I've got a Thinkpad A21p, and I regularily drag it around with me everywhere. It's got the 1600x1200, 15" screen. It's got some special shell over the screen, and the whole thing clamps together rather well. I'm, on the whole, incredibly happy with it. Though, I wish IBM would've omitted the floppy drive or at least, replaced the floppy drive with a zip drive. The screen hinges are tough, much tougher than the PowerBook hinges I've seen - also, the edge of the screen wraps down and catches with the side of the base, so it is less likely to stress the hinges with lateral movement (being jarred around.). I did put a piece of duct tape over the bottom dock-mate port. I felt that the flaps could be moved aside too easily, and didn't want anything getting in there. Oh, my thinkpad doesn't seem to have a fan that I know about, and if you use it for 30 hours straight, it gets warm on the bottom, but, I could easily leave it on my lap, not like some of the powerbooks.
Anyways, about wear and tear. The corners of the laptop are now showing some brassing. Otherwise, I just have a roll of foam in the bottom of my backpack's laptop pouch, to kushion the beast when I set my bag down.
I've had this thing since mid-summer, and it's still fine. I'm confident it could handle a reasonable drop if the lid was shut, but, otherwise, I'd feel safer with one of those iBooks with a polycarbonate shell. I take it everywhere - the law library reading room here at the university of michigan is a great place to study; I lug it there every day. Also, I tend to keep a few sheets of paper caught between the keyboard and the screen - and, no problems yet. The screen, again, is gorgeous. On the whole, I like the machine much better than the competing Dell models. It's definately sturdier and tougher, and, IMHO, a better engineered product.
Btw, I also like the Thinkpad T series, and was considering one - but, I got this A21p used, and was sold on the screen. Now, I'd be keen on a T22 or something with the 1400x1050 14.1", though. At the time, the best was a 1024x768, I believe.
With laptops, I've found that what matters most is the amount of ram, the size of the screen, and the battery life. I personally, couldn't care less if my laptop was 1ghz, 850 mhz, or 400 mhz. I speedstep my processor as slow as it can go. Hell, I use linux vga-console a lot.
The new iBook's tempt me for this reason - great battery life, OSX, decent screen, and industrial case.
I was not impressed with the Inspiron 3500's I worked with this summer. They were being used in the field by pipeline engineers, and were regularly coming back with the shit knocked out of them.I get the feeling my thinkpad would have fared better. Some of our execs were using 8000 series, and they were decent, but might as well have beeen desktop machines - they never left the office. The people with 5000's seemed to take them back and forth from home with enough regularity. (This might be that the people with the more expensiev laptops didn't do work at home, too.)
Someone mentioned that the 8000's had poor build quality - I agree. Though, I liked the 5000's - they were tight. I'd definately recommend T-series IBM replacements though, if I could.
Oh, yeah, I also had the pleasure of using a sony 505JS for a while, and loved it. The screen was small, but, the whole package was tiny. If owned one, I'd be tempted to mill out a box to keep it in, if I could keep it in something roughly the size of my calc text, I'd be happy and confident enough to throw it around recklessly. Granted, putting it in a thick metal box would defeat it's light weight, but I'd have the option between insane ruggedness and portability. If I ever replace my A21, I'd look at these.
What's wrong with that building? The 1st and 3rd floors had little redeeming value, but, the rest of it was good. In fact, I spent a lot of time in that building. It's a good building, a solid building - It's not falling down any time soon.
Aight, sorry - I do agree with you, and have a stack of IBM DDYS disks to show for it - BUT, you said some things that were wrong:
Just about all PCI ide controllers can use DMA, which cuts the cpu intensiveness down to almost the scsi-level.
Further, Not all drives have to have their own individual interrupt - this depends on the ide chip and how they are arranged on the pci daughterboard, or on the motherboard. (interrupt sharing, etc.) Promise chips will use one interrupt for two interfaces.
SCSI does offer a whole slew of advantages, disconnect, command queueing, etc. These are advantages in a RAID setup. IDE does suck, but not because it's cpu intensive or gobbles interrupts.
Instead of listening to vague stories about someone's hairdresser's uncle.. why don't you learn something about the technology? 802.11b uses frequency hopping and direct sequence spread spectrum transmission techniques to avoid interference. The sequence correlation in the DSSS will make your transmissions distinct from those of the cordless phone. Even large bursts of RF, like microwave ovens, will not significantly impair the 802.11b functionality.
First off, I suggest you get a college education - it's an amazing asset. Try taking a signals course, even a simple one that covers filters, z-transforms, and DFT's. You'll hit a topic called sliding correlation, which is the basis of transmission in most 2.4 ghz equipment. We hit this my freshman year.
Do your homework and make your own decisions - don't let others make them for you.
(I'm not trying to troll, but - this person is asking for it! They're not bothering to investigate the issue and complaining about it. Laziness is one thing, but to complain about being lazy is disgusting.)
Somehow, another U of M, the University of Michigan, manages to do this for about 40K students. I don't know what your operating budget is, but, It's certainly not outside the scope of a university. Oh, our IFS space is backed up regularily and you can submit requests for files from tape; so, it's not just some mirrored FC cabinet.
Re:Why keep re-inventing SCSI? - its the cables
on
Firewire and Linux?
·
· Score: 1
>> When you dig down into the SBP-2 layer of IEEE-1394 you will find that it is SCSI. SCSI commands and responses are used for the mass storage device on IEEE-1394. The only thing that is different is the physical and low level signal transmission. So, at the software level (once you get above the lowest level packet sender/receiver) there is no difference from scsi.
