I'm told there was also software that would drive the head stepper motor in the Commodore 1541 with varying-width pulses, to play music.
I remember programs which would play music through old dot matrix pin printers. Alternating the striking and non-striking of paper in a PCM fashion could even render very noisy sampled sound. This type of technique was also used for the PC speakers of the time (which were designed to only play tones of square wave), to play digitized sounds. This technique eventually ended up in games.
I remember first running a program written by a guy who wanted to prove the technique. It played him singing "Daisy Bell (A Bicycle Built for Two)". A song which was played by HAL in 2001 A Space Odyssey.
Re:Does not higher density mean higher risk?
on
Hard Drive Window
·
· Score: 1
Higher density means the heads fly closer to the platter and small dust particles would 'crash' the head. This means the head will hit the particle and either drag it, or bump over it. Kinda like hitting a body with your car. I wouldn't worry too much, both the heads and platters are covered with a diamond like coating. Also, any 'floating' particle would be spun off when the disk started to spin up. I once took apart a working 6G WD drive and the inside looked like my car's brake pads with all of the dust in the drive.
I have taken apart lots of drives and only ever saw one with obvious circular scuffing all around the drive. Every other drive had to my eye perfect mirror finishes.
As I have stated elsewhere, I took apart a drive with growing bad sectors and within SECONDS of being opened, the drive was DESTROYED. It initially worked and while open and spinning it died completely, returning no data at all (forever trying to read).
I would not be telling people to not worry. Every drive I have ever seen has had warning stickers which stated that opening or removal of any sticker (some cover holes in the casing) will void the warrantee. Those warnings are there for a reason! Cigarette smoke particles can kill a drive (I'm not talking about cigarette ash, I'm talking about the particles which make up the smoke!), imagine what household dust can do. And this was true for drives of 15 years ago. Modern drives have much lower flight heights, making the situation worse. If I ever accidentally scratched through or punctured a drive hole sticker, I would frantically image the drive, then zero it out and then turf it (or maybe pull it apart and keep the nice looking platters).
VooDoo 2 was a great chipset for its time. Held up with newer games well past its prime.
Actually, it didn't hold up well at all. It was severely hampered by the PCI bus and the "AGP" Voodoo cards were not being used in AGP mode, rather some PCI compatibility mode.
The fact is, that GPU focused games of the time were being made to at least run well on the older Voodoo2 cards, because there were so many out there. The game houses were targetting thier market and thus slowing development due to the Voodoo2. 3Dfx accelerated 3D game technologies and then severely slowed it down by choosing to stick with PCI.
Here, take a look at a Matrox G200 (!) kicking the shit out of a Voodoo2 SLI setup! This situation, where a MUCH slower 3D card is killing the "awesome" Voodoo2 SLI because the Voodoo2's are: 1/ PCI and 2/ storing the SAME texture data in each of the cards memory because 12MB+12MB does not give you 24MB in an SLI setup (the data MUST be mirrored, except for the frame buffer).
"S3's own `mon2.dm2' demo, using their own `newS3' map is using a whole lot of really large textures, making the scenery look very detailed and pretty. It would have been a shame if Savage3D would not score well in it, but this would almost have happened. In the latest driver I received, the `AA" setting needs to be enabled to run mon2.dm2 at a decent speed. Anyway, in mon2.dm2 the Savage3D scores higher and higher the more test runs you do. I decided to put down the result of the second run. Matrox' G200 looks very good at mon2.dm2 as well, obviously due to its AGP 2x interface. It reaches about 66% of the Savage3D scores and this without texture compression. The coolest thing are the TNT results however. TNT scores better than Savage3D although TNT does not use texture compression!!! Only with a Celeron 266 TNT looks a little slower than Savage3D, which is another proof for the CPU dependency of TNT in Quake 2. 3Dfx products have yet to learn what AGP means, so the mon2.dm2 is a serious threat to them. 22 fps scored by a $500 Voodoo2 SLI combo looks a bit sad, doesn't it? Banshee looks even worse, its `AGP-interface' only consists of the mechanical connector for the AGP slot."
