Slashdot Mirror
Auto Industry's Fastest Processor Is 128Mhz
afabbro writes "GM stated that the 2011 Buick Regal will have the auto industry's fastest processor: 128Mhz, and 3MB of flash. 'Three meg of flash memory and 128MHz clock speed doesn't sound like a lot in terms of computing power until you consider the environment these controllers have to live in. Our controllers are made to operate reliably up to 260 degrees (127C) and down to -40 degrees (-40C) for the life of the vehicle.'"
128MHZ for a rugged CPU for automotive use is a good thing, but clock speed is just one of many factors. TFA was a tad light on information and worded as an ad (which is to be expected from GM's press website), but other than just mentioning vague details and the fact that Freescale made it, this doesn't really mean much without factoring in other details.
Will this mean the 2011 Regal will be leaps and bounds over the 2010? Yes. How much is debatable.
Will this matter in the total scheme of automotive technology? Not really. ECMs have been improving each year, so the 2011 Regal may have a bump in the control CPU's clock speed, but perhaps some other car maker would have a different architecture in place (multiple modules controlling different functions such as PATS/antitheft, O2 sensor, fuel sensor [1], etc.)
Will other car companies have improvements in their technology? Assuredly. Ford has some new engines going in the mainstream line of vehicles. Other vehicle makers may be bringing diesels to the US.
The big question in all of this: Is there a car example I can go on here?
[1]: I'm sure all cars in the US will eventually be going Flex-Fuel (talk about bumping gasoline from 10% to 15% is happening in some places here in the US), so having the circuitry in place to handle varying amounts of ethanol will be crucial.
That in the early 90's that would have been a top end pc, and it's still probably more computing power than the space shuttle.
Environment similar to mil spec, durability like industrial, prices like consumer products.
The real "Libtards" are the Libertarians!
There are digital signal processors in production for years which go up to 333MHz, like ADI SHARCs. BMW had one in their E70 X5 audio system. I am sure there are more examples. I will check at work tomorrow but I am sure there are also higher speed C out there.
Timo's Audio Software http://www.esseraudio.com
128 MHz should be enough for every car.
rewriting history since 2109
Fancy Mars Spacecraft.
Launched several years ago with a faster processor (133MHz!), more storage, and a worse environment.
my car's built in GPS is probably more than 128mhz so I don't think that's the highest automotive CPU
Today while I was filling up my 2003 Corolla with gas, a guy drove up to the next pump in his 1952 MG convertible. Which gets 30MPG. My Corolla gets 27MPG.
--
make install -not war
Those conversions are correct you idiot. Durrr math is hard.
Obligatory:
Yo dawg, we heard you like to drive, so we replaced your 128 MHz GM CPU with an Intel Core i7-980X 3.6 gHz processor so you can simulate driving while you drive.
Only -40? How do new cars do in REAL cold, anyway?
-40F is equal to -40C
Considering the ECU is, most often, mounted in the engine bay, both are temperatures the controller could experience.
So... 3 years or 25 000 miles...
3 years or 25 000 miles...
No. But you should hire someone to find it first.
most of the ones i've seen have been mounted inside the car, either in some kick panel or under the seat/floorboard... hondas/mazdas anyway
durrr it would be inappropriate to exclude one of the figures. They both must be present for consistencies sake. Communication is hard durrrrr. fuck head.
Rob Malda has an opening for your cock. Two in fact.
Do you even lift?
These aren't the 'roids you're looking for.
128mhz and 3mb stockage ? Does this mean that there will be no Flash support on the 2011 Buick Regal ?
No money left for decent parts.
Name one new US car with a carburetor.
Will it run Crysis?
I agree with AC - name one current (2010 or 2011) American vehicle (I'll even grant trucks, including the over the road tractors!) with a carburetor. As a matter of fact, name one made since 1999 ...
Sure a car can have computers, even my 93 Dodge has some kind of controller.
Hey we could have another TCP/IP type revolution here... add layers and gain functionality.
What is needed is to connect existing automobiles to an autonomous vehicle interface. The autonomous vehicle interface would provide a connection point for an autonomous vehicle computer to be attached.
The autonomous vehicle computer would query the interface device and get a description response of all the controls and sensors available. The interface would organize and scale the data available from the car. The interface would convert autonomous vehicle computer data into the signals expected by the devices attached to the automobile.
Think of the autonomous vehicle interface as an Arduino Mega that is wired into whatever the vehicle has available. The interface would be like interaction with a python interpreter.
