The Untold Story of the Invention of the Game Cartridge 60
harrymcc writes In 1973, an obscure company which had been making electronic cash registers looked for a new business opportunity. It ended up inventing the game cartridge--an innovation which kickstarted a billion-dollar industry and helped establish videogames as a creative medium. The story has never been told until now, but over at Fast Company, Benj Edwards chronicles the fascinating tale, based on interviews with the engineers responsible for the feat back in the mid-1970s.
Re: (Score:2)
Has anyone actually successfully used vinyl to store data?
What else would you possibly store on vinyl?
Re: (Score:2)
What else would you possibly store on vinyl?
A cold glass of water? You wouldn't want to warp the finish on your coffee table, would you?
Selectavision (Score:4, Informative)
What else would you possibly store on vinyl?
Superior analog video, of course:
http://en.wikipedia.org/wiki/C... [wikipedia.org]
https://www.youtube.com/watch?... [youtube.com]
Re: (Score:2)
I actually have one of those. My grandma found it in her sister's attic when cleaning the place out and figured I'd like it. I thought it was a laserdisc player when she described it to me, so I was really surprised to find that it was actually a vinyl-based video format.
Video quality isn't much worse than '80s VHS.
Re:Obligatory Joke (Score:4, Funny)
We'd probably have records with twenty parallel grooves and reading heads with 40 or more needles.
Re: (Score:3)
We'd probably have records with twenty parallel grooves and reading heads with 40 or more needles.
I'm an old fart. I worked on a Burroughs B3500 back in the 70s. The "Disk Module" was a 10MB cabinet about 4 ft by 4ft by 2ft. It had 4 platters and there was one head per track. Seek time was non-existent but rotational delay was a bitch.
Re:Obligatory Joke (Score:5, Informative)
Yes, actually...
"In 1986, the new BASICODE 3 standard was developed. The most important additions were routines for simple monochrome graphics, reading and writing data from within programs and sound output. BASICODE 3 made BASICODE popular in the computer scene of the GDR, and from 1989 onward BASICODE programs were transmitted via radio throughout the GDR. Also, a book was published which included a vinyl record with Bascoders for all computers common in the GDR. The last revision of BASICODE, which featured color graphics, was released as BASICODE 3C in 1991."
http://en.wikipedia.org/wiki/B... [wikipedia.org]
In the UK during the early 1980's, pop star Chris Sievey released a 7" single record where side B was the program code in audio format for the Sinclair ZX81 microcomputer. You plugged in your turntable's output into the ZX-81 "loaded" the record into memory, flipped over the record, played the music on Side A while running the program which gave you a "music video" while the song played. It was very innovative at the time:
https://www.youtube.com/watch?... [youtube.com]
Re: (Score:2)
http://blog.modernmechanix.com... [modernmechanix.com]
It was a pretty good idea, considering that software was loaded from audio cassettes back in the day, and unlike a cassette, a flexible record can easily be distributed in an unpackaged magazine.
And we are back to them again... (Score:3)
USB Flash Drives are Cartridges reborn.
Re:And we are back to them again... (Score:5, Informative)
Not exactly. A flashdrive is a serial bus peripheral that communicates using an encoded format.
A cartridge is typically little more than an exposed parallel logic bus that directly maps to the console's main memory. (Atari 2600 carts were literally a ROM chip, with its leads exposed on the card edge. Nothing else. http://www.hardwaresecrets.com... [hardwaresecrets.com] )
NES and SNES carts had special purpose chips on them to help page the address of the ROM inside the main CPU's memory space, and even a few special co-processors on occasion, but were still just a direct tie-in to the memory bus.
A DIMM and a cartridge have more in common than a USB stick does with a cartridge.
Re: (Score:2)
And apparently the site I linked does not like hard links. Meh. Whatever. Just image search for "2600 cartridge interior", you will find it.
Re: (Score:2)
You CRIMINAL!
But actually you could just copy+paste the link to your web browser, defeating the site's referrer check.
Re: (Score:2)
Re: (Score:2)
Strange; seems to work for me.
Re: (Score:2)
You CRIMINAL!
But actually you could just copy+paste the link to your web browser, defeating the site's referrer check.
With the new firefox <meta name="referrer"> tag, criminal sites like /. will soon hotlink images anywhere without fear of consequences!
https://blog.mozilla.org/secur... [mozilla.org]
Re: (Score:1)
And apparently the site I linked does not like hard links. Meh. Whatever. Just image search for "2600 cartridge interior", you will find it.
Here's a link to one of their articles which shows the contents of a cartridge [hardwaresecrets.com]. Hopefully that'll work?
Re:And we are back to them again... (Score:4, Insightful)
Yes a USB Flash Drive is more complex than a Cartridge, being that it is going threw a Universal Bus and is Read/Write (However Zelda use to be able to save data).
