DescentBB

Should I try it? I'd like to try it just to see if it speeds up at all, but if it's potentially harmful, I'll steer clear.

I believe iPods and most other MP3 players that show up as mass storage devices in windows use the fat32 file system and can be defragmented if you want. Odds are it isn't that badly fragmented though. Also if it is a flash based player then defragmenting is irrelevant, it only applies to the hard disk based players.

Maybe I'll back it up and give it a shot...

there really isn't any point.

you can't directly access the area that stores the music on the ipod, and fragmentation on external drives isn't an issues that would demand attention.

fliptw wrote:you can't directly access the area that stores the music on the ipod

Huh? The music is in the (hidden) iPod_Control folder.

Krom wrote:I believe iPods and most other MP3 players that show up as mass storage devices in windows use the fat32 file system and can be defragmented if you want. Odds are it isn't that badly fragmented though. Also if it is a flash based player then defragmenting is irrelevant, it only applies to the hard disk based players.

And trying to defragment flash-based players would really wear the chips down (wear levelling).

There is no \"seeking time\" on flash memory, so defragmentation doesn't make much sense.

Anyway the address space of the device controller does not nessecarily reflect the physical ordering.

My iPod is a hard drive based iPod. 5g, 30gb.

heftig wrote:

Krom wrote:I believe iPods and most other MP3 players that show up as mass storage devices in windows use the fat32 file system and can be defragmented if you want. Odds are it isn't that badly fragmented though. Also if it is a flash based player then defragmenting is irrelevant, it only applies to the hard disk based players.

And trying to defragment flash-based players would really wear the chips down (wear levelling).

Munk wrote:There is no "seeking time" on flash memory, so defragmentation doesn't make much sense.

Anyway the address space of the device controller does not nessecarily reflect the physical ordering.

So much knowledge yet lacking some very important "practical" understanding.

Guys, if it's flash memory based or hard disk based there is one thing that consistently holds true: Computer based disk storage is really pretty "brain dead" because it uses very little if any planning when storing a file. When storing a file, the computer device uses the first available sector regardless of how many consecutive sectors are available until either it finishes storing the file or comes to an already in use sector. If it comes to an already in use sector then it drops a "marker/pointer" and then skips to the next available sector and continues storing the file. This continues until the file is stored. Adding files and then removing others results in sectors between files becoming available. This is how fragmentation occurs. The file is never stored on a unfragmented block of sectors unless 1) the first available block of sectors is large for the entire file in one block, or 2) the programing for storing the file is specifically programmed to force the file to be stored in one block sectors without any fragmentation (very, very, very few programs if any are designed in this fashion). Therefore, fragmentation occurs on hard disks, flash memory, thumb drives, iPods, and even the RAM in your computer.

However, defragmenting only provides noticable performance benefit for disk based storage devices and slower storage mediums. Defragmenting can improve performance on some rather slow thumb drives, but not many.

So why doesn't defragmentation show much benefit on the flash memory devices (and devices like them)? Because reading/writing to memory chips is extremely fast when compared to disk based systems. The performance hit of accessing a fragmented file in flash memory is miniscule when compared to the performance hit when accessing a fragmented file on a disk based system. ... but the fragmentation still exists.

On any device fragmentation can lead to errors when the fragmentation reaches extreme levels. On flash memory devices the extreme level is much higher than disk based systems. That is another reason why defragging a flash memory device can seem irrelevant.

On most computer systems, if you can access the device as a drive, then you should be able to defrag it. (insert appropriate disclaimer, basically not all devices are designed the same).

The all-time best and most effective defrag method is to completely remove the files from the devices and then copy the files back on. This method will give you the best defrag, but depending on how the interface of your device is designed you may or may not be able to do that or it may be totally impractical to do it. That is why programs to effect a defrag exist for your computers because your OS usually can't be copied off and back on with any with any ease.

As for "wearing the chips down" ... that isn't the case. Chips don't "wear down", they either work or don't work (like flipping a switch on and off). Use them more frequently and yes it will probably fail sooner. But "wear down"? Nope. Go learn more about how electronics work.

Yes, there is "seek time" with flash devices, but it's extremely small.

Last of all, lest I forget...
Krom is correct that your iPod or MP3 player is probably not fragmented much at all... unless you've been adding and removing files a lot. Even then, the speed that chips read/write data make it nearly irrelevant. The disk based players will see the benefit much more than chip based players.

TechPro wrote:2) the programing for storing the file is specifically programmed to force the file to be stored in one block sectors without any fragmentation (very, very, very few programs if any are designed in this fashion).

Actually, pretty much all modern file systems do this. For example exFAT, HPFS and NTFS. They all try to store the files without fragmentation, unless that is not possible.

