Posts by John Smith 19 16330 publicly visible posts • joined 10 Jun 2009
Is there any business where that would be greeted by not much more than a "So what?"
In (US) car terms that would be a car that can do an average 174 000 mpg. Call it a top up every 50 years.
Thumbs up for a clever hack that could be retro-fitted to other mfg's design flows.
The UK has some amazing engineering skills that could supply goods or services to customers anywhere. That back street lockup might actually be doing half its business with South Korea for example.
AFAIK their attitudes seems to be roughly "We know all our customers and they know us so we put something up there because we felt we had to." I'd call it parochial, but I'm not sure how many of them could spell it.
Which means when they next big customer they have never dealt with before goes looking for a new supplier (probably these day with a web search for a quick look) they will be passed completely by.
Fapbook is not a choice. They don't plan parties or organise holidays.
"Now, if your simulation accuracy is 1 ms (and that is rather coarse as 0.1 ms is not unheard of) - you need to do 1000 * number_of_synapses * N (N=2) floating point reads, same number of writes - and several multiplications and additions for every single synapse. Even for a bee brain-size, that is many terabytes per second of I/O. And >that< is the biggest problem of large-scale bilogically-plausible neural networks."
Interesting to put some brackets on what kind of hardware you're looking at.
For "neuron" density you would seem to want some kind of fine grained parallelism. You might look at some of the ICL DAP processors of the early 70's. SIMD mostly but with some controlled instruction bypass at the processor level. I suspect the connectivity (especially) the fan out is the big problem. All done long before any kind of PAL architecture was available.
Perhaps an architecture where every node is the same but and can support message passing but whose intermediate results and node identifier can migrate. In time highly connected codes also become physically close nodes within the array.
That's about the highest frequency brain signals picked up.
So everything humans do is down to a) Massive parallelism b) Huge fanout/fanin in the neurons (up to 10 000:1 Vs maybe 10:1 in silicon).
AFAIK still the only serious neural net computer effort was WIZARD, which is going back a bit.
Interesting the first 4 posts were all moderator deleted.
"man thumbs down obviously dont watch the movies and the liquid nitrogen effect and the shock watch that make terminators and guns all explode, there no difference to a metal hull and chasis weilded together"
Oh darn. I only learned my physics and engineering in a classroom, not a cinema.
If you look at where this applies, mostly deep space probes they tend to consist of insulated boxes to keep the electronics around room temperature and the rest of the structure goes to ambient. It's not an issue.
Did you also learn your computer security watching "Swordfish"? I had a friend who was a big film fan and a CS grad. Her comment was "ROTFLMFAO."
"Frankly, the scientists at NASA don't know what to do with more power than this, it will require a paradigm shift for them. "
Yes power is a scarce resource on probes and if this goes ahead the power budget could grow by at least an order of magnitude. It will take time to process that and consider options. I hope people fly sooner rather than later.
"The surface power guys are generally looking at 40 to 100 kWe systems, the nice thing here is they aren't as concerned with mass, so we can still go with simple technology. The end goal will be MW class systems to drive large EP systems, like VASIMR."
My sense is that if it were possible to design a module in the 40Kw range that could be clustered while operating in a range of gravity fields (down to orbital) one basic design could be produced, giving economies of scale. I'd guess the number (or type) of radiators would change depending on where they would be used. A 5 unit cluster could power a VASIMR test unit. If demonstrated that would change the yardstick for round trips to Mars from 18 months to 2 1/2, possibly with substantial effects on viable mission architectures, especially if the payload was big enough to accommodate a single power unit and its associated landing gear.
"This experiment has rekindled a lot of interest at NASA, they had lost faith (and rightfully so) that we could do any real reactor testing anymore. "
I don't think the fact this is a live reactor can be over emphasized. Given the upcoming funding situation I hope you'll be able to establish and maintain some momentum in the effort, both within LANL and NASA.
"I have a site that steps through my views of nuclear power in space at spacenuke.blogspot.com"
I've seen it. I'm not sure if it's my browser security or what but I found a lot of broken links.
