'Pruned' Microchips Twice As Fast and Efficient 127
Zothecula writes "If you had to use a commuting bicycle in a race, you would probably set about removing the kickstand, fenders, racks and lights to make the thing as fast and efficient as possible. When engineers at Houston's Rice University are developing small, fast, energy-efficient chips for use in devices like hearing aids, it turns out they do pretty much the same thing. The removal of portions of circuits that aren't essential to the task at hand is known as 'probabilistic pruning,' and it results in chips that are twice as fast, use half the power, and are half the size of conventional chips."
Hm (Score:5, Insightful)
It's news that removing unnecessary parts of a circuit make it more efficient? Really?
Re:Hm (Score:5, Insightful)
I think the news is they developed a heuristic of least used parts of a chip, slapped on a tiny emulator so functions don't fail, and call it a day.
For example, Chip $foo has functions A B C D E & F. E is used on average once every gigaflop, so using the CPU/other functions, they implement E and cut out all parts for E.
Re: (Score:1)
It's called optimising and has been going on for decades (AFAIK for emulation of operations see x86 processors which emulate cisc instructions). Its just not worth the effort in most cases since producing optimised chips costs a lot more than simply using mass produced generic chips.
Re: (Score:3)
Hell, engineers have been doing that for at least 50 years. Big names are IBM with the S/360, DEC with the PDP-11 and VAX and Intel with the 8086
It was the only way they could make CISC w/o chips the size of Andre The Giant's thumb. Micro-ops made them obsolete in X86 CPUs in the late 1990s.
Re:Hm (Score:5, Funny)
The best part is that this can be applied iteratively. Once E is eliminated there's a new "least used" function which can be eliminated. By extension, any CPU can ultimately be pruned down to a single NOP instruction, with the entire rest of the instruction set emulated in software.
Re: (Score:1)
At least you could get down to the case of an infinite line of rocks, with "read rock", "write rock", "move left", and "move right" instructions. Everything else is just icing on top of that.
Re: (Score:3, Funny)
"By extension, any CPU can ultimately be pruned down to a single NOP instruction, with the entire rest of the instruction set emulated in software."
Replace the CPU with a solar cell.
Not only can it perform NOPs at the same rate as any pruned CPU, but it also runs at a negative current draw when you open the lid (or if you have enough illuminated 'bling' fans installed).
Re:Hm (Score:5, Interesting)
Re: (Score:1)
I know you were going for a joke, but that actually does work with "Subtract and branch if less than or equal to zero" and a couple other instructions. Obviously not practical, but see http://en.wikipedia.org/wiki/One_instruction_set_computer [wikipedia.org]
Re: (Score:2)
For example, Chip $foo has functions A B C D E & F. E is used on average once every gigaflop, so using the CPU/other functions, they implement E and cut out all parts for E.
The best part is that this can be applied iteratively. Once E is eliminated there's a new "least used" function which can be eliminated. By extension, any CPU can ultimately be pruned down to a single NOP instruction, with the entire rest of the instruction set emulated in software.
Not quite to THAT degree, but Mark Twain suggested something along those lines: http://www.plainlanguage.gov/examples/humor/marktwain.cfm [plainlanguage.gov] I'll leave it to the reader to provide a coding sample to implement this.
Re: (Score:2)
I know a few people on Facebook that have in fact started to adopt Twain's logical spelling system. Lolcats appear to be fans too.
RISC redeux (Score:2)
An analogy comes to mind: some kids buy a car and pull off pieces until it doesn't
Re: (Score:2)
Well RISC is still around in the form of ARM cpus...
Re: (Score:2)
Re: (Score:2)
Re: (Score:2)
I'd rather be a duct tape engineer than the guy complaining about them.
Re: (Score:2)
This seems relevant to Sony, who got bashed for removing the Emotion Engine [wikipedia.org]. Maybe, depending on how many PS2 games were played versus PS3 games, they were actually saving us an insignificant amount of money by removing a rarely used bit of circuitry.
