Thanks CanadaRAM - very interesting and informative post about SSDs.
As to the Activity monitor numbers - I'm pretty sure I copied the second number wrong - it should have been 281.3 MB
Right now page ins are just over 1 MB and page outs and swap is zero.
I remember perviously that the 2.33 GB page ins really struck me as being rather high, the other two numbers were in the MB range.
But now thinking about it, the old numbers seem perfectly normal with page outs at 281 MB.
Yeah, 2 GB ins and 200 MB outs would be a normal ratio. As I mentioned, the absolute numbers mean nothing other than your machine has been running a longer or shorter time since the last reset. It'd be like saying "my odometer is 126,345 KM, do I need to fill my gas tank?" when what matters more is how long its been since your last fill.
__________________
CanadaRAM supplies RAM memory, drives and upgrades to Mac owners all over Canada http://www.canadaram.com
Thanks for the info CanadaRAM! It's certainly helping me understand a lot more.
So for a visual, the cells can act kind of like a pixel in a monitor, if you think of the image being sent as being the same as kbs being saved, where smaller pixels result in a better experience for the user, and where some die faster than others. If one dies, the others will continue to work, and the energy no longer goes to the dead one, and the picture is still spread across the entire monitor. We can see the one is not working, if we take the time to look, but it doesn't necessarily affect our overall experience until many begin to fade/die. Would that be a good analogy?
I'm wondering if anyone who has been using the air since 2010 or and SSD for a few years has experienced any of this slowing down they talk about. Or maybe someone who repairs macs for a living, if they've witnessed it at all?
Quote:
Originally Posted by eMacMan
Since we are talking Tiger on a PPC computer, disabling Stoplight altogether may significantly improve performance. You can do this in Onyx or I can post the Terminal commands.
That said, in Tiger disabling Stoplight also disables the built in search features. No reason this should be so. Earlier versions of OS X, OS 9 and even OS 8.6 all allowed content indexing, AKA Spotlight, which was disabled by default. All search parameters except file content worked just fine with the earlier systems.
If you do chose to disable it, FindAnyFile and EasyFind will do the trick though not nearly as elegantly as in Panther.
Thanks for the suggestion Ehmacman. I had Spotlight crippled as soon as I experienced it, only allowing it to search for a very limited amount of documents only. I didn't disable it entirely, because I read it would affect my Finder somehow, which I use exclusively, because it works, and I can actually find things with it, unlike Spotlight (Tiger version anyway).
P.S. I think you meant to put this in the other thread. This one is about the NEW possibilities.
__________________
12" Powerbook G4 1.5! It's small,shiney, beautiful, and it has brains too!
Tascam US-122 Interface
Fear is but a wall of illusion to that which will make you happy.
So for a visual, the cells can act kind of like a pixel in a monitor, if you think of the image being sent as being the same as kbs being saved, where smaller pixels result in a better experience for the user, and where some die faster than others. If one dies, the others will continue to work, and the energy no longer goes to the dead one, and the picture is still spread across the entire monitor. We can see the one is not working, if we take the time to look, but it doesn't necessarily affect our overall experience until many begin to fade/die. Would that be a good analogy?
Doesn't quite gel for me as an analogy -- if the controller does its job correctly, then the mapping out of bad cells is entirely invisible to the user and the operating system of the computer. The data is not lost (as it would be in a dead pixel on a screen) and there is no perceptible difference (until the day some time in the future when the controller runs out of spare cells to map into the drive space)
__________________
CanadaRAM supplies RAM memory, drives and upgrades to Mac owners all over Canada http://www.canadaram.com
The more I read about SSDs the more confused I get.
I thought by "spare cells" the SSD manufacturers meant "spare cells" so I was trying to find out how many spare cells are typically on an SSD.
Didn't get anywhere with either a hard number or a percentage when articles and discussions talk about spare cells and their use with SSDs.
But reading about it, it no sounds as if there are no actual "spare cells" as in additional cells than the advertised drive capacity - the "spare cells" are the cells not used simply because the drive is not 100% full - so If one has a 100GB drive and it's 80% full, the 20% not used are considered spare cells and will be used to replace cells that have "died"
Is that actually true; if not, why does no drive manufacturer I could find spell out the number of spare cells?
The more I read about SSDs the more confused I get.
I thought by "spare cells" the SSD manufacturers meant "spare cells" so I was trying to find out how many spare cells are typically on an SSD.
Didn't get anywhere with either a hard number or a percentage when articles and discussions talk about spare cells and their use with SSDs.
But reading about it, it no sounds as if there are no actual "spare cells" as in additional cells than the advertised drive capacity - the "spare cells" are the cells not used simply because the drive is not 100% full - so If one has a 100GB drive and it's 80% full, the 20% not used are considered spare cells and will be used to replace cells that have "died"
Is that actually true; if not, why does no drive manufacturer I could find spell out the number of spare cells?