I find it completely shocking that 3Dfx had the talent to design such a great video card and then completely fail to see the value of AGP over PCI. That failure ultimately killed them. It gave their competitors a window of oportunity to make lots of money to put into more R&D while 3Dfx played catch-up. 3Dfx couldn't, it was too late, they died.
Games never looked that great in 16 bit colour anyway.
It worked like that for several more months until I got bored and tried to lube the spindle with drop of WD-40.
Stiction in HDD's was usually between the heads and platters, not the spindle.
Putting WD40 on the spindle may have allowed some to flick out over the platters thanks to centrifugal force, when the heads hit this much denser than air liquid. BANG! Buggered heads and possibly some platter surface area.
Sure, a lot of people (myself included) have run old drives with the cover off when ready to chuck them anyway or in an emergency have opened a drive, repaired it and had it survive long enough to get the data off safely.
Many years ago I did this with a Brand Technologies 200MB IDE drive which had growing defects. Once I opened it, it still worked as baddly as before, but SECONDS later the whole drive was full of defects. No sector would return data. I used to work in a military clean room and not even there would I open a HDD which was working fine. The distance between the heads and the platters when moving are WAY to small for even the clean room I worked in.
This is not to say that there are not clean rooms good enough, but certainly nobodys home or office type workplace is good enough. Cigarette SMOKE particles will damage a drive. Seriously small.
the problem I was referring to is not which track you are on, but being in the "center" of a given track.
Oh right. Yes that would seem to pose a big problem. Have you heard the odd noises some of these new big drives make periodically? I bought 2 120GB Seagates (yeah, not big any more) and every now and then they make this high pitched noise which Seagate claims is head position calibration. I guess this is due to the extremely tight tolerances now being used. I wonder if they can do that for one arm, if it means that they could actually make 2 arms feasible? Or perhaps this calibration movement is too small to allow calibration width of a whole track wide?
I don't know. For high I/O I'd rather see cheap RAM drives and systems with lots of RAM to avoid slower busses anyway.
Yes, it probably would be overly complex. However, in regards to head placement accuracy between two seperate arms, today modern drive heads are guided by servo location information on the actual disks themselves. So it might not be too hard to achieve.
And the rest of us plan for failure by using RAID in addition to backups (and system images).
I couldn't agree more. I worry about the people who use RAID *as* their only "backup" and then find out the hard way that it does not protect from most forms of user error.
I wasn't saying he was impressed by Windows 3.0. I was saying that Windows 3.0 allowed him to see that the mouse resolution of those old Microsoft mousies was not terribly blocky like using them in text mode might have alluded.
I've questioned the usefulness of hdd cache compared to OS main memory cache before on slashdot and gotten flamed. Unfortunately I've still never seen any benchmark that convinced me of whether large onboard cache really helps, or just helps results on benchmarks which intentionally avoid OS disk caching. If anybody has some hard info, post a link.
I don't have a link at the moment with any hard info. But I did recently test re-reading a 1GB file in FreeBSD 6.0 Release on my AMD XP2800+ with 2GB DDR ram...
came out to about 588 MB/s.
(that's 2^20 MB/s and not the bogus 10^6 MB/s.)
I might be able to dig up a link which proves that larger on-drive caches can actually hurt performance, since I did see a benchmark at Storagereview which showed this in two almost exactly the same WD drives which only differed in cache size (same model drives, except one was a "Special Edition" with larger cache only)....
The transfer rate during normal use is 20 MB/sec and has been for 10 years. No-one has ever made a hard drive go faster during normal use. Sequential reads from outer tracks have gotten up to 60MB/sec and bus speeds have allowed sequential RAID accesses from outer tracks to get up to 250 MB/sec but why pay for something advertized to perform at a level you'll only see 1% of the time?
Where did you pull this 20 MB/sec figure? And what is this "normal use" for which you speak?
