Just hope that you got big enough car to fit a bicycle in it (or attach to it safely), because you won't afford gas to go to work, and you still need to goto gym and car:)
Well, i know gas is way cheaper in US (~1/4th of price here) but still.
cover the plaza with solar panels:D:D Use the solar panel generated electricity half for ACs and whatever else is needed on the plaza, rest sold back to network;)
If you are worth your weight as a developer, you've already done model isolation layer where your all queries would be, thus it's not that hard to rewrite the queries. If this was to be expected, you've made it far simpler already.
In any case, i don't see the anti-MySQL points. I've tried Postgres once - that was enough, i'm not going back to it. It was weird as shit, required some weird conundrums for permissions and DBs changes, didn't seem to be properly isolating but more like hacked together to support more than 1 DB, with 1 set of perms per server - no, i don't mean it didn't support, that's just the way it felt.
If you got 2 choices, otherwise equal for what you need now - always choose the simpler one. Postgres definitely is not the simpler one.
- Insert Windows 7 Pro/Enterprise/Whatever higher end blend you prefer
- Go for a coffee
- Install programss to users preference Done.:)
Windows updates works seamlessly on the BG, only thing which might be annoying is that if you have your computer on 24/7 and expect it to be all programs running, ready for use in the morning (WU might reboot). Windows provides working Firewall, and "something" defender for some malware (So install AVG/whatever you prefer, if user clicks OK/YES to everything), by default defrags nightly etc.
All those things you mention are ultimately not needed (Well, maybe use ninite to install the choice of apps).
Re:it already is almost dead due to ISP's
on
BitTorrent Turns 10
·
· Score: 1
no, that is incorrect. Most http servers actually does not limit the data rate per connection, address or by any other means neither.
If you get better speeds with multiple connections, it most likely means your connection just sucks.
You can almost certainly overvolt an electric motor, unless it's already at it's peak RPM rating. Especially true with brushless.
I would bet the motors used in HDDs would run fine for years even at 36V, assuming they are 12V. Thus tripling the RPM rating. Just avoid stopping & starting the platters often (highest peak of power used). Question is can platters do it. HDDs are meant to work for years upon years, so they should be working at rather low end of their potential capability, in terms of wear & tear plus heat. Heat is the killer for electric motors.
I guess, at it's simplest just increase the motor voltage might actually work. Unless the controller electronics are relying on being at exactly correct RPM rating.
Could be even rather simply to test, regularly PSU supplies upto 12.35V, change this to car battery charger which will give you 13.6V or thereabouts, see what happens:)
Motor replacement is not viable option, unless you got a lab to work in.
Replacing with 15k RPM hardware -> why not just use the 15k RPM drive then?;) (well platter size...)
RAID1 requires n+n drives, not n+n. RAID5 is n+1, RAID6 is n+2. All HW raid solutions except RAID0 has serious performance degradation, even most of the time HW RAID0 gives marginal benefit.
SW raid (Linux) RAID0 gives *almost* linear performance scale upto 4 drives at least. RAID10 in Linux works in a sane way vs. HW RAID10, it actually gives almost same performance benefits as RAID0, while offering the mirrored redundancy.
RAID1, RAID5, RAID6 in simplified terms could offer better performance than single drive, but the performance is just way too degraded due to techniques involved. If i recall right RAID1 is synchronous, so both drives to need it written before going on for next thing, and reading won't be spread over both disks neither.
Sure, they give some benefit, but RAID0 actually gives performance.
In your case, you need RAID0 type of arrangement for the managing of the heads actually, if you want tto see actual performance gains worth anything.
Now you ask, why HW Raid is being sold for enterprises for very high cost, if they have weak performance? To skim more HW sales. HP & Dell servers with more than 4 drives have the way weakest I/O performance i've seen. HP Enterprise drives weak performance is only rivalled by Seagate ES.2. Their designs are intentionally made to weaken the performance. ie. 6 drive chassis uses only a single S-ATA link, 12 drive chassis atleast used 2. Their RAID cards are also insanely slow. RAID10 is made the way it gives you no perf. benefit (compare linux SW RAID10).
That way they can sell you an reaaally expensive SAN, reeeaaaally expensive SAS 15k 2.5" drives with SSD like capacities at even higher price. Still you are being capped most likely by the transport layer to quite slow speeds.
