Become a MacRumors Supporter for $50/year with no ads, ability to filter front page stories, and private forums.
Clive At Five said:
no no no...

Low end full-size iPod: two 20 GB one-inch drives; Bigger battery.

-Clive
How about this:
Keep iPod flash as lowest-cost iPod
Make tiny iPod Micro of 4-5 GB
Turn iPod mini into 20 GB
Up reg iPods to 40, 60 GB
vPod for video 100+ GB--- These smaller hard drives will create some intensely competitive video players :eek: , although that could be good for us.... :rolleyes:
 
YAY

yay..that means i can get my higher capacity iPod mini in a couple of years..i dont want 20 gb but like 8/9 gb mini is PERFECT!
man, i wish they put one a 8gb mini in like 3 months...PPLLLLEEEAAASSEEEE.
 
SiliconAddict said:
With a 4800 RPM drive? I can't even imagine that the thing would be anything other then glacier slow. I'd be gnawing my arm off in frusteration before the OS was even loaded.

-SiliconAddict

One additional advantage of getting the bits so much closer together by standing them up on the disk is the seek and transfer times decrease inversly with the increased density. Meaning that a 4800rpm drive that has twice the density of another 4800rpm drive has about half the seek and transfer time - more bits over the head at a given time.

This has the additional benefit of not needing to up the speed, therefore preserving battery times. But what we're looking at is a 10x-100x increase in density as 100grains that are used to record a bit can be 'stood up' and therefore take 1% the space (ideal world, of course). So that can mean that it's conceivable that a drive could get the same performance as today's while running at 48rpm (barring all that space, of course).

I'd love to see the battery time on that.
 
madmaxmedia said:
I don't believe there are any HD-based camcorders (at least consumer level.)
If you include the prosumer market, there are 720p cameras from JVC under $2000 and 1080i 3-CCD cameras from Sony at around $3000 and $4500. They record in the HDV format which is editable on a computer. Basically it's High-Def MPEG-2 in a transport stream that is recorded to tape. You could wrap MPEG-4 in a transport stream and write to tape, but that's not really the same things as recording H.264 AVC HP directly on your camcorder.

The Apple products transcode HDV only in the sense that they pull out the raw MPEG-2 information from the transport stream and throw it into a new wrapper. They are not transcoding the video from MPEG-2 to some other compression scheme like H.264 as the capture from the camera and stream the data to disk.
 
max_altitude said:
Sounds awesome!!! They're getting smaller and smaller all the time, before we know it they'll be microscopic!
Unfortunately, there is a point of diminishing returns.

Even if the bit-density keeps on growing at these rates, there is a practical minimum size for a spinning platter. This is because you can't reduce the size of the spindle by the same amount you reduce the size of everything else. As the platter gets smaller, the percentage of it that is usable (not cut out to make room for the spindle) gets smaller.

I've read a few articles on the subject, and from what I gather, drives are currently at their practical lower-limit (with Toshiba's 0.85" drive.)

We probably won't see drives get any smaller than this, even though we will probably see their capacities continue to increase for at least a few more years.

Of course, if we develop something practical that doesn't involve spinning platters (like really high-density flash, or data crystals or something else) then all bets are off.
 
Diatribe said:
The thing I really want is a 300GB notebook HD and a 150GB iPod.
Because then I could go Lossless only. That would be nice.
I guess with this announcement these two are not far away. 2 years max. Niiiiiiice :D

Not unless they massively increase the ram, or you want much less battery time on the IPOD.

And really I want to have a DJ option on my ipod where I can scratch and beat match with my scroll wheel.
 
Porchland said:
Someone out there who does video tell me: How big a harddrive would a camcorder need to record an hour of SD? HD?
Others have posted typical densities for SD/DV/HDV/etc.

It's probably also worth noting that store-bought DVDs hold approximately 10G (assuming they're dual-layer.) They typically hold between 2 and 4 hours of movie (depending on lots of different production factors, including video complexity and the number/type of sound tracks.)

