Because AMD Athlon/Phenom CPUs have the memory controller integrated into the CPU, the CPU (not the motherboard) actually dictates what type of RAM you can use.
For all the desktop class AMD Athlon/Phenom CPUs, you can use un-buffered ECC memory. Just make sure it's not buffered or registered. You need an Opteron to use buffered or registered memory.
If you want an Intel processor, you have to use a Xeon (and the right mobo) to use ECC memory.
Thin film manufacturers like FirstSolar have already announced that they are producing panels below $1/watt. NanoSolar is very likely there as well.
Silicon panels have dropped a huge amount in the past 6-9 months with the weak economy. Wholesale prices are close to $2/watt, and you can buy panels retail at $3/watt.
Yes, the rest of the costs to install a system (inverter, wiring, mounts, labor) are quickly becoming the largest cost in a solar system, so expect to see drops in price there as well.
While that may be acceptable for casual use, for enterprise use, a UPS is not enough to guarantee your data safety as power losses even in enterprise situations are surprisingly common.
For non-critical servers or desktops, sure - but in that case, it's easier to simply increase the amount of RAM in the system and disable fsyncs.
For customers who only want/need a 100GB of storage, SSDs are the way to go. They do currently cost a lot more than rotating storage, but a SSD makes a HUGE difference in the apparent performance of many day-to-day tasks.
A good 120GB SSD like the OCZ Agility costs about $300 compared to $40 for a 160 SATA drive so the price premium is huge.
BTW - I'm not sure why you say drives smaller than 300GB are hard to find - or why your customers complain about it. NewEgg has a ton of drives smaller than that with the smallest at 80GB. Pricing looks like this:
It's pretty clear that it's simply not cost effective to produce a drive cheaper than $35. When you can pay another $15 and get 6x the storage space and a drive that is faster as well, it's pretty much a no brainer. If they really only need a 100GB drive, you should consider short-stroking a larger drive (just create a 100GB partition instead of using the full disk) as that will significantly improve random read/write performance due to shorter seek distances.
Battery backed up (BBU) RAID controllers with volatile RAM cache are very common in the server market because of the huge performance increase of small random writes.
The RAM cache lets the controller cache writes and then send them to the disk in batches while performing write combining so multiple small writes get turned into larger writes reducing the number of disk seeks required to store the data. Also think of the case where your controller has a 512MB cache and you write 200MB to disk. The controller can say OK as soon as it's written to RAM (fraction of a second) where your typical fast disk these days will take 2 seconds.
Without having a battery to back up the volatile RAM cache, you could lose a lot of data if the server lost power, but with it, you go go at least a couple days without losing data.
So now, let's replace that 512MB BBU RAM cache with a 16GB SLC SSD. You won't quite get the burst speed of the BBU RAM controller, but in sustained server loads performance should be a lot better. The SSD will also be able to store a lot more data for reads. If the controller is smart and only uses the SSD for caching random read patterns, you could get close to SSD performance for a lot of workloads but still have 1TB of disk storage.
I know people who own Priuses, though none that actually get 50mpg at highway speed in real life.
With the 2nd Gen Prius (2004-2009), 50mph is only seen if you keep speeds under 70mph on the highway.
For that matter, very rarely do Prius owners even achieve 40mpg at highway speed.
Bull - every Prius owner I know easily exceeds 40mpg at highway speed. In my experience, the only time I can get worse than 40mpg is at highway speeds of 85mph+.
And I have no reason to doubt that at the speeds many people would like to drive on the highway (70-90mph) the Prius will be lucky to even get 30mpg while the VW will still be in the 40+ range or better.
More bull. In my testing with a 2008 Prius (and others have backed this up):
So frankly, the only way you're getting worse than 30mpg in a Prius is if you have it floored and are doing over 100mph. I have no doubt that a diesel will match the Prius in mpg at high (75mph+) highway speeds. But let's not forget the energy content advantage diesel has over gasoline as well as the higher exhaust emissions.
I'm not all that impressed, either at this point, but I do have to take issue with a couple of your points.
The current model VW's that have diesel options (Jetta, Golf, Beetle) can average 50mpg all day long with 4 adults and the AC turned on.