Oh, and the best 4 hdd performance i've seen to date using 7200rpm drives, was on SW RAID0/RAID10, on a computer build on cheapest AMD HW for that performance level. Same drives on a intel machine performed weaker, tho that might just be because it was a HP server. (Phenom X4 vs. Quad Xeon).
Weakest atleast 2HDD performance was on a HP 12 drive server. It was even weaker than a 6 drive server. It had 12x2Tb drives, and it performed weaker than a cheap Core2Duo with 2x750Gb Cudas! The DC kept on looking into it for a week, updated all firmware, tried different OSs etc. but performance still sucked. They ended up giving me several machines in replacement for the same price (Same CPUs, more ram on each, 4 drives on each).
Best Price / Performance is with 750Gb 7200.12 Cuda's by a huge margin, but you can't get them anywhere anymore (new that is). 500Gb Cudas come next (unbelievably yes). For larger your only option is WD Blacks. Altho, i do hear Hitachi Deskstars do give a decent performance too.
Don't believe reviews btw. 250Gb WD Blue was actually supposed to have damn good perf/cost ratio, but they ended up costing about 80-90% of a 500Gb Cuda, with 60% the performance, with way higher variation and higher degradation depending upon position in the drive. 500Gb Blue & Black i still got to test.
We got hundreds of 500Gb & 750Gb Barracudas online. Annual failure rate is about 4%, but did peak at 7.38% at one point.
Larger seagates are the worst drives ever, starting from 1Tb. Their sustained maximum contiguous read spead is about 22M/s only, if you are lucky! (We've tried ES.2 and Constellation only if i recall right, both high end drives, Constellation meant for enterprise only, and ES.2 if i recall right was basicly 'cuda meant for RAID required envs like video editing, for which their sustained rate is far too low)
So counting for 0% err correction loss (in real world atleast 10%), maximum speed constantly, meaning every receiving end SMTP answers fast etc. means that each message was 132.71 bytes on average. round it to 133bytes, you don't have room for much, even with from a@b.com to b@b.com and subject a you use 40bytes...
Now that's not real world, real world is more like sender paul@warner,com, recipient mark28@aol.com subject Great Deals, takes 65 bytes.
So you have 68 bytes for your message. And that is with 0% error correct. More likely your maximum budget is 112.8bytes average per message, or 3.06k/s.
So i'm calling you on your load of shit:)
Maybe, just maybe, 25k messages an hour, leaving a budget of about 440 bytes per message.
Now, every mailer uses a queue spool and discreet agents to do the actual sending, if i recall right. Well atleast in a case sending cannot be done immediately it's first queued up. That would mean 27.7 seeks average per second for writing. 36.1ms per seek + write of the max 200bytes. Then another for read. At least that portion can be done on that era HW:)
Bloody Finland and medical care which does not work, roads which are full of potholes without warnings, great distances and very expensive gas, (immensively high taxes in multiple layers, food which costs too much for joe average to get anything better than the cheapest, public transportation which takes hours for relatively short distances to use,) witty intelligent politicians who gives away our tax money, smoking hot Johanna Tukiainen
We operate hundreds of Seagate drives and our annual failure rate is rather low. WD drives are giving grief tho dropping constantly out of RAID etc.
Almost every seagate drive that fails is years old already, and been in 24/7 usage in a server.
Hell, i even got some seagates getting quite an abuse on RAID, and still no failures despite they get occasionally kicked, are stacked on top of each other with only mounting being the cabling etc. Tho i am expecting them to fail at ANYTIME, they are getting that bad of an abuse.
I would suspect your usage pattern is causing some major abuse to the drives, operating near magnetic sources? Vibration? Do not have external drives on your desk, as it's going to have multiple types of vibration (speakers, typing on keyboard etc.). Or maybe you just keep dropping them while they are running?
We operate about over 180 seagate drives ranging from older Barracuda 500Gb to ES.2 1.5Tb, and we are quite satisfied. At one point we had 7.6% annual failure rate, but it only happened with older Cudas, which were at least 2.5years old at the time or so. Quite old drives in anycase. With newer drives we have very low annual failure rate right now.
