When we learn about the internals of a microprocessor, we are shown a diagram that resembles the 8-bit devices of the 1970s. There will be an ALU, a program counter, a set of registers, and address and data line decoders. Most of us never go significantly further into the nuances of more modern processors because there is no need. All a processor needs to be is a black box, unless it has particularly sparked your interest or you are working in bare-metal assembly language.
We imagine our simple microprocessor as built from logic gates, and indeed there have been many projects on these pages that create working processors from piles of 74 series chips. But just occasionally a project comes along that reminds us there is more than one way to build a computer, and our subject today is just such a moment. [Olivier Bailleux] has created his “Gray-1”, a processor whose only active components are memory chips, both ROM and RAM.
The clever part comes with the descriptions of how the ROMs are used to recreate the different functions of the processor, through careful programming. Some functions such as registers for example use loops, in which some of the address lines are driven from the data lines to maintain the ROM at a set location. The name of the computer comes from its program counter, which counts in Gray code.
The full processor implements a RISC architecture, and there is a simulator to allow code development without a physical unit. The write-up is both comprehensive and accessible, and makes a fascinating read.
It’s safe to say this is the only processor we’ve seen with this novel approach to architecture. Some more conventional previous features though have been an effort to create a processor entirely from NAND gates, and another made from 74 logic.
As it turns out, the answer is not 42, it’s 42.3 — thousand. That’s how many discrete transistors spread across the 30 m2 room housing this massive computation machine. [James Newman’s] Megaprocessor, a seriously enlarged version of a microprocessor, is a project we’ve been following with awe as it took shape over the last couple of years.
[James] documented his work in great detail, and by doing so, took us on a journey through the inner workings of microprocessors. His monumental machine is now finished, and it’s the ultimate answer to how a processor – and pretty much everything that contains a processor – works.
Continue reading “42,300 Transistor Megaprocessor Is Complete”
I made a bee line for one booth in particular at this year’s Bay Area Maker Faire; our friend [Eric Schlaepfer] had his MOnSter 6502 on display. If you missed it last week, the unveiling of a 6502 built from discrete transistors lit the Internet afire. At that point, the board was not fully operational but [Eric’s] perseverance paid off because it had no problem whatsoever blinking out verification code at his booth.
I interviewed [Eric] in the video below about the design process. It’s not surprising to hear that he was initially trying to prove that this couldn’t be done. Unable to do so, there was nothing left to do but devote almost six-months of his free time to completing the design, layout, and assembly.
What I’m most impressed about (besides just pulling it off in the first place) is the level of perfection [Eric] achieved in his design. He has virtually no errors whatsoever. In the video you’ll hear him discuss an issue with pull-up/pull-down components which did smoke some of the transistors. The solution is an in-line resistor on each of the replacement transistors. This was difficult to photograph but you can make out the soldering trick above where the 3-pin MOSFET is propped up with it’s pair of legs on the board, and the single leg in the air. The added resistor to fix the issue connects that airborne leg to its PCB pad. Other than this, there was no other routing to correct. Incredible.
The huge schematic binder includes a centerfold — literally. One of the most difficult pieces of the puzzle was working out the decode ROM. What folds out of this binder doesn’t even look like a schematic at first glance, but take a closer look (warning, 8 MB image). Every component in that grid was placed manually.
I had been expecting to see some tube-based goodness from [Eric] this year. That’s because I loved his work on Flappy Bird on a green CRT in 2014, and Battlezone on a tube with a hand-wound yoke last year. But I’m glad he stepped away from the tubes and created this marvelous specimen of engineering.
Moore’s Law states the number of transistors on an integrated circuit will double about every two years. This law, coined by Intel and Fairchild founder [Gordon Moore] has been a truism since it’s introduction in 1965. Since the introduction of the Intel 4004 in 1971, to the Pentiums of 1993, and the Skylake processors introduced last month, the law has mostly held true.
The law, however, promises exponential growth in linear time. This is a promise that is ultimately unsustainable. This is not an article that considers the future roadblocks that will end [Moore]’s observation, but an article that says the expectations of Moore’s Law have already ended. It ended quietly, sometime around 2005, and we will never again see the time when transistor density, or faster processors, more capable graphics cards, and higher density memories will double in capability biannually.
Continue reading “Exponential Growth In Linear Time: The End Of Moore’s Law”
[Josh] hit the same issue we’ve faced before: cable modems don’t match a form factor and usually don’t make themselves easy to mount on something. We could complain about routers as well, but at least most of those have keyhole slots so you can hang them on some screws. Inspiration struck and he fabricated his own rack-mount adapter for it. Velcro holds it in place, with a cutout bezel to see the status lights and an added fan to keep things cool.
Here’s a pair of strange but possibly interesting ones that were sent in separately. The first is an analysis of how much energy short-run CNC prototyping consumes versus traditional manufacturing. The other is an article that [Liz] wrote about getting started with CNC mill bits. She says she compiled all that she learned as she was getting started in the field and wants to save others the effort.
This one goes back several years, but who doesn’t love to hear about a voice-controlled wheelchair?
So you can solder QFN parts but you can’t hammer a nail straight into a piece of wood? The answer, friend, is a laser guided hammer. Someone hire this [Andybot] person, because the solution to the problem shows the ability to out-think an interesting dilemma: how do you put a laser in a hammer head and still use it to hit things?
We’ve seen a lot of these long-range WiFi hacks over the years. This one is worth looking at because of the work done to create an outdoor mount that will stand the test of time.
And finally, we’re still really fond of this 2-bit paper processor that helps you wrap your brain around what’s going on with those silicon wafers that rule our everyday lives. [glomCo] liked it as well, and actually coded an emulator so that you can play with it without printing anything out on paper. We think it takes away some of the fun, but what an excellent programming exercise!
This is a microscopic photograph of an 8085 processor die. [Ken Shirriff] uses the image in his explanation of how the ALU works. It is only capable of five basic operations: ADD, OR, XOR, AND, and SHIFT-RIGHT. [Ken] mentions that the lack of SHIFT-LEFT is made up for by adding the number to itself which has the effect of multiplying a number by two; the same mathematical function performed by a shift operation.
His post details the gate arrangement for each ALU operation. This is clear and easy to follow, and was based on reverse engineering work already done by a team who meticulously decapped and photographed the dies.
Not long ago this explanation would have been voodoo to us. But we worked our way through The Elements of Computing Systems text-book by following the online Nand to Tetris course. It really demystifies the inner working of a chip like the 8085.
Now if you really want to understand this ALU you’ll build it for yourself inside of Minecraft.
Personally we find this Ball-Grid Array chip-swap rather horrifying. But if you want to beef up the processor on your 701 Eee PC this is what you’ll need to go through. Not only did [Red Fathom] upgrade to a 1.6 GHz chip, but he managed to get the computer to boot up with the new hardware in place.
BGAs are notoriously hard to solder. This hack pulls it off using just a hot air gun. [Red Fathom] heats the board from the underside until the solder melts and he can pluck off the old chip. He then uses a solder braid and iron to remove extra solder from the footprint. After a little cleanup with a cotton swab and some flux he plops in an Intel Pentium M LV 778. It doesn’t look like he added any solder after the cleaning process. Perhaps he’s relying on the small amount left on the tinned pads of the board?
After the break you can see the soldering process and a video of the new processor booting Xandros.
Continue reading “Swapping out Eee PC BGA chip for 1.6 GHz upgrade”