TechPro wrote:Defragmenting can improve performance on some rather slow thumb drives, but not many.

More precisely, HDD thumb drives. But those things are rare.

USB 1.1 thumb drives are slow. That doesn't mean they would benefit from defragmentation.

TechPro wrote:On any device fragmentation can lead to errors when the fragmentation reaches extreme levels. On flash memory devices the extreme level is much higher than disk based systems. That is another reason why defragging a flash memory device can seem irrelevant.

I fail to see how disk fragmentation could lead to errors.

Possibly if the maximum number of extents per file was to be exceeded. I could not find any information on this, so I assume this number is high enough not to be exceeded in any realistic scenario.

Even if, this would completely independent of the memory device type. Fragmentation is a problem of the file system, not the hardware itself.

TechPro wrote:The all-time best and most effective defrag method is to completely remove the files from the devices and then copy the files back on. This method will give you the best defrag, but depending on how the interface of your device is designed you may or may not be able to do that or it may be totally impractical to do it. That is why programs to effect a defrag exist for your computers because your OS usually can't be copied off and back on with any with any ease.

Actually, modern defragmentation programs can do this just fine, down to completely reordering the files on the disk for most efficient access (you won't be a able to do this by simply copying). The only thing they can't move are locked files.

TechPro wrote: As for "wearing the chips down" ... that isn't the case. Chips don't "wear down", they either work or don't work (like flipping a switch on and off). Use them more frequently and yes it will probably fail sooner. But "wear down"? Nope.
Go learn more about how electronics work.

Each EEPROM and flash media segment can fail on its own, usually after 10000 to 1000000 erase cycles. The whole chip will not simply stop working.

Possibly the media has got spare segments that are used instead of failed segments (just like hard drives have got spare sectors).

TechPro wrote:Yes, there is "seek time" with flash devices, but it's extremely small.

I don't think flash devices have a "seek time". There might be a delay before data is returned, but it makes absolutely no difference how the data is distributed. Flash media hasn't got any moving parts like hard disk have. There are no heads that need to seek the data.

heftig wrote:

TechPro wrote:2) the programing for storing the file is specifically programmed to force the file to be stored in one block sectors without any fragmentation (very, very, very few programs if any are designed in this fashion).

Actually, pretty much all modern file systems do this. For example exFAT, HPFS and NTFS. They all try to store the files without fragmentation, unless that is not possible.

Not true. exFAT (designed for Windows CE where NTFS not feasible) does not prevent fragmentation. HPFS does lower the frequency of fragmentation, but it does allow it to happen anyway. NTFS was claimed by Microsoft to minimize fragmentation but any serious user of NTFS files systems knows the NTFS systems fragment (and do it a lot, that's why Microsoft included their defrag utility).

heftig wrote:

TechPro wrote:Defragmenting can improve performance on some rather slow thumb drives, but not many.

More precisely, HDD thumb drives. But those things are rare.

Nope, not limited to HDD thumb drives. I have five 1gb and bigger thumb drives that can prove you wrong.

heftig wrote:USB 1.1 thumb drives are slow. That doesn't mean they would benefit from defragmentation.

Doesn't mean they wouldn't either.

heftig wrote:

TechPro wrote:On any device fragmentation can lead to errors when the fragmentation reaches extreme levels. On flash memory devices the extreme level is much higher than disk based systems. That is another reason why defragging a flash memory device can seem irrelevant.

I fail to see how disk fragmentation could lead to errors.

Possibly if the maximum number of extents per file was to be exceeded. I could not find any information on this, so I assume this number is high enough not to be exceeded in any realistic scenario.

Even if, this would completely independent of the memory device type. Fragmentation is a problem of the file system, not the hardware itself.

I see you don't have much experience repairing end user's computers or you would have seen instances where high fragmentation leads to errors (not just delays)... So I'll let that slide.

You're right, fragmentation can be avoided through the file system and is NOT a fault of the hardware HOWEVER anytime you have hardware motion involved (like in a hard disk drive), the frequency of error is exponentially higher. This is why flash memory is almost completely free of errors due to fragmentation.

heftig wrote:

TechPro wrote:The all-time best and most effective defrag method is to completely remove the files from the devices and then copy the files back on. This method will give you the best defrag, but depending on how the interface of your device is designed you may or may not be able to do that or it may be totally impractical to do it. That is why programs to effect a defrag exist for your computers because your OS usually can't be copied off and back on with any with any ease.

Actually, modern defragmentation programs can do this just fine, down to completely reordering the files on the disk for most efficient access (you won't be a able to do this by simply copying). The only thing they can't move are locked files.