For some of the things I do, I do actually see about 58 MB/sec on a not new 120GB Seagate.
I find this static 20 MB/sec figure very strange. Meanwhile, I find faster disks to be very much noticably faster during use than slower disks.
I am aware of those standards. However I refuse to adhere to them outside of my work life and will always provide a disclaimer when refering to data sizes on the net.
Standards boards have made some VERY VERY STUPID decisions in the past. And this is one of them.
For a very long time MB, GB, etc was actually meaningful in a computer science sense, since modern digital computers use... well... you know... BINARY!?!?!
2^ is much more meaningful in Comp Sci and digital binary systems than 10^ is. 10^ does not FIT well, much less fit perfectly like 2^ does.
I would like to meet the complete and utter retards at the standards board who thunk up this pure fucking genius idea. "Hey, I know! Lets take a unit of measurement which FITS PERFECTLY and has been used for DECADES and COMPLETELY FUCK IT AND THE INDUSTRY UP so as to pander to CORPORATE DESIRES!"
I call on all geeks who care, to refuse to adhere to that bullshit "standard", continue to use the old proper 2^ based standard and provide a disclaimer to what they mean and the ridiculous reason for why the disclaimer is needed.
MiB should be 10^6 and MB should have REMAINED 2^20.
I assumed he was talking about having multiple servo/arms to employ multiple read/write heads. I've been wondering for a long time if we would ever see this in SCSI drives, to drive up their high I/O even more. High speed SCSI drives already use physically smaller disks. Having seperate read/write heads on seperate servo/arms at opposite ends of the drive could be a really good thing.
It would of course increase complexity though and thus reduce durability.
40C or less: 5 years 41C to 45C: 4.5 years 46C to 50C: 4 years 51C to 55C: 3.5 years 56C to 60C: 3 years 61C and more "Strengthen cooling power so that the DE surface temperature is 60C or less".
Doesn't do any good if the drive has crashed. I will no longer buy WD; crappy quality.
A word of advice from someone who was hurt by a flakey 340MB Maxtor many years ago. If you do this, eventually you will never want to buy ANY drive again. I have come across poor quality drives that died quickly with:
Maxtor IBM Hitachi Fujitsu Toshiba WD Seagate
Every drive manufacturer has its share of bad batches or makes a tech mistake which has big consequences. Some people get hurt and then complain in real life and on the net and then replace that drive with some other manufacterer, get lucky with that drive and proclaim how great that company is. Stick around long enough and you will come full circle and hopefully by that time you'll have realised that a good backup procedure should be as critical to you as your data is. You should also hopefully realise by then that overall, they're mostly close to each other in quality.
I used to sing the praises of Seagate and WD and complain about Maxtor and then find LOTS of people who agreed and about the same amount of people who disagreed.
In 1991, I had a 200MB "Brand Technologies" IDE hard drive. When people at my college at the time were talking about what sorts of computers they had at home, they were sceptical to non-believing that I actually had a 200MB drive! Which from memory transfered data at about 600kbytes/sec. ; ) That drive only lasted about a year and I'd never heard of "Brand Technologies" ever again. Hmmm, maybe I should have bought that 105MB Seagate.
I remember an Amiga 500 owner being amazed when I told him I had 4MB RAM.
PS, pron at 1024x768 256 colours was pretty awesome in 1991. I can remember watching images getting written into the frame buffer. You could actually see it draw it was so slow. ; )
Oh then there was that time that an Amiga 500 owner was gloating about the really shitty mouse resolution (Microsoft) he was witnessing in class. I let him gloat for a while and then pointed out that we were in text mode and the blocky jumping he was seeing was due to that and not the mouse resolution. Going into Windows 3.0 shut him up real quick on that issue.
2. 8 to 16 MB hard drive memory cache. With the memory cache that big many hard drive access operations are quite a bit faster.
Actually, increasing on-drive cache sizes have shown to DECREASE access times.