BTW Apologies. The US reactor in space was the SNAP 10. I don't think they eve got to a SNAP20.
"you dont need to release excess heat, you can just circulate it around the hull and interior, and stop the -200c killing people when going further away from the sun"
The -270c temperature applies to object which are in shadow from only major bright objects (like the Sun, Earth etc). In full sunlight outside surfaces are exposed to roughly 1300W/m^2 and the outside surfaces hit +200c.
Any object in space is inside a vacuum flask called the universe. The only way to transfer heat is by radiation. What happens is heat buildup inside an object happens until a) The temperature rises enough for enough heat to be radiated away that it reaches an equilibrium temperature (Stefan Boltzman law) b)The heat producing mechanism can't release any more. In the case of humans that would be when they die of heat stroke, literally boiled in the bag.
A neat description of this can be found in "Have spacesuit will travel" and "The Forever War."
"Still doesn't solve two problems "
Nor was it meant to
"First, you need to be able to release excess heat,"
In point of fact that is an issue with all space systems. The word I think you're looking for is "coupling" as in coupling the heat being generated to the environment. It's why the Shuttle had 2 big radiators inside it's payload bay doors and the ISS has a big one as well. Without it the crew will literally cook themselves. This shot of the US's only space tested reactor is instructive.
http://en.wikipedia.org/wiki/File:SNAP_10A_Space_Nuclear_Power_Plant.jpg
The black thing at the top of the cone is the reactor. The rest is radiator. Radiator design is a key element of any space system, even communications satellites.
"but electromagnetic propulsion still needs something to "throw" like a hydrogen supply. "
You might like to look up "ion engine"or "Hall thruster" to see what is currently available. The standard propellant is Xenon in high pressure tanks.
"We're making inroads at pure-energy propulsion, but what's being produced so far is still far too inefficient for prime time."
That's a maybe we're making progress. While some results are promising no one is near a fielded system.
"You are a smart man!"
Not a universal view, but I have my moments.
"This is not a fantastic technical breakthrough, "
Perhaps not, but it breaks the "never been done/can't be done" loop. I'm not completely certain that a reactor driven Stirling engine has never been run before but if not then it is the start of the art, Like Whittles W1.
"The big value was to show we could do a real nuclear reactor demonstration for short time and relative low cost, and then build from there. "
The fact this is a live nuclear reactor is very impressive in the timescale, given the much stricter H&S regime that needs to be complied with. Watching some old film of a guy machining a bar of metal in a lath that turned out to be one of the SNAP20 nuclear fuel pins without being in a hot cell was err interesting.
The only additional information I found on another site was that the test was 24w(e). can you confirm this?. I could not find anything on the NASA NTRS server (lots on fission of course but nothing with "flattop" fission) about the programme..
The 100Kw design seems (potentially) to be quite a handy size package which could be dropped into a lot of systems for use pretty much anywhere in the solar system. I'll wish you luck with its development.
NASA have been at work on a next generation RTG using Stirling engines and a Pu package they call the "General Purpose Heat Source."
It looks like they had a few Stirling spare, knocked up a few heat pipes compatible with the reactor and it was game on.
Another site claims this thing was putting out 24W, while the Stirling RTG (misnomer as it does not *have* thermoelectric elements) can push 500W(e), as did the only US reactor the SNAP20.
The SNAP20 used "Peltier" or TE elements with a 1.9% efficiency. Current ones hit about 6% and Stirlings about 24-26% (but with the risk of moving parts). The big benefits of a reactor a)throttling, as unlike an RTG, which starts at maximum and has to be designed to have enough power at the destination to get the job done they can be throttled down in cruise (or even left off till the destination before first switch on) b) A variety of nuclear fuels are possible for the reactor, whereas the GPHS is designed for only one kind of Pu fuel, and its in short supply in the US. Europe seems to be standardising on the stuff found in smoke detectors.
Down side is like all reactors it will need a minimum critical mass which is typically kept low by using highly enriched (70%+ IIRC) or bomb grade fuel.
