Spinning disks have enough rotational momentum to keep spinning (which keeps the heads floating) for long enough to park the heads via a weak spring, with zero electricity. A head crash doesn't corrupt a few sectors so much as cause catastrphic damage - that disk would likely never read another sector again. Properly-functionioning spinning disks haven't had issues with random data loss on power failure for at least a decade now.
And your impression of cheap SSDs is dead, flat wrong. They're cheap - every unnecessary part is left off to save money. And we've all (all of us who pay attention) known for years that SSDs (even some with power fail protection) will lose data (even bits which it has reported to have sync'd) on power loss.
A UPS is not enough, if you need to have your data, you need multiple layers of backup, and an SSD must have some method of writing out voltatile data (mostly internal metadata, not cache) before it shuts down.
Properly-functionioning spinning disks haven't had issues with random data loss on power failure for at least a decade now.
Source?
And your impression of cheap SSDs is dead, flat wrong. They're cheap - every unnecessary part is left off to save money. And we've all (all of us who pay attention) known for years that SSDs (even some with power fail protection) will lose data (even bits which it has reported to have sync'd) on power loss.
I think you misread what I wrote. I wrote that I would expect cheap SSDs to "survive a power loss event without being bricked." I did not write that they would retain all data, which seems to be what you are arguing against.
I have heard rumors that some cheap SSDs do not honor the SATA SYNC command. Unfortunately I do not have a reliable source for this theory, do you?
A UPS is not enough, if you need to have your data, you need multiple layers of backup, and an SSD must have some method of writing out voltatile data (mostly internal metadata, not cache) before it shuts down.
I don't think anyone is arguing that a UPS is a replacement for backups.
That bug was not as prominent as you make it sound - I was unable to reproduce the issue (or any other issue) across tens of thousands of reboots on the buggy firmware (running Linux and a quality SAS HBA.) The circumstances to produce a corruption were much more rare than simply "power loss," and many users with "safe" platforms could easily expect 0.5% AFR.
Even with "risky" OS and controller combinations, there was an element of probability involved, so most (probably almost all) power loss events would not hit the bug.
Plus the SSD320 was difficult to obtain back then, and reasonable operators upgraded to the firmware version with this bug fixed, so only a small percentage of the units were ever even vulnerable.
The "super capacitors" (almost nobody uses actual super capacitors after early models discovered that super capacitor lifetime at server temperature was inadequate) are just a low-pass filter - they usually only keep the drive online for a couple dozen milliseconds after main power goes down.
Most reasonable SSDs do not write cache at all, but thanks to the wear-leveling issues, they need to have a sector-mapping table to keep track of where each sector actually lives. That table takes many, many updates, and since it's usually stored in some form of a tree, it's expensive to save to media which does not support directly overwriting data (IE NAND, which requires a relatively long erase operation to become writable.) This table is typically what is lost during power events, and it is not usually written out when you sync a write.
So what happens is you write, sync, get an ack, lose power, reboot, and magically that sync'd data is either corrupt, or, even worse, it's regained the value it had before your last write, with no indication that there is a problem. This can cause some extremely interesting bugs.
And also make sure no UPS is ever loaded over ~40%, so when one of the pair fails the other can handle the full load. And always set them up in isolated pairs, with no equipment plugged into a pair member and a non-member, to prevent cascading failures.
I've seen a few different models of 2.5" Enterprise SAS drives go through a similar test (as a control,) and they were perfectly fine with no corruption after around 2 weeks of reboots at approximately 10 minute intervals.
It's the SSDs that I worry about - some store their own firmware on the same NAND as the User's data, and allow it to become corrupted during power-loss events. Several cheap models won't last more than a couple days before they simply drop off the SATA bus and never come back.
Sounds like you might be interested in this function that I keep in my bashrc, which prints a random line from Uncyclopedia (note, output may contain NSFW language and mangled unicode characters):
mknoise(){
curl -Ls http://uncyclopedia.wikia.com/wiki/Special:Random/ | sed -n '/<p>/,/<\/p>/p' \
| sed -e 's/<[^>]*>//g' -e '/^*$/d' -e '/article/g' | \
perl -MList::Util=shuffle -e 'print shuffle(<STDIN>);' | sed -n '/^.\{35\}/p' | head -n1
}
My issue with this is that the pit stops will be so long and variable (it's not mentioned in the article, but the driver must sprint 100m to switch cars, and a 6 point harness is non-trivial to buckle and tighten,) that it will nullify any sort of on-track advantage, which is the actual interesting part of racing.