There was a classic example with a WD product which also came in a special edition with much larger cache. I think it was 2MB versus 8MB. Exact same drive, except the "special edition" had the larger cache... oh and slightly slower access times.
Apparently this comes down to the management of larger on-drive caches taking longer to complete per read request.
For best performance, on-drive buffer/caches should not be any larger than they need to be to act most effectively as buffers. Especially since a large on-drive cache is not much good when system RAM is being used as a buffer cache, at MUCH higher speeds. At the end of the day, on-drive caches are good to a point, but making them larger than they NEED to be just hurts performance. It is of course, good for marketting though. ; )
According to the article, the 250GB bursts hit 171 MB/s, so actually it would be hindered by SATA1. Burst speed isn't my #1 consideration anyways, but it's something.
Drive quoted burst speed comes from or to on drive cache anyway. That cache is good to help the drive sustain it's highest read transfer rates through read-ahead (when the OS comes back for the next block, it has already been read from disk) and also an OS can send small writes to the drive faster. But in practice this mostly just helps a disk to meet it's highest sustained transfer rates. The burst speed sounds good and 16MB cache sounds good, but in these modern times, when we use OS' which use free memory as buffer/caches, we have a LOT of memory and that memory is REALLY FAST, on-drive caches are mostly being used as buffers. As far a caching goes, they don't really get used all that much, since re-reading a block will almost always come from system RAM before it comes from drive cache RAM. Sure it is true that the read-ahead caching on the drive is caching, but in practice it is mostly used as a buffer.
Are there some other SCSI drives with higher performance now?
From a practical point of view or from a meaningless burst speed point of view due to large on-drive caches and fast busses?
I have a Fujitsu SCSI320 drive which sustains about 94MB/s at the beginning of the disk, which slowly tapers off to about 64MB/s at the end of the disk. That is faster than the raptor and this SCSI drive is also faster than the raptor in other aspects like read service times, I/O rates, etc.
There have been Fujitsu, Maxtor, Seagate and Hitachi SCSI drives faster than the raptor for a long time. The Maxtor Atlas 15K II is really fast.
In fact, as far as sustained reads and writes go, access times and sustained I/O, has SCSI EVER lost the top spot?
Remember though that working directly on a flash device (and repeatedly writing to it, very noticeable if you're suffering from heavy Savitis and can't prevent your fingers from reaching the CTRL+S keys every 15s or so) could heavily lower it's life expectancy.
I wonder how well thumb drives stack up against CF cards, as far as durability goes. I have been using San Disk and Lexar CF cards in some firewalls (OpenBSD, pf) for more than 6 months without issue. I use noatime and softdeps on each partition. I know of other people who have done the same for years.
San Disk released a flash memory durability calculation document which is interesting. So I wonder if thumb drives use wear leveling too?
Although that link implies this is regarding MMC, the document looks exactly like the CF related document, which is currently a 404. The document also seems to be regarding flash memory generically, so maybe San Disk thumb drives at least do employ wear leveling?
> What 15.4" inch laptop does 1920x1200 on the built in LCD? I have not seen that res on anything less than 17".
This was even an option on the Dell Inspiron 8600 two years ago. Today, you can configure the Inspiron 6000 with that resolution (WUXGA) for less than $800.
I'm ECSTATIC about this news - as long as the rumored ability to dual-boot Windows XP is a reality.
With Yonah, it can be even better than the traditional sense of dual-booting. Since Yonah does virtualization in hardware, you should be able to run OSX and XP at the same time and switch between them while they both continue to run.
Combine that with the fact that the Yonah will be dual core and WOW!
Anyone want to buy a Sony VAIO VGN-A49GP when the new Powerbook comes out? Please? I'm gunna need some cash. ; )
Slashdot Mirror
I'm told there was also software that would drive the head stepper motor in the Commodore 1541 with varying-width pulses, to play music.
I remember programs which would play music through old dot matrix pin printers. Alternating the striking and non-striking of paper in a PCM fashion could even render very noisy sampled sound. This type of technique was also used for the PC speakers of the time (which were designed to only play tones of square wave), to play digitized sounds. This technique eventually ended up in games.