But compared to other forms of storage after cartridges downfall after the Nintendo 64. The USB Flash brings up many of the core Ideas of the Cartridge. A device that has no moving parts that is made to be constantly inserted and removed.
The Ford Model T is also very quite different than a Ford Mustang. But still they are classified as cars.
Re:And we are back to them again... (Score:4, Insightful)
Clearly, you do not understand the implications of what I said.
USB storage requires an entire operating system to even access. It requires a kernel OS process, a driver for the specific USB chipset attached to the system, and yet another driver for the filesystem stored on the device. In addition to this, it has to have multitasking capabilities to switch between reading and decoding the data on the device, and executing the read program code.
In stark contrast, a game cartridge requires no OS at all. The CPU simply jumps to the address location where the cartridge's ROM chip is logically assigned, and the CPU treats it exactly like it was fetching from RAM. The game is the ONLY thing running.
The better comparison would be "USB is the new data cassette!"
You can put all kinds of crazy things inside a cartridge that you simply could never put inside a USB stick. Things like additional RAM that runs at full speed, Whole new CPUs, an additional peripheral bus connector-- you name it.
In theory, a (crazy) person could take an old SNES console, and with a very purpose built cartridge, turn it into a playstation.
Really, the comparison you are drawing does not really reflect what a cartridge actually is.
That's why the Nintendo PlayStation died (Score:3)
In stark contrast, a game cartridge requires no OS at all. The CPU simply jumps to the address location where the cartridge's ROM chip is logically assigned, and the CPU treats it exactly like it was fetching from RAM.
That was true until roughly the Nintendo DS. It uses a seek and read protocol, similar to parallel ATA or SD, to read 512-byte blocks of game data from the Game Card. Even the N64 and GBA used a seek and read protocol, but the console's memory controller abstracted it away from the CPU with wait states.
In theory, a (crazy) person could take an old SNES console, and with a very purpose built cartridge, turn it into a playstation.
I don't see how. There isn't enough bandwidth over the Super NES cartridge bus to refill video memory every frame. That's part of why the agreement between Nintendo and Sony to make the Super NES CD periphera
Re: (Score:2)
I said "in theory". The devil's in the details.
still, I wonder how far one could get coopting both the EXT connector on the bottom, and the cartridge slot. Use the EXT connecter for IO (including video and SPU channel) and use the cartridge slot for the snes-side program code, ram access, and return communications. A hardware mpeg decoder in the cartridge slot, with the mpeg stream coming in on the ext port or something maybe. Given that the screen resolution isnt so big to begin with, a few hundred kbps w
Re: (Score:2)
I wonder how far one could get coopting both the EXT connector on the bottom, and the cartridge slot.
Copy bandwidth between the A bus (main memory and the cartridge) and the the B bus (the PPU and the edge connector) is 2.7 MB/s, but then you have to copy it back out to video memory. You could put the copy source on the cart A bus, but then you still get only 2.7 MB/s to fill video memory. You can write to memory only during vertical blanking, which is about 6 KiB worth of copying. The Super Game Boy accessory got away with it because a 160x144 pixel frame at 2bpp fits into 6 KiB.
in theory, one could do this on a cartridge, and never hope to do so over USB.
My USB TV tuner begs to di
Re: (Score:2)
Re: (Score:1)
PS1 beat Saturn because Sony had no 32X (Score:2)
You'd put the A/V out on the cartridge containing the PlayStation
At that point, you're just using the Super NES as a power supply and controller reader.
Maybe you could even go all 32X
Better yet, go all Saturn, dropping the old hardware entirely because an S-Video output and SPI inputs are probably cheaper than all that circuitry to interface with the legacy console. That's what led Sony to success with PlayStation: it skipped the 32X stage.
Re: (Score:2)
That's because the Nintendo DS doesn't use cartridges, it's more like a memory card...
Re: (Score:2)
I don't see how. There isn't enough bandwidth over the Super NES cartridge bus to refill video memory every frame.
There is, but not quite at the full frame size. This is why starfox has borders, it actually does this.
Another cartridge that did this was the Super Gameboy. There is a literal entire gameboy chipset in there, and it is copying the video output to the snes in realtime (again, not the full screen though, there's a border)
Re: (Score:2)
Which are all found on any modern game console. Nintendo have been using SD cards in a modified case for years now in the DS.
Re: (Score:2)
With large volume flash storage becoming cheap, far in excess of that of DVD and even bluray, why not use them as game storage.
Because mass produced pressed discs are still cheaper to manufacture than flash storage. a 32GB SD card costs 20 bucks, a 50GB blu-ray RW disc is 3.50
Games could be saved onto the flash drive, the hard drive or both,
Where have you been the past 9 years. You can, if you want, go digital and install games from PSN to a PS3 hard drive and never bother with a disc. On the PS4, games are installed to the hard drive, even if you use the disc. (if you buy the disc version the game will do a disc check on start but that's all you need it for)
The PS vita, uses a combination of
Re: (Score:1)
USB storage requires an entire operating system to even access.