Now the autonomous vehicle computer, think of that as a Linux netbook running a variety of programs for the car. There could be add on sensor devices attached to the autonomous vehicle computer. Like a GPS, a data cell phone and a 802.g wireless connection to nearby vehicles and 802.g radio equipped traffic devices.
http://lessco2essay.blogspot.com/2010/11/proposal-for-autonomous-vehicle.html
So what to do with such a modification: Completely end drunk driving accidents. Reduce the kWh per 100 passenger-kilometres. Do aggressive dynamic insured and paid ride sharing.
Dramatically reduce distracted driving damage. Reduce direct fuel use by coasting up to stoplights. All of this with existing vehicles.
http://www.inference.phy.cam.ac.uk/withouthotair/c20/page_118.shtml
The exciting part of doing AVT autonomous vehicle technology like this is: It is not proprietary and locked up in islands of make and model specific functionality. Some aspects of AVT can be backported to most older cars and real energy and safety benefits accrued.
Three meg of flash memory and 128MHz clock speed doesn't sound like a lot in terms of computing power
Guess that depends on your point of view, a car travelling 360Km/Hr is travelling 100m/s, so in a millisecond travels 10cm or about 4 inches. Assuming one instruction per clock cycle you can do a lot of useful stuff with 128,000 instructions, or put another way probably about one million for every revolution of the wheel
3MB of FLASH is huge as well when you aren't loading a lot of crap like multimedia, not that it would run Linux but I just took a look at the last kernel I built for an embedded platform and it came in under 2MB with quite a generous set of modules loaded.
I overclocked mine to 350 using an extra heatsink.
Since we are talking about GM, I guess they could put in an uncooled Athlon XP. That would best match the CPU MTBF to the useful life of the vehicle.
Watch this talk ... it is eye opening ...
http://win-dms-ms1.caltech.edu/five/Viewer/?peid=f2da15a3da764d1aa5d968568b739069
x
You guys have cocks? http://www.wired.com/bodyhack/2006/12/indian_men_too_/
Where's the blu-ray disk player that has thousands of mp3's in a disposable format?
And I just ran out of points. darn
-1 ignorant. If memory serves, the nissan GT-R processor is 20MHz RISC.
The Kruger Dunning explains most post on
Ugh, watched the video on that site.. some people either need to prepare what they are going to say, or just do retakes.. Also the interference mentioned in the clip sounds like bull, a psp is not going to interfere with an iphone and viceversa.
How ever it's pretty cool that we have tough freescale processors, wonder why they didn't use arm.
as it turns out.
http://www.metacafe.com/watch/607965/how_cold_a_car_can_handle/
carburetors
No.
I wondered before why ECM/PCMs etc. have to be under the hood. Why can't they be on the dashboard side of the firewall?
Genuine question here: why do these things need to be so rugged? Why can't they just slot under the dashboard where the environment (I'd hope) is a little more comfortable?
I'm sure those were not for the engine. There are cars with 10 or more ECUs, from which only 1 is for the engine.
Summarising, a modern turbodiesel is inherently about 25% more efficient than an equally modern gasoline engine. With old and crude designs like, say, carb hemi V8s, the Diesel has more like a 2:1 advantage. The remaining 5% comes from the fuel.
Sheesh, kids today. Get off my lawn.
From scarped cliff or quarried stone she cries "A thousand types are gone, I care for nothing, no not one."
I am an Automotive engineer. I work on infotainment components. This story is 100% wrong. 128Mhz, what BS
You dont need a high frequency processor to run an ECU.. The complexity of software like win 3.11 running on a 25mhz 486sx is staggering compared to those ECUs. They are built to be safe and reliable, not to be powerful.
If you need more than 128mhz you're doing it wrong.
As a comparison... Kollsnes Gas Processing plant has ABB MFP controllers for control of for instance huge 30mw compressors. These controllers run at... wait for it... 25mhz. They're so reliable that we have never had one go down since 1996. They're not powerful, but they're daaaymn reliable... So reliable that it is hard to get the customer to upgrade to newer and more powerful units because it -just- -works-.
My 1989 Citroen XM had six or seven ECUs depending on whether or not you count the heater and aircon ECU separately - engine, ABS, suspension (hydraulic suspension with electronic stabilisation), heater, aircon, central locking (which also dealt with windows, courtesy lights and immobiliser) and the dashboard and instrumentation controller. That doesn't include all the little motor drivers, sensor amplifiers and assorted other little boxes dotted around the vehicle.
Surprisingly enough, although much of it was somewhat ahead of the technology curve it was pretty reliable once you replaced the rather poor quality earthing connectors in the engine bay - the suspension ECU tries to switch about 10A for the stiffness electrovalve through the same earth tag as everything else, which usually resulted in *something* locking up and misbehaving.