Re: (Score:3, Informative)
I suppose everyone in the world is not already as intelligent and knowledgeable as you and might possibly find this interesting. Perhaps you could find it in your heart to allow the less intelligent and knowledgeable the opportunity to discover something like this so that then they will be as intelligent and knowledgeable as you. :-)
Re:Hm (Score:4, Informative)
I actually know next to nothing about electronic circuits. I think the most complex thing I ever built was a metronome with variable tempo.
I would expect that almost everyone here would know this kind of stuff already...
He's merely pointing out (Score:2)
Is that so wrong?
Re:Hm (Score:4, Interesting)
They aren't cutting out entire blocks of ASIC circuitry in a Boolean keep or remove decision. They are sacrificing precision by reducing transistor count, and doing so in a somewhat heuristic approach in order to limit the loss of precision. Their algorithm will explore the worst case and best case of each arithmetic operation in order to achieve this. Not too different from the MiniMax approach to playing Chess or other board games.
Re: (Score:3)
Re: (Score:2)
On the other hand, a bicycle analogy is green...and therefore much more politically correct than your carbon-spewing analogy...chuckle...
Re: (Score:2)
Been doing that since I bought my very first car - oh, wait. Slashdot. Car analogies. Crap. Here come the mods from hell . . . .
How about a pizza analogy. That slice isn't doing anything; yum.
Re: (Score:1)
The article is confusing, but it sounds like they're sacrificing correctness. They heuristically pick out circuits that give a fairly low error rate, cut it out, and just write the software to deal with the new errors that crop up.
Re: (Score:2)
Re: (Score:2)
Except they didn't just make a broad statement that such a thing should/could be possible. Any dummy could do that. They're actually *doing* it.
Re: (Score:1)
http://en.wikipedia.org/wiki/Reduced_instruction_set_computer#Early_RISC [wikipedia.org]
RISC was wrong (Score:3)
Yes, you are right. This is the spirit of RISC.
But RISC was wrong. RISC resulted from a study of what instruction were actually used by typical applications that were compiled with standard compilers. This is like studying what railroad tracks are used and concluding that rail travel would be optimum if certain tracks were eliminated and others improved. This conclusion is wrong, because it assumes that existing rails include all optimal paths. In actuality, there might be paths that do not currently have r
Re: (Score:1)
This is like studying what railroad tracks are used and concluding that rail travel would be optimum if certain tracks were eliminated and others improved.
And in case anyone thinks this sounds like a good idea, it's been tried [wikipedia.org], and essentially destroyed the network.
Re: (Score:3)
RISC resulted from a study of what instruction were actually used by typical applications that were compiled with standard compilers.
Reduced instruction set computing, or RISC (pronounced /rsk/), is a CPU design strategy based on the insight that simplified (as opposed to complex) instructions can provide higher performance if this simplicity enables much faster execution of each instruction.http://en.wikipedia.org/wiki/Reduced_instruction_set_computing [wikipedia.org]
The point of RISC wasn't to reduce the instruction set to those most commonly used, although the concept of "make the common case fast" resulted in some of that. The main goal was to reduce the complexity of what each instruction does so that any instruction will execute in the same number of clock cycles as any other instruction. By enforcing instruction execution time, this opened up the ability to pipeline instructions easily. In fact, this model was used in the Pentium Pro. and Pentium Core architec
Re: (Score:2)
Yes, your points are true. I didn't really explain myself well.
I understand that a goal of RISC was to reduce execution time: the path through the PLA. And I had forgot about equalizing instruction times. Thanks for reminding me of that: it has been a couple of decades.... ;-)
What I am really saying is that there was - and still is - a disconnect between hardware design and software design. Yes, hardware was ill-matched to languages, but the reverse was (and is) true as well: lots of things that programmers
Re: (Score:1)
Re: (Score:2)
Yes, indeed. But that is because the CISC chips contains a grab-bag of instructions, many of which were not needed. What is needed is the right set of instructions, not a minimal set.