Under certain conditions (steady state cruising on the highway at moderate speeds), the current TDIs may get 50mpg, but it seems that in real life fuel economy is really around 40mpg on average. (see fueleconomy.gov)
I for one would rather start with a diesel and tune it to get 70mpg without a trunk full of batteries.
Good luck with that. It doesn't appear to be possible unless you limit operation to 45-50mph cruising in an aerodynamic vehicle.
Please prove me wrong and make a prediction about the climate, such as, 'It won't snow in London next year".
You are still confusing climate with weather. Please, the post you replied to again, but here is a concrete example for you.
In London in winter, the current climate indicates that the temperature will normally be between 0*-10*C with light rainfall and rarely do temperatures dip below freezing or does it snow.
If it snows in London next winter, that will be a day of bad weather!
We can barely predict tomorrow's weather with any degree of accuracy (in some areas even that's stretching it), but we are supposed to believe any scientist that claims to have GLOBAL weather licked?
Trying to predict the weather is a LOT more difficult than predicting climate. Predicting the weather compared to climate is like predicting the output of a Random Number Generator over the next 14 iterations compared to the output of a RNG over the next 1000 or more.
Over 14 iterations the RNG (assuming 0-1 range), the average is likely to vary significantly from 0.5 and predicting which side of 0.5 it will fall is impossible, but over 1000 iterations, it will be very close to 0.5.
Although they aren't yet in stock, zipzoomfly is already listing the price at $223.25 (though you can't preorder).
Nice! If they actually end up selling at that price at launch, I will be impressed.
I have a Vertex and while the performance has been great, it doesn't seem to be very mature compared to regular disks.
For example, I've personally had these problems with it:
1. Firmware flash tool doesn't work on all computers. Have to remove it and move it to another computer to flash it. 2. Have seen the drive lock up and basically disappear a couple of times now. Have to power down and up to recover.
I've never seen those types of issues with any other regular IDE/SATA drive (though I guess Seagate did have a few problems with some of their drives a little while ago), and I wouldn't expect the Intel drive to have those issues.
Had the Intel been available for only slightly more when I got mine, I would have opted for the Intel over the Vertex. If OCZ/Indilinx isn't able to resolve these problems, I will likely stay away from them in the future.
I suspect we'll see the 2nd gen X-25M launch at prices similar to the current X-25M, and then drop down to the $225/80GB that you can get them in 1,000 unit quantities over the next couple months.
The competition for these Intel drives is at least 2-3x behind in random IOPs. Too bad the streaming write performance didn't go up significantly, because that's the only place where the Intel drives lag behind their competition.
My rough, but optimistic calculations showed that a Corolla (an efficient compact car) which is about the same weight as a Tesla and probably similar in overall aerodynamic drag (Tesla has smaller frontal area, but is less aerodynamic than a Corolla) is at best 25% efficient overall.
Cars have been steadily getting more efficient over time, the fleet age is close to 10 years old and the fleet averages less than 20 mpg. Is it that much of a leap to suggest that the fleet is less efficient than a Corolla? Did you miss my note that the fleet is likely to consume more than 270 Wh/mile?
I keep on seeing 30% efficiency in a "typical" automobile, but I think that is being generous by measuring the engine as it's peak and not average efficiency when in use in your typical automobile.
So let's take the most efficient gas-electric hybrid on the market today, the Prius which is rated at 50mpg. Given that one gallon of gas contains 33.6 kWh of energy, we calculate that the Prius burns about 672 Wh/mile. It's commonly quoted that a typical car needs about 225-300 Wh/mile. For example, the Tesla is quoted as using 270 Wh/mile. The Prius is slightly heavier, but more aerodynamic. But the Tesla is also able to capture more energy through regenerative braking. So let's use 270 Wh/mile. Note that I'm assuming the Tesla is 100% efficient (which it isn't, but is probably around 90%) which makes my numbers below optimistic.
The puts the Prius at 40% efficient - which is absolutely the best available right now.
A Corolla which averages 30mpg - is only 25% efficient.
And the fleet average fuel economy is only 20mpg. So the fleet on average is probably only about 15-20% efficient (taking into account that the fleet probably uses more than 270 Wh/mile.
That said - your 100 kWh estimate for an ultra efficient vehicle that can go 400 miles seems to be right on target using my numbers above.