Luxury. We used to have to get out of the lake at six o'clock in the morning, clean the lake, eat a handful of gravel, work twenty hour day at mill for tuppence a month, come home, and Dad would thrash us to sleep with a broken bottle, if we were lucky!
Meanwhile I'm working on a micro-controller project that runs at 500Hz (not kilo, just hertz).
If you keep the code tight and hand-craft it, 128Mhz is blindingly fast.
This seems pretty fast for a rugged processor but speed isn't everything.
Is it going to turn the world upside down? Probably not. Is it faster? Yes.
Is this a revolutionary new invention? No, speeds have been getting faster.
+5 please
Our controllers are made to operate reliably up to 260 degrees (127C) and down to -40 degrees (-40C) for the life of the vehicle.'"
More like a different universe, really. Chips like the PIC18F line (from Microchip) are still widely used. These are 8-bit processors, running at a few 10s of MHz (at most).
I currently work in the military avionics field; a 1 GHz Celeron is still pretty bleeding edge. A big part of the issue is heat -- fans are problematic, especially in harsh environments. You need something with low enough power dissipation that can be sealed inside a box and passively cooled.
These processors run real time software, there isn't likely to be a big bulky OS in there. The speed of the CPU doesn't really make much difference as it won't be multitasking. As long as it can respond in time then that's all that matters.
Often the biggest problem with the software in ECUs is all the varying laws in each country. So the software is often a compromise, this is why "chipping" your car can improve performance.
Considering a DVD with mp3s already can hold thousands (probably an *average* person's audio library would fit on one single-layer DVD).
Secondly, I thought optical media for MP3 storage would be great and so I got a headunit some years ago explicitly with MP3 DVD capability. The problem when compared to even USB flash is the seek time is noticeably slow. Even if you operationally did all sorts of tricks (e.g. prebuffer the beginning of the 'next' track), you won't hit all the cases of navigation and nothing mitigates the initial spin-up and index time.
Considering that USB flash is relatively inexpensive, focusing on home-written optical media would be a waste. BluRay may have a place as a player for publisher provided movies, but publishers will never sell a large audio collection in one chunk to warrant anything other than uncompressed CD on their end.
XML is like violence. If it doesn't solve the problem, use more.
You can easily run EFI on a processor in the 8-30MHz range. It just goes to shot you that the code bloat we see prevalently in the computer software industry is starting to move to embedded applications, which were formerly the last bastion of efficient, purpose-built code.
Well I hope it doesn't run Windows otherwise drivers may get the dreaded "Blue Windscreen of Death". Although if that happens, it will be the "Blue mixed with Blood Windscreen of Death"
Heh.
Other plants can be used to make ethanol, but it's not being done widely
Sure it is. Brazil has been producing efficient sugarcane-based ethanol for decades, and now accounts for almost 40% of the world's ethanol fuel production. Not that it matters much to the US, because of the quotas and massive tariffs to protect the crappy corn ethanol industry...
Corn is being subsidized for reasons other than ethanol. Most animal feed is corn-based, and cheap corn means cheap meat (which Americans, on average, love). If you cut the subsidies for corn, feed becomes more expensive, and then so does everyone's steaks and burgers. And we can't have that. Check out The Omnivore's Dilemma by Michael Polan.
US corn is so subsidized that it's actually cheaper for Mexicans to import it and use it as their staple instead of growing their own. It's one of the few things that's "cheaper" to make in the US than Mexico--think about that.
Agricultural subsidies are huge in the US (and many other Western countries as well).
It probably runs OSEK OS if it's in a car. OSEK/VDX is a software platform agreed on by several automotive companys. Specification of the single components is free to everyone. And in contrast to the documentation on the Linux kernel it's complete. You could sit down and recreate the OS and other software just from these docs.
You americans and your silly cars.....
Meanwhile out in the howling wastes populated by nothing but foreign ghosts, a VW Golf will give you a 74mpg combined cycle (83mpg extra-urban, 60mpg around town). Without any mucking about with Hybrid powertrains. Hell, even a BMW X3 SUV will give 50mpg combined cycle.
And for the record, many ECU's tend to use the microcontroller just as an interface to the CANBUS, with the heavy lifting done on automotive grade DSP's (eg. a TI dm642 is quite common clocked at 400MHz and up, and has been for years), so this article is just lame beyond belief.
It's amazing when I tell people that you can't just put a computer in any environment and it will work, In a car we have one of the hardest computer operating environments for a computer system, so much RF and motor noise and conditions which cause computers to fail. The Automotive controls have to work in environments so far beyond a normal computer operating environment it's amazing. So 128 Mhz and 3 MB flash, is actually a really good system.