Good work to get to a nuclear demonstration (not a simulator element) in 6 months and maybe the start of something quite impressive.
" I don't know what constraints there are on the flight profile, but I expect a relatively low-speed climb to high altitude and then a turn onto the "launch" heading for the run to Mach 5 and the extreme switch-over altitude."
The takeoff speed is roughly M0.5.
You might like to revise your idea of the flight profile.
"The major benefit of this is that the cooler tech is not limited to just Space launchers."
No the major benefit of this is it brings an HTOL SSTO vehicle a good deal nearer to being flight tested. That's their core business.
REL rate hypersonic cruise as tougher than launch. People have described cruise as like continuous re entry. It's a very tough environment. Whereas most REL tests have been run for about 10mins which REL say covers the whole of the launch cycle. LAPCAT cruise is measured in hours
REL know they need bigger partners and building an "industrial consortium" to do Skylon and the bulk of SABRE (IE all the stuff downstream of the pre cooler) is a key task.
"I hope the UK and European authorities are providing Reaction Engines with the required level of security against industrial and defence espionage."
Well in an ideal world the "Greatest advance since the jet engine" should probably rate a high end private security firm staffed by hyper alert ex SF types, attack dogs and a SoA alarm system guarding a TEMPEST shielded segregated computer network.
But I think they'll probably be sticking with the PC's they got on a block buy at PC World and the ex PC partnered with Lucky the faithful (three legged) Labrador for dog patrol on the perimeter.
All joking aside I also hope they will have put some money into both their physical and IT security.
That said I wonder if people realize how difficult real industrial espionage is.
Keep in mind the Soviet union did this with Concorde. Despite detailed access to the plans (and the whole resources of the KGB to bulglarise, bribe and blackmail) and the entire aeronautical establishment on call they still failed to realize the critical importance of the wings shape in delivering its handling and fuel efficiency, especially at low speed.
"Cooling the air down wasn't that much of a challenge,"
That's a bit of an understatement.
"the really neat trick (which as far as I know hasn't been revealed) is that the thing didn't freeze up into a solid mass of ice as the water in the air froze."
Correct . "Frost control" (and liquifying air, rather than just making it very cold) sank the 1950s LACE concepts.
They've given a few hints but I suspect their model on this will be the Whitehead torpedo whose "Special Sauce" (c Lewis Page) was its (all mechanical) attitude sensing control system which was held as a "trade secret" and users were trained through the 19century equivalent of NDA's.
"Avez vooz thoughte zate ze Concorde was doomed the moment we got the French on board?"
Actually the UK govt wanted to cancel Concord (or Concorde depending where and when you were talking).
At times it was only the clause that required both governments to cancel it that kept the programme running.
However the French did "contribute" the need to build 3 generations of prototypes before they accepted that a) Less than 100 seats was a complete non-starter b) You needed at least London/New York range.
Presumably that lesson will not be forgotten the next time round. BTW REL's M5 airliner proposal is on their site under LAPCAT.
Had they got to a 17th Concorde there were plans to up the range to Frankfurt/New York, without afterburner (people forget Concorde did routine "super cruise" for decades while the US Military tried to get it to work).
"if only the British Government had been able to come up with a fiver."
Well it was renew the patent or pay the gas bill. GE, P&W, Bell Aircraft are all jolly glad Frank chose the sensible course.
Of course in 1936
"But Wing Cmdr be reasonable. Our finest scientific minds assure me that you'd need so much cast iron your design would be too heavy."
Bill Gunston describes the antics around UK aircraft and jet engine quite well.
"Come on UK Gov, stump up the cash! You owe Alan Bond and the public for cancelling HOTOL!"
You'll find REL would prefer they did not stump up the money.
It's complicated. Many of the senior staff lived through Concorde. They know that large scale govt investment brings endless bureaucracy and funding instability (see NASA RE SLS, Orion/SM etc).
They also acknowledge that HOTOL "would not have worked" (Direct quote from Alan Bond). However in understanding why they got to Skylon. See "The 3 Rocketeers" film for more.