I remember first running a program written by a guy who wanted to prove the technique. It played him singing "Daisy Bell (A Bicycle Built for Two)". A song which was played by HAL in 2001 A Space Odyssey.
Higher density means the heads fly closer to the platter and small dust particles would 'crash' the head. This means the head will hit the particle and either drag it, or bump over it. Kinda like hitting a body with your car. I wouldn't worry too much, both the heads and platters are covered with a diamond like coating. Also, any 'floating' particle would be spun off when the disk started to spin up. I once took apart a working 6G WD drive and the inside looked like my car's brake pads with all of the dust in the drive.
I have taken apart lots of drives and only ever saw one with obvious circular scuffing all around the drive. Every other drive had to my eye perfect mirror finishes.
As I have stated elsewhere, I took apart a drive with growing bad sectors and within SECONDS of being opened, the drive was DESTROYED. It initially worked and while open and spinning it died completely, returning no data at all (forever trying to read).
I would not be telling people to not worry. Every drive I have ever seen has had warning stickers which stated that opening or removal of any sticker (some cover holes in the casing) will void the warrantee. Those warnings are there for a reason! Cigarette smoke particles can kill a drive (I'm not talking about cigarette ash, I'm talking about the particles which make up the smoke!), imagine what household dust can do. And this was true for drives of 15 years ago. Modern drives have much lower flight heights, making the situation worse. If I ever accidentally scratched through or punctured a drive hole sticker, I would frantically image the drive, then zero it out and then turf it (or maybe pull it apart and keep the nice looking platters).
VooDoo 2 was a great chipset for its time. Held up with newer games well past its prime.
Actually, it didn't hold up well at all. It was severely hampered by the PCI bus and the "AGP" Voodoo cards were not being used in AGP mode, rather some PCI compatibility mode.
The fact is, that GPU focused games of the time were being made to at least run well on the older Voodoo2 cards, because there were so many out there. The game houses were targetting thier market and thus slowing development due to the Voodoo2. 3Dfx accelerated 3D game technologies and then severely slowed it down by choosing to stick with PCI.
Here, take a look at a Matrox G200 (!) kicking the shit out of a Voodoo2 SLI setup! This situation, where a MUCH slower 3D card is killing the "awesome" Voodoo2 SLI because the Voodoo2's are: 1/ PCI and 2/ storing the SAME texture data in each of the cards memory because 12MB+12MB does not give you 24MB in an SLI setup (the data MUST be mirrored, except for the frame buffer).
"S3's own `mon2.dm2' demo, using their own `newS3' map is using a whole lot of really large textures, making the scenery look very detailed and pretty. It would have been a shame if Savage3D would not score well in it, but this would almost have happened. In the latest driver I received, the `AA" setting needs to be enabled to run mon2.dm2 at a decent speed. Anyway, in mon2.dm2 the Savage3D scores higher and higher the more test runs you do. I decided to put down the result of the second run. Matrox' G200 looks very good at mon2.dm2 as well, obviously due to its AGP 2x interface. It reaches about 66% of the Savage3D scores and this without texture compression. The coolest thing are the TNT results however. TNT scores better than Savage3D although TNT does not use texture compression!!! Only with a Celeron 266 TNT looks a little slower than Savage3D, which is another proof for the CPU dependency of TNT in Quake 2. 3Dfx products have yet to learn what AGP means, so the mon2.dm2 is a serious threat to them. 22 fps scored by a $500 Voodoo2 SLI combo looks a bit sad, doesn't it? Banshee looks even worse, its `AGP-interface' only consists of the mechanical connector for the AGP slot."
I find it completely shocking that 3Dfx had the talent to design such a great video card and then completely fail to see the value of AGP over PCI. That failure ultimately killed them. It gave their competitors a window of oportunity to make lots of money to put into more R&D while 3Dfx played catch-up. 3Dfx couldn't, it was too late, they died.
