Because OpenBSD isn't available in a user friendly (read idiot proof).iso download?
You want to install OpenBSD (on x86 for one example)? Do any one of these...
1/ Download all the files for x86. Should be about only 150Mb including X. Omit downloading any files which start with an 'x' if you don't want to use X. Burn these files to a CD and specify to your CD burning software of choice, that cdrom37.fs should be the boot image. It is a 2.88Mb bootable floppy image. Install by booting the resultant CDR. Real simple.
2/ Download an appropriate 1.44Mb floppy image, dd it to a floppy, boot from it and perform a network install from the Internet. Real simple. The bootable CDR is prefered though, because you may have to try installing a few times if OpenBSD is new to you.
If you have EVER burnt an x86 bootable CDR with ANY burning software, then making an OpenBSD bootable install CD is super simple.
Anyone who is not capable of burning a bootable OpenBSD CDR, should probably not be using anything outside of the Microsoft or Apple world, because it is an absolute no-brainer.
BTW, they do provide.iso's since machines have started to be sold without floppy drives. The.iso's are pretty much an image of a bootable CD made with not much more than the cdrom37.fs as the boot image. You could burn that to a CDR, also burn a seperate CDR with your arch specific.tgz files and then after booting from the bootable CDR, when you are prompted for where to get the install files, insert the CD with the.tgz's. If you need to do it this way because you can't figure out how to do it properly, you should probably get a new hobby/job. As the supplied.iso's are pretty much for network installs on machines without floppies.
BTW, if you can't manage any of this, then security is probably the least of your problems, in which case you could just use any number of the unofficial OpenBSD iso images that are scattered around the net. OpenBSD is not the kind of OS where you would want to be lazy as far as learning the system goes. So steer clear of it if you are not willing to put in the small amount of effort required in making yourself a bootable iso.
I'd be concerned that setting up 128K blocks would be counter-prodcutive, since you'd have a hard time getting your your blocks to align perfectly with the drive's hardware blocks. After all, you lose a few K here for the partition table, another few bytes there for filesystem info...
Your results are not indicative of flash performance - CF is simply not that fast. I frequently get 10MB/sec with my USB 2.0 SD card reader and generic PQI 1GB SD card.
What are you talking about? You say "not that fast" and then tell me you get faster results?
I compared one flash device against one HDD device on the same system with a shitty IDE controller. I don't claim to have the fastest flash device around or that this is a good test for what is coming. But what it should be is an even test on the same system between a very small sample. So far I have always found flash devices to be slower than HDD devices. That is my point. You say you get 10MB/s with flash, I say I get 40MB/s with 7200RPM disk. But...
Flash can be *very* fast. Remember, you can interleave many flash chips using RAID-like techniques without the cost of having multiple disk assemblies.
I never said flash can't be fast. I'm well aware that just about any group of storage devices can be interleaved and I would not bet against flash being made really fast. Interleaving or not.
I'll be testing soon on a faster x86 system with a decent IDE controller and more appropriate OS. Lexar, San Disk and 7200 RPM HDD direct to IDE, USB2 and Firewire 400.
I'm not trying to predict the future here. Just looking at the current state of what's relatively current with the minimal sample I have.
I'm glad you linked to that pdf, I was going to go looking for it as I remember it having some interesting theoretical case studies which reflected the fact that modern CF cards are good enough to be used as system disks in embedded applications or firewalls at least. I know of people who have been running OpenBSD out of CF for years. With cards which are not as durable as these modern cards with wear levelling built in.
From that pdf, San Disk claim "a data logging operation using a 1GB card where a 4kB file is updated every five seconds", yielding a CF lifetime of 317 years!
The same but using 4kb blocks, causing less sequential writes due to expected fragmentation, will yield a life of about 80 years.
I'm not too worried about the life of my firewall. ; )
I usually expect a lifetime from real HDD's of about 5 years, seeing some go past 10 and since manufacturers of flash are no doubt continuing to increase the durability of flash memory I expect that flash could exceed traditional HDD's in life and performance. I have been waiting for that day to come! Imagine main storage in your laptop which can be powered down often and up very quickly, with no great power surge required from the batteries. Plus with killer random IO and potentially very fast sequential transfer speed. Lighter, quieter and cooler...
With advances in reduced CPU power, more efficient backlighting and solid state non-volatile memory, we might actually see notebooks getting real full business days from batteries to be the norm. And desktops which put less burden on the environment.
Let's say you write one file once and only read it from then on. The access time would get updated each time, wearing out a single block faster than the rest.
noatime is your friend for that. My OpenBSD firewall is running from a 1GB SanDisk Ultra II CF card in a Sun Ultra 10. I'm using noatime and soft-updates for all slices on the CF and apparently noatime alone is enough to use modern CF cards as system disks.
