Wow a 300 gig storage disk

[oldman420]

http://news.com.com/Maxell+focuses+on+h...aphic+storage/2100-1015_3-5973868.html



Information storage media company Maxell has said it will launch its first holographic storage products in September 2006. The first removable drive will have a capacity of 300GB and a throughput of 160mbps
This will revolutionize storage IMO.
Imagine keeping your whole system on a cd sizeed disk.

 

[patentman]

Originally posted by: oldman420
This will revolutionize storage IMO.

Storage density yes.... sotrage technology, that remains to be seen. I remind you that current hdd storage costs fractions of a cent per megabyte. Holographic storage is likely to be significantly more expensive then that for a good while until manufacturing methods etc. are perfected.

Also, while I'm not certain as to how the throughpout numbers of a conventuional hdd and holographic storage stack up, even if holographic has conventional HDD's beat in throughput, that doesn't say anything about access time. In other words, performance of holograp[hic storage might be less than a convention hdd, even though throughpuit is equivalent or better.

That said, holographic storage is cool technology and is a good replacement for tape backups in the short term. As the tech develops your prediction may ultimately prove to be true. But I say it will be at least 5-10 years before conventional HDD's start to be displaced.

 

[oldman420]

Yes I agree as to the newness of the technology but if you are using light to access multiple bits simultaneously across a 3d structure I think the bus to the system would be the biggest challenge fire wire maybe?
BTW that 160 is just the write speed the r/w performance is way higher.
this is the beginning of a bigger revolution which will end with data cubes with terabyte capacity and gigabyte throughput, if you have a larger structure you can in theory access more bits simultaneously

 

[Matthias99]

Originally posted by: oldman420
Yes I agree as to the newness of the technology but if you are using light to access multiple bits simultaneously across a 3d structure I think the bus to the system would be the biggest challenge fire wire maybe?
BTW that 160 is just the write speed the r/w performance is way higher.
this is the beginning of a bigger revolution which will end with data cubes with terabyte capacity and gigabyte throughput, if you have a larger structure you can in theory access more bits simultaneously

160Mbps is only 20MBps, and is significantly slower than most desktop hard drives (to say nothing of 10/15KRPM SCSI drives). I'm not seeing where they say the read/write performance is "way higher".

You can do a lot of things "in theory", but doing them in practice, and for a reasonable amount of money, is another story entirely. Holographic storage has been "just around the corner" for a good decade. Basically, I'll believe it when I see it on a store shelf.

 

[Gunslinger08]

It would take about 4 hours and 15 minutes to write a full disk... that's just sick.

 

[pm]

I didn't see any mention of improvements in latency in the article. Latency to optical disks has never been all that good compared to hard disks... optical disks have been an order of magnitude slower in terms of access latency compared to hard disks for as long as there have been optical drives to compare against. Without a massive improvement in access latency, I don't see it replacing hard disks any time soon.

 

[sdifox]

pm, I wonder if the access latency issue has to do with the fact that optical disks are removable and not as rigid as say a HD platter? If you are not on a fixed spindle, it is not safe to spin at high speed. Also, the lase assembly is mostly plastic, compared to the HD head assembly's metal. You cannot move the laser assembly at the same speed as the HD head when it's much weaker than the latter. HD platter is also much denser than current optical disk media. All those things increase latency.

 

[aLeoN]

So will this use less electricity and/or weigh less? I can see use in laptops..

 

[patentman]

Originally posted by: sdifox
pm, I wonder if the access latency issue has to do with the fact that optical disks are removable and not as rigid as say a HD platter? If you are not on a fixed spindle, it is not safe to spin at high speed. Also, the lase assembly is mostly plastic, compared to the HD head assembly's metal. You cannot move the laser assembly at the same speed as the HD head when it's much weaker than the latter. HD platter is also much denser than current optical disk media. All those things increase latency.

Factors influencing platter speed of hard drives has been discussed quite a bunch. See this thread, in particular the posts by klaviernista (my other alias on this site) near the bottom of the thread.

Access time in traditional (CD's, Laser discs etc.) optical media is limited by many of the same physical factors that limit the speed of hard discs, e.g. the lack of perfectly flat and balanced disc substrates that are affordable. Also, (though I'm sure lots of people in this forum already know this) data in most traditional optical media is recorded as a series of land and pits (the surface looks something like this: -_-_-_----_-_-_---....). Data is read from the disc by a laser having a defined wavelength. Basically what happens is that when the laser is shined on the disc, it is either reflected or scattered depending on whether the laser hits the bottom of a pit or the surface of a land. When the disc is spun, the reflected signal is detected by a detector and transformed into a digital signal.

Because the laser must be able to reach the bottom of the pits, the size of the pits must be at least the same if not slightly larger then the wavelength of the laser. Thus, atypical cd, which uses a 780nm red laser, will have a minimum piut diameter of 780nm or slightly above 780nm. While most people realize that this limits the storage density of conventional CD's, it also serves as a limit on the speed at which data can be read from the cd. In other words, a CD having 780nm pits and spun at 500RPM will be read slower then a CD having, say, 500nm pits that is spun at 500 RPM.

