Hyperdrive anyone really try it?

This is nothing new, and I doubt that their product is affordable.

Just make a ram disk and use that. Same difference.
 
Why they keep developing these pieces of hardware without finding a way of getting around the silly attachment to ATA/IDE and now even SATA interface restrictions is simply beyond me. Seems pretty stupid to create a big RAMdisk (which is pretty much what it is) and then slap a stranglehold on such a thing by forcing it to fall back to antiquated (yes I said antiquated) transfer interfaces - hell even SATA II simply pales in comparison to what a RAMdisk can do.

Someone someday will invent one of these things and it won't be hamstringed by "hard drive" interfaces. Even calling it "the future of the hard drive" is just idiotic in my opinion. Surely someone out there knows how to create such a thing, right?

For me, I'll just keep using a proper RAMdisk for various environmental variables, browser caches, temporary files, etc. It's like a Raptor on crank multiplied by the speed of light squared. :D
 
Why they keep developing these pieces of hardware without finding a way of getting around the silly attachment to ATA/IDE and now even SATA interface restrictions is simply beyond me. Seems pretty stupid to create a big RAMdisk (which is pretty much what it is) and then slap a stranglehold on such a thing by forcing it to fall back to antiquated (yes I said antiquated) transfer interfaces - hell even SATA II simply pales in comparison to what a RAMdisk can do.

Someone someday will invent one of these things and it won't be hamstringed by "hard drive" interfaces. Even calling it "the future of the hard drive" is just idiotic in my opinion. Surely someone out there knows how to create such a thing, right?

For me, I'll just keep using a proper RAMdisk for various environmental variables, browser caches, temporary files, etc. It's like a Raptor on crank multiplied by the speed of light squared. :D


Yea, seems like if someone is willing to drop $3000 (?) per every 16 gigs then they should be able to have some better type of interface.


PCI-E interface? Just it in slot, cable to box. That would be fast.

Firewire? Is that faster then SATA II?

320 UW SCSI?
 
Its funny, all over the place they keep saying its oh so faster than a raptor, when raptors are not the fastest drives you can get. Plus for a 16gb model, 2300US, on a sata interface? Common now.

Plus a overall downside of ramdrives that need to hold data, they have to be connected to a power source when the pc is powered off, (partially a imo downside)
 
Why they keep developing these pieces of hardware without finding a way of getting around the silly attachment to ATA/IDE and now even SATA interface restrictions is simply beyond me.

Because people are interested in improving the performance of the slowest component in the system, the harddrive, be it to boot a machine faster or load BF2 more quickly.
Seems pretty stupid to create a big RAMdisk (which is pretty much what it is) and then slap a stranglehold on such a thing by forcing it to fall back to antiquated (yes I said antiquated) transfer interfaces - hell even SATA II simply pales in comparison to what a RAMdisk can do.

There are two major differences between a RAMdisk and the hyperdrive/ iRAM:
  • a RAMdisk requires a RAMdisk driver, which isn't ``stock'' in Windows. As such RAMdisks cannot be used as root partitions
  • RAMdisks are usually not battery backed. a flick of the reset switch or shutting down the PC will erase the contents. This behavior makes it next to impossible to use a RAMdisk as root partitions or to improve BF2 loading times, since the data would have to be loaded each time the PC boots.

Someone someday will invent one of these things and it won't be hamstringed by "hard drive" interfaces. Even calling it "the future of the hard drive" is just idiotic in my opinion. Surely someone out there knows how to create such a thing, right?
I agree that this is certainly not the future of harddrives. And yes, it'd be somewhat smart to offer these devices with multiple SATA ports to minimize the interface limitation. Have the device give the option to use up-to four SATA ports with 1/4 of the total capacity. That was the user that wants to maximize the benefit can utilize four of his on-board SATA ports (surely people that spend $2800 on a HDD have a ``lastest gen'' motherboard) to have ~1.2 GiB/s of bandwidth to their device, which should be a piece of cake for DDR to provide.
 
It would be faster I believe it comes with a battery backup to hold data and it recommends hooking it up to a ups device to hold it up past the 6 hours it is said to hold info.

Wouldn't it be awsome to have such a thing intergrated into the mobo and have it interface at ram speed not sata speed?
 
I would love a PCI-E x16 add in card that had 16GB of DDR2-533 and a battery backup. Just dump it to hard drive once per day or per week for a backup.
 
If they gave it a faster interface and a cheaper price i'd be all over it.

Basically this is what gigabyte is doing with their iram2... but at a much more affordable price.

$2,300 for this unit with out any memory is insane for any end user, thats what? 32gb ram in 4 gig sticks... OUCH.
 
Back
Top