I would still like to look into 128kb blocks though...
It's not USB that's the problem (although, that IS with a different chipset on the drive end. The FAIR comparison would be to use a CF to IDE adaptor, and benchmark a high-speed CF card versus an IDE HDD in a USB HDD enclosure.
I have a Lexar 512MB 80x CF card handy, with CF-IDE converter, an open USB2/Firewire HDD enclosure and a 40GB 7200RPM Maxtor HDD.
due to the large sector/page size of flash, you end up doing multiple writes when you think you are doing a single one. For example, if you write to a file in 4 chunks, 32K at a time, it uses up 4 of your writes. It might be possible to remove this with intelligent caching, but you're gonna need a lot of RAM for the caching.
That is interesting. I know modern CF cards employ wear levelling within the cards themselves, but I was previously leaning on BSD's soft-updates and noatime to prolong to life of the CF cards in a few servers and firewalls. Maybe some slices should also have FFS file systems with a 128kb block size to limit any block to no more than one file. I'm off to see if FFS in OpenBSD can use a block size of 128kb. Assuming this would actually cause 128kb block sized writes?
Honestly, this is just an idea that isn't ready yet. Flash is too slow to write right now. The life is decent. Reads work well.
Yes I must say that I find the high speed flash cards I use, which are directly plugged into my motherboards IDE controllers via passive IDE-CF converters (no electronics, just different connectors at either ends of the converters), are slow to write.
I have Lexar Pro series 80X CF's and San Disk Ultra II CF's. The Lexar's are WAY faster than the San Disk's BTW. But still they're pretty slow. I use them in Sun Ultra 10's and 5's, which have really slow IDE interfaces, but the CF's are slower than these and are also slow in various PC's I've tried them in.
# dd bs=64k count=1024 if=/dev/zero of=/tmp/64MB.bin 1024+0 records in 1024+0 records out 67108864 bytes transferred in 54.374 secs (1,234,192 bytes/sec)
# dd bs=128k count=512 if=/dev/zero of=/tmp/64MB.bin 512+0 records in 512+0 records out 67108864 bytes transferred in 54.322 secs (1,235,385 bytes/sec)
# dd bs=64k count=1024 if=/dev/zero of=/var/log/pf/64MB.bin 1024+0 records in 1024+0 records out 67108864 bytes transferred in 4.043 secs (16,594,884 bytes/sec)
# dd bs=128k count=512 if=/dev/zero of=/var/log/pf/64MB.bin 512+0 records in 512+0 records out 67108864 bytes transferred in 3.980 secs (16,858,338 bytes/sec)
Here I have written 64Mb to a San Disk Ultra II, first with 64kb chunks and then with 128kb. I did this because I usually use 64kb chunks and thought that 128kb would be faster for CF. Turned out to not be so, unless dd is writing 128kb chunks but the writes are being committed to the file system at no greater than the filesystems block size at a time.
The two last dd's are the same thing but being written to a 120GB Seagate 7200 RPM PATA drive. Both the CF and HDD are directly connected to the on board IDE controllers of a Sun Ultra 10, each seperately as masters without slaves.
The IDE controllers on Sun Ultra 5/10 motherboards are garbage BTW. They can get the best out of the CF cards I have, but certainly not the Seagate HDD, which does about 2-3 times better transfer rates in some of my PC's.
I feel compelled now to do these tests again but to raw devices instead of to files within a filesystem. If the results were that 128kb block size filesystem is likely to be many times faster than a filesystem with much smaller blocks, I might be inclined to build filesystems with 128kb blocks on my CF based machines. But I think I would be limited to 1024 files on a 128Mb slice. Which might not be workable.
Let's not forget that the hard disk has a buffer to isolate the latency, and likely uses prefetching to keep it full. I'd be interested in knowing if anything like that would help the thumb drive.
On a related note, some company (I think it was Lexar?) makes flash cards for digital cameras which includes some large fast buffer to allow cameras to burst writes off to the card at high speed, so that while the camera is taking the next shot, the card can simultaneously be commiting the buffered data to non-volatile flash memory. So that pro digital cameras can have faster sustained continuous shooting.
Please refrain from using the played out jokes seen elsewhere on Slashdot, you would do particularly well not mentioning "in soviet russia...", "xyz is dying" and "imagine a beowulf cluster of these!" in the future
Don't forget Natalie and those grits which are said to be hot.