IIRC REL estimate 15% of their budget has been from various govts. The rest has been (very) private funding. Govt endorsement yes, govt only (or govt majority) funding no.
The Isp difference alone should stop anyone starting a new solid launch vehicle today.
The "simplicity" and cheapness are both very doubtful once you factor in the support costs of moving round large lumps of explosive.
Mfg is relatively simple but diagnosing problems is tough.
Japan has historical connections to solids and I think their first orbital vehicle was a solid but anyone targeting this tech is looking for ballistic missile tech, not a launcher. Looks like they now have it.
The MIT Media Lab had a project called "things that think" either incorporating processors in objects or roll to roll printing electronics.
IIRC they had a "memory element" patent (or licensed it form Zerox PARC) for a EEPROM element using carbon (or silicon?) granules as the storage element.
Beware US patent trolls on the sniff.
"'m not really sure why being able to monitor jet engines in realtime will save money. I would assume the airplane already routes the data in question (and far more) to the black box, so presumably the data is already being collected. How is having it in realtime saving money?"
You'd be wrong. Airlines carry 3 types of recorder. Maintenance, flight (the black box) and the cockpit voice recorder. The latter 2 (I'm not sure if they are combined into 1 "box") are designed to survive a crash and drop into an ocean.
AFAIK data items are set by law and actually quite limited. Any other parameters are at the mfg discretion.
IIRC BEA started carrying data recorders back in the early 70's which allowed them to re-schedule maintenance and keep more aircraft flying longer (BEA seem to have been pretty innovative in IT). These were not designed to be crashworthy. It's very handy if they do as they carry many more parameters.
"My mate with Derby connections says the smarter flight systems do already do this, but I've heard from two sources that the CCGT systems in power stations etc are controlled by little more than bog standard PLCs and the site operators can twiddle with the settings to suit their personal preferences in any given set of circumstances."
I guessed aircraft would be up for anything that cut their fuel bills.
As for PLC's, robust for control but not exactly the system of choice for an optimisation algorithm.
might be an opportunity there for a service (I thought RR did quite a lot of the stationary gas powered generators sets as well) but running an app on a PC in the site office seems just as possible.
"Weather is Weather, my understanding is that the Jetsream is South of where it normally sits - nothing to do with climate change. "
How about if stays South of where it's normally at?
Or it moves South of where it's normally at every year from now on?
Any one year it's weather.
Every year (or every year at the same time)
What's that?
"He told me he is growing several different types of wheat. Most notably spelt, but also emmer. "It grows everywhere, do not require antifungal treatment, no pesticides and no fertilization". I"
If it gets the job done that's fine. The joker in the pack seems to be hardy wheats don't have the yield of the hybrids.
If you're a small holder just growing for yourself that's fine but as a commercial taking your wheat to a market as a product getting 2x the yield for the same land means 2x the cash. The question is would your loose all your increased profit in pesticide costs?
You might like to look at the history of plant breeding. Breeders were improving yields long before genetic engineering. Indeed the theory was laid with studies in pea species.
Those in 3rd world countries seeking to improve their countries crop yields/pesticide resistance.
Global agrichemicals companies seeking to cross sell (our corn is resistant to our super duper combined pesticide/herbicide/window cleaner and it's only 250% more expensive than the regular stuff, but its infertile so you'll need another batch next year).
which would be fine *if* those genes didn't have a nasty habit of leaking into the other species.
Except they do.
"I worked at a company where they saught to allegiate the problems by having reviews every month. "
"Allegiate?" I am not familiar with this concept.
"My boss told him that in reviews that I intimidated my direct manager in the frequent reviews to get good scores.
It was absolutely insane. I don't think I've intimidated anyone in my life. "
Me either. Indeed at one point my supervisor at the time said "You don't scare me."
Odd thing to say really.....
"" Well, just like the appraisal systems under discussion, El Reg Badge Bonanza is just another system to be gamed. I couldn't possibly suggest you go create 100 throw-away email addresses at Gmail and similar sites, and then create 100 El Reg accounts, and then give yourself 100 upvotes to get the Ultimate El Reg Shiny! "
But if it's so easy why do you want it?