Bottom line, don't bother worrying about it too much. And all storage drives are going to die at some point.
Both HDDs and SSDs come with "spare" block or cells, and when they get used up and nothing left to replace any bad blocks, the drive at that point going to become unreliable and replacement is basically the best option and just get along with things.
For a quick mathematical estimation and knowing how and what size the memory chips are usually manufactured, ie: a 120 GB SSD can't actually be manufactured to that size, and is most likely manufactured with a total or 128 GB capacity chips, so some basic math says it has 8 GB of "spare" cells - when new.
Maybe someone can correct me if I'm wrong.
As for a HDD's "spare" blocks I don't know and I'll bet you never thought about it.
As I understand it, some will "map out" any bad blocks, as will any Mac's Disk Utility when using the 'zero out' option, and any 'spare' blocks will be made available, but how many are available I have no idea.
All I do know is when any partitioning/formatting/zeroing-out is done and the drive is still slow and acting goofy and/or unreliably, it's time to toss the drive.
And don't forget to pull a kaput HDD drive apart before recycling it and salvage the super strong magnets that can be quite useful.
All I'm trying to figure out is if an SSD is inherently less reliable and has a shorter life than a conventional hard drive.
I have actually never had a hard drive fail on me yet, the oldest one was in a G4 tower that is 12 years old,
Capacity is the pits of course, it's also IDE, but it still works fine not that I use it very much at all.
The G4 is sitting mostly in the corner - all I use it for is to transfer old family VHS tapes onto DVDs and that's because I bought a PCI card years ago to do that.
SSDs definitely have a finite life but none of the SSD manufacturers want to talk about it.
From what I have read, it also seems that the larger the SSD gets the shorter the life since the cells get smaller.
Hmmm... maybe you could go on an adventure like Don Quixote of La Mancha and seek your answer.
All I know is that ALL data can be lost regardless of device used, and if important, must be backed up. Period!!
From my experience, the older SCSI and later PATA/IDE options were the most reliable, and with the later higher density SATA HDD and SDD and small flash thumb drives, a bit less reliable.
Bottom line IMHO, don't worry about it too much and use whatever drive source and backup you want or need for your use, and always have a backup regardless and carry on.
Life's too short as it is and not worth worrying about something that just might be just a bit "inherently less reliable"!!
And just an extra (and unfounded) thought, the SDDs are basically all electronic and with no moving mechanical parts that are usually the first to fail with HDDs I'd suggest, like all mechanical computer stuff.
I do back up regularly and I'm not worrying...I just like to know what I'm buying.
Quote:
And just an extra (and unfounded) thought, the SDDs are basically all electronic and with no moving mechanical parts that are usually the first to fail with HDDs I'd suggest, like all mechanical computer stuff.
This is what one would expect but on the three external drive failures I have had it was the bridge each time, ie the electronics, not the hard drive in the enclosure.
There was also that misconception floating around that an SSD requires a lot less power than a conventional hard drive (another ehMac thread) and it turned out that this wasn't true at all.
I'm not going to spend any more time trying to find answers about SSDs, I'm just not going to buy one any time soon.
The more I read about SSDs the more confused I get.
I thought by "spare cells" the SSD manufacturers meant "spare cells" so I was trying to find out how many spare cells are typically on an SSD.
Didn't get anywhere with either a hard number or a percentage when articles and discussions talk about spare cells and their use with SSDs.
But reading about it, it no sounds as if there are no actual "spare cells" as in additional cells than the advertised drive capacity - the "spare cells" are the cells not used simply because the drive is not 100% full - so If one has a 100GB drive and it's 80% full, the 20% not used are considered spare cells and will be used to replace cells that have "died"
Is that actually true; if not, why does no drive manufacturer I could find spell out the number of spare cells?
Whatever you are reading is inaccurate
Typical overprovisioning is 7% - 10%, so on a 120 GB drive there would be 8 GB or more of overprovisioning.
Overprovisioning is the amount of available cells beyond the advertised and formatted capacity. There may be 128 GB of chips on the SSD, but the controller is programmed to format to 120, reserving 8. You cannot use this extra space and you cannot see it.
It is not because the drive is not full. You can fill it with the whole 120 GB and you will still have 8 GB of spare cells.
If you have never had a hard drive fail in 12 years, then buy a lottery ticket, you are one of the luckiest computer owners on the planet. The average is about 3 - 4 years, some sooner, and some later. Your experience is on the far end of the Bell curve and can't be used as a precedent.
__________________
CanadaRAM supplies RAM memory, drives and upgrades to Mac owners all over Canada http://www.canadaram.com
Linear Mode