"Truly, the only people who deserve a complete helping of blame are the bad people. Let's not forget that they're the ones doing this to us. They're the ones who are annoying an entire planet. They're the ones who are costing us billions of dollars a year to secure our systems against them. They're the ones who place their desire for fun and/or profit ahead of everyone on earth's desire for peace and the right to privacy."
Is this a little pointless? The fact is that complex systems (not just talking about computer systems) are rarely going to be perfect and bad people will always try to exploit the exploitable for their own gain. Is there any point in complaining? There will always be bad people, there will always be exploitable systems. Why whine? The hackers validate and keep alive an industry and frankly I think that it can be enjoyable on both sides.
Whining won't get rid of bad people and won't fix the Worlds exploitable systems. The arms race will not stop. I see hackers on both sides as being a necessary evil and I respect hackers for their talent, regardless of what side they are on.
In short, don't count on them. If you want a "good enough" PowerBook with many luxury features from Bluetooth 2.0+EDR to two-finger scrolling, ambient light sensors that illuminate your keyboard and dim your screen at the same time, etc., get it now.
I'd love to get one now, but what I do is very memory intensive. I'd like to just get a Powerbook and access an Opteron remotely or something similar since I guess even the fastest notebooks are not going to match a really fast desktop. But often a customer of mine will hand me a CDR or DVDR and ask me to extract and analyse what is on it there and then..
I'll take a look at Samsung and see if they're available here.
As for upcoming PowerBook CPUs before the Intel switch, the two likely candidates are the 7448, a slightly better-cached and better-MHzd version of the current 7447B. While the 8641D will be out some time 2006 and comes with two cores, making it interesting, it probably won't have a better front-side bus system either.
I thought that I had read (a year or so ago) that the new Freescale G4's had FSB's fast enough to fully exploit DDR RAM. I thought they'd be ready by now. That was where my hope had been.
So my hopes are dashed? I wan't a powerful notebook and had hoped for a Powerbook. Thinkpads are so expensive on the high end but I like their quality after Powerbooks.
If you wanted to make the BSDs relavent, perhaps you should not have focused on these narrow bits of the industry: security, portability, etc and instead focused on the larger picture, which linux has been doing since day one.
Focus on the larger picture?
Actually, the BSD's do tend to think for the long term and lay down good foundations. Look at the Linux world, where major systems change dramatically not only between major versions, but even within a single stable branch. Linux firewalling (between majors), the VM system chop and change (within a "stable" branch), etc?
I don't beleive it is true that Linux looks at the greater picture. I think the truth is that Linux has many more developers and a large percentage of them are still mostly in experimental tinker mode.
I enjoy the BSD's for the platform being stable. Not stable code, stable design. Linux disapoints me when the VM system changes suddenly when we're not expecting it and Linus makes comments (just as bad as Theo's) to the effect of "I don't know it well / have not looked" but "my system is better / I don't see good points in the other system".
When Theo has an OS that I can purchase software for, run openGL windows games on, and download an iso for, burn and be up and running in 30 minutes, then he can talk.
Then "he" can talk? Talk about what? Did you mean "we" can talk?
If OpenGL is important to you, avoid OpenBSD. You don't need to download an iso for OpenBSD, because you can download it and easily burn one under 150MB download for any arch. You could also download a floppy image and perform an internet or network install from it. Installs including X, take 5 mins or less from CD, once you know what you are doing.
Every now and again, i'm tempted to give it a shot, then he makes another outrageous talking out of his arse comment, having a cry about other people, and i'm reminded of why I've tried just about every other OS i can get my hands on, other than NetBSD as yet:D
Maybe you should try OpenBSD and rate it on it's own value. It's a great system.
I had 3 spare x86 boxes, and tinkered with all three OSs for at least 6 months.
I've been using OpenBSD since 2.5. A lot has changed since then and 3 boxes are not a large sample size. My experience is the opposite of yours.
Sure, the BSDs have better design, I agree, and I would love to run them, but if I'm limited to 10% of the available hardware
This is absolute garbage. 10%? Bullshit.
and every time I need a new NIC I have to snoop around the store looking for that one magic NIC with the right chipset revision
There are a lot of crappy NIC's out there. There are also some very good NIC's which also happen to be cheap. 3Com are mostly crap. Intel's tend to be great. OpenBSD supports high quality cards well. Do you buy any card as long as it just works? Or do you seek out good cards which are cheap and well supported?
well I consider that a larger burden freedom-wise than MS places on its users.
If you choose hardware first and then consider software to go with it, then you are putting the cart before the horse. I prefer to provide solid solutions by sticking with high quality software and purchasing good quality, well supported hardware to go with that software. Finding good hardware to run OpenBSD is easy. In fact I mostly find OpenBSD to run well on random computers which I try to run it on.