Also it might have something to do with the ratio of downvotes to upvotes . A 100% approval rating is very unlikely and some saddo is probably using some sort of down voting script already.
"I'm a programmer. I wrote code. I hope to..um..write more code?"
But imagine this conversation.
"boss. John (moustache twitching) we've noticed you use structured constructs in your code"
"me That's what we were taught to use in the latest version of the language"
"boss. We normally use GOTO's for compatibility with the old machine"
"me. Err right. If then else bad, GOTO good."
"boss.OK then, carry on."
This is neither an UL nor a funny story. It happened in the early 90's.
"where survival depends on you not being in the bottom 10% and any chance of a bonus means you have to be at least in the median group ... and they wonder why morale is shot to hell?"
But the beatings must continue till the morale improves.
Do you think that management might have lost sight of the purpose of all this?
Back then various LSI conference proceedings were full of architecture to link simple (usually bit serial) processors to individual storage chunks down to the bit level, promising massive processor/memory bandwidth. Image processing was a popular theme but rock strata reconstruction (IE looking for oil) was in there too.
IIRC only the ICL design made it into production. I think N. Carolina @ Chapel Hill were keen on it as well.
So the density has improved, the chunks got bigger and the processors will presumably be full 32 bits.
Didn't happen then. Not convinced it'll happen now.
But yes, it could play crysis.
Because the press told them to be.
They should stop wetting themselves about what the press (IE Murdoch) knows
It's interesting that "The Press" has suddenly discovered an interest in "freedom" (albeit mostly their own) while the Communications Interception Programme rolls on.
A bit f***ing late in my opion.
I refer to the whiny, confrontation avoiding nonentity who insists on down voting every post I make.
I guess you worked up a script to do it. Could any human be that anally clenched to waste their time?
Well done you. You must be so proud of yourself.
Have I offended you in some way? Hold opinions you don't agree with? Replied to you in a way you found hurtful? None of that's going to change unless you communicate with me instead at me.
I doubt you will contact me. Despite probably being several thousand miles away from you and incapable of offering any physical threat to you you're still too scared of that much communication with another human being.
I'm not angry or impressed, just bored. That you make me your target just shows how pathetic you are.
Fail because that is what you are.
"Those folks with flawed genetics are now free to continue procreating and possibly spreading hereditary borne issues such as autoimmune, heart disease, and other known problems."
Fine words and very practical.
Perhaps we should start with Stephen Hawking. He's had a fortune in medical care and what's he ever done for society or our understanding of the universe or...
Oh wait..
In a world where you societies survival was solely based on physical ability you might have a point.
The city state of Sparta went out of business a few millenia ago.
"Yes having read the private emails of scientists we now know it's not a hoax."
Hoax. No. FUBAR of epic proportions in data management yes (as documented in the harryreadme file).
Take a bunch of data sets which are not aligned with each other.
Run through sequence of badly coded programs (unrecorded sequence, setting and weather the program replaced some of the data with data hard coded in the program) plot on a graph, have it peer reviewed by some friends, get it published and hold a press conference.
BTW I actually think there is a serious AGW threat to the world but the travesty of the scientific method seen in the development process means I'm b***ered if this information proves a damm thing.
It is not mandatory that you always show your working but it is that you can if asked. If you can't that's what we lay people call "making some s**t up."
"The immediate problem, as I see it, is not that we need to find a solution to a rapidly expanding population but rather that we"
You might start by verifying that there is an accelerating global population.
IIRC that's not what is happening in the G7 countries for a start. You might then go on to places like Japan, Singapore and China (obviously a special case) and see what has happened there with rising living standards.
"I read an interesting theory the other day, in which it was suggested that technology growth goes hand in hand with population growth. "
It depends on the infrastructure.
10 billion educated literate and well fed people could explode across the solar system. 8 billion starving illiterates whose governments have washed their hands of them won't.
The Register 
Biting the hand that feeds IT
Copyright. All rights reserved © 1998–2024