An overlooked aspect of this point is that languages are poorly matched to what programmers need to do. My favorite example is linked lists. Linked lists are so widespread and yet they are not built into languages. That means that hardware cannot provide built-in support for something that is widely used.
Before asking the quest
Re: (Score:1)
Re: (Score:2)
Re: (Score:2)
Gee, that would be like... the DSP or the GPU!
Exciting Discovery!!!111 (Score:1)
Re: (Score:2)
It's news that doing so at the expense of increasing the error rate is such a significant net win for selected applications.
Re: (Score:2)
Indeed. Pruning - makes your processor get up and go!
Re: (Score:1)
Now all we need is fertilizer for our processors. Then we prune them and make them grow!
Then I'll graft bits and pieces of other processor and chips and make my own type of chip!
But then, the bugs get at it and I'll have to spray poison on it....hmmmm.....maybe I could make organic chips?! Think about it: computer chips made out of carbon! The hippie computer scientists will pay a premium for those!
I'll be rich!
Training wheels (Score:2, Funny)
I'll be removing the training wheels off my Harley this afternoon... thanks to this article I can be badass and efficient
Re: (Score:3)
I'll be removing the training wheels off my Harley this afternoon... thanks to this article I can be badass and efficient
Harley and efficient are two words that just don't go together. Bad and Harley might, but badass and Harley certainly don't either. Harleys only seem to appeal to mid-life wannabe posers that can afford them and like riding loud, leaking, unreliable motorcycles. The smart riders buy Japanese for the reliability and don't worry about trying to impress (but actually annoy) their neighbors with a sputtering Harley.
Re: (Score:2)
Re: (Score:2)
Sorry, I don't own a crotch rocket. I consider high-end race bikes on the street to be just a useless as a poser Harley. My last ride was a restored 1978 KZ650, and my next will likely be a dual-sport.
Let me guess, you own a Harley and it took you this long to actually read the comments?
Re: (Score:2)
Re: (Score:2)
Don't own a Harley. Just hate crotch rockets, and I brush up on comments after weekends. Besides, obnoxiously loud bikes are better - cars are more likely to know you're there if your pipes are loud. Less chance of death is good.
Except that statistically, loud pipes don't save lives. Probably not a direct cause, but some studies show bikes with loud pipes are more likely to be involved in an accident (probably speaks to inexperience more than anything). They aren't even loud unless you're behind them, which means you're just pissing off the people behind you and giving yourself a false sense of safety. Besides if the guys with loud pipes were really doing it to be safer, they should probably start by wearing a helmet - something
Re: (Score:2)
Then again, the harley riders I know also refused to even take the bike out on the street until they had l
As Charlie Sheen would say, (Score:1)
Well DUH....
Now my computer won't start up! (Score:5, Funny)
Re: (Score:2)
Slashdot cynics are right again (Score:5, Insightful)
Someone's going to chime in and say that the naysayers are oversimplifying or denigrating this because they didn't think of it, but I think the quote below says enough.
Uh, no, Professor, I don't believe it is.
Re: (Score:3)
That was my thought, that sort of thinking led Cyrix and AMD to lose serious ground to Intel when they failed to recognize the implications of the FPU to gaming. They were already behind, but this was a pretty significant loss for them.
Plus, wasn't this one of the ideas behind RISC?
Re: (Score:3)
The big idea behind RISC was to design simpler processors in the first place so you could ramp up the clock rates. This is taking an existing chip and getting rid of portions of it. They're in the same ballpark, but not quite the same thing. Modern x86 chips were designed to be RISC with a CISC-to-RISC instruction decoder place in front, so it's kind of like that.
Re: (Score:2)
No, RISC was an attempt to design a minimal instruction set under the assumption that a smaller instruction set leads to smaller CPUs that can be pipelined more efficiently.
This is going at it from the other side—taking an existing instruction set and determining which instructions are least frequently used and can be microcoded or software-emulated to allow for smaller chip design.