Stop yapping Theo and go write some firewire drivers, or whatever technology came out 5 years ago that your system still doesn't support.
Theo and Co. do not write OpenBSD for you, they write it for themselves. If they don't want or need firewire, they won't spend time writing software for it. The lack of support of some device is not indicative of a lack of OpenBSD quality. It is indicative of either a lack of interest or a lack of open documentation.
If you don't like it, don't use it. But please don't exagerate the weak points.
Considering most NDAs have a clause that says you're not even allowed to admit that the NDA exists in the first place, it's hopeless to ask me this, but I can tell you this: I'm a 21-year old student from the North West of Germany, far far away from the headquarters of Apple in Cupertino, and unlikely to have inside knowledge.
I've been under a few NDA's, one of them was under a foreign company who never met me face to face. Apple though, is so careful that I guess this would not fly with them.
I'm not a US citizen either, but my country is one of USA's "bitches" it seems and their laws are coming to a town near me soon. : ( A few nights ago, an advertisement on TV warned that distributing a new release game would carry a penalty as high as 2 years in prison!
My GF still wants a Powerbook, the idea of an Apple has grown on her. She loves my mini. We're waiting for the next round of Powerbooks. I would really like full speed access to the supplied DDR RAM though.
Personally, I never want someone who works with me to shrug of failure. I want to understand why they failed and learn from it.
When I say "shrug off failure", I mean that their failure itself is not a big deal to them and they quickly move on to understanding their failure and fixing it, instead of spending time worrying about it and trying to make it look less of a failure or pointing the finger elsewhere.
Their failing, not the failure itself. Of course the failure is important to understand and learn from.
Yeah, some people won't make the cut, but making a judgement along the lines of "17 years of people and 2 qualify as exceptional" is a harsh standard to impose.
Maybe I'm harsh then. Exceptional is exceptional after all. Maybe my idea of the word is too strong for your liking. When I state exceptional, I am talking about the full package. Someone who is brilliant and reasonable for example. Someone who is truely gifted yet willing to consider an alternative idea or opinion.
I don't see it as black-and-white, I see a majority of plodders, just turning up to work to pay the bills. Some loud arseholes who are "it", talk a lot but don't really contribute much more than the "safe-bet pre-packaged half-solution" which almost fits and then boast about how wonderful they are. Then there are the nice folks who care about the job but don't do it all that well. And then there are those few exceptionals who not only manage to put up with the rest, but also see the real issues and fix them in creative ways that have the bullshit artists desperately searching for some way to save face and the plodders complaining that this makes everyone else look bad (and the nice folks congratulating).
There is a very bad problem in IT where I live, where people protect their knowledge and if they're lacking they bullshit. Those people are not exceptional and unfortunately I see the majority either dull and many being full of crap.
This is really how I see it. Maybe things are great in your part of the World, but in my part of the World the mundane rule and protect each other. I'm quite sick of being verbally shot down by morons when I know I have an excellent, unique solution which actually fits while using current infrastructure (and my track record show this time and time again when I actually get given the chance). It is true that I have difficulty with these people, but my difficulty does not make them any less mundane or insecure.
I do need to work on my people skills to learn to manage these types, but I think my problems are nowhere near as bad as theirs and at least I am aware and willing to face my shortcomings. A bullshit artist is a bullshit artist and plodder is a plodder. They are not what they are because of me, but I do see them for what they are.
Maybe I'm right, maybe I'm not. But I useta work at a software company, and I've seen some of the silly things people have been labeled 'incompetent' for.
Fair enough. I guess (and I should know, working for the crowd I work for), that money, time, exaustion and expectations can make anyone angry when things don't go as planned.
I have no idea how the Microsoft devs are treated, so I can't blame them without knowing them. I left more than one job because a manager did not understand or appreciate the issues they're supposed to be managing.
You probably should not make such sweeping generalizations. There are many personality types among people who are very effective at very complex tasks.
You are absolutely correct. I was too hasty with that post. Need sleep......
Real engineers fix problems, they don't get emotional.
This is so true. I've worked with many people in IT and communications over the past 17 years, in financial, military and educational institutions from desktop support to reverse engineering. People who get emotional when challenged or proven wrong are putting their ego before the problem. Their ego becomes the biggest problem and the real problem they're getting paid to fix tends to get fixed in a way that makes them look good, which might not actually be the technically better way.
The most exceptional people I have worked with, shrugged failure off and carried on with fixing things or making them better. The loudest people don't know shit and cover it up with fast talking. It seems the quiet, well educated people who are comfortable with themselves are the ones who make the biggest differences.