Hell, i'm expecting some failures at home tho! I got 4 Barracudas (3xfew months old 7200.12 500G, +years old 7200.11 500G which already showed signs of potential failure) just stacked on top of each other in a RAID array, only insulation being the antistatic wrappers, and the only mounting being the cabling. I even manage to occasionally bump the drives as they are on bottom shelf of my desk, and they are running at over 50C now and then. Yet, this setup has been working for the past 4months flawlessly now. New drives usually have higher failure rate (less than 6month drives have almost same failure rate as 3+ yo drives due to manufacturing defects). What i'm trying to say: They can take quite an abuse too, and still work reliably.
Maybe i'm lucky, but few hundred drives lucky?...
WD drives we have tho... Quite different story! They keep dropping out of RAID with no apparent reason, and failure rate seems higher. We don't have WD drives in the hundreds tho, so there's no proper statistical probability to be calculated here. Also we have only 2Tb models in use.
I got my hands on a few Hitachi 3Tb drives for testing, anxious to get to see how they perform!
It would make their live EASIER. One step is removed: That is determining passwords from the hash. I would say that makes their lives significantly easier.
you are speaking downsides of a carb. Fuel injection (which is computerized) solves all of those problems.
With a cable you can choose not to stick it directly to bottom but actually let the engine "wake up" even with carb.
Changing out air filter does jack shit, unless the one being replaced is overtly dirty. The stock air filter is often better than any "freeflow aftermarket" air filter by itself as the only modification. Air filter if designed properly at factory is NOT a restrictor, quite the contrary. For example in a few V8 BMWs we put in otherwise stock engine a huge aftermarket K&N cone filter, end result was less power, higher consumption. Only nicer sounds for those who enjoy intake sounds. Stock air filter was replaced quite soon. Same thing happens to most other cars as well. The problem is that most common installation changes from cold air intake to sucking in hot air, or changes the total length of intake by shortening it, thus moving power more to the higher revs.
Whole exhaust, can do difference, but may actually again make things worse. It depends heavily on the exhaust manifold.
To get good power out of otherwise stock engine, an "more complete" solution is needed, in generally something like: * larger exhaust. For higher powered 1.6L engine: 2", 2L engine: 2.5", 3L engine 3.5" and so forth. Rough figure. With lower power output, decrease the size a little bit. (Higher velocity of exhaust gas, thus causing sucking effect) * Free flow intake filter ONLY if there is a reason to why stock might be restricting, and ONLY if you can make it get ONLY cold air (use the stock filter casing, or build a "intake box") OR in case stock is sucking in hot air but only when including cold air intake * New HIGH QUALITY spark plugs (ie. Bosch Super4 for an regular engine) * High quality new oil * Ignition timing * Possible new spark plug wires, and other wearing parts of ignition * New oil filter ** Checking proper engine electrical grounding (All the way from battery to engine. Preferrably one grounding wire to block, and one to cylinder head) * Coolant, with glycol * Valve clearance alignment (only if engine needs one) * Possibly better ignition coil (for stronger spark. Depends upon the engine's limitations)
along with HEAVILY REVVING the engine after doing everything that. You need multiple 100% load accelerations to clear some of the crap accumulated, the cylinder pressure & heat has to go very high temporally. So regularly drive the engine for normal operation temperature. Then full acceleration and really quick gear change, all the way to rev limiter on 2nd and then let it "engine brake" for a while (say if rev limit at 6½k RPM, engine brake down until 5k RPM), cruise for a moment on higher gear (like normal driving), and redo multiple times. Also do for higher gears if possible somewhere in your vicinity. The engine braking portion is also crucial, it will cause very low pressure in the cylinder. It should clear some of the soot/deposits. The danger here is if the car is from longtime "woman's use", and never really revved more than like 2½k for a veeery long time (like 100k) very large pieces of soot might get loose and cause valve damage, if that happens, the engine was pretty much needing a complete overhaul in any case. Some of the deposits should clear from cylinder and exhaust ports like this, but only often done high revving will have a good effect. A nice side effect is that it might help the piston rings to seal better if the engine is rarely revved high.
More exhausting solution is to remove cylinder head, intake, exhaust, clean it all up, realign exhaust and intake to cylinder head, get valve seams worked with multi-angle seams, clean up everything properly, change all gaskets, seals etc. and the engine easily becomes better than new (Depending upon mass mfg tolerances used). At that time you can also opt for better flowing intake or exhaust manifold of small size, not racing, but properly built fo
Easy with throttle cable: Just use the throttle differently.
