The other day I was told that a pace-maker some company is making will use an asynchronous processor. (I think) It scares me to think that the "Its usually pretty quick, sometmes it can take an infinate ammount of time to complete an operation but it will finish" philosophy is being tried out in pace-makers.
I've seen this on "Tommorows World" (its a British geek thing) a few years back. Im supprised that it didnt get hit by the tommorows world curse. Anyway, Its like using SRAM reather than constantly refreshing DRAM.
A few months ago I was talking to someone who was working on using the transistors on LCDs to do actual computation. The problem was thet there is a hude varioation in speed between individual transistors. Making the worst case delay and clock distribution too painful to make the system usefull for anything.
They were at the Async confrence and they were very intrested in doing everything asynchronously. It makes sence as implementing the logic in asynchronous circuits solves both problems.
I think RIAA is too keen to kill the networks that are slowly killing themselves. Take gnutella which when you search for a song you will get several different names for the same song, some other song wrongly labled, a few more truncated files and the rest are hosts which have been turned off days ago. There is no point RIAA attacking now when the networks are a mess. They shoud save their main thrust for when these problems are fixed. In the mean time publisize these problems and that its more hassle than its worth.
AMD has got a killer processor with trhe hammer but the current sample versions have a tiny cache. This is because AMD never had good luck with production flaws and their yeild is usually low (So I hear). So they are forced to making smaller chips. Now IBM have had some silly sized silicon experience. They are way way ahead of the others. especially as the hammer is a dual ISA chip so it would need a copy of each library for both architectures. Thus a need for a doubled cache. I remember a colegue coming back from Hot Chips and there being a presentation by some people of the so called giant chips they theoreticly could make and IBM just blew them out the water with some 8 metal layer 5x5 cm monstrosety (numbers are prob. way out).
Well if there is a company which has X,000 employees the statistical chances of the %'s being well out and the firm choosing their employees regardless of race and sex, are very very low.
Errr what? I work in a research group where we spend £2000 on a 900MHz Sun when compared with £600 for a 2GHz PC its appauling. Yes MHz != MHz so we did loads of comparison tests and still they are well behind.
Sun should no way drop solaris. The only reasom most people buy suns nowdays is because they can run thir cad software. And the cad companies only release their software on suns is because they know that not that many hackers can get their hands on them.
SGI like all server companies cannot drop their customer base suddenly and leave then stranded. They will carry on developing IRIX and MIPS systems and softly let them die. Just like HP with Alpha read this (funny) also shame. Yeah they have done a lot of stuff for Linux and the boxes do contain lots of their own stuff but its a large step down from making their fully own hardware (very novel stuff) and Irix down to using Itanics and doing Linux hacking. When a company gets down to that level and gives up their uniqueness then they are no longer special (in my hart anyway). Afterall there is no end of companies about to release multi intanic systems.
Re:SGI is finally making some new products
on
New SGI Altix 3000
·
· Score: 2, Insightful
I dont think you can call this any way new. They are simply packaging other peoples (intel + linux) products into boxes (Yes I know: not as easy as it sounds).
You can see SGI is getting desperate. They have dropped their MIPS systems and IRIX. They dont want to aim be the best around when they can just packahe other peoples stuff. Soon you will be able to see SGI PCs running a microso~1 OS and made with all standard parts. shame
I went to manchester and I had a great time. I became an alcoholic and had lots of "para-homosexual" activities. I think it depends where you go. I think the point of throwing you out to the real world during your second and third years is to ensure you are prepared for life. Also it depends where you go. Where did you go to in "England" (prenounced UK).
I demonstrate at Manchester University and there are people I know would be better off if they went straight to a job. Some people are planning to be HTML writers and have no desire to learn about computer architecture. They are wasting 3 years of their life during which they could get vital experience of a real job. People coming out of university cant get jobs because emplyers think they will want to move onto something better very soon after.
Whale researchers have found what the annoying sound that keeps interrupting their recordings of Minke Whales. The sonar pings have been desturbing whase scientists for years. The source was unknown until recently when a submarine surfaced shortly after destroying a research crews sound recording.
Please no more top 10 of anything anymore. Please?
on
Biggest IP cases of 2002
·
· Score: 1, Offtopic
I am getting tired of "Playboy's biggest breasts of the year" etc
You still cannot get past some limmiting factors like speed of light and the absolute minnimum structure size.
What the Japanease are looking into will be very large chips. Diamonds are the only good way to get a good yeilds of these. But still when you have a 10x10mm 100 GHz chip it takes several clock cycles to get some information from one side of the chip to the other.
Normal design methodologies will no longer work in the near future just like they are starting to get difficult now. (Moore's Law slowing down)
Slashdot Mirror
The other day I was told that a pace-maker some company is making will use an asynchronous processor. (I think)
It scares me to think that the "Its usually pretty quick, sometmes it can take an infinate ammount of time to complete an operation but it will finish" philosophy is being tried out in pace-makers.