The batteries that will be going into modern EVs do not include acid and lead - they primarily include Lithium which is fairly benign. The CEO of BYD which manufactures a LOT of Lithium batteries (chances are if you have a Lithium battery for one of your electronic gadgets, it's made by BYD) went so far as to drink a glass of the electrolyte that goes into their batteries to show that they are safe (not that I'd recommend that to anyone!)
Even in NiMH batteries which is the primary type of battery currently used in hybrids today, the Nickel used is not substantial and is recyclable if/when the battery dies.
Me too. I wonder if having done a good deal of "playing" with 1/10th scale RC cars over the years has helped. Certainly looking at the batteries in the RC cars from 15 years ago, when 1.2Ah NiCd SubC cells were top of the line and now 4-5Ah NiMH SubC cells are common and Lithium cells are another 50-100% better than that depending on the chemistry and especially once you ditch the cylindrical format and look at prismatic cells.
I must admit that I do appreciate a nice sounding internal combustion engine as well. There is something very pleasing about hearing one run through the gears - can't quite put my finger on it. But I'd also much rather not hear someone else making that noise, especially when I'm at home or work concentrating!
I'm an EV fan (and would gladly take a Tesla over a Ferrari), but comparing the Tesla to a F430 is a bit like comparing apples to oranges. It's probably better to compare it to the Elise/Exige from which it's chassis is based off of (let's use the higher performance Exige model):
Tesla numbers on the left, Exige numbers on the right:
So the Exige basically wipes the Telsa in every test except for 0-60 which is a wash (where the flat torque of an electric motor pays off) and efficiency where an EV blows gasoline cars out of the water.
All the extra weight of the Tesla comes from the battery pack which weighs about 900 lbs. No doubt if the battery pack was half the weight and if the gearing was a bit higher, it'd be able to match the Exige in performance, but we've still got at least a couple years before batteries reach that kind of energy/power density.
FWIW, I'd still take the Tesla over the Exige if I could afford either. Being able to refuel every night at home is a killer feature. Not to mention being able to refuel with renewable energy and having no tailpipe emissions.
TFA isn't talking about charging your car with 365 kW at your house. At your house, you'll charge with whatever power you have (typically 240V, 30-50A or about 10 kW).
The fast charging is for when you're taking a long trip and want to get going on your way as soon as possible.
Making batteries does not cause huge amounts of pollution. Where do you get your information from?
And as far as recycling rates, 98% of lead-acid batteries are recycled. Automobiles as a whole are recycled at a rate about 95%. Why would any other batteries used in future EVs be any different?
1000 amps is still a lot of current - seems like they could downsize the size of the conductor a good deal by upping the voltage going into the vehicle and converting it down to the proper voltage closer to the batteries.
Exactly. Current parallel hybrids work around conversion losses by:
1. Using a CVT to keep the engine in the most efficient operating range when it's running. 2. Using high efficiency inverters, generators and motors (2nd Gen Prius inverters/generators/motors are 90% efficient over a very wide operating range). 3. Recovering energy any time your foot is off the gas pedal and the car is moving.
A serial hybrid has the drawback of requiring the motor/batteries capable of producing maximum propulsion power which significant increases their cost. In today's hybrids maximum load is shared by the engine.
A serial hybrid makes the most sense when you are planning on using a large battery pack and plugging it in to avoid running the petrol engine in the first place.
Fuel efficiency is only really comparable on the highway. Hybrids tend to blow diesels away in the city. So yeah, if you do a lot of highway driving, a clean diesel is a good choice. You can even get one in the US now - the Jetta. They've been selling very well.
Just don't forget that diesel has an inherent ~15% advantage over gasoline cars because of the higher energy (and CO2) content per gallon.
Diesel also enjoys a hefty price advantage over in Europe over gasoline. Diesel is slightly cheaper than gas right now in the US, but it hasn't always been.
BTW, not only is soot a harmful byproduct from diesels, NOx emissions also tend to be significantly higher. Only recently in the US have soot and NOx emissions from diesels sold in the US come close to matching their gasoline powered counterparts.
Also, mileage in particular is noted as mediocre for a hybrid in this review - and isn't that pretty much the defining characteristic for any hybrid?
Seems like your mileage may vary as with all cars. With hybrids, it seems to vary a bit more as there were more than a few comments where people are regularly exceeding the EPA numbers.