You know, accelerator has more than the 3 positions you seem to think there is: full stop, full acceleration, maintain speed.
It actually has practically unlimited different positions, which all have more or less linear effect for all practical purposes on the power produced. You can start burning tires, or you can extremely slowly limp up to speed. You can also vary the speed of the change of throttle positions to have the desired effect, you can "slowly wakeup" the engine if it's one of those which respond slowly, or goto full output for the revs immediately or whatever in the middle.
Now, the electronics is doing that for you, and you are being ignorant and clueless about what is actually happening, and therefore loosing lots of potential savings in fuel consumption by not actually knowing what's going on.
Throttle cables do not snap, snag, stick neither, unless on a maintenance it was damaged, improperly installed or otherwise. Cables are actually extremely durable.
I've seen faulty cables causing snagging, or even getting stuck. But in both cases reason was that it was omitted, and in both cases the owner knew the throttle cable was damaged, so it's actually a car owner fault.
Computers have their places in cars, but interfering with driver input is not one of them.
I have owned a lot of cars, all of which had mechanical throttle cables. I'm a hobbyist, i go driving on a track, build engines, do bodywork myself etc. To date on any car i've driven or a close friend has owned a significant period of time, only one had problems with the cable: Which was broken but he decided to try to use it anyway.
Electrical throttle has given me grief many times tho. On a mondeo V6 it infact limited the performance heavily being broken. One car i drove had no feeling in the accelerator: Infact it was so bad, that the car's owner drove without shoes so that she could have even a idea if she's touching the accelerator or not. And i can see why after driving the car for a while (I could not feel the pedals either through my shoes). Other cars had laggy acceleration, or non-linear acceleration pedal effect.
I've personally never had trouble with mechanical throttle cable, but have had with electrical. Infact, the butterfly valve axle or accelerator assembly otherwise i've seen to be weaker than the cable itself (i've broken accelerator attachment assembly, and a butterfly valve axle), but breakage of those parts is caused by incorrect accelerator pedal movement limiter adjustment.
Now brake override: STUPID idea. Insanely stupid idea.
Closing throttle for traction: Again stupid idea. Any idea how hard it is to get anywhere in winter with just anykind of traction control? You've probably never driven during winter. (Oh and you can do it with simple solenoid valve if you have to, on a mechanical system)
In any case, electric accelerator introduces lots of new BREAKING, WEARING parts, lag and cost due to the extra parts. A simple potentiometer is a wearing part too. Good example is Logitech Momo, the utmost highest quality of the date for wheel + pedals for simulations and gaming. I drove enough to brake the potentiometer:)
Pretty much anything which interferes with driver control is actually quite bad idea, it takes the responsibility from driving from driver to the car, causes extra problems, and does not allow the driver to do what he/she wants. If you can't control your car without them: You should not drive a car.
All this interference with driver decisions for most part is simply because people don't care enough to learn to actually control a car. A moving object which regular weights at least a 1000kg, and moves at high speeds, thus having large kinetic energy. And people simply won't bother to learn how to brake or steer. No wonder we have so many accidents.
The interference is only a good idea for actual assisted driving, if it would be otherwise too dangerous or you could not drive. For example elderly or disabled people. But like it is now, it's mainly used to do the driving, and in itself causes average driver's to loose all idea how an car actually handles, and therefore causing innecessary accidents.
ABS systems for example are FAR FAR FAR too slow and moronic to work on slippery surface (snow, ice, bad tires on wet tarmac).
Traction control often causes a car to get stuck because the traction control removes all power from the wheels, or does not allow any slip as is required for maximal friction (a way too little known characteristic of rubber is that to a degree friction actually increases when slipping, over static friction).
So my suggestion to you: Before you start lecturing, maybe you should gather some kind of ACTUAL driving experience.
Slashdot Mirror
Just hope that you got big enough car to fit a bicycle in it (or attach to it safely), because you won't afford gas to go to work, and you still need to goto gym and car :)
Well, i know gas is way cheaper in US (~1/4th of price here) but still.
cover the plaza with solar panels :D :D ;)
Use the solar panel generated electricity half for ACs and whatever else is needed on the plaza, rest sold back to network
with mist there might be temp cooling, but the problem is that it makes the air more humid, which makes it for more than the moment even worse.
overtime the surroundings will become more humid.
