5 ms·

Just rebranding or remarketing something isn't the same as actual innovation. The home computers of the 1980s, such as the C64, were fun in part because they r

by motters 16y ago

Just rebranding or remarketing something isn't the same as actual innovation. The home computers of the 1980s, such as the C64, were fun in part because they represented something genuinely new and innovative in technology. Taking some PC hardware and slapping a Commodore sticker onto it isn't quite the same.

rbanffy 16y ago
In 1982, you could build a computer with 4-bit color at 320x200 graphics, a 3 voice synthesizer, without any software to run on it at the right price and have a huge hit. You just can't do it now. If you build a computer that uses anything but an x86 processor, you won't be able to sell it to people who like Windows. You could go for 1080p screens, but many high-end home PCs surpass that. You could run a complete Linux desktop, but that would face some user resistance. It kills me but there seems to be no space for innovation in the desktop PC market.
- anigbrowl 16y ago
  Perhaps it's that barriers to entry on the net are so high. I think you could usefully sell a $30 computer (simple to program, cheap to make, flexible enough to be useful) in places like Africa etc to people who could not afford to have one otherwise but even a basic web browsing experience needs a lot of CPU cycles nowadays. Of course there are text-based browsers, but you might as well go with the $100 price point like OLPC and just look for subsidies. I have heard nothing about how programmable those are, for young hackers in poor countries, I hope there's access to python, js, or similar. Closer to home, I would have quite liked to use the Playstation 3 as a computer, to the point that I learned quite a lot about the Cell chip inside it and started thinking about Linux distros again. But Sony made it non-expandable (boo), with no access to the GPU (WTF), were late with driver support (argh) and finally ditched the OS loader option entirely. They briefly hinted at giving away their 3d modeling engine but changed their minds about that too. Now, I know they recoup the hardware development costs from royalties on the games and have an investment to protect, but with a little more patience, support, and faith in their own engineering instincts they would have been rewarded with PS3s popping up up at every science fair. People are nostalgic about their 80s computers out of all proportion to their quality because they were easy to do things with if you wanted. I do think there's still room for lightweight and hackable products. One which missed the mark by going down the x86 route and overloading price and spec was the eee keyboard PC (http://www.tomshardware.com/news/eee-eee-keyboard-eee-pc-wireless-hdmi-atom,10416.html http://www.tomshardware.com/news/eee-eee-keyboard-eee-pc-wir...). Imagine if it had been $99-200 and run android or so!
  - rbanffy 16y ago
    > I have heard nothing about how programmable those are They are excellent. They come with a Red Hat OS, with Python, Smalltalk and lots of goodies. > But Sony made it non-expandable (boo) They needed to do it because they sell (sold) it at a loss. They want you to buy games, not to browse the web or crunch numbers. I would pay for an expandable, non-subsidized PS3 (extra geek points if they resurrect the News name). I bet it wouldn't be that expensive now. And I wouldn't ask for PS3 compatibility.
    - anigbrowl 16y ago
      If you can make a credible application (eg academic or sw publishing involvement) to http://develop.scee.net/ http://develop.scee.net/ my understanding is that a test unit + SDK is under $2k. It won't play commercially available games/DVDs/blu-rays or dump the hypervisor, but I doubt that's why you want it. Twice the super-fast RAM (or maybe expansion slots for same), RSX access & Linux-compatibility. As it seems quite secure there is a limited trade in used units, which I've seen for under $1000. If you can live without access to the graphics core, older Linux-compatible models can be had cheap on eBay, like $3-400. Sadly the Cell doesn't seem to have much future outside of industrial-scale computing.
- Someone 16y ago
  That 'huge hit' of 1982 would be a Microsoft Kin today. Also, only a few 'you' managed to make a big hit. It wasn't that easy. In fact, many winners of that time already had some other product (Commodore, Atari) to build upon. Today, I think the main barrier is economies of scale; it takes a much larger investment to enter the market at a competitive price point (for example, Apple sells a petabyte of flash memory on the opening week of a new iPhone launch, and probably even an order of magnitude more.) because of this innovation almost has to happen in markets where supply is (still) so limited that the big boys do not want to compete (current examples include color e-ink and roll-up displays)
  - rbanffy 16y ago
    The huge hit sold a couple million units. That's about a million times more than the Kin. The C64 was a hit because if offered much more than its competition did at the time at a lower price. 64K, 200x320 and 3-voice was pretty high-end 8-bit system by then. But I agree - to have a monster hit these days you have to move a couple million boxes a month. That's seriously hard.

