCD sniffing to capture data from old devices, reimagined the resistor color code, revisited the magic of Polaroid instant cameras, and took a trip down television’s memory lane. But it wasn’t all old stuff — there’s flat-packing a sphere with math, spraying a fine finish on 3D printed parts, a DRM-free label printer, and a look at what’s inside that smartphone in your pocket — including some really weird optics.
Check out the links below if you want to follow along, and as always, tell us what you think about this episode in the comments below!
Episode 167 Show Notes:
News This Week:
- We couldn’t come up with anything. But at times like these, no news is good news!
What’s that Sound?
- Have a listen to the secret sound in this week’s episode. If you can figure out what device it comes from, or even if you just want to take a wild guess, enter here!
Interesting Hacks of the Week:
- Will MiSTer Fool You Into Learning FPGAs?
- A (Nearly All) New Commodore 64
- This Spherical Lamp’s Pieces Ship Flat, Thanks To Math
- Keenan Crane’s surface cutting paper. (And the rest of his research.)
- All The Sticky Labels You Could Ever Need: No DRM, Just Masking Tape
- Theory, Practice, And Ducted Fans
- Retrotechtacular: How Television Worked In The 1950s
- Reddy Kilowatt, the unofficial “Retrotechtacular” mascot
Quick Hacks:
-
- Elliot’s Picks:
- Dan’s Picks:
That Polaroid article is from 2010 ? Or do you have a visual transcript of something new in the current podcast, for the hearing impaired?
The 2010 article is just further reference for last weeks article right above it.
Dan Maloney’s allergy to FPGAs on the C64 was interesting. I understand where he’s coming from. Because you cannot see the physical logical connectivity whereas old design syou can easily relate this to a schematic he feels more comfortable.
For me coming from the PCB design land for real world items where i laid out large logic TTL and CMOS designs (late 1980s’) this was becoming more and more painful. When we came to re-spin or revise for new products and add features or fix things a new PCB layout was required, tested and verified. We also had all of the associated signal integrity problems large logic interconnects (Ground bounce, overshoot, undershoot, terminations) all requiring ever more work. The PCB was also larger than we would like as products were trying to get smaller and old TTL and CMOS is just plain power hungry!
When i moved to designing Network cards (1990’s) from standard logic all of the logic IC’s got mopped up into multiple FPGA devices and eventually one single FPGA. That was the real beauty of FPGAs. Mop up all the logic you can at the time and leave the physical analog interface. Now even some the analog interfaces can be mopped up. Any consequential revisions or fixes were mostly on the FPGA as long as we got the surrounding analog support part correct. Yes there are still signal interconnect problems but they are better understood. The PCB’s then got ever smaller area and less power hungry.
I think you will find all retro devices in the future will have to be in some FPGA or be emulated in Microprocessor just from lack of old silicon and general production economics.
Maybe one day we will see someone say i have emulated my android phone on the latest ARM/Intel silicon chip and inside that i am playing space invaders on a C64 emulator. I will be allergic then…
I feel it’s missing the point of vintage computing to munge an entire vintage computer into an FPGA because running the original hardware is, for me at least, fun and part of the attraction is that I can get into the data and address bus between the chips and tinker with it (sure, yes I can do Verilog/VHDL and ‘analyse’ it but it’s not as much fun)
I foresee a widening market for replacement custom chips built around FPGA or CPLD chips and I don’t consider that to be much of a departure from the original designs as you could fit such ‘parts’ into an original board mixed in with whatever original parts you have.
The analogue stuff is a challenge though, for instance, as far as I know there’s no replacement for a 6581 SID chip that’s not a CPU emulatiing the function.
I do remember seeing programmable analogue arrays though, I wonder…
I think everyone gets into something like retrocomputing for different reasons, but I totally get the idea of the physicality, especially when you want to be tweaking around with the thiing.
Check out the Logic Noise series, where I’m following the now 50 year old tradition of messing around with logic chips to make square wavey synth noises on breadboards. Would the same be possible in FPGA fabric? Sure, but…
If you’ve got chips on a breadboard, rewiring things up on the fly is part of the fun — making discoveries. Making the same “re-wirings” in an FPGA would be a minute-long synthesize and flash cycle: no fun. Hands-on and immediate is essential for creativity and fortuitous discovery.
And going from the breadboard / circuit diagram to the Verilog in Mister Retro Wolf’s videos was really great for comprehension, if you understand the old logic chips. So something there was also key.
But would I rather lay out a computer system in Verilog than on a breadboard, if it’s a get-it-right-once-and-move-on type project? Probably. Fewer moving parts to mess up, quicker build.
I also think that the emulate-anything nature of the FPGA is relevant and interesting for retro-gaming / retro-computing. And I _do_ really like the way things like busses and interfaces are simply defined, and then that’s the implementation done. So for building a (retro)computer, FPGA all the way. For me. :)
(BTW: I might have misspoke in the video — the board that the Mister FPGA project is based on is a buy-it, closed-source dev kit. I found this out later as I looked deeper into the project. Minor bummer, IMO, but working a port to another platform would be doable if one were sufficiently motivated — there’s no shortage of cool and powerful open FPGA dev boards.)