Actually VW sold a dual clutch DSG transmission in the 2003 R32, years before the GTR hit the market, and for a fraction of the price. All other things the same, a dual clutch is not much faster than a properly-operated manually-clutched gearbox, but in the case of the GTR, we'll never be able to make that comparison.
>All other things the same, a dual clutch is not much faster than a properly-operated manually-clutched gearbox
Formula One drivers in the early 90s were the only ones really able to come close, with shift times between 150 and 200ms. The average person is more around 300 - 400ms when shifting aggressively. You can be faster than 300 - 400ms, but not with 100% consistency, especially on downshifts, and especially with some transmissions with odd clutch engagement points or shift levers that require more deliberate effort.
A modern DCT, like Porsche's 2nd generation PDK, has a shift time of 150ms. Porsche's 3rd generations are even faster -- the PDK-S in the new GT3 is 100ms. The version in the 918 Spyder is 50ms. All of these are perfectly rev-matched, every time.
I've driven many tremendously good manual transmissions. However, generation two PDK is good enough that after four test drives (two with a 6-speed, two with PDK + Sport Chrono) I've actually decided to order my Cayman S with PDK since it's unquestionably better. The only thing I wish is it had the "burnout/donut mode" of the 3rd gen.
I'm a little surprised. I figure most people buying a Cayman S are buying it for fun. I've never driven with a DCT, but most people who have that I've heard from generally say that manual is more fun, while DCT is better in every other respect.
Did you find DCT to be at least as fun as manual? I think I'd miss the feeling of nailing the revs perfectly on a downshift while braking for a turn on the track.
People buy Porsches for fun or track duty. Some for both, in which I'm in the latter crowd. I autocross my cars, and do DEs -- my current MX-5 has been on at least 13 different race courses in the country, including some of the very best (Watkins Glen, Mid-Ohio, Road Atlanta, Road America, Infineon, Laguna Seca, Sebring, VIR, Lime Rock). It's also my daily when I need to use a car to commute.
As for the Cayman S, it's a damn serious car. My load out hits 60 in 4.1-4.2s with launch control, and the 1/4 mile in around 12.6-12.7 @ 110mph (and 175mph in ~49 seconds). A set of Michelin Pilot Super Sports will get it to hold around 1.1g of lateral grip (sustained, not peak). You can't go anywhere near its limits on public roads, except in a straight line. If you want sideways fun, there's the MX-5 and BRZ/FR-S for exactly that.
In a more powerful car (trust me, 325HP is still a lot in a 2900lb car!), DCTs give you a nice quick immediate jolt from the power surging in almost immediately. It keeps a wonderful engine note (and boy does a Cayman S with sport exhaust make some noise) uninterrupted and always on boil. Once Porsche changes its PDK shift levers to those like the GT3 (pull to upshift), those who want to think they're in a racing car with a sequential box can. If I could get PDK-S in a Cayman S, I would and would mostly use the shift lever.
I find most detractors of a good DCT haven't driven a car with one. Additionally, not all manuals are great to begin with, which can be frustrating when you're trying to be quick. The MX-5 is ridiculously good for what the car cost. The Type 997 911 had a great 6-speed as well. My Lancer Evolution VIII MR had an extremely heavy clutch, and I'd say the last few BMW M cars have had a heavier clutch than I'd prefer (and dev money was mostly spent on DCTs). Corvette Z06s and Vipers (last drove a third gen/ZB) require a very deliberate use of the shift lever.
Thanks for the reply :) Hope it's a blast. The Cayman S is my #1 dream car. (I currently drive a Z4 3.0si coupe and take it up to Brainerd International once or twice a year.)
The cayman/s has very high limits that you wont reach on a public road. If you are driving it solely on public roads, the PDK is actually great. Its good in stop and go traffic, making that occasional/daily commute to the office much more pleasant.
> a dual clutch is not much faster than a properly-operated manually-clutched gearbox
That's not correct.