by motters 16y ago

rbanffy 16y ago

In 1982, you could build a computer with 4-bit color at 320x200 graphics, a 3 voice synthesizer, without any software to run on it at the right price and have a huge hit. You just can't do it now. If you build a computer that uses anything but an x86 processor, you won't be able to sell it to people who like Windows. You could go for 1080p screens, but many high-end home PCs surpass that. You could run a complete Linux desktop, but that would face some user resistance. It kills me but there seems to be no space for innovation in the desktop PC market.

anigbrowl 16y ago
Perhaps it's that barriers to entry on the net are so high. I think you could usefully sell a $30 computer (simple to program, cheap to make, flexible enough to be useful) in places like Africa etc to people who could not afford to have one otherwise but even a basic web browsing experience needs a lot of CPU cycles nowadays. Of course there are text-based browsers, but you might as well go with the $100 price point like OLPC and just look for subsidies. I have heard nothing about how programmable those are, for young hackers in poor countries, I hope there's access to python, js, or similar. Closer to home, I would have quite liked to use the Playstation 3 as a computer, to the point that I learned quite a lot about the Cell chip inside it and started thinking about Linux distros again. But Sony made it non-expandable (boo), with no access to the GPU (WTF), were late with driver support (argh) and finally ditched the OS loader option entirely. They briefly hinted at giving away their 3d modeling engine but changed their minds about that too. Now, I know they recoup the hardware development costs from royalties on the games and have an investment to protect, but with a little more patience, support, and faith in their own engineering instincts they would have been rewarded with PS3s popping up up at every science fair. People are nostalgic about their 80s computers out of all proportion to their quality because they were easy to do things with if you wanted. I do think there's still room for lightweight and hackable products. One which missed the mark by going down the x86 route and overloading price and spec was the eee keyboard PC (http://www.tomshardware.com/news/eee-eee-keyboard-eee-pc-wireless-hdmi-atom,10416.html http://www.tomshardware.com/news/eee-eee-keyboard-eee-pc-wir...). Imagine if it had been $99-200 and run android or so!
- rbanffy 16y ago
  > I have heard nothing about how programmable those are They are excellent. They come with a Red Hat OS, with Python, Smalltalk and lots of goodies. > But Sony made it non-expandable (boo) They needed to do it because they sell (sold) it at a loss. They want you to buy games, not to browse the web or crunch numbers. I would pay for an expandable, non-subsidized PS3 (extra geek points if they resurrect the News name). I bet it wouldn't be that expensive now. And I wouldn't ask for PS3 compatibility.
  - anigbrowl 16y ago
    If you can make a credible application (eg academic or sw publishing involvement) to http://develop.scee.net/ http://develop.scee.net/ my understanding is that a test unit + SDK is under $2k. It won't play commercially available games/DVDs/blu-rays or dump the hypervisor, but I doubt that's why you want it. Twice the super-fast RAM (or maybe expansion slots for same), RSX access & Linux-compatibility. As it seems quite secure there is a limited trade in used units, which I've seen for under $1000. If you can live without access to the graphics core, older Linux-compatible models can be had cheap on eBay, like $3-400. Sadly the Cell doesn't seem to have much future outside of industrial-scale computing.
Someone 16y ago
That 'huge hit' of 1982 would be a Microsoft Kin today. Also, only a few 'you' managed to make a big hit. It wasn't that easy. In fact, many winners of that time already had some other product (Commodore, Atari) to build upon. Today, I think the main barrier is economies of scale; it takes a much larger investment to enter the market at a competitive price point (for example, Apple sells a petabyte of flash memory on the opening week of a new iPhone launch, and probably even an order of magnitude more.) because of this innovation almost has to happen in markets where supply is (still) so limited that the big boys do not want to compete (current examples include color e-ink and roll-up displays)
- rbanffy 16y ago
  The huge hit sold a couple million units. That's about a million times more than the Kin. The C64 was a hit because if offered much more than its competition did at the time at a lower price. 64K, 200x320 and 3-voice was pretty high-end 8-bit system by then. But I agree - to have a monster hit these days you have to move a couple million boxes a month. That's seriously hard.