Unfortunately, in the past 17 years, only two people in my mind stand out to be the exceptional people, the rest are all competing in a bullshit competition with each other or are otherwise mediocre.
Slashdot Mirror
Because OpenBSD isn't available in a user friendly (read idiot proof) .iso download?
.iso's since machines have started to be sold without floppy drives. The .iso's are pretty much an image of a bootable CD made with not much more than the cdrom37.fs as the boot image. You could burn that to a CDR, also burn a seperate CDR with your arch specific .tgz files and then after booting from the bootable CDR, when you are prompted for where to get the install files, insert the CD with the .tgz's. If you need to do it this way because you can't figure out how to do it properly, you should probably get a new hobby/job. As the supplied .iso's are pretty much for network installs on machines without floppies.
You want to install OpenBSD (on x86 for one example)? Do any one of these...
1/ Download all the files for x86. Should be about only 150Mb including X. Omit downloading any files which start with an 'x' if you don't want to use X. Burn these files to a CD and specify to your CD burning software of choice, that cdrom37.fs should be the boot image. It is a 2.88Mb bootable floppy image. Install by booting the resultant CDR. Real simple.
2/ Download an appropriate 1.44Mb floppy image, dd it to a floppy, boot from it and perform a network install from the Internet. Real simple. The bootable CDR is prefered though, because you may have to try installing a few times if OpenBSD is new to you.
If you have EVER burnt an x86 bootable CDR with ANY burning software, then making an OpenBSD bootable install CD is super simple.
Anyone who is not capable of burning a bootable OpenBSD CDR, should probably not be using anything outside of the Microsoft or Apple world, because it is an absolute no-brainer.
BTW, they do provide
BTW, if you can't manage any of this, then security is probably the least of your problems, in which case you could just use any number of the unofficial OpenBSD iso images that are scattered around the net. OpenBSD is not the kind of OS where you would want to be lazy as far as learning the system goes. So steer clear of it if you are not willing to put in the small amount of effort required in making yourself a bootable iso.
I'd be concerned that setting up 128K blocks would be counter-prodcutive, since you'd have a hard time getting your your blocks to align perfectly with the drive's hardware blocks. After all, you lose a few K here for the partition table, another few bytes there for filesystem info...
Very true. That sucks then.
He was saying the Compact Flash interface is the slow part.
Oh right. Appologies all round then.
Your results are not indicative of flash performance - CF is simply not that fast. I frequently get 10MB/sec with my USB 2.0 SD card reader and generic PQI 1GB SD card.
What are you talking about? You say "not that fast" and then tell me you get faster results?
I compared one flash device against one HDD device on the same system with a shitty IDE controller. I don't claim to have the fastest flash device around or that this is a good test for what is coming. But what it should be is an even test on the same system between a very small sample. So far I have always found flash devices to be slower than HDD devices. That is my point. You say you get 10MB/s with flash, I say I get 40MB/s with 7200RPM disk. But...
Flash can be *very* fast. Remember, you can interleave many flash chips using RAID-like techniques without the cost of having multiple disk assemblies.
I never said flash can't be fast. I'm well aware that just about any group of storage devices can be interleaved and I would not bet against flash being made really fast. Interleaving or not.
I'll be testing soon on a faster x86 system with a decent IDE controller and more appropriate OS. Lexar, San Disk and 7200 RPM HDD direct to IDE, USB2 and Firewire 400.
I'm not trying to predict the future here. Just looking at the current state of what's relatively current with the minimal sample I have.
I'm glad you linked to that pdf, I was going to go looking for it as I remember it having some interesting theoretical case studies which reflected the fact that modern CF cards are good enough to be used as system disks in embedded applications or firewalls at least. I know of people who have been running OpenBSD out of CF for years. With cards which are not as durable as these modern cards with wear levelling built in.
From that pdf, San Disk claim "a data logging operation using a 1GB card where a 4kB file is updated every five seconds", yielding a CF lifetime of 317 years!
The same but using 4kb blocks, causing less sequential writes due to expected fragmentation, will yield a life of about 80 years.
I'm not too worried about the life of my firewall. ; )
I usually expect a lifetime from real HDD's of about 5 years, seeing some go past 10 and since manufacturers of flash are no doubt continuing to increase the durability of flash memory I expect that flash could exceed traditional HDD's in life and performance. I have been waiting for that day to come! Imagine main storage in your laptop which can be powered down often and up very quickly, with no great power surge required from the batteries. Plus with killer random IO and potentially very fast sequential transfer speed. Lighter, quieter and cooler...