With the above in mind, the pits of a CD are equivalent to 1 bit of information, and in most modern magnetic discs (such as those used in conventional HDD), 1 grain=1 bit. Hoever, the size of the magnetic grains used in conventional HDD's is on the order of 10-50nm, almost 1/16th the size of the average pit in a CD. Thus, assuming that the detectors for optical media and magnetic media are equally efficient, a hard drive platter spun at the same rate as a conventional optical disc will be read at a much higher rate.

On another note, I don;t know where you got the idea that hard drives are more rigid then optical discs. The prevailing substrate for most hard drives and optical discs right now is made from either polycarbonate or glass. As the substrate is what imparts most of the rigidity to the disc, and the substrates of both media are typically the same, I'm not certain why you think that one is more rigid then the other. Certainly the substrates of hard discs are manufactured to more exacting standards then optical discs, but that has nothing to do with their rigidity really.

edit: and yes, I realize their are many types of polycarbonate and glass. When I say substrates of magentic and optical media are typically the same material, I mean that they are typically the "exact" same material.

 

[sdifox]

Originally posted by: patentman

Originally posted by: sdifox
pm, I wonder if the access latency issue has to do with the fact that optical disks are removable and not as rigid as say a HD platter? If you are not on a fixed spindle, it is not safe to spin at high speed. Also, the lase assembly is mostly plastic, compared to the HD head assembly's metal. You cannot move the laser assembly at the same speed as the HD head when it's much weaker than the latter. HD platter is also much denser than current optical disk media. All those things increase latency.

Factors influencing platter speed of hard drives has been discussed quite a bunch. See this thread, in particular the posts by klaviernista (my other alias on this site) near the bottom of the thread.

Access time in traditional (CD's, Laser discs etc.) optical media is limited by many of the same physical factors that limit the speed of hard discs, e.g. the lack of perfectly flat and balanced disc substrates that are affordable. Also, (though I'm sure lots of people in this forum already know this) data in most traditional optical media is recorded as a series of land and pits (the surface looks something like this: -_-_-_----_-_-_---....). Data is read from the disc by a laser having a defined wavelength. Basically what happens is that when the laser is shined on the disc, it is either reflected or scattered depending on whether the laser hits the bottom of a pit or the surface of a land. When the disc is spun, the reflected signal is detected by a detector and transformed into a digital signal.

Because the laser must be able to reach the bottom of the pits, the size of the pits must be at least the same if not slightly larger then the wavelength of the laser. Thus, atypical cd, which uses a 780nm red laser, will have a minimum piut diameter of 780nm or slightly above 780nm. While most people realize that this limits the storage density of conventional CD's, it also serves as a limit on the speed at which data can be read from the cd. In other words, a CD having 780nm pits and spun at 500RPM will be read slower then a CD having, say, 500nm pits that is spun at 500 RPM.

With the above in mind, the pits of a CD are equivalent to 1 bit of information, and in most modern magnetic discs (such as those used in conventional HDD), 1 grain=1 bit. Hoever, the size of the magnetic grains used in conventional HDD's is on the order of 10-50nm, almost 1/16th the size of the average pit in a CD. Thus, assuming that the detectors for optical media and magnetic media are equally efficient, a hard drive platter spun at the same rate as a conventional optical disc will be read at a much higher rate.

On another note, I don;t know where you got the idea that hard drives are more rigid then optical discs. The prevailing substrate for most hard drives and optical discs right now is made from either polycarbonate or glass. As the substrate is what imparts most of the rigidity to the disc, and the substrates of both media are typically the same, I'm not certain why you think that one is more rigid then the other. Certainly the substrates of hard discs are manufactured to more exacting standards then optical discs, but that has nothing to do with their rigidity really.

edit: and yes, I realize their are many types of polycarbonate and glass. When I say substrates of magentic and optical media are typically the same material, I mean that they are typically the "exact" same material.

Even if the CD and the HD platter is made of the same material, the HD is still mounted on a spindle, which increases rigility, has to be more that cd clamp. I am not aware that HDD have switched to policarbonate, I was under the impression that they are mostly glass.
Even if they are policarbonate, I sure hope they are more rigid than the ones used on CD, DVD, etc. The amount of wobble optical disc produces is crazy. It's a wonder they work at all

We are discussing latency, not transfer rate. I did say HD is much denser than optical medial.

 

[TSS]

160 mbps is fast if compared to media we have now. i belive DVD 16x is 22,xx MBps, so a bit faster, but the 1600mbps is just 1x speed.. they will find a way to speed that up.

compared to HDDs their still slow yes, but since this is really falls under next gen DVD/s or optical media, the 2 dont really compare do they? im pretty sure we will get optical HDD's one day but i'd imagine the technology beeing a bit more further along then it is right now.

what i wonder though, is why there is still 1 laser in those things? wouldnt it be an easy way to increase read rates by just adding another laser? afterall, there seem to be 2 or more of everything these days (dual-core, SLI, RAID). sure itll cost more but wouldnt the cost be justified by performance, at least at some point? a cd recorder costs about 20 bucks these days, even if a 2nd laser would double that price, i wouldnt mind that for a 100x CD player/burner.