Games never looked that great in 16 bit colour anyway.
It worked like that for several more months until I got bored and tried to lube the spindle with drop of WD-40.
Stiction in HDD's was usually between the heads and platters, not the spindle.
Putting WD40 on the spindle may have allowed some to flick out over the platters thanks to centrifugal force, when the heads hit this much denser than air liquid. BANG! Buggered heads and possibly some platter surface area.
Sure, a lot of people (myself included) have run old drives with the cover off when ready to chuck them anyway or in an emergency have opened a drive, repaired it and had it survive long enough to get the data off safely.
Many years ago I did this with a Brand Technologies 200MB IDE drive which had growing defects. Once I opened it, it still worked as baddly as before, but SECONDS later the whole drive was full of defects. No sector would return data. I used to work in a military clean room and not even there would I open a HDD which was working fine. The distance between the heads and the platters when moving are WAY to small for even the clean room I worked in.
This is not to say that there are not clean rooms good enough, but certainly nobodys home or office type workplace is good enough. Cigarette SMOKE particles will damage a drive. Seriously small.
Are all our moondust belong to them?
the problem I was referring to is not which track you are on, but being in the "center" of a given track.
Oh right. Yes that would seem to pose a big problem. Have you heard the odd noises some of these new big drives make periodically? I bought 2 120GB Seagates (yeah, not big any more) and every now and then they make this high pitched noise which Seagate claims is head position calibration. I guess this is due to the extremely tight tolerances now being used. I wonder if they can do that for one arm, if it means that they could actually make 2 arms feasible? Or perhaps this calibration movement is too small to allow calibration width of a whole track wide?
I don't know. For high I/O I'd rather see cheap RAM drives and systems with lots of RAM to avoid slower busses anyway.
Yes, it probably would be overly complex. However, in regards to head placement accuracy between two seperate arms, today modern drive heads are guided by servo location information on the actual disks themselves. So it might not be too hard to achieve.
And the rest of us plan for failure by using RAID in addition to backups (and system images).
I couldn't agree more. I worry about the people who use RAID *as* their only "backup" and then find out the hard way that it does not protect from most forms of user error.
I wasn't saying he was impressed by Windows 3.0. I was saying that Windows 3.0 allowed him to see that the mouse resolution of those old Microsoft mousies was not terribly blocky like using them in text mode might have alluded.
I've questioned the usefulness of hdd cache compared to OS main memory cache before on slashdot and gotten flamed. Unfortunately I've still never seen any benchmark that convinced me of whether large onboard cache really helps, or just helps results on benchmarks which intentionally avoid OS disk caching. If anybody has some hard info, post a link.
I don't have a link at the moment with any hard info. But I did recently test re-reading a 1GB file in FreeBSD 6.0 Release on my AMD XP2800+ with 2GB DDR ram...
came out to about 588 MB/s.
(that's 2^20 MB/s and not the bogus 10^6 MB/s.)
I might be able to dig up a link which proves that larger on-drive caches can actually hurt performance, since I did see a benchmark at Storagereview which showed this in two almost exactly the same WD drives which only differed in cache size (same model drives, except one was a "Special Edition" with larger cache only)....
The transfer rate during normal use is 20 MB/sec and has been for 10 years. No-one has ever made a hard drive go faster during normal use. Sequential reads from outer tracks have gotten up to 60MB/sec and bus speeds have allowed sequential RAID accesses from outer tracks to get up to 250 MB/sec but why pay for something advertized to perform at a level you'll only see 1% of the time?
Where did you pull this 20 MB/sec figure? And what is this "normal use" for which you speak?
For some of the things I do, I do actually see about 58 MB/sec on a not new 120GB Seagate.
I find this static 20 MB/sec figure very strange. Meanwhile, I find faster disks to be very much noticably faster during use than slower disks.
Uh, no. Fastest almost any drive can transfer data off the platters is about 60-70MB/sec, and that's only the very tip-top of the line drives.