With advances in reduced CPU power, more efficient backlighting and solid state non-volatile memory, we might actually see notebooks getting real full business days from batteries to be the norm. And desktops which put less burden on the environment.
Let's say you write one file once and only read it from then on. The access time would get updated each time, wearing out a single block faster than the rest.
noatime is your friend for that. My OpenBSD firewall is running from a 1GB SanDisk Ultra II CF card in a Sun Ultra 10. I'm using noatime and soft-updates for all slices on the CF and apparently noatime alone is enough to use modern CF cards as system disks.
I would still like to look into 128kb blocks though...
It's not USB that's the problem (although, that IS with a different chipset on the drive end. The FAIR comparison would be to use a CF to IDE adaptor, and benchmark a high-speed CF card versus an IDE HDD in a USB HDD enclosure.
I have a Lexar 512MB 80x CF card handy, with CF-IDE converter, an open USB2/Firewire HDD enclosure and a 40GB 7200RPM Maxtor HDD.
Watch this space! ; )
That is interesting. I know modern CF cards employ wear levelling within the cards themselves, but I was previously leaning on BSD's soft-updates and noatime to prolong to life of the CF cards in a few servers and firewalls. Maybe some slices should also have FFS file systems with a 128kb block size to limit any block to no more than one file. I'm off to see if FFS in OpenBSD can use a block size of 128kb. Assuming this would actually cause 128kb block sized writes?
Honestly, this is just an idea that isn't ready yet. Flash is too slow to write right now. The life is decent. Reads work well.
Yes I must say that I find the high speed flash cards I use, which are directly plugged into my motherboards IDE controllers via passive IDE-CF converters (no electronics, just different connectors at either ends of the converters), are slow to write.
I have Lexar Pro series 80X CF's and San Disk Ultra II CF's. The Lexar's are WAY faster than the San Disk's BTW. But still they're pretty slow. I use them in Sun Ultra 10's and 5's, which have really slow IDE interfaces, but the CF's are slower than these and are also slow in various PC's I've tried them in.
Here I have written 64Mb to a San Disk Ultra II, first with 64kb chunks and then with 128kb. I did this because I usually use 64kb chunks and thought that 128kb would be faster for CF. Turned out to not be so, unless dd is writing 128kb chunks but the writes are being committed to the file system at no greater than the filesystems block size at a time.
The two last dd's are the same thing but being written to a 120GB Seagate 7200 RPM PATA drive. Both the CF and HDD are directly connected to the on board IDE controllers of a Sun Ultra 10, each seperately as masters without slaves.
The IDE controllers on Sun Ultra 5/10 motherboards are garbage BTW. They can get the best out of the CF cards I have, but certainly not the Seagate HDD, which does about 2-3 times better transfer rates in some of my PC's.
I feel compelled now to do these tests again but to raw devices instead of to files within a filesystem. If the results were that 128kb block size filesystem is likely to be many times faster than a filesystem with much smaller blocks, I might be inclined to build filesystems with 128kb blocks on my CF based machines. But I think I would be limited to 1024 files on a 128Mb slice. Which might not be workable.
Let's not forget that the hard disk has a buffer to isolate the latency, and likely uses prefetching to keep it full. I'd be interested in knowing if anything like that would help the thumb drive.
On a related note, some company (I think it was Lexar?) makes flash cards for digital cameras which includes some large fast buffer to allow cameras to burst writes off to the card at high speed, so that while the camera is taking the next shot, the card can simultaneously be commiting the buffered data to non-volatile flash memory. So that pro digital cameras can have faster sustained continuous shooting.
What's happening Flash?
Gordon's alive!!!
Please refrain from using the played out jokes seen elsewhere on Slashdot, you would do particularly well not mentioning "in soviet russia...", "xyz is dying" and "imagine a beowulf cluster of these!" in the future
Don't forget Natalie and those grits which are said to be hot.
"Truly, the only people who deserve a complete helping of blame are the bad people. Let's not forget that they're the ones doing this to us. They're the ones who are annoying an entire planet. They're the ones who are costing us billions of dollars a year to secure our systems against them. They're the ones who place their desire for fun and/or profit ahead of everyone on earth's desire for peace and the right to privacy."
Is this a little pointless? The fact is that complex systems (not just talking about computer systems) are rarely going to be perfect and bad people will always try to exploit the exploitable for their own gain. Is there any point in complaining? There will always be bad people, there will always be exploitable systems. Why whine? The hackers validate and keep alive an industry and frankly I think that it can be enjoyable on both sides.