However it might work, as long as new air comes in which is not that humid. :)
If you are worth your weight as a developer, you've already done model isolation layer where your all queries would be, thus it's not that hard to rewrite the queries. If this was to be expected, you've made it far simpler already.
In any case, i don't see the anti-MySQL points. I've tried Postgres once - that was enough, i'm not going back to it. It was weird as shit, required some weird conundrums for permissions and DBs changes, didn't seem to be properly isolating but more like hacked together to support more than 1 DB, with 1 set of perms per server - no, i don't mean it didn't support, that's just the way it felt.
If you got 2 choices, otherwise equal for what you need now - always choose the simpler one. Postgres definitely is not the simpler one.
I've got you an solution to all of that:
- Insert Windows 7 Pro/Enterprise/Whatever higher end blend you prefer :)
- Go for a coffee
- Install programss to users preference
Done.
Windows updates works seamlessly on the BG, only thing which might be annoying is that if you have your computer on 24/7 and expect it to be all programs running, ready for use in the morning (WU might reboot).
Windows provides working Firewall, and "something" defender for some malware (So install AVG/whatever you prefer, if user clicks OK/YES to everything), by default defrags nightly etc.
All those things you mention are ultimately not needed (Well, maybe use ninite to install the choice of apps).
no, that is incorrect. Most http servers actually does not limit the data rate per connection, address or by any other means neither.
If you get better speeds with multiple connections, it most likely means your connection just sucks.
You can almost certainly overvolt an electric motor, unless it's already at it's peak RPM rating. Especially true with brushless.
I would bet the motors used in HDDs would run fine for years even at 36V, assuming they are 12V. Thus tripling the RPM rating. Just avoid stopping & starting the platters often (highest peak of power used). Question is can platters do it. HDDs are meant to work for years upon years, so they should be working at rather low end of their potential capability, in terms of wear & tear plus heat. Heat is the killer for electric motors.
I guess, at it's simplest just increase the motor voltage might actually work. Unless the controller electronics are relying on being at exactly correct RPM rating.
Could be even rather simply to test, regularly PSU supplies upto 12.35V, change this to car battery charger which will give you 13.6V or thereabouts, see what happens :)
Motor replacement is not viable option, unless you got a lab to work in.
Replacing with 15k RPM hardware -> why not just use the 15k RPM drive then? ;) (well platter size ...)
RAID1 requires n+n drives, not n+n. RAID5 is n+1, RAID6 is n+2.
All HW raid solutions except RAID0 has serious performance degradation, even most of the time HW RAID0 gives marginal benefit.
SW raid (Linux) RAID0 gives *almost* linear performance scale upto 4 drives at least. RAID10 in Linux works in a sane way vs. HW RAID10, it actually gives almost same performance benefits as RAID0, while offering the mirrored redundancy.
RAID1, RAID5, RAID6 in simplified terms could offer better performance than single drive, but the performance is just way too degraded due to techniques involved. If i recall right RAID1 is synchronous, so both drives to need it written before going on for next thing, and reading won't be spread over both disks neither.
Sure, they give some benefit, but RAID0 actually gives performance.
In your case, you need RAID0 type of arrangement for the managing of the heads actually, if you want tto see actual performance gains worth anything.
Now you ask, why HW Raid is being sold for enterprises for very high cost, if they have weak performance? To skim more HW sales.
HP & Dell servers with more than 4 drives have the way weakest I/O performance i've seen. HP Enterprise drives weak performance is only rivalled by Seagate ES.2. Their designs are intentionally made to weaken the performance. ie. 6 drive chassis uses only a single S-ATA link, 12 drive chassis atleast used 2. Their RAID cards are also insanely slow. RAID10 is made the way it gives you no perf. benefit (compare linux SW RAID10).
That way they can sell you an reaaally expensive SAN, reeeaaaally expensive SAS 15k 2.5" drives with SSD like capacities at even higher price. Still you are being capped most likely by the transport layer to quite slow speeds.
Oh, and the best 4 hdd performance i've seen to date using 7200rpm drives, was on SW RAID0/RAID10, on a computer build on cheapest AMD HW for that performance level. Same drives on a intel machine performed weaker, tho that might just be because it was a HP server. (Phenom X4 vs. Quad Xeon).