I've seen this on "Tommorows World" (its a British geek thing) a few years back. Im supprised that it didnt get hit by the tommorows world curse.
Anyway, Its like using SRAM reather than constantly refreshing DRAM.
A few months ago I was talking to someone who was working on using the transistors on LCDs to do actual computation. The problem was thet there is a hude varioation in speed between individual transistors. Making the worst case delay and clock distribution too painful to make the system usefull for anything.
They were at the Async confrence and they were very intrested in doing everything asynchronously. It makes sence as implementing the logic in asynchronous circuits solves both problems.
Well in Elite anyway.
Ah well... Lonely life.
I think RIAA is too keen to kill the networks that are slowly killing themselves. Take gnutella which when you search for a song you will get several different names for the same song, some other song wrongly labled, a few more truncated files and the rest are hosts which have been turned off days ago.
There is no point RIAA attacking now when the networks are a mess. They shoud save their main thrust for when these problems are fixed. In the mean time publisize these problems and that its more hassle than its worth.
AMD has got a killer processor with trhe hammer but the current sample versions have a tiny cache. This is because AMD never had good luck with production flaws and their yeild is usually low (So I hear). So they are forced to making smaller chips. Now IBM have had some silly sized silicon experience. They are way way ahead of the others. especially as the hammer is a dual ISA chip so it would need a copy of each library for both architectures. Thus a need for a doubled cache.
I remember a colegue coming back from Hot Chips and there being a presentation by some people of the so called giant chips they theoreticly could make and IBM just blew them out the water with some 8 metal layer 5x5 cm monstrosety (numbers are prob. way out).
Well if there is a company which has X,000 employees the statistical chances of the %'s being well out and the firm choosing their employees regardless of race and sex, are very very low.
Yeah and I could have a couple IPX/IPC/Sparc1 boxes too. They still wont run cadence.
Errr what? I work in a research group where we spend £2000 on a 900MHz Sun when compared with £600 for a 2GHz PC its appauling. Yes MHz != MHz so we did loads of comparison tests and still they are well behind.
Sun should no way drop solaris. The only reasom most people buy suns nowdays is because they can run thir cad software. And the cad companies only release their software on suns is because they know that not that many hackers can get their hands on them.
SGI like all server companies cannot drop their customer base suddenly and leave then stranded. They will carry on developing IRIX and MIPS systems and softly let them die. Just like HP with Alpha read this (funny) also shame.
Yeah they have done a lot of stuff for Linux and the boxes do contain lots of their own stuff but its a large step down from making their fully own hardware (very novel stuff) and Irix down to using Itanics and doing Linux hacking.
When a company gets down to that level and gives up their uniqueness then they are no longer special (in my hart anyway).
Afterall there is no end of companies about to release multi intanic systems.
I dont think you can call this any way new. They are simply packaging other peoples (intel + linux) products into boxes (Yes I know: not as easy as it sounds).
You can see SGI is getting desperate. They have dropped their MIPS systems and IRIX. They dont want to aim be the best around when they can just packahe other peoples stuff. Soon you will be able to see SGI PCs running a microso~1 OS and made with all standard parts. shame
I went to manchester and I had a great time. I became an alcoholic and had lots of "para-homosexual" activities. I think it depends where you go. I think the point of throwing you out to the real world during your second and third years is to ensure you are prepared for life. Also it depends where you go. Where did you go to in "England" (prenounced UK).
I demonstrate at Manchester University and there are people I know would be better off if they went straight to a job. Some people are planning to be HTML writers and have no desire to learn about computer architecture. They are wasting 3 years of their life during which they could get vital experience of a real job. People coming out of university cant get jobs because emplyers think they will want to move onto something better very soon after.
Whale researchers have found what the annoying sound that keeps interrupting their recordings of Minke Whales. The sonar pings have been desturbing whase scientists for years. The source was unknown until recently when a submarine surfaced shortly after destroying a research crews sound recording.
I am getting tired of
"Playboy's biggest breasts of the year"
etc
What happened to the quickies?
I want them back.
They were great news and cool stuff nibbits.
A little agressive?
Actually it took be a few minutes to find the article.
This story is an example of someone who actually tried to do something simmilar.
Its a fantasticly strange and scary story.
What they should do is to advertise the event using popups.
What do you case it in?
What kind of wires do you attach?
Yes Diamonds are better than Silicon but:
You still cannot get past some limmiting factors like speed of light and the absolute minnimum structure size.
What the Japanease are looking into will be very large chips. Diamonds are the only good way to get a good yeilds of these. But still when you have a 10x10mm 100 GHz chip it takes several clock cycles to get some information from one side of the chip to the other.
Normal design methodologies will no longer work in the near future just like they are starting to get difficult now. (Moore's Law slowing down)
It isnt a MIPS.
Its a CRIS
It only says its a 100 MIPS processor. (Million Instructions Per Sec)
The MIPS has 32 registers.