Whining won't get rid of bad people and won't fix the Worlds exploitable systems. The arms race will not stop. I see hackers on both sides as being a necessary evil and I respect hackers for their talent, regardless of what side they are on.
Thanks for those links.
In short, don't count on them. If you want a "good enough" PowerBook with many luxury features from Bluetooth 2.0+EDR to two-finger scrolling, ambient light sensors that illuminate your keyboard and dim your screen at the same time, etc., get it now.
I'd love to get one now, but what I do is very memory intensive. I'd like to just get a Powerbook and access an Opteron remotely or something similar since I guess even the fastest notebooks are not going to match a really fast desktop. But often a customer of mine will hand me a CDR or DVDR and ask me to extract and analyse what is on it there and then..
I'll take a look at Samsung and see if they're available here.
As for upcoming PowerBook CPUs before the Intel switch, the two likely candidates are the 7448, a slightly better-cached and better-MHzd version of the current 7447B. While the 8641D will be out some time 2006 and comes with two cores, making it interesting, it probably won't have a better front-side bus system either.
I thought that I had read (a year or so ago) that the new Freescale G4's had FSB's fast enough to fully exploit DDR RAM. I thought they'd be ready by now. That was where my hope had been.
So my hopes are dashed? I wan't a powerful notebook and had hoped for a Powerbook. Thinkpads are so expensive on the high end but I like their quality after Powerbooks.
If you wanted to make the BSDs relavent, perhaps you should not have focused on these narrow bits of the industry: security, portability, etc and instead focused on the larger picture, which linux has been doing since day one.
Focus on the larger picture?
Actually, the BSD's do tend to think for the long term and lay down good foundations. Look at the Linux world, where major systems change dramatically not only between major versions, but even within a single stable branch. Linux firewalling (between majors), the VM system chop and change (within a "stable" branch), etc?
I don't beleive it is true that Linux looks at the greater picture. I think the truth is that Linux has many more developers and a large percentage of them are still mostly in experimental tinker mode.
I enjoy the BSD's for the platform being stable. Not stable code, stable design. Linux disapoints me when the VM system changes suddenly when we're not expecting it and Linus makes comments (just as bad as Theo's) to the effect of "I don't know it well / have not looked" but "my system is better / I don't see good points in the other system".
When Theo has an OS that I can purchase software for, run openGL windows games on, and download an iso for, burn and be up and running in 30 minutes, then he can talk.
:D
Then "he" can talk? Talk about what? Did you mean "we" can talk?
If OpenGL is important to you, avoid OpenBSD. You don't need to download an iso for OpenBSD, because you can download it and easily burn one under 150MB download for any arch. You could also download a floppy image and perform an internet or network install from it. Installs including X, take 5 mins or less from CD, once you know what you are doing.
Every now and again, i'm tempted to give it a shot, then he makes another outrageous talking out of his arse comment, having a cry about other people, and i'm reminded of why I've tried just about every other OS i can get my hands on, other than NetBSD as yet
Maybe you should try OpenBSD and rate it on it's own value. It's a great system.
Honestly, do you think he'd be so bitter if he had a girlfriend?
Theo lives with his girlfriend and some cats.
For what it's worth. It's just a silly joke though really.
I had 3 spare x86 boxes, and tinkered with all three OSs for at least 6 months.
I've been using OpenBSD since 2.5. A lot has changed since then and 3 boxes are not a large sample size. My experience is the opposite of yours.
Sure, the BSDs have better design, I agree, and I would love to run them, but if I'm limited to 10% of the available hardware
This is absolute garbage. 10%? Bullshit.
and every time I need a new NIC I have to snoop around the store looking for that one magic NIC with the right chipset revision
There are a lot of crappy NIC's out there. There are also some very good NIC's which also happen to be cheap. 3Com are mostly crap. Intel's tend to be great. OpenBSD supports high quality cards well. Do you buy any card as long as it just works? Or do you seek out good cards which are cheap and well supported?
well I consider that a larger burden freedom-wise than MS places on its users.
If you choose hardware first and then consider software to go with it, then you are putting the cart before the horse. I prefer to provide solid solutions by sticking with high quality software and purchasing good quality, well supported hardware to go with that software. Finding good hardware to run OpenBSD is easy. In fact I mostly find OpenBSD to run well on random computers which I try to run it on.
Stop yapping Theo and go write some firewire drivers, or whatever technology came out 5 years ago that your system still doesn't support.
Theo and Co. do not write OpenBSD for you, they write it for themselves. If they don't want or need firewire, they won't spend time writing software for it. The lack of support of some device is not indicative of a lack of OpenBSD quality. It is indicative of either a lack of interest or a lack of open documentation.