Weakest atleast 2HDD performance was on a HP 12 drive server. It was even weaker than a 6 drive server. It had 12x2Tb drives, and it performed weaker than a cheap Core2Duo with 2x750Gb Cudas! The DC kept on looking into it for a week, updated all firmware, tried different OSs etc. but performance still sucked. They ended up giving me several machines in replacement for the same price (Same CPUs, more ram on each, 4 drives on each).
Best Price / Performance is with 750Gb 7200.12 Cuda's by a huge margin, but you can't get them anywhere anymore (new that is). 500Gb Cudas come next (unbelievably yes). For larger your only option is WD Blacks. Altho, i do hear Hitachi Deskstars do give a decent performance too.
Don't believe reviews btw. 250Gb WD Blue was actually supposed to have damn good perf/cost ratio, but they ended up costing about 80-90% of a 500Gb Cuda, with 60% the performance, with way higher variation and higher degradation depending upon position in the drive. 500Gb Blue & Black i still got to test.
We got hundreds of 500Gb & 750Gb Barracudas online. Annual failure rate is about 4%, but did peak at 7.38% at one point. Larger seagates are the worst drives ever, starting from 1Tb. Their sustained maximum contiguous read spead is about 22M/s only, if you are lucky! (We've tried ES.2 and Constellation only if i recall right, both high end drives, Constellation meant for enterprise only, and ES.2 if i recall right was basicly 'cuda meant for RAID required envs like video editing, for which their sustained rate is far too low)
1 BTC is indefinitely divisable, so that will not be a problem.
So counting for 0% err correction loss (in real world atleast 10%), maximum speed constantly, meaning every receiving end SMTP answers fast etc. means that each message was 132.71 bytes on average. round it to 133bytes, you don't have room for much, even with from a@b.com to b@b.com and subject a you use 40bytes ...
Now that's not real world, real world is more like sender paul@warner,com, recipient mark28@aol.com subject Great Deals, takes 65 bytes.
So you have 68 bytes for your message. And that is with 0% error correct. More likely your maximum budget is 112.8bytes average per message, or 3.06k/s.
So i'm calling you on your load of shit :)
Maybe, just maybe, 25k messages an hour, leaving a budget of about 440 bytes per message.
Now, every mailer uses a queue spool and discreet agents to do the actual sending, if i recall right. Well atleast in a case sending cannot be done immediately it's first queued up. That would mean 27.7 seeks average per second for writing. 36.1ms per seek + write of the max 200bytes. Then another for read. At least that portion can be done on that era HW :)
Cupcakes in pipes, what an wrapping!
Special hero heavy metal version shot in lapland: http://www.youtube.com/watch?v=_M6px6Ynm90
Freemium is often very excellent :)
Let's not forget http://www.youtube.com/watch?v=B5ZfQMOHB5k ;)
Yes, hell froze over
actually fresh water is a bit scarce in some places, such as UAE & there abouts. But yeah... Not exactly that scarce :)
It's gizmag ... Prepare to be annoyed with ads
Bloody Finland and medical care which does not work, roads which are full of potholes without warnings, great distances and very expensive gas, (immensively high taxes in multiple layers, food which costs too much for joe average to get anything better than the cheapest, public transportation which takes hours for relatively short distances to use,) witty intelligent politicians who gives away our tax money, smoking hot Johanna Tukiainen
We operate hundreds of Seagate drives and our annual failure rate is rather low. WD drives are giving grief tho dropping constantly out of RAID etc.
Almost every seagate drive that fails is years old already, and been in 24/7 usage in a server.
Hell, i even got some seagates getting quite an abuse on RAID, and still no failures despite they get occasionally kicked, are stacked on top of each other with only mounting being the cabling etc. Tho i am expecting them to fail at ANYTIME, they are getting that bad of an abuse.
I would suspect your usage pattern is causing some major abuse to the drives, operating near magnetic sources? Vibration? Do not have external drives on your desk, as it's going to have multiple types of vibration (speakers, typing on keyboard etc.). Or maybe you just keep dropping them while they are running?
well you are unfortunate.