The subtle difference is that this isn't changing the ISA. It is just doi
Re: (Score:2)
Really what is happening is they are taking an existing design and trying to automagically modify it for their purposes instead of simply engineering a processor that does what they need. I'm pretty sure that if someone sat down and did a clean-sheet design for a hearing-aid processor, you would do even better than TFA's method. The question becomes cost. Do you have a computer program do some pruning based on a probabilistic heuristic of some kind, or do you pay an engineer to actually design an efficie
That's fine ... (Score:5, Interesting)
... for a specific application, like a hearing aid. Not so good for microprocessors intended for general purpose use (broad markets).
If you have sufficient market volume, you can afford to produce some sort of 'application specific integrate circuit'. Hmm, an ASIC. Now there's a novel idea (putting on jacket to make a dash to the patent office).
Madman Muntz famous(and rich)for this last century (Score:5, Interesting)
From Wikipedia entry on Madman Muntz:
Re:Madman Muntz famous(and rich)for this last cent (Score:4, Insightful)
Yeah, I'm not that crazy about that idea.
From what I gather the components being removed are most likely resistors and capacitors. And sure, some can be probably removed, if you don't mind ending up with a noisy power supply and too much current going to various parts.
So you're left with a device that kind of works, but that may mysteriously stop working in a few months.
Re: (Score:3)
He was probably responsible for those TVs I had when I was young which would lose sync when conditions weren't perfect, as in: either the Sun or the Moon were up.
Re: (Score:1)
this is the result of engineers who don't know how to design.
if you are _designing_ something, why in the world are you going to put in something you don't need ?
I spend my days trying to come up with clever ideas to NOT use things. Simple as possible and no simpler.
Madman Muntz went bankrupt (Score:4, Informative)
What you didn't mention is that "Muntz admitted his business lost $1,457,000 from April to August 1953,[28] and although he tried to reorganize, Muntz TV filed bankruptcy and went out of business in 1959" (from the same Wikipedia article)
You see, engineers don't sprinkle components at random. Every component in an electronic circuit is there for a reason. If something can be removed, what you have is a defective specification, maybe your circuit is designed to perform a function that's not often used, maybe it's designed to function in a situation that never happens. In that case you can ask the engineer to redesign for looser specifications.
Removing components at random is just stupid.
Re: (Score:3)
Muntz's designs were for areas with strong signals, so he put in fewer gain stages than competitors. Where competitors put in adjustments so that a set could be retuned to account for aging, Muntz did not. His sets were good for a few months or maybe years, but when components drifted due to age, the components had to be replaced or the circuits trimmed by adding parallel or serial devices to compensate for the drift.
In tube or discrete transistor circuits, it's good design practice to use resistive cathode
Re: (Score:2)
Long ago, I had a demonstration radio in class.
One day a student challenged the function of component after component which I then removed.
It turned out later that he had been shocked and wanted company.
Re: (Score:2)
Sounds a bit like how Woz built the first Apple computers, finding ways to do more with less.
erm (Score:5, Insightful)
"I believe this is the first time someone has taken an integrated circuit and said, 'Let's get rid of the part that we don't need,'"
I believe this to be a basic part of design.
Wait until US insurance companies chime in. (Score:1)
Not what hearing aid processors need (Score:1)
I'm hearing impaired from birth (23 yrs). Just got my newest pair last week (previous pair is 5 years old, but working perfectly).
In several ways, this new pair is an upgrade.... but in one key way, I fucking hate these things. Both the previous and current hearing aids are digital (my previous pairs were analog). With the older pair there is a two second delay between turning the hearing aid on and hearing stuff. The new pair has a minimum of 6 seconds...
So if I need to scratch the inside of my ear qui
Re: (Score:2)
I'm hearing impaired from birth (23 yrs). Just got my newest pair last week (previous pair is 5 years old, but working perfectly).
In several ways, this new pair is an upgrade.... but in one key way, I fucking hate these things. Both the previous and current hearing aids are digital (my previous pairs were analog). With the older pair there is a two second delay between turning the hearing aid on and hearing stuff. The new pair has a minimum of 6 seconds...