Today's camcorders don't provide the same compression ratio/quality as studio-produced DVDs, but I don't think it would be too much of a stretch to assume that such cameras will be available in two years when these new drives start shipping.
 
shamino said:
Others have posted typical densities for SD/DV/HDV/etc.

It's probably also worth noting that store-bought DVDs hold approximately 10G (assuming they're dual-layer.) They typically hold between 2 and 4 hours of movie (depending on lots of different production factors, including video complexity and the number/type of sound tracks.)

Today's camcorders don't provide the same compression ratio/quality as studio-produced DVDs, but I don't think it would be too much of a stretch to assume that such cameras will be available in two years when these new drives start shipping.

Also, using a newer codec will significantly change the compression/quality ratio. MPEG2 (used in DVD's) is far behind the times in terms of the quality you can achieve at any given bit rate.

So for example a hard drive-based camcorder recording in H.264 would be able to store more than 2 and 4 hours of movie on a 10GB hard drive (at quality equivalent to DVD, at least in terms of compression.)
 
patrick0brien said:
-SiliconAddict

One additional advantage of getting the bits so much closer together by standing them up on the disk is the seek and transfer times decrease inversly with the increased density. Meaning that a 4800rpm drive that has twice the density of another 4800rpm drive has about half the seek and transfer time - more bits over the head at a given time.

This has the additional benefit of not needing to up the speed, therefore preserving battery times. But what we're looking at is a 10x-100x increase in density as 100grains that are used to record a bit can be 'stood up' and therefore take 1% the space (ideal world, of course). So that can mean that it's conceivable that a drive could get the same performance as today's while running at 48rpm (barring all that space, of course).

I'd love to see the battery time on that.


Patrick, I think you're confusing some entirely different aspects here...the actual vertical stacking will cut the needed space for all grains in about half making an immediate jump possible of twice the data capacity of longitudinal recording harddisks. Then you have the aspect of interference called superparamagnetism that will be less of a problem with the vertical stacking, making another 5fold increase in density possible untill they hit the magnetic wall again. So a twofold increase in two years time and about a tenfold increase in about ten years time. I'd settle with that. But first things first; if they'd just roll that #$%& 120GB notebookdrive from their assemblylines anytime soon...
 
Seek speeds

patrick0brien said:
-SiliconAddict

One additional advantage of getting the bits so much closer together by standing them up on the disk is the seek and transfer times decrease inversly with the increased density. Meaning that a 4800rpm drive that has twice the density of another 4800rpm drive has about half the seek and transfer time - more bits over the head at a given time.

This has the additional benefit of not needing to up the speed, therefore preserving battery times. But what we're looking at is a 10x-100x increase in density as 100grains that are used to record a bit can be 'stood up' and therefore take 1% the space (ideal world, of course). So that can mean that it's conceivable that a drive could get the same performance as today's while running at 48rpm (barring all that space, of course).

I'd love to see the battery time on that.

Unfortunately that's not how disks work. While it's true that the datarate will be greatly increased by having a greater surface density, the seek times are directly proportional to disk rotation velocity.
The seek time is the time it takes for the rotating platter to reach the read-head (let's not count the time it takes for the head to change tracks on the platter for the following examples).
4800rpm is 80 rotations/sec, which means that in the worst case, the platter has to do a full rotation for the data to once again be under the head. This is 1/80 sec = 12.5ms.
5400rpm: about 11ms
7200rpm: about 8.3ms

Your 48rpm drive would have, unfortunately, a 1.25 second seek time. While the sustained transfer rate would be greater indeed (and might be useful for sustained linear reads and writes like video recording/playing) the second you had a fragment of file located elsewhere, you'd get an instant penalty in speed, and a huge one at that...
 
-minimax

I think we're burning the same candle at two ends - with different terms and angles.

