Actually all the AVIs play in Quicktime, but for some reason the aspect ratio is wrong on some of them (16:9 plays as 4:3). Perian gives you pretty much every codec from VLC.
The problem for you is probably that you don't change the depth that you focus for purely stereographic 3D displays. If you try to focus on an object that would be further away the effect is lost.
It had uses, but I think it did complicate more than it helped.
Example of use:
I have a lot of AVIs, and a handful of them only play in Quicktime, or only play in VLC. I changed the creator code on those files so they open in specific player.
I've yet to be impressed by any of them, for any use, with any hardware.
I've yet to be impressed by your comment, which contains no reason for your opinion.
My point is there are no benchmarks given, only opinion, and bitching about other people's opinions is pointless.
Benchmarking responsiveness is pretty straight forward; anything reacting to user-input slower than ~150ms is perceived by humans as slower than real-time. Nobody seems to care enough to actually benchmark the responsiveness of any of these schedulers, so I don't see how people like Con can be taken seriously.
I really doubt it's the space requirements. The cooling system is likely going to be several times the size of the computer.
You have a couple facts wrong.
They only have 4 nodes with these drives
Each node has 16 DDR2 slots that hold 4GB sticks
They aren't maxing out the RAM slots on each node and they seem to be relying on these IO nodes to increase performance. I'd like to know how/why and this article doesn't explain anything.
SATA SSDs have write-leveling implemented in the disk controller. If the drive was primarily empty then moving the file to a new block every time it was written would be trivial. I don't read German, but unless the disk was first filled to capacity, then test wasn't very useful in determining the reliability of SSDs.
I doubt this scheduler will give much of an improvement... it's likely it will never make it into a mainline distro since it's hard to tell if it actually helps. He fails to provide any kind of benchmark for proving this scheduler is better.
You do have to give Hans Reiser credit for putting the work in to maintain his code... Reiser4 was/is one of the most cutting edge file-systems available. If he hadn't murdered his wife we might not even be looking to EXT4 and BTRFS.
"Hard drives are still the most cost-effective way of hanging on to data," Handy said. But for scientific research and financial services, the results are driven by speed, which makes SSDs makes worth the investment.
Why is the super computer ever being turned off? Why not just add more RAM?
SSD is cheaper than DDR ( ~$3/GB vs ~$8/GB ), but also ~100 times slower.
That makes sense only if you classify pain and discomfort as separate things instead of levels on a scale. What is your basis for believing you could make an animal that can be discomforted but unable to feel pain?
The other thing this does is bypass the "slow" SATA interface. We have laptop SSD drives that saturate SATA 3.0 and newer drives should be able to saturate the upcoming SATA 6.0. I don't know what kind of bandwidth is going to be available on this new flash slot, but I hope it's a LOT.
Lower bounds on SLC Flash is on the order of 100K erases. If nothing is doing write-leveling, then just changing the access-time on certain files can burn out flash drives in a relatively short time.
Having the OS write data to disk as fast as it can doesn't guarantee data integrity. If the OS crashes while a file is being written, then at least the updates to that file are lost. If the OS crashes as the journal is being written then your file system will be inconsistent and possibly toast.
If a 16GB Brainwood used a revolving cache, where any data not already in flash was read from disk and written over the oldest data in flash, then you would see very few erase cycles per day per block. You would need to do more than 16GB of disk IO to eat up one of the 100K erase cycles.
With intelligent cache techniques you should be able to get the erase-cycle count for each block very low.
I think the population growth question is probably the most interesting question for the future. Everything is centered around the next generation being bigger than the current generation, but the growth simply can't continue unless life expectancy starts plummeting. Something has going to give... I hope I live to see what it is.
I think going back to a 2,000,000,000 world population with local farming might be an option. We will need to see what kind of horrors happen when we run out of room.
That makes more sense, I'm more used to OSX than Windows, so the OSXish interface on Windows's iTunes didn't bother me. Running a Windows apps in OSX via Wine is harsh...
How about people actually have to prove they can operate a vehicle before operating it in public! Forcing people to retake driving tests every few years isn't excessive government intervention.
But seriously, people need to be careful not to be run-over anytime they are near a public place... we have a LOT of drunks, old people, texters, and asshats on the road.
Slashdot Mirror
Actually all the AVIs play in Quicktime, but for some reason the aspect ratio is wrong on some of them (16:9 plays as 4:3). Perian gives you pretty much every codec from VLC.
http://perian.org/
The problem for you is probably that you don't change the depth that you focus for purely stereographic 3D displays. If you try to focus on an object that would be further away the effect is lost.
Computer generated holograms are getting better, but compressing holographic video is still a long way away. http://www.engadget.com/2007/08/31/researchers-develop-a-360-degree-holographic-display/
Ya, I finally read the article and realized you can still do the "open with" thing... I always figured that it just changed the creator code.
Anyhoo, this is definitely a better setup, but where is the "open with" info/metadata stored?
It had uses, but I think it did complicate more than it helped.
Example of use:
I have a lot of AVIs, and a handful of them only play in Quicktime, or only play in VLC. I changed the creator code on those files so they open in specific player.
Long-running simulations can run completely awry if one of the DIMMs dies part-way in.
Being able to record snapshots for later reuse or verification helps ensure the correctness of the simulation.