Actually, top of the line SCSI drives are actually pushing 100MB/s.
Yes, that is 2^20 Megabytes and not the bullshit 10^6 MB.
IEEE Standards Bearer, January 1997
I am aware of those standards. However I refuse to adhere to them outside of my work life and will always provide a disclaimer when refering to data sizes on the net.
Standards boards have made some VERY VERY STUPID decisions in the past. And this is one of them.
For a very long time MB, GB, etc was actually meaningful in a computer science sense, since modern digital computers use... well... you know... BINARY!?!?!
2^ is much more meaningful in Comp Sci and digital binary systems than 10^ is. 10^ does not FIT well, much less fit perfectly like 2^ does.
I would like to meet the complete and utter retards at the standards board who thunk up this pure fucking genius idea. "Hey, I know! Lets take a unit of measurement which FITS PERFECTLY and has been used for DECADES and COMPLETELY FUCK IT AND THE INDUSTRY UP so as to pander to CORPORATE DESIRES!"
I call on all geeks who care, to refuse to adhere to that bullshit "standard", continue to use the old proper 2^ based standard and provide a disclaimer to what they mean and the ridiculous reason for why the disclaimer is needed.
MiB should be 10^6 and
MB should have REMAINED 2^20.
It's called "seek time".
I assumed he was talking about having multiple servo/arms to employ multiple read/write heads. I've been wondering for a long time if we would ever see this in SCSI drives, to drive up their high I/O even more. High speed SCSI drives already use physically smaller disks. Having seperate read/write heads on seperate servo/arms at opposite ends of the drive could be a really good thing.
It would of course increase complexity though and thus reduce durability.
I recently purchased a Fujitsu SCSI320 drive which had some interesting entries in one of the support pdf's, regarding expected lifetime versus heat:
Doesn't do any good if the drive has crashed. I will no longer buy WD; crappy quality.
A word of advice from someone who was hurt by a flakey 340MB Maxtor many years ago. If you do this, eventually you will never want to buy ANY drive again. I have come across poor quality drives that died quickly with:
Maxtor
IBM
Hitachi
Fujitsu
Toshiba
WD
Seagate
Every drive manufacturer has its share of bad batches or makes a tech mistake which has big consequences. Some people get hurt and then complain in real life and on the net and then replace that drive with some other manufacterer, get lucky with that drive and proclaim how great that company is. Stick around long enough and you will come full circle and hopefully by that time you'll have realised that a good backup procedure should be as critical to you as your data is. You should also hopefully realise by then that overall, they're mostly close to each other in quality.
I used to sing the praises of Seagate and WD and complain about Maxtor and then find LOTS of people who agreed and about the same amount of people who disagreed.
In 1991, I had a 200MB "Brand Technologies" IDE hard drive. When people at my college at the time were talking about what sorts of computers they had at home, they were sceptical to non-believing that I actually had a 200MB drive! Which from memory transfered data at about 600kbytes/sec. ; ) That drive only lasted about a year and I'd never heard of "Brand Technologies" ever again. Hmmm, maybe I should have bought that 105MB Seagate.
I remember an Amiga 500 owner being amazed when I told him I had 4MB RAM.
PS, pron at 1024x768 256 colours was pretty awesome in 1991. I can remember watching images getting written into the frame buffer. You could actually see it draw it was so slow. ; )
Oh then there was that time that an Amiga 500 owner was gloating about the really shitty mouse resolution (Microsoft) he was witnessing in class. I let him gloat for a while and then pointed out that we were in text mode and the blocky jumping he was seeing was due to that and not the mouse resolution. Going into Windows 3.0 shut him up real quick on that issue.
2. 8 to 16 MB hard drive memory cache. With the memory cache that big many hard drive access operations are quite a bit faster.
Actually, increasing on-drive cache sizes have shown to DECREASE access times.
There was a classic example with a WD product which also came in a special edition with much larger cache. I think it was 2MB versus 8MB. Exact same drive, except the "special edition" had the larger cache... oh and slightly slower access times.