Slashdot Mirror
Stop cherry-picking, 1998 was an abnormally warm year due to a number of factors.
'Global warming stopped in 1998'--Only if you flagrantly cherry pick
Because AMD Athlon/Phenom CPUs have the memory controller integrated into the CPU, the CPU (not the motherboard) actually dictates what type of RAM you can use.
For all the desktop class AMD Athlon/Phenom CPUs, you can use un-buffered ECC memory. Just make sure it's not buffered or registered. You need an Opteron to use buffered or registered memory.
If you want an Intel processor, you have to use a Xeon (and the right mobo) to use ECC memory.
He's talking about panels only.
Thin film manufacturers like FirstSolar have already announced that they are producing panels below $1/watt. NanoSolar is very likely there as well.
Silicon panels have dropped a huge amount in the past 6-9 months with the weak economy. Wholesale prices are close to $2/watt, and you can buy panels retail at $3/watt.
Yes, the rest of the costs to install a system (inverter, wiring, mounts, labor) are quickly becoming the largest cost in a solar system, so expect to see drops in price there as well.
While that may be acceptable for casual use, for enterprise use, a UPS is not enough to guarantee your data safety as power losses even in enterprise situations are surprisingly common.
For non-critical servers or desktops, sure - but in that case, it's easier to simply increase the amount of RAM in the system and disable fsyncs.
You mean, kind of like this?
Viking Arxcis-NV
And:
Adaptec RAID 5805Z
Looks like all the Adaptec 5Z series cards have the "battery less" RAM->flash backup feature.
For customers who only want/need a 100GB of storage, SSDs are the way to go. They do currently cost a lot more than rotating storage, but a SSD makes a HUGE difference in the apparent performance of many day-to-day tasks.
A good 120GB SSD like the OCZ Agility costs about $300 compared to $40 for a 160 SATA drive so the price premium is huge.
BTW - I'm not sure why you say drives smaller than 300GB are hard to find - or why your customers complain about it. NewEgg has a ton of drives smaller than that with the smallest at 80GB. Pricing looks like this:
80GB - $35 - $0.50/GB
120GB - $40 - $0.33/GB
250GB - $45 - $0.18/GB
320GB - $50 - $0.15/GB
500GB - $50 - $0.10/GB
It's pretty clear that it's simply not cost effective to produce a drive cheaper than $35. When you can pay another $15 and get 6x the storage space and a drive that is faster as well, it's pretty much a no brainer. If they really only need a 100GB drive, you should consider short-stroking a larger drive (just create a 100GB partition instead of using the full disk) as that will significantly improve random read/write performance due to shorter seek distances.
Which brings up an interesting design thought:
Battery backed up (BBU) RAID controllers with volatile RAM cache are very common in the server market because of the huge performance increase of small random writes.
The RAM cache lets the controller cache writes and then send them to the disk in batches while performing write combining so multiple small writes get turned into larger writes reducing the number of disk seeks required to store the data. Also think of the case where your controller has a 512MB cache and you write 200MB to disk. The controller can say OK as soon as it's written to RAM (fraction of a second) where your typical fast disk these days will take 2 seconds.
Without having a battery to back up the volatile RAM cache, you could lose a lot of data if the server lost power, but with it, you go go at least a couple days without losing data.
So now, let's replace that 512MB BBU RAM cache with a 16GB SLC SSD. You won't quite get the burst speed of the BBU RAM controller, but in sustained server loads performance should be a lot better. The SSD will also be able to store a lot more data for reads. If the controller is smart and only uses the SSD for caching random read patterns, you could get close to SSD performance for a lot of workloads but still have 1TB of disk storage.
With the 2nd Gen Prius (2004-2009), 50mph is only seen if you keep speeds under 70mph on the highway.
Bull - every Prius owner I know easily exceeds 40mpg at highway speed. In my experience, the only time I can get worse than 40mpg is at highway speeds of 85mph+.
More bull. In my testing with a 2008 Prius (and others have backed this up):
70mph = ~48mpg
75mph = ~45mpg
80mph = ~43mpg
85mph = ~40mpg
90mph = ~37mpg
95mph = ~34mpg
So frankly, the only way you're getting worse than 30mpg in a Prius is if you have it floored and are doing over 100mph. I have no doubt that a diesel will match the Prius in mpg at high (75mph+) highway speeds. But let's not forget the energy content advantage diesel has over gasoline as well as the higher exhaust emissions.
