News for hackers. Stuff that matters.
Re:Fifty or Sixty Years of Processor Development for This? by nootropicat on Wednesday April 04, 2018 @03:03AM (#16752629)
Imagine asics for everything. I mean everything, like implementing a javascript engine directly. The energy efficiency could go up 2-3 orders of magnitude (looking at the difference in bitcoin mining between gpus and asics).

Old gaming consoles had cartridges (with memory); I can imagine a future in which complex software is transported in the same manner, except cartridges contain specialized asics. Or perhaps a step forward - a chip making device in every home, an equivalent of sorts of burning music to cd.

Re:Fifty or Sixty Years of Processor Development for This? by _chris_ on Wednesday April 04, 2018 @03:25AM (#16752726)
I assume you meant to reference this, but just to make it clear, those old gaming console cartridges sometimes DID have special chips in them. Much to the horror of emulator writers a decade later, hehe.

Wiki: https://en.wikipedia.org/wiki/List_of_Super_NES_enhancement_...

Re:Fifty or Sixty Years of Processor Development for This? by Tobba_ on Wednesday April 04, 2018 @05:12AM (#16753195)
Personally I'd predict the opposite (and current) pattern, creating generic chips which can replace many ASICs / less generic chips. If you can produce a chip which can replace 10 other low/medium-volume designs, economical scaling will win out as long as you're not adding too much overhead. This has been driving FPGAs forward for quite some time, although they're inherently pretty inefficient in terms of die size. Plus, it also provides some logistical advantages in terms of supply chain fragility.

See also: how ridiculously cheap microcontrollers have gotten, and the current messy DRAM pricing (high-capacity chips used in phones sometimes ending up cheaper than low-performance/capacity chips).

Re:Fifty or Sixty Years of Processor Development for This? by majewsky on Wednesday April 04, 2018 @04:45AM (#16753085)
> Old gaming consoles had cartridges (with memory); I can imagine a future in which complex software is transported in the same manner, except cartridges contain specialized asics.

I think if that scenario happens, the user won't notice because those ASICs will be deployed in "the cloud".

Re:Fifty or Sixty Years of Processor Development for This? by iainmerrick on Wednesday April 04, 2018 @03:13AM (#16752669)
I mean everything, like implementing a javascript engine directly.

In that case, rather than a Javascript engine, wouldn't you have an ASIC for the script itself?

Re:Fifty or Sixty Years of Processor Development for This? by yjftsjthsd-h on Wednesday April 04, 2018 @08:18AM (#16754000)
You wouldn't want a new chip per website. Maybe hardware handling of a standard library, though.
Re:Fifty or Sixty Years of Processor Development for This? by AstralStorm on Wednesday April 04, 2018 @12:10PM (#16756096)
Or even just of the expensive operations like synchronization, context switches, DMA to cache from network interface...

Wait. We have most of that already. :)

Re:Fifty or Sixty Years of Processor Development for This? by hyperpallium on Wednesday April 04, 2018 @04:04AM (#16752919)
Energy efficiency is important (in a server/bitminer and in mobile), but do you have figures for (1) performance; (2) cost?

Custom asics would be expensive now, but perhaps... exponential... cost decrease with sufficient demand. A related approach is programmable asics. I've heard this was researched decades ago, but (I presume), silicon was so cheap, so to speak, it wasn't worth it then.

Today, with "peak silicon", perhaps all these discarded tailings will be picked over. To mix mining metaphors.

Re:Fifty or Sixty Years of Processor Development for This? by nootropicat on Wednesday April 04, 2018 @04:48AM (#16753096)
The only relevant figures I know of are from mining bitcoin. [0].

>A related approach is programmable asics. I've heard this was researched decades ago

What do you mean, if not fpgas?

[0] https://en.bitcoin.it/wiki/Mining_hardware_comparison https://en.bitcoin.it/wiki/Non-specialized_hardware_comparis...

Re:Fifty or Sixty Years of Processor Development for This? by hyperpallium on Wednesday April 04, 2018 @06:02AM (#16753378)
thanks.

wait, we're looking at 18,000,000 Mhash/s vs 2568 Mhash/s Like... 4 orders of magnitude?

And, 11,000 Mhash/J vs 3 Mhash/J, over 3 ordera of mag, just like you said.

yah, fpgas