Nope. Computers can boot off USB. But of course the technicalities of how this is done are different.
Re: (Score:2)
What does this mean? Do you mean basically bank-switching ROMs? (Very very very very rough analogy -- manual VM.) That way, you could get more than 64K of ROM on a NES..
If so, not sure what you mean by it on the SNES though. The 65816 has a 16 meg address space.. None of the SNES cartridges were anywhere near that, were they?
Re: (Score:2)
yeah it basically just means bank switching roms.
emulators(wildcarts have lists of what styles they support.
dunno what's the deal with snes carts. maybe they saved on pins/traces.
Re: (Score:1)
Re: (Score:2)
Re: (Score:3)
MOS technology built most of the cartridge ROMS (Score:5, Informative)
MOS Technology did a big business in manufacturing ROMs for Atari's cartridges (both the 2600 and 400/800 /XE line). They also made the 6502 variants used in the 2600 and 400/800/XE and in Atari's main competitors (Apple and Commodore).
Eventually MOS was purchased by Commodore and stopped making ROMs but cranked out the 6502s and SID chips.
For some reason they never got around to making a followup to the 6502 and let the next generation business go to Motorola. Greed does that.
Another fun fact: The original VCS games were programmed on a PDP-11 using a cross-assembler (!) and soon enough Atari upgraded to a VAX. When a game was finished they sent program tape to MOS who made the metal mask. The ROMs were pre-processed up to the metal deposition step. Then the final metal pattern was defined by whatever program was being written to ROM. This is one reason how MOS made them so cheaply: they mass produced ROM blanks and then programmed them with a single mask. I talked with an Atari old-timer about the process a couple of years ago. Great stories.
Re: (Score:3)
Re: (Score:2)
I hate to jump on you for this because you made an excellently informative post but what exactly did greed do? Greed seems to be a buzzword for any failure in business today, at least to people who prattle on about how anti-establishment they are while taking every advantage of corporate output.
Jack Tramiel (the owner of Commodore) sucked as much cash out of the business while not providing even a shoestring business to develop more advanced integrated circuits. They limped along for a while based on the heroic efforts of a few creative engineers but Commodore Semiconductor Group (CSG) collapsed because Jack Tramiel treated CSG like his own personal bank account.
Once he left to buy Atari in 1984 CSG was too far behind in the microprocessor wars to field another chip.
So, in this case it was greed.
Re: (Score:2)
They caught up a little by purchasing Amiga Corporation in 1984, but those used Motorola CPUs along with a family of custom integrated circuits that were developed well into the 1990s. Amiga was partially financed by Atari until Tramiel bought Atari.
Better than replacing socketed UV-EPROM chips! (Score:2, Interesting)
I worked with lots of embedded systems in the '80's that stored their program on banks of UV-EPROM chips that we'd have to upgrade using a a chip-puller (screwdriver) and a chip eraser and burner.
The sockets would eventually fail, requiring repair/replacement of the entire CPU board (about a square foot in size).
When we saw our first game system with replaceable ROM cartridges, there was much forehead slapping, since we were already using board-edge connectors (on a bus very similar to the S-100).
Re: (Score:2)
Were those sockets made by Texas Instruments? They made edge wipe sockets which failed over time whether the chips were regularly replaced or not.
We had a Channel F! (Score:2)
As a kid my family had a Channel F and a boatload of cartridges for it -- plus 2 built-in games! The last one I remember getting was "Galactic Space Wars", and then the Atari 2600 showed up thus relegating the Channel F to the closet.
The graphics weren't great, and the games were something that a beginner could code up on an Atari 800 of the same era in BASIC. But they were fun enough. The Channel F did have a really unusual controller: the joystick could be moved about in a normal axis (up,down,left,rig
Re: (Score:1)
We had one too.
Imagine playing a video game via a Bop-It(tm); that will give you the proper experience.
Thankfully the Atari 2600 and VIC-20 arrived soon thereafter #andtherestishistory
Hacking the Atari (Score:4, Interesting)
Re: (Score:1)
Yes, the Channel F was important, and its cartridges made the Channel F the first programmable game console. No need to paint it as way more innovative than it was. The good people at Magnavox (no, not Baer personally) invented the idea that you could plug a card into your game console and it would change which game you are playing.
So the distinction is that the Channel F's games were in the cartridge and the Odyssey's were not? Were these jumpers used to select a game built in, or were they changing the parameters of operation in effect to create each game? If the latter is true, then the Odyssey cartridge would in fact contain game programming. It would also seem there would be a limit to how many such unique games could be made.