I'm not all that impressed, either at this point, but I do have to take issue with a couple of your points.
Under certain conditions (steady state cruising on the highway at moderate speeds), the current TDIs may get 50mpg, but it seems that in real life fuel economy is really around 40mpg on average. (see fueleconomy.gov)
Good luck with that. It doesn't appear to be possible unless you limit operation to 45-50mph cruising in an aerodynamic vehicle.
You are still confusing climate with weather. Please, the post you replied to again, but here is a concrete example for you.
In London in winter, the current climate indicates that the temperature will normally be between 0*-10*C with light rainfall and rarely do temperatures dip below freezing or does it snow.
If it snows in London next winter, that will be a day of bad weather!
Weather is not climate.
Trying to predict the weather is a LOT more difficult than predicting climate. Predicting the weather compared to climate is like predicting the output of a Random Number Generator over the next 14 iterations compared to the output of a RNG over the next 1000 or more.
Over 14 iterations the RNG (assuming 0-1 range), the average is likely to vary significantly from 0.5 and predicting which side of 0.5 it will fall is impossible, but over 1000 iterations, it will be very close to 0.5.
Nice! If they actually end up selling at that price at launch, I will be impressed.
I have a Vertex and while the performance has been great, it doesn't seem to be very mature compared to regular disks.
For example, I've personally had these problems with it:
1. Firmware flash tool doesn't work on all computers. Have to remove it and move it to another computer to flash it.
2. Have seen the drive lock up and basically disappear a couple of times now. Have to power down and up to recover.
I've never seen those types of issues with any other regular IDE/SATA drive (though I guess Seagate did have a few problems with some of their drives a little while ago), and I wouldn't expect the Intel drive to have those issues.
Had the Intel been available for only slightly more when I got mine, I would have opted for the Intel over the Vertex. If OCZ/Indilinx isn't able to resolve these problems, I will likely stay away from them in the future.
I suspect we'll see the 2nd gen X-25M launch at prices similar to the current X-25M, and then drop down to the $225/80GB that you can get them in 1,000 unit quantities over the next couple months.
The competition for these Intel drives is at least 2-3x behind in random IOPs. Too bad the streaming write performance didn't go up significantly, because that's the only place where the Intel drives lag behind their competition.
My rough, but optimistic calculations showed that a Corolla (an efficient compact car) which is about the same weight as a Tesla and probably similar in overall aerodynamic drag (Tesla has smaller frontal area, but is less aerodynamic than a Corolla) is at best 25% efficient overall.
Cars have been steadily getting more efficient over time, the fleet age is close to 10 years old and the fleet averages less than 20 mpg. Is it that much of a leap to suggest that the fleet is less efficient than a Corolla? Did you miss my note that the fleet is likely to consume more than 270 Wh/mile?
I keep on seeing 30% efficiency in a "typical" automobile, but I think that is being generous by measuring the engine as it's peak and not average efficiency when in use in your typical automobile.
So let's take the most efficient gas-electric hybrid on the market today, the Prius which is rated at 50mpg. Given that one gallon of gas contains 33.6 kWh of energy, we calculate that the Prius burns about 672 Wh/mile. It's commonly quoted that a typical car needs about 225-300 Wh/mile. For example, the Tesla is quoted as using 270 Wh/mile. The Prius is slightly heavier, but more aerodynamic. But the Tesla is also able to capture more energy through regenerative braking. So let's use 270 Wh/mile. Note that I'm assuming the Tesla is 100% efficient (which it isn't, but is probably around 90%) which makes my numbers below optimistic.
The puts the Prius at 40% efficient - which is absolutely the best available right now.
A Corolla which averages 30mpg - is only 25% efficient.
And the fleet average fuel economy is only 20mpg. So the fleet on average is probably only about 15-20% efficient (taking into account that the fleet probably uses more than 270 Wh/mile.
That said - your 100 kWh estimate for an ultra efficient vehicle that can go 400 miles seems to be right on target using my numbers above.