127 to -40 is a problem. Actually it's a pretty serious problem. Well 127 on the high side is probably manageable, depending on where exactly the processor is in the engine, most inhabited places only rarely get to 50-60C, and then if you leave the vehicle out in the sun it's only 80 or 90. But about 7 years ago we had a cold snap that hit basically the great lakes, and the St. Lawrence area and it was consistently below -40 for more than two weeks, and that's in an area with ~25 or 30 million people (and by extension 25 or 30 million cars since it's north america). And that sort of thing isn't all that uncommon in a lot of canada or the US. So that's not even accounting for anywhere else in the world. I'm sure parts can exceed tolerance for a while, but I'd hate find out the 12 CPU's in my car all need to be replaced because I left it out overnight in -45 C weather.
And for this anemic POS computer they charge something like 2 grand. What a RIPOFF!
Today while I was filling up my 2003 Corolla with gas, a guy drove up to the next pump in his 1952 MG convertible. Which gets 30MPG. My Corolla gets 27MPG.
So you are comparing a an MG that gets maybe 50HP, has a top speed of 77mph and does 0-60 in 18.2 seconds with a Corolla that even in its mildest trim gets 95HP which is at minimum roughly double the fuel economy per available horsepower. Could you come up with a slightly worse comparison? Maybe a Prius to an M1 Abrams Tank? It is perfectly possible and even easy to find cars using modern technology to get much higher MPG. However there are other issues to consider because MPG isn't the only consideration. Safety, reliability, comfort, entertainment, intended use and other features are all trade offs engineers need to make to satisfy both legal and customer requirements. There are very good reasons we don't make cars like a '52 MG anymore.
This reminds me of the stupid comparisons of the cost of a gallon of diesel fuel with a gallon of gasoline. Diesel has about 20% more energy per unit of volume so PER UNIT OF HORSEPOWER, diesel is more efficient under many (though not all) circumstances. Simply saying diesel is more expensive per gallon really tells you nothing. The useful comparison is cost per unit of horsepower, i.e., how much power was generated per dollar. Grossly simplified, at the current prices near me diesel would have to be over $0.60/gallon more expensive to make gasoline a better deal per unit of horsepower for a typical automobile.
For extreme conditions look to the Oil & Gas Industry. The tools they drop into wells are rated -40 C to 175 C. That is operating temperature. The tools have to work at that temperature for 8, 12, or even 20 hours without failure. TI is developing a line of high reliability microcontrollers that are rated to 205 C.
The OnStar module found in most GM cars has an automotive-qualified 400MHz PPC 603e-based core and the latest versions have 16MB (or more) of flash and RAM.
-Splat
I've seen all kinds of cars and tractors start in temperatures getting near or below -40 degrees. Some times that meant the transmission got busted.
As a teenager in northern Canada, I learned that you need to warm up the transmission as well as the engine in extreme cold. A friend of my dad's forgot this lesson and and had to replace his car's automatic transmission.
In extreme cold, you can protect your transmission by putting it in neutral for a few minutes. This gets the transmission oil moving (and warming) without engaging more delicate mechanical parts. Do not leave an automatic transmission in "Park".
BTW - While several minutes of idling in neutral during EXTREME cold conditions are required to warm the transmission, 90 seconds of idling is all your engine needs. Any extra idling time is for only for the driver's comfort (i.e. warms up the cars interior )
-- "At Microsoft, quality is job 1.1" -- PC Magazine, Nov. 1994
At the time I am writing this, nobody is still imagining a beowulf cluster of these processors!!! Are you sleeping ?!?
What the "article" describes is standard for high-grade automotive MCUs. 128MHz may be new in engine controllers (although I doubt by much), but it is not new in cars. I work on production automotive MCUs for antilock brakes and chassis control that go up to 160MHz and >200MHz is in the pipeline. I don't think we sell anything with 3 MB of flash (which might be the more exciting part), but that's just a matter of what people are willing to pay. The operating conditions (-40C to 125C, etc.) are bog-standard for this application. It doesn't sound like there's anything new here -- they just made a new ECM and are patting themselves on the back. It may be better than anything else by some metric, but they didn't say what that metric really is.
That being said, it is nice to see something concrete about what their engineering group is up to rather than some empty fluff about new Buicks being able to cure cancer or something.
Visit the
Do not leave an automatic transmission in "Park".
I take it you mean, start the engine while in "P" and then switch to "N"?
Thanks for the tip. It rarely gets below -20C here where I am in tropical Toronto but it's always good to be nice to the mechanisms.