Apparently this comes down to the management of larger on-drive caches taking longer to complete per read request.
For best performance, on-drive buffer/caches should not be any larger than they need to be to act most effectively as buffers. Especially since a large on-drive cache is not much good when system RAM is being used as a buffer cache, at MUCH higher speeds. At the end of the day, on-drive caches are good to a point, but making them larger than they NEED to be just hurts performance. It is of course, good for marketting though. ; )
According to the article, the 250GB bursts hit 171 MB/s, so actually it would be hindered by SATA1. Burst speed isn't my #1 consideration anyways, but it's something.
Drive quoted burst speed comes from or to on drive cache anyway. That cache is good to help the drive sustain it's highest read transfer rates through read-ahead (when the OS comes back for the next block, it has already been read from disk) and also an OS can send small writes to the drive faster. But in practice this mostly just helps a disk to meet it's highest sustained transfer rates. The burst speed sounds good and 16MB cache sounds good, but in these modern times, when we use OS' which use free memory as buffer/caches, we have a LOT of memory and that memory is REALLY FAST, on-drive caches are mostly being used as buffers. As far a caching goes, they don't really get used all that much, since re-reading a block will almost always come from system RAM before it comes from drive cache RAM. Sure it is true that the read-ahead caching on the drive is caching, but in practice it is mostly used as a buffer.
Are there some other SCSI drives with higher performance now?
From a practical point of view or from a meaningless burst speed point of view due to large on-drive caches and fast busses?
I have a Fujitsu SCSI320 drive which sustains about 94MB/s at the beginning of the disk, which slowly tapers off to about 64MB/s at the end of the disk. That is faster than the raptor and this SCSI drive is also faster than the raptor in other aspects like read service times, I/O rates, etc.
There have been Fujitsu, Maxtor, Seagate and Hitachi SCSI drives faster than the raptor for a long time. The Maxtor Atlas 15K II is really fast.
In fact, as far as sustained reads and writes go, access times and sustained I/O, has SCSI EVER lost the top spot?
Remember though that working directly on a flash device (and repeatedly writing to it, very noticeable if you're suffering from heavy Savitis and can't prevent your fingers from reaching the CTRL+S keys every 15s or so) could heavily lower it's life expectancy.
I wonder how well thumb drives stack up against CF cards, as far as durability goes. I have been using San Disk and Lexar CF cards in some firewalls (OpenBSD, pf) for more than 6 months without issue. I use noatime and softdeps on each partition. I know of other people who have done the same for years.
San Disk released a flash memory durability calculation document which is interesting. So I wonder if thumb drives use wear leveling too?
Although that link implies this is regarding MMC, the document looks exactly like the CF related document, which is currently a 404. The document also seems to be regarding flash memory generically, so maybe San Disk thumb drives at least do employ wear leveling?
Anyone?
> What 15.4" inch laptop does 1920x1200 on the built in LCD? I have not seen that res on anything less than 17".
This was even an option on the Dell Inspiron 8600 two years ago. Today, you can configure the Inspiron 6000 with that resolution (WUXGA) for less than $800.
So it was. Wow.
I'm ECSTATIC about this news - as long as the rumored ability to dual-boot Windows XP is a reality.
With Yonah, it can be even better than the traditional sense of dual-booting. Since Yonah does virtualization in hardware, you should be able to run OSX and XP at the same time and switch between them while they both continue to run.
Combine that with the fact that the Yonah will be dual core and WOW!
Anyone want to buy a Sony VAIO VGN-A49GP when the new Powerbook comes out? Please? I'm gunna need some cash. ; )
I was running 1920x1200 on my 15.4" laptop
What 15.4" inch laptop does 1920x1200 on the built in LCD? I have not seen that res on anything less than 17".
Jezzus
I was actually aiming for a "Funny", but anyway. It seems you have some very serious issues.
If I said "Linux", it would not matter would it?