You're right that modern defragmentation programs can do a good job ... but because "locked" files also include files that are actively in use by the system (considered "locked" because they are "in use) cannot be defragmented ... That is why a moving the files off (removing them) and then copying them back is extemely effective. You can only do that without the system loaded (usually you'd use a different drive to run the system from temporarily). That's when you can get the system, locked, and "in use" files defragmented.

heftig wrote:

TechPro wrote: As for "wearing the chips down" ... that isn't the case. Chips don't "wear down", they either work or don't work (like flipping a switch on and off). Use them more frequently and yes it will probably fail sooner. But "wear down"? Nope.
Go learn more about how electronics work.

Each EEPROM and flash media segment can fail on its own, usually after 10000 to 1000000 erase cycles. The whole chip will not simply stop working.

Possibly the media has got spare segments that are used instead of failed segments (just like hard drives have got spare sectors).

A flash media segment that fails would be just like getting bad sectors on a hard disk, and in the case of flash media would be comparable to chip failure. And yes, chips DO fail and simply stop working.

heftig wrote:

TechPro wrote:Yes, there is "seek time" with flash devices, but it's extremely small.

I don't think flash devices have a "seek time". There might be a delay before data is returned, but it makes absolutely no difference how the data is distributed. Flash media hasn't got any moving parts like hard disk have. There are no heads that need to seek the data.

"Seek time" is simply the time it takes to seek for the requested data. It does not matter if it's all electronic or a hard disk drive. There is still a measurable amount of "seek time". In hard disk drives, the seek time is much much longer because it includes the time to move the read/write heads in the drive. Since flash memory does not have any mechanical parts, the time to seek the data is very, very small, to the point that is seems to not exist. It still exists.

OK, I'm done pissin' now.

TechPro wrote:Not true. exFAT (designed for Windows CE where NTFS not feasible) does not prevent fragmentation. HPFS does lower the frequency of fragmentation, but it does allow it to happen anyway. NTFS was claimed by Microsoft to minimize fragmentation but any serious user of NTFS files systems knows the NTFS systems fragment (and do it a lot, that's why Microsoft included their defrag utility).

They do try to store files as a single fragment. No "brain dead" writing to the first cluster found.

TechPro wrote:I see you don't have much experience repairing end user's computers or you would have seen instances where high fragmentation leads to errors (not just delays)... So I'll let that slide.

You're right, fragmentation can be avoided through the file system and is NOT a fault of the hardware HOWEVER anytime you have hardware motion involved (like in a hard disk drive), the frequency of error is exponentially higher. This is why flash memory is almost completely free of errors due to fragmentation.

But that would imply the disk is damaged, wouldn't it? If not, where would the errors come from?

Eek. All I really wanted was a \"Yes, go ahead, it won't hurt,\" or a \"No, avoid it, the iPod will explode\" really.

TechPro wrote:I see you don't have much experience repairing end user's computers or you would have seen instances where high fragmentation leads to errors (not just delays)... So I'll let that slide.

Short of a mechanism explaining how these errors arise, this isn't very convincing. What kind of errors are you talking about anyways? And on which file systems?

Only errors I have seen from fragmentation are buffer underrun errors on old cd burners.

Although severe fragmentation could theoretically cause problems with some time critical programs during heavy disk activity if they can't get some required chunk of data fast enough. But then again that situation would only happen in heavy multitasking and even a perfectly defragmented hard drive would still have the same problem.

Also techpro, there are programs available that can defrag at boot time so the system can fully dismount the volume and defrag EVERY file, even the MFT and page files. No need to manually copy or disk clone the drive, since that takes too long and isn't as effective as you would think anyway.

a fragmented drive thats low on free space might expose errors from manufacturing... I'd guess.

If the file system was in an inconsistent state or otherwise damaged(fatXX anyone?) would make defrag apps throw hissy fits - they hate broken file systems.

I've seen computers that have not once been defragmented after as many as four years of use, and other than being slow there have never been any errors that could be attributed to fragmentation.

So yes or no? That's really all I need.

I do transfer files to and from it a lot, if that's any helpful info.

no.

You might want to from time to time (like every third blue moon or something like that) ...

If it's an iPod ('classic' design) then it has a hard disk ... yes, you would be wise to defrag it from time to time, but it does not need be frequent at all.

The iPod Nano, iPod Shuffle, and iPod Touch are all flash memory based. A defrag would provide hardly any benefit at all, so little you will probably not even be able to measure it. Don't worry about it. You don't need to defrag them.

Close enough to the Yes/No ?

Linky to iPod specs

Yes. Thank you. I hope Apple comes out with a 30gb flash-based iPod. Longer life, less boot time, and better battery performance.