We operate about over 180 seagate drives ranging from older Barracuda 500Gb to ES.2 1.5Tb, and we are quite satisfied. At one point we had 7.6% annual failure rate, but it only happened with older Cudas, which were at least 2.5years old at the time or so. Quite old drives in anycase. With newer drives we have very low annual failure rate right now.
Hell, i'm expecting some failures at home tho! I got 4 Barracudas (3xfew months old 7200.12 500G, +years old 7200.11 500G which already showed signs of potential failure) just stacked on top of each other in a RAID array, only insulation being the antistatic wrappers, and the only mounting being the cabling. I even manage to occasionally bump the drives as they are on bottom shelf of my desk, and they are running at over 50C now and then. Yet, this setup has been working for the past 4months flawlessly now. New drives usually have higher failure rate (less than 6month drives have almost same failure rate as 3+ yo drives due to manufacturing defects).
What i'm trying to say: They can take quite an abuse too, and still work reliably.
Maybe i'm lucky, but few hundred drives lucky? ...
WD drives we have tho... Quite different story! They keep dropping out of RAID with no apparent reason, and failure rate seems higher. We don't have WD drives in the hundreds tho, so there's no proper statistical probability to be calculated here. Also we have only 2Tb models in use.
I got my hands on a few Hitachi 3Tb drives for testing, anxious to get to see how they perform!
It would make their live EASIER. One step is removed: That is determining passwords from the hash.
I would say that makes their lives significantly easier.
you are speaking downsides of a carb. Fuel injection (which is computerized) solves all of those problems.
With a cable you can choose not to stick it directly to bottom but actually let the engine "wake up" even with carb.
Changing out air filter does jack shit, unless the one being replaced is overtly dirty. The stock air filter is often better than any "freeflow aftermarket" air filter by itself as the only modification. Air filter if designed properly at factory is NOT a restrictor, quite the contrary. For example in a few V8 BMWs we put in otherwise stock engine a huge aftermarket K&N cone filter, end result was less power, higher consumption. Only nicer sounds for those who enjoy intake sounds. Stock air filter was replaced quite soon. Same thing happens to most other cars as well.
The problem is that most common installation changes from cold air intake to sucking in hot air, or changes the total length of intake by shortening it, thus moving power more to the higher revs.
Whole exhaust, can do difference, but may actually again make things worse. It depends heavily on the exhaust manifold.
To get good power out of otherwise stock engine, an "more complete" solution is needed, in generally something like:
* larger exhaust. For higher powered 1.6L engine: 2", 2L engine: 2.5", 3L engine 3.5" and so forth. Rough figure. With lower power output, decrease the size a little bit. (Higher velocity of exhaust gas, thus causing sucking effect)
* Free flow intake filter ONLY if there is a reason to why stock might be restricting, and ONLY if you can make it get ONLY cold air (use the stock filter casing, or build a "intake box") OR in case stock is sucking in hot air but only when including cold air intake
* New HIGH QUALITY spark plugs (ie. Bosch Super4 for an regular engine)
* High quality new oil
* Ignition timing
* Possible new spark plug wires, and other wearing parts of ignition
* New oil filter
** Checking proper engine electrical grounding (All the way from battery to engine. Preferrably one grounding wire to block, and one to cylinder head)
* Coolant, with glycol
* Valve clearance alignment (only if engine needs one)
* Possibly better ignition coil (for stronger spark. Depends upon the engine's limitations)
along with HEAVILY REVVING the engine after doing everything that. You need multiple 100% load accelerations to clear some of the crap accumulated, the cylinder pressure & heat has to go very high temporally.
So regularly drive the engine for normal operation temperature. Then full acceleration and really quick gear change, all the way to rev limiter on 2nd and then let it "engine brake" for a while (say if rev limit at 6½k RPM, engine brake down until 5k RPM), cruise for a moment on higher gear (like normal driving), and redo multiple times. Also do for higher gears if possible somewhere in your vicinity.
The engine braking portion is also crucial, it will cause very low pressure in the cylinder.
It should clear some of the soot/deposits. The danger here is if the car is from longtime "woman's use", and never really revved more than like 2½k for a veeery long time (like 100k) very large pieces of soot might get loose and cause valve damage, if that happens, the engine was pretty much needing a complete overhaul in any case.
Some of the deposits should clear from cylinder and exhaust ports like this, but only often done high revving will have a good effect. A nice side effect is that it might help the piston rings to seal better if the engine is rarely revved high.