So if I need to scratch the inside of my ear quickly, I can do that in a second. Then I wait another 6 before I can hear again. Similarly, if I'm working without hearing aids in (relaxing, comfortable etc) and someone says something to me, I now have to wait almost 10 seconds before I can have them repeat what was said, then reply.
See the problem?
Hearing aid engineers are doing a lot of this work wrong (or for the wrong market, aka old people). Battery life is fine (a bit more than 2 weeks). I don't use the shitty auto-background-blocking programs on the hearing aid, either.
Guess I wasn't logged in when I posted this. Replies to this one will get my attention instead :)
Re: (Score:2)
The new pair has a minimum of 6 seconds...
Just wait until Microsoft ports Vista to these :-(.
So if I need to scratch the inside of my ear quickly, I can do that in a second. Then I wait another 6 before I can hear again.
So, why don't these things have some sort of 'suspend' mode, where they can wake up quickly?
If battery life is the issue, why don't hearing aids come with rechargeable batteries? Take them out at night. Plug them into a charging adapter and they'll be topped of (with a capacity of a few days) the next morning. I used to have a tiny FM radio with a rechargeable battery and charging adapter. The adapter itself could either be plugged into a wall wart or powe
Re: (Score:2)
I would LOVE to track me down a HA engineer at Phonak. I think they're all in Europe though.
Suspend mode would be nice. I sort of can do it, e.g. put the hearing aid in "tele-coil" mode which is used for phone conversations (doesn't use microphone -> won't be any feedback loop = noise). But you can only cycle through the programs in one direction, and there's 5 of them, so that's 5 button presses. A second toggle button like the volume control would be nice (aka back and forward), but apparently they
Re: (Score:2)
But you can only cycle through the programs in one direction, and there's 5 of them, so that's 5 button presses. A second toggle button like the volume control would be nice (aka back and forward), but apparently they choose a single push button to make it easier for people with limited dexterity.
Why don't these things have BlueTooth? Not just for the obvious cell phone connectivity. But imagine an app on your phone/PDA/whatever that talks to the hearing aid(s) and toggles them between modes, has nice big buttons and sliders on a GUI and can display some info about the hearing aid state.
Re: (Score:2)
Bluetooth is nowhere near low-power enough. The hearing aids can actually communicate via a lower power (less info of course) wireless method that I don't know much about. Such allows both hearing aids to change modes simultaneously, but that's not a feature I want to be restrained by, and so had my audiologist disable it.
I have wires that plug directly into an add-on boot on my hearing aids. I have zero problem listening to music.
Plus I've played with bluetooth enough to find it to be unreliable way too
Re: (Score:1)
Do you actually use all 5 modes ? If not just have them reprogrammed so ever other mode is t-coil. Hearing aids seem to be continuously improving but over the years they have removed two things I miss 1) the volume knob, 2) an actual off switch.
Re: (Score:2)
I have uses for 5 modes. I need to go back in and get the progs switched around a bit. The intended configuration will be:
1) Default 2) telecoil 3) active background blocking weak 4) active background blocking strong 5) direct audio input only (aka the cables mentioned in another branch of the thread)
In addition these new aids auto detect sound and switch to microphone + direct audio input. (so previously this config was it's own programming slot)
Re: (Score:1)
Re: (Score:1)
And I would hope the measurements would be in microamperes. Or that the D cell that powers it is a standard one, not an expensive lithium D cell...
It's the Analogy, Stupid... (Score:2)
If I had to use a commuter bike that I could modify on a race, I'd be thinking about changing the gear ratio before dropping a marginal amount of weight.
Re: (Score:1)
Are you making the assumption of a single gear commuter bike ??
My commuter class bike ( often refereed to as hybrid ) has 15 speeds/gears. I use , perhaps 3 or 4 of them .. But the range is there , both high and low.
What exactly would you change ?? give it a higher high gear ?? A lower low ?? How would that be better then loosing 10 pounds of crap ??