Thats why the average PC is a steaming pile of dog crap compared to what is actually used in important applications. Believe it or not, keeping errors low is not exactly high on the list at Monkey$haft.
I posted this elsewhere first- in a discussion about how automakers complain that the MPG targets set for them are unrealistic- but it seems to fit here:
The 3rd most fuel-efficient car in the last 30 years, after the Prius and the Insight, was the 26-year-old Chevy Sprint (see http://www.popularmechanics.com/cars/news/fuel-economy/epa-fuel-efficient-cars-chevy-sprint ). Overall, despite all the progress in engineering over the last 30 years, fuel efficiency hasn't increased all that much on average.
Advances in efficiency are offset by increased curb weight due to overdone crashworthiness standards and a crashworthiness arms race. If _your_ vehicle has a lot of mass between you and the front of the car, you're safer in a head-on collision. But if _everybody's_ cars have that increased mass, then you're not really safer.
Other reasons for a lack of progress include people's demands for higher acceleration performance (leading to big, heavy engines which use a lot of fuel) and other amenities. The automakers are right that the demand for higher efficiency isn't strong enough to outweigh costs and the demands for other amenities, and the only way this is likely to change is if fuel prices increase.
Fuel taxes and more balanced safety standards (recognizing safety v. efficiency is generally a tradeoff and so we really _don't_ want the safest cars possible) are the only things the government can do that will have a reliable and sizable effect in increasing average fleet efficiency. Just yelling at auto manufacturers for selling people what they want to buy isn't going to help.
LOL yeah how dumb is that they should really proofread their stuff before they look stupid
It should be
260 degrees (127C) and down to -40 degrees (-173C)
One guy defending his 'eevil gas-guzzler muscle car' because it held together longer, thus with lower manufacturing costs
Well, he probably figured it's really cool - I wonder if the environmental numbers are even better with a *compact* older car
I listen to both RIAA and non-RIAA stuff if I like the music, tangential business/politics nonwithstanding.
A old style carb? or a new style "injected mixing box"? Very few new US cars are really direct into the cylinder injection.
All of the above was encrypted with a Quad ROT-13 method. Unauthorized decryption is in violation of the DMCA.
Also, that guy sounded like the kind of guy who's really enthusiastic about maintaining his vehicle, which has got to help longevity - I wonder if those kinds of gearheads could get sufficiently excited about maintaining $old_compact.
-
No worries about replying to the sig; it's there for a reason. :)
I have indeed heard the RIAA 'horror stories'.
Seems you're thinking along the lines that if we ignore the RIAA today (even if we think there's _some_ value to particular RIAA artists), its costs and problems would be eliminated or reduced in the future? I see clear costs and unclear benefits in *completely* ditching them.
I simply don't feel the furor common around here, by taking it case-by-case, I end up with a mixture of RIAA and indie.
I listen to both RIAA and non-RIAA stuff if I like the music, tangential business/politics nonwithstanding.
You can start all modern engines coupled to an automatic transmission in neutral (N).
Gives a new meaning to "blood boiling". Considering also child deaths in hot cars, I would say a solar powered fan would be a better investment.
What about manual transmissions? Any special considerations for them in extreme cold?
> Also, that guy sounded like the kind of guy who's really enthusiastic about maintaining his vehicle, which has got to help longevity - I wonder if those kinds of gearheads could get sufficiently excited about maintaining $old_compact.
Unlikely.
> I simply don't feel the furor common around here, by taking it case-by-case, I end up with a mixture of RIAA and indie.
Actually, you don't take it case-by-case, you just decided to ignore the issue. Which is fine, but if it quacks, name it duck. I happen to think it's better to not support the big labels, if possible, in the foreseeable future.
I live in Saskatchewan. When it's' really cold out, you want to idle the engine for a short amount of time and then start moving slowly.
It's better for the engine to be running with a load on it, and by starting off slowly you minimize the stress on the other components as they fight against cold fluids. Transmission fluid, brake fluid, power steering fluid, suspension oil, etc....they all need to warm up.
This means that for the first 10 minutes or so I stay on residential streets, avoid sharp turns, avoid bumps as much as possible, baby the stick when shifting, etc.
Also, if you leave the heater off for the first while the engine warms up faster. You can also block off the front grille to minimize the airflow through the rad.
It's not good for the engine, but it can be done. It's easier with full synthetic oil.
Yes, I *do* ignore the label issue, because I simply don't care. By 'case-by-case', I consider on a case-by-case basis whether I like individual artists, regardless of what label type they chose to go with.
I listen to both RIAA and non-RIAA stuff if I like the music, tangential business/politics nonwithstanding.
Do you think they have the turbo button somewhere hidden? Overclock that mother!