The batteries that will be going into modern EVs do not include acid and lead - they primarily include Lithium which is fairly benign. The CEO of BYD which manufactures a LOT of Lithium batteries (chances are if you have a Lithium battery for one of your electronic gadgets, it's made by BYD) went so far as to drink a glass of the electrolyte that goes into their batteries to show that they are safe (not that I'd recommend that to anyone!)
Even in NiMH batteries which is the primary type of battery currently used in hybrids today, the Nickel used is not substantial and is recyclable if/when the battery dies.
Me too. I wonder if having done a good deal of "playing" with 1/10th scale RC cars over the years has helped. Certainly looking at the batteries in the RC cars from 15 years ago, when 1.2Ah NiCd SubC cells were top of the line and now 4-5Ah NiMH SubC cells are common and Lithium cells are another 50-100% better than that depending on the chemistry and especially once you ditch the cylindrical format and look at prismatic cells.
I must admit that I do appreciate a nice sounding internal combustion engine as well. There is something very pleasing about hearing one run through the gears - can't quite put my finger on it. But I'd also much rather not hear someone else making that noise, especially when I'm at home or work concentrating!
I'm an EV fan (and would gladly take a Tesla over a Ferrari), but comparing the Tesla to a F430 is a bit like comparing apples to oranges. It's probably better to compare it to the Elise/Exige from which it's chassis is based off of (let's use the higher performance Exige model):
Tesla numbers on the left, Exige numbers on the right:
So the Exige basically wipes the Telsa in every test except for 0-60 which is a wash (where the flat torque of an electric motor pays off) and efficiency where an EV blows gasoline cars out of the water.
All the extra weight of the Tesla comes from the battery pack which weighs about 900 lbs. No doubt if the battery pack was half the weight and if the gearing was a bit higher, it'd be able to match the Exige in performance, but we've still got at least a couple years before batteries reach that kind of energy/power density.
FWIW, I'd still take the Tesla over the Exige if I could afford either. Being able to refuel every night at home is a killer feature. Not to mention being able to refuel with renewable energy and having no tailpipe emissions.
TFA isn't talking about charging your car with 365 kW at your house. At your house, you'll charge with whatever power you have (typically 240V, 30-50A or about 10 kW).
The fast charging is for when you're taking a long trip and want to get going on your way as soon as possible.
Not much, only about 7%.
Making batteries does not cause huge amounts of pollution. Where do you get your information from?
And as far as recycling rates, 98% of lead-acid batteries are recycled. Automobiles as a whole are recycled at a rate about 95%. Why would any other batteries used in future EVs be any different?
1000 amps is still a lot of current - seems like they could downsize the size of the conductor a good deal by upping the voltage going into the vehicle and converting it down to the proper voltage closer to the batteries.
Exactly. Current parallel hybrids work around conversion losses by:
1. Using a CVT to keep the engine in the most efficient operating range when it's running.
2. Using high efficiency inverters, generators and motors (2nd Gen Prius inverters/generators/motors are 90% efficient over a very wide operating range).
3. Recovering energy any time your foot is off the gas pedal and the car is moving.
A serial hybrid has the drawback of requiring the motor/batteries capable of producing maximum propulsion power which significant increases their cost. In today's hybrids maximum load is shared by the engine.
A serial hybrid makes the most sense when you are planning on using a large battery pack and plugging it in to avoid running the petrol engine in the first place.
Fuel efficiency is only really comparable on the highway. Hybrids tend to blow diesels away in the city. So yeah, if you do a lot of highway driving, a clean diesel is a good choice. You can even get one in the US now - the Jetta. They've been selling very well.
Just don't forget that diesel has an inherent ~15% advantage over gasoline cars because of the higher energy (and CO2) content per gallon.
Diesel also enjoys a hefty price advantage over in Europe over gasoline. Diesel is slightly cheaper than gas right now in the US, but it hasn't always been.
BTW, not only is soot a harmful byproduct from diesels, NOx emissions also tend to be significantly higher. Only recently in the US have soot and NOx emissions from diesels sold in the US come close to matching their gasoline powered counterparts.
Seems like your mileage may vary as with all cars. With hybrids, it seems to vary a bit more as there were more than a few comments where people are regularly exceeding the EPA numbers.
Looking at the customer entered numbers at fueleconomy.gov seems to indicate that they aren't alone.
Know of any other 3700+lb cars that can fairly easily beat 40mpg? There aren't even that many small cars that can do that!