Your comment is not making sense to me. Please elaborate.
Re: (Score:2)
Take my road bike, for example. At top gear, I have roughly a 5:1 gear ratio. The manufacturer's comparable mountain bike has roughly a 4:1 gear ratio. The bikes' weights aren't that far off, and I would imagine the weight would really come into play on very long and/or stage races. A bike of comparable weight is still going to be hamstrung by the gear ratio. Even in the unlikely event that I get the commuter bike lighter than the competetion's racing bikes, they are going to have a more favorable sele
Re: (Score:1)
Yea ..Ok .. I understand what you are saying , However Its still not making sense as to why you would concern yourself with the extremes of the gear range , before you think about getting rid of ten pounds of dead weight.
The range of the commuter gears ( numbers used as per your comment ) may be from 1:4 to 4:1 .. Your race class machine may be from 1:5 to 5:1
Without knowing the type of race to be riden ( As no details have been given about that ) .. You are saying that having the extra range between 1:4 a
also (Score:1)
also they are twice as good at doing half the nothing. They run empty infinite loops at half the power too.
New Idea? Or just a twist on an old one? (Score:1)
Re: (Score:1)
The FPU was an expensive low-yield section of the circuit on the 486 processor. Often, the 486sx parts were ones with defects in the FPU section, so they just disabled the FPU and sold them cheap.
Article is incomplete (Score:2)
There is obviously more to the technology, but the "unnecessary" information has been pruned in order to make the article tighter and more accessible to the masses. Unfortunately, they removed all the bits that separate the approach from the engineering norm so it no longer functions as News for Nerds.
It's about the Error Rate. (Score:2)
The key part they are removing is error detection and correction. They are creating chips which have an ~8% chance of producing an incorrect result. Supposedly hearing aids will accept a 10% error rate, so it is a good trade off.
These aren't "redundant" parts, they're parts which prevent errors from happening. It's just that in some applications they don't care about errors.
It's like looking at the various floating point bugs and going, "meh, close enough". Sucks for a spreadsheet, but if all you care a
timothy is at it again (or how stupid is he?) (Score:2)
I don't think Timothy is a nerd. I don't even think he knows what one is.
I'm sorry if this offends you tim, but from what i've seen, you are really fucking lame. You post up the stupidest fucking articles possible.
What, does your 8 year old cousin do this for you?
I understand the concept of "practice makes perfect" but you should realize that there is some things we aren't good at, and this job you do, here at slashdot, isn't for you. You suck at it. You haven't gotten better, you've gotten worse.
Re: (Score:2)
Homer was right! (Score:1)
Why can removing components make chips faster? (Score:1)
Digital chips are roughly comprised of memory (flip flops) with logic in between. On each clock cycle the logic takes data from one piece of (input) memory, transforms it in some way and stores it in some other (output) memory.
One of the primary limitations on the speed of the chip is the longest path. The length of a path is roughly a function of the physical length of the path that the data takes from input to output memory and the number/type of logic gates in between. The speed of the chip is roughly
Re: (Score:1)
Dude, Nobody, but NOBODY does not do logic elimination. Some cheeseball routers dont do efficiency analysis, but on CPUs for core logic, driven by the ever increasing need for speed, they all do it. After an auto router takes a few passes at it, then a human inspects and corrects their work, then several others look over that work. Some times they even make steppings significantly faster! (There is one case of AMD doing this ). Later steppings are faster because they have spent more time running simulations
passwords reset on laptop (Score:1)
Reminds me of a joke (Score:2)
Patient I'm losing my hearing
Doctor (After checking him for anything serious) Don't worry it's part of the aging process.
Patient Can you do anything for me?
Doctor We can get you a hearing aid.
Patient How much will that cost?
Doctor Well, we have two models. One costs $1000 and the other costs $5.
Patient What's the difference between the two.
Doctor One is a sophisticated minitiarized amplifier assembled, customized to fit your ear, and tuned by highly trained technicians. The other is