More exhausting solution is to remove cylinder head, intake, exhaust, clean it all up, realign exhaust and intake to cylinder head, get valve seams worked with multi-angle seams, clean up everything properly, change all gaskets, seals etc. and the engine easily becomes better than new (Depending upon mass mfg tolerances used). At that time you can also opt for better flowing intake or exhaust manifold of small size, not racing, but properly built fo
Easy with throttle cable: Just use the throttle differently.
You know, accelerator has more than the 3 positions you seem to think there is: full stop, full acceleration, maintain speed.
It actually has practically unlimited different positions, which all have more or less linear effect for all practical purposes on the power produced. You can start burning tires, or you can extremely slowly limp up to speed.
You can also vary the speed of the change of throttle positions to have the desired effect, you can "slowly wakeup" the engine if it's one of those which respond slowly, or goto full output for the revs immediately or whatever in the middle.
Now, the electronics is doing that for you, and you are being ignorant and clueless about what is actually happening, and therefore loosing lots of potential savings in fuel consumption by not actually knowing what's going on.
Can do that with cable: Driver decides.
Throttle cables do not snap, snag, stick neither, unless on a maintenance it was damaged, improperly installed or otherwise. Cables are actually extremely durable.
I've seen faulty cables causing snagging, or even getting stuck. But in both cases reason was that it was omitted, and in both cases the owner knew the throttle cable was damaged, so it's actually a car owner fault.
Computers have their places in cars, but interfering with driver input is not one of them.
I have owned a lot of cars, all of which had mechanical throttle cables. I'm a hobbyist, i go driving on a track, build engines, do bodywork myself etc.
To date on any car i've driven or a close friend has owned a significant period of time, only one had problems with the cable: Which was broken but he decided to try to use it anyway.
Electrical throttle has given me grief many times tho. On a mondeo V6 it infact limited the performance heavily being broken. One car i drove had no feeling in the accelerator: Infact it was so bad, that the car's owner drove without shoes so that she could have even a idea if she's touching the accelerator or not. And i can see why after driving the car for a while (I could not feel the pedals either through my shoes). Other cars had laggy acceleration, or non-linear acceleration pedal effect.
I've personally never had trouble with mechanical throttle cable, but have had with electrical. Infact, the butterfly valve axle or accelerator assembly otherwise i've seen to be weaker than the cable itself (i've broken accelerator attachment assembly, and a butterfly valve axle), but breakage of those parts is caused by incorrect accelerator pedal movement limiter adjustment.
Now brake override: STUPID idea. Insanely stupid idea.
Closing throttle for traction: Again stupid idea. Any idea how hard it is to get anywhere in winter with just anykind of traction control? You've probably never driven during winter. (Oh and you can do it with simple solenoid valve if you have to, on a mechanical system)
In any case, electric accelerator introduces lots of new BREAKING, WEARING parts, lag and cost due to the extra parts. A simple potentiometer is a wearing part too. Good example is Logitech Momo, the utmost highest quality of the date for wheel + pedals for simulations and gaming. I drove enough to brake the potentiometer :)
Pretty much anything which interferes with driver control is actually quite bad idea, it takes the responsibility from driving from driver to the car, causes extra problems, and does not allow the driver to do what he/she wants. If you can't control your car without them: You should not drive a car.
All this interference with driver decisions for most part is simply because people don't care enough to learn to actually control a car. A moving object which regular weights at least a 1000kg, and moves at high speeds, thus having large kinetic energy. And people simply won't bother to learn how to brake or steer. No wonder we have so many accidents.
The interference is only a good idea for actual assisted driving, if it would be otherwise too dangerous or you could not drive. For example elderly or disabled people. But like it is now, it's mainly used to do the driving, and in itself causes average driver's to loose all idea how an car actually handles, and therefore causing innecessary accidents.
ABS systems for example are FAR FAR FAR too slow and moronic to work on slippery surface (snow, ice, bad tires on wet tarmac).
Traction control often causes a car to get stuck because the traction control removes all power from the wheels, or does not allow any slip as is required for maximal friction (a way too little known characteristic of rubber is that to a degree friction actually increases when slipping, over static friction).
So my suggestion to you: Before you start lecturing, maybe you should gather some kind of ACTUAL driving experience.