-tych0

Oo. You're right, sorry about that, My mistake for including seek times. One would need to change the properties of the read/write head and slider actions to make seek times decrease over a given rotation...
 
an unexciting breakthrough

hd storage as we have known it is merely making the next logical jump in its road to obsolescence, this only means bigger storage for devices where you will never touch the drive. don't forget folks, audio professionals are many years ahead of video in adopting fancy new storage techniques and have done it much faster once again. solid state (flash) media and blu-ray are the wave of the future in video, just look at the pro audio move away from DAT towards DVD and solid state that started a few years ago. sony proved them right when they released their mpeg/imx camera that stores directly on blu-ray discs (look for compatibility in fcp5). tape feed mechanisms are expensive to manufacture and have moving (aka unreliable) parts, something that hard disks are unable to replace. filming can't grind to a halt when people need to download from their camera's drive to another source, whether you're on a film set or watching your child's graduation. not to mention that modular hard drive systems are relatively expensive to manufacture.

these hard disks are going to be used in mp3 players, laptops, and desktops. camcorders will adopt flash media or mini blu-ray for consumer use because it is exceedingly cheap to manufacture and has more than enough storage for the compressed signal most people need. it will also be easier for people to store flash disks as 'tapes' with their old footage on it; there is no market for video cameras that can't keep a copy permanently stored on a cartridge of some kind (tape, dvd, etc) and it gives the consumer far too much access to moving parts to allow them to touch a hard drive.
 
Kagetenshi said:
Wrong. That's just a lame attempt to redefine it in response to the hard drive industry being idiots; with regard to storage space, those are all based on powers of 2.

~J

No. For all units, Kilo has always meant 1000, Mega 1000000 and Giga 1000000000 in SI.
 
Kagetenshi said:
In SI, yes. Anything measured in bits has never used SI.

~J

bits, bytes are not in SI. But the IEC created the 'bi' names because of the prefixes in SI, not because 'storage industry is bad'.

The names you mentioned are all in the IEC standard, and they are powers of 10.

BTW, it's not yodabyte. It's yottabyte.
 
I am waiting for Bluetooth and 0.85" drive before I buy the iPod mini.
 
i can't wait until there is an iPod thinkwheel

all you have to do is think and it scrolls right to your song
 
weldon said:
The Apple products transcode HDV only in the sense that they pull out the raw MPEG-2 information from the transport stream and throw it into a new wrapper. They are not transcoding the video from MPEG-2 to some other compression scheme like H.264 as the capture from the camera and stream the data to disk.
The Apple Intermediate Codec is indeed a distinct compression scheme. It introduces artifacts not present on the camera original files. The quality is generally good and mostly unnoticeable in moving video, but you'll see the difference between a frame grab from AIC and a frame grab from the .m2t file.
 
alfismoney said:
Solid state (flash) media ... the wave of the future in video.

Uh-huh. Well, I'll just replace the 400+ gig in my Mac with flash media then... flash media has a long long way to go in terms of read/write speed and capacity.

You will see these drives everywhere long before flash media becomes a viable proposition for video work, esp. broadcast/HD video.
 
-And to tack on to what Blue Velvet is saying,

I have an issue with the whole blu-ray analogy as 1. I don't know of any drives you can buy yet, 2, I certainly don't know of and Writiable Blu-ray drives you can find, 3. all of the Video and Audio pros I know use Hard drives and avoid writable optical media for several reasons: (a) one must encode in-stream, unlike an, HD's ability to swap, (b) It has been discovered that writable media is not good for permanent archiving as the photographic laver fades after a few years - faster if exposed to light.

Strangely enough the most reliable media is still the Hard drive.

Additionally, most of thos afor-mentioned Video and Audio guys use RAIDs to store and edit upon, and use LaCie D2's for archiving due to the metal shell that is a very effective Farraday Cage.
 
In an iPod = Boring.
In an iPalmMac = WOW!
I want my home to be with me all the time and not have to lug around my PowerBook. Then when I want to use my MacTablet or Mac PowerBook (or heaven forbid, PowerMac) I just slip the iPalmMac into it's dock in the larger machine and we have unity!
 
Register on MacRumors! This sidebar will go away, and you'll see fewer ads.