Sure, but mechanical disks make more sense for storing snapshots. They have 768GB of RAM and 4TB of SSDs in their cluster.
I've yet to be impressed by any of them, for any use, with any hardware.
I've yet to be impressed by your comment, which contains no reason for your opinion.
My point is there are no benchmarks given, only opinion, and bitching about other people's opinions is pointless.
Benchmarking responsiveness is pretty straight forward; anything reacting to user-input slower than ~150ms is perceived by humans as slower than real-time. Nobody seems to care enough to actually benchmark the responsiveness of any of these schedulers, so I don't see how people like Con can be taken seriously.
You have a couple facts wrong.
They aren't maxing out the RAM slots on each node and they seem to be relying on these IO nodes to increase performance. I'd like to know how/why and this article doesn't explain anything.
DRAM/DDR drives aren't anything new; hooking up 1TB of DDR would be expensive, but so is 1TB of X25-E drives
http://techreport.com/articles.x/16255/12
http://www.ddrdrive.com/
SATA SSDs have write-leveling implemented in the disk controller. If the drive was primarily empty then moving the file to a new block every time it was written would be trivial. I don't read German, but unless the disk was first filled to capacity, then test wasn't very useful in determining the reliability of SSDs.
True, I'm sure they have a valid technical reason, but the article completely fails to point out what that would be.
If the "special I/O nodes" are connected using a 10Gb network, then the 4x3Gb SATA drives would let them saturate the network bandwidth.
Con is saying that his schedulers give the user the impression that the system is more responsive without providing any hard benchmarks.
Whether a person is impressed or not is the only benchmark that Con seems to care about...
Linux on low-end hardware is not being ignored:
The next major release of Ubuntu will be faster on low-end hardware.
You do have to give Hans Reiser credit for putting the work in to maintain his code... Reiser4 was/is one of the most cutting edge file-systems available. If he hadn't murdered his wife we might not even be looking to EXT4 and BTRFS.
The article says it's using intel SSDs hooked up via SATA, which come with the regular 3 year disk warranty.
"Hard drives are still the most cost-effective way of hanging on to data," Handy said. But for scientific research and financial services, the results are driven by speed, which makes SSDs makes worth the investment.
Why is the super computer ever being turned off? Why not just add more RAM?
SSD is cheaper than DDR ( ~$3/GB vs ~$8/GB ), but also ~100 times slower.
That makes sense only if you classify pain and discomfort as separate things instead of levels on a scale. What is your basis for believing you could make an animal that can be discomforted but unable to feel pain?
Ever throught about being restrained and had someone pull your fingernails out ?
Without pain, what would keep a toddler from pulling his fingernails out? (Besides that it is pretty difficult).
The main thing this would do that battery backed up DRAM wouldn't do is allow for quick boot and hibernate, which is something the enterprise people generally don't care about. The flash looks like it will be replaceable via a dimm-like slot. http://news.cnet.com/8301-13924_3-10258748-64.html and http://www.hardware.info/en-UK/news/ymiclpqWwpyaaJY/Computex09_Intel_P55_motherboard_gallery/
The other thing this does is bypass the "slow" SATA interface. We have laptop SSD drives that saturate SATA 3.0 and newer drives should be able to saturate the upcoming SATA 6.0. I don't know what kind of bandwidth is going to be available on this new flash slot, but I hope it's a LOT.
Yes, it looks like they are planning to use a dimm-like slot.
http://www.hardware.info/en-UK/news/ymiclpqWwpyaaJY/Computex09_Intel_P55_motherboard_gallery/
Lower bounds on SLC Flash is on the order of 100K erases. If nothing is doing write-leveling, then just changing the access-time on certain files can burn out flash drives in a relatively short time.
Having the OS write data to disk as fast as it can doesn't guarantee data integrity. If the OS crashes while a file is being written, then at least the updates to that file are lost. If the OS crashes as the journal is being written then your file system will be inconsistent and possibly toast.
If a 16GB Brainwood used a revolving cache, where any data not already in flash was read from disk and written over the oldest data in flash, then you would see very few erase cycles per day per block. You would need to do more than 16GB of disk IO to eat up one of the 100K erase cycles.
With intelligent cache techniques you should be able to get the erase-cycle count for each block very low.
Reminds me of Futurama.
Fry: How can you people be so blase? Here you are in the year 3000 or so, yet you just sit around like it's the boring time I came from.
Farnsworth: Boring? Wasn't that the period when they cracked the human genome and boy bands roamed the Earth?
I think the population growth question is probably the most interesting question for the future. Everything is centered around the next generation being bigger than the current generation, but the growth simply can't continue unless life expectancy starts plummeting. Something has going to give... I hope I live to see what it is.
I think going back to a 2,000,000,000 world population with local farming might be an option. We will need to see what kind of horrors happen when we run out of room.
That makes more sense, I'm more used to OSX than Windows, so the OSXish interface on Windows's iTunes didn't bother me. Running a Windows apps in OSX via Wine is harsh...
How about people actually have to prove they can operate a vehicle before operating it in public! Forcing people to retake driving tests every few years isn't excessive government intervention.
But seriously, people need to be careful not to be run-over anytime they are near a public place... we have a LOT of drunks, old people, texters, and asshats on the road.
care to elaborate on this progress?