When the GT-R first came out I read a great article about why it was so damn quick down the 1/4 compared to super cars that are 2-6x more expensive. The reasons were:
1. Most of them dyno at around 480hp AT THE WHEELS. So it's making more power than Nissan claims.
2. It's the first car to get all the electronics (stability control, etc. etc.) good enough to actually make it faster, not just get in the way
3. The dual-clutch gearbox changes were so fast, they did a comparison of how long the GT-R is in neutral during the 1/4 compared to other supercars with regular manual gearboxes. I remember the time difference being in the .4-.8 second range. That was the #1 reason it was so fast - it spends an extra 0.5 seconds actually putting power to the road, which is obviously huge in a ~12 sec 1/4.
I suspect that a big limitation in shift speed is often the angular mass of the crankshaft and flywheel - that energy has to go somewhere in an upshift, and doing that quickly on a typical car with wide gearing and a heavy flywheel puts an incredible amount of stress on things. Gear ratios cancel out in the real world, since with close ratios you're also shifting more often.
I would expect something like the GT-R to have much less crank/flywheel angular mass than the average car, as well as beefed up transmission components. Even so, IIRC they had to dial things back a bit because transmissions were blowing up.
DCTs get rid of the delay from matching the input and output shafts of a conventional manual. Very helpful, but not a cure-all. My car (automated manual, non-DCT) spends much more time matching the engine to the input shaft than the input to the output shaft.
You seem to have ignored my "All other things the same" qualifier - that means you have to assume that power, weight (except for the difference between the gearboxes,) suspension, environment, tires, and everything else, is identical for the quoted part of my sentence to apply.
I don't see how your first two points are relevant to what I said, or your original post. You can love the GTR all you want, I don't mind, and I never claimed that there was anything wrong with it.
As for point 3, if you were to compare to a theoretical manual transmission GTR, that half a second of extra power would still only net you a couple tenths advantage, since acceleration and aerodynamic drag are nonlinear.
And I stand by my claim that a couple tenths on a 12 second quarter mile is 'not much faster.'
> I don't see how your first two points are relevant to what I said
They're not, I just wanted to include them for the curious.
> You can love the GTR all you want, I don't mind, and I never claimed that there was anything wrong with it.
It's an amazing car, but it's not like I have one or even want one.
>As for point 3, if you were to compare to a theoretical manual transmission GTR, that half a second of extra power would still only net you a couple tenths advantage, since acceleration and aerodynamic drag are nonlinear. And I stand by my claim that a couple tenths on a 12 second quarter mile is 'not much faster.'
You've never tried to build a 12 second car have you? A couple of tens is HUGE for a well behaved road car with a sound system, air conditioning and creature comforts. You could easily spend tens of thousands on a drag car in an attempt to get that. (I just looked it up, Motor Trend ran an 11.6, which is F-A-S-T)
> with a sound system, air conditioning and creature comforts
Reminds me of that episode with May saying something on the lines of "We're going at 400kmph... and we've got air-conditioning and radio". Now, that is serious engineering.
Since we're talking robo-manual transmissions, I think you guys would appreciate Koenigsegg's clever design. They use a wet clutch (as a brake) on the input shaft of their transaxle to quickly match the speed required by the next gear. End result is a lighter weight single clutch transaxle that shifts just as quickly as the DSG/PDK dual clutch designs. Let's just ignore the 1.5M price tag.
Yep, these kept going off on the (crowded) train this morning, and I thought they were some sort of fire alarm. It didn't help that everyone assumed the noise was not coming from their phone (because they'd never heard it before or found a configuration page for it,) so they mostly just let it go off while looking around for the source. The whole ride was more than a bit surreal.
In California, it doesn't matter, since it's already illegal to give speeding tickets from a manned airplane (even though they still have plenty of signs to warn you that they might do just that.)
And your impression of cheap SSDs is dead, flat wrong. They're cheap - every unnecessary part is left off to save money. And we've all (all of us who pay attention) known for years that SSDs (even some with power fail protection) will lose data (even bits which it has reported to have sync'd) on power loss.
A UPS is not enough, if you need to have your data, you need multiple layers of backup, and an SSD must have some method of writing out voltatile data (mostly internal metadata, not cache) before it shuts down.