If you don't like it, don't use it. But please don't exagerate the weak points.
Considering most NDAs have a clause that says you're not even allowed to admit that the NDA exists in the first place, it's hopeless to ask me this, but I can tell you this: I'm a 21-year old student from the North West of Germany, far far away from the headquarters of Apple in Cupertino, and unlikely to have inside knowledge.
I've been under a few NDA's, one of them was under a foreign company who never met me face to face. Apple though, is so careful that I guess this would not fly with them.
I'm not a US citizen either, but my country is one of USA's "bitches" it seems and their laws are coming to a town near me soon. : ( A few nights ago, an advertisement on TV warned that distributing a new release game would carry a penalty as high as 2 years in prison!
My GF still wants a Powerbook, the idea of an Apple has grown on her. She loves my mini. We're waiting for the next round of Powerbooks. I would really like full speed access to the supplied DDR RAM though.
Personally, I never want someone who works with me to shrug of failure. I want to understand why they failed and learn from it.
When I say "shrug off failure", I mean that their failure itself is not a big deal to them and they quickly move on to understanding their failure and fixing it, instead of spending time worrying about it and trying to make it look less of a failure or pointing the finger elsewhere.
Their failing, not the failure itself. Of course the failure is important to understand and learn from.
Yeah, some people won't make the cut, but making a judgement along the lines of "17 years of people and 2 qualify as exceptional" is a harsh standard to impose.
Maybe I'm harsh then. Exceptional is exceptional after all. Maybe my idea of the word is too strong for your liking. When I state exceptional, I am talking about the full package. Someone who is brilliant and reasonable for example. Someone who is truely gifted yet willing to consider an alternative idea or opinion.
I don't see it as black-and-white, I see a majority of plodders, just turning up to work to pay the bills. Some loud arseholes who are "it", talk a lot but don't really contribute much more than the "safe-bet pre-packaged half-solution" which almost fits and then boast about how wonderful they are. Then there are the nice folks who care about the job but don't do it all that well. And then there are those few exceptionals who not only manage to put up with the rest, but also see the real issues and fix them in creative ways that have the bullshit artists desperately searching for some way to save face and the plodders complaining that this makes everyone else look bad (and the nice folks congratulating).
There is a very bad problem in IT where I live, where people protect their knowledge and if they're lacking they bullshit. Those people are not exceptional and unfortunately I see the majority either dull and many being full of crap.
This is really how I see it. Maybe things are great in your part of the World, but in my part of the World the mundane rule and protect each other. I'm quite sick of being verbally shot down by morons when I know I have an excellent, unique solution which actually fits while using current infrastructure (and my track record show this time and time again when I actually get given the chance). It is true that I have difficulty with these people, but my difficulty does not make them any less mundane or insecure.
I do need to work on my people skills to learn to manage these types, but I think my problems are nowhere near as bad as theirs and at least I am aware and willing to face my shortcomings. A bullshit artist is a bullshit artist and plodder is a plodder. They are not what they are because of me, but I do see them for what they are.
Maybe I'm right, maybe I'm not. But I useta work at a software company, and I've seen some of the silly things people have been labeled 'incompetent' for.
Fair enough. I guess (and I should know, working for the crowd I work for), that money, time, exaustion and expectations can make anyone angry when things don't go as planned.
I have no idea how the Microsoft devs are treated, so I can't blame them without knowing them. I left more than one job because a manager did not understand or appreciate the issues they're supposed to be managing.
I want my software written by George! Google for the entire story if you are interested - it is one of my all-time favorites (written in 1957).
Wow, you have me wanting to read more. I normally read technical books only, but I'm compelled to read more. Thanks.
You probably should not make such sweeping generalizations. There are many personality types among people who are very effective at very complex tasks.
You are absolutely correct. I was too hasty with that post. Need sleep......
Real engineers fix problems, they don't get emotional.
This is so true. I've worked with many people in IT and communications over the past 17 years, in financial, military and educational institutions from desktop support to reverse engineering. People who get emotional when challenged or proven wrong are putting their ego before the problem. Their ego becomes the biggest problem and the real problem they're getting paid to fix tends to get fixed in a way that makes them look good, which might not actually be the technically better way.
The most exceptional people I have worked with, shrugged failure off and carried on with fixing things or making them better. The loudest people don't know shit and cover it up with fast talking. It seems the quiet, well educated people who are comfortable with themselves are the ones who make the biggest differences.
Unfortunately, in the past 17 years, only two people in my mind stand out to be the exceptional people, the rest are all competing in a bullshit competition with each other or are otherwise mediocre.