Hey - funny that.. you know, ATAPI, FibreChannel, and USB all use SCSI commands too?
>> At the physical level you get to trade a 50 or 68 pin connector and cable for a 6 or 4 pin connector and cable. The controller chips probably cost about the same in volume, maybe a couple of bucks different. A good SCSI cable (and don't mess with bad ones) is $50. A good firewire cable is $7.
You're also trading 100, 160, or 200 MegaBytes/s (ata-100,ultra-160,2 gig FC) for 50 MegaBytes/s.. And if minimizing the number of conductors is your goal, FC over copper only uses two LVPECL pairs..
>> There is your reason. A $300 disk is $350 with SCSI and $308 with Firewire. (I added a dollar for the $0.50 license fee on the ports at each end of the cable.:-) A 12% cost savings will win in the end.
Since u160 scsi is almost three times as fast, with firewire, you get almost a 70% saving in usable bandwidth!
>> Non-tangibles such as easy configuration, the ability to pile a dump truck load of disks on a single interface, and not becoming ensnared in a wriggling mass of cables are just nice bonuses.
What a bonus! Not only is your bus markedly slower, it supports more devices, so you can take advantage of that absent bandwidth!
Honestly - I'm not trolling. (honest) - I'm trying to add perspective.. '1394 is nifty, and certainly better than USB, but, it is by no means a replacement for ATA, SCSI, or FC. Sure, It's cheaper, sure it uses cheaper cable.. but, you get a cheaper technology. Like dad always said, 'You get what you pay for.' Me? I'll pay the $42 extra for my U160 drives.
>For the KVM switcher that needed 9 volts, I tried using the 7 volt level, which almost worked, but not well enough, so I took the 12VDC line, dropped a variable resistor on in a voltage adjustment (not rheostat) config, measured with the DVM for the proper voltage (a little over 9VDC), then tried it - it worked fine. I let it run for a while, then felt it for heat - not too big a deal, ran cool. I could have also dropped the voltage to 9 volts over a few diodes as well, or if I wanted to be really cool - use a variable regulator circuit. But this particular device didn't need it.
Uhh. Your voltage will vary with your load - it suits you to use a voltage regulator. Use an LM7809 and attach it to your 12Volt line and ground. You should be able to get these from jameco or even radio shack.
They really aren't criticizing.NET languages, software, or architecture. Microsoft is positioning it's passport system to collect phenomenal amounts of information about people without their knowledge. Hence, the attack on privacy.
Honestly - the open source operating system community stands to benefit from Itanium more than Hammer. Kernel extensions for itanium and hammer are already underway, with gcc targets already developed. There are few 64-bit architectures that are being actively developed anymore. Alpha and MIPS are basically EOL, only UltraSparc survives. If Intel has a fast, efficient 64 bit processor, we can expect Linux to immediately benefit, recompile everything and you're gold. Hammer includes a set of 64 bit extensions to ia-32. IA-64, by all reports, is much better designed. With the ability to recompile all code for native ia-64 bytecode, people don't need to bother with ia-32 emulation. x86 should've died more than a decade ago; unfortunately, it seems that in general the worst technology seems to be most fundamentally embraced.
I think that your needs are best solved with the ucsimm. Its relatively cheap ($210), however, it already runs uCLinux, has 8 mb ram, 2 mb flash, and all the development tools set out. That means more of your time will be spent developing your application, rather than hacking up TI hardware.;) Check out the uclinux website.
Well. 6.1 megapixels through software-interpolation. The CCD is not really 6.1 megapixels. Also, the camera is based off of the Nikon N60 body. IMHO, an unfortunate choice. The N60 doesn't support many of the really novel features of the newer lenses coming out, like AFS and VR. Its' autofocus also leaves much room for improvement. A better choice to have built this camera on would have been the Nikon F100, N70, or N90s. It is, though, still cheaper than the Nikon D1 or a Kodak Hybrid.
It is beautiful. Its a professional quality Digital Camera, takes Nikon's F-Mount Autofocus lenses. This isn't a camera with a cheap, small CCD. Its not a full 35mm size ccd, but, its still a really really good. And, Only $6,000. Btw, you'd have to spend a good $2000 on lenses, too, like the 17-35f2.8, 50f1.8, and 80-200f2.8 AFS. Nikon Website's Propaganda Or, You could get a Kodak DCS 620, which has a full-frame 35mm ccd. Its ungodly sharp and crisp. Only, what, $20,000? (This too, uses Nikon Lenses.) Nikon lenses are quality. Yeah, They're not Zeiss lenses, but, They're not shabby either, better than that Tamron/Canon/Fuji crap. The reason I point out these two cameras, is because often times I see people buying just by the numbers. They want to maximize the amount of pixels, and minimize the cost. Now, In my humble estimation, 1024x768 sharp color-balanced pixels is better than 16000x12000 pixels that have been software-interpolated from some crappy ccd.. I really want a D1, and may purchase on this summer. It'll use all of my nikon lenses, and is a really really nice camera. Though, I'm really tempted to wait until they come out with a Digital camera which has the same frame size as a my film cameras. So, until then, I shall stay with the best digital photography solution: Nikon film camera, Fujichrome (velvia, provia, astia, whatever.), and a film-scanner. The Fujichrome has a lot more resolution than a ccd. However, The CCD's they put in slide-scanners take minutes to go over the whole slide, thus, they do a much better job than the CCD's in a regular digital camera.
In regards to silent operation.. I have a Fujitsu M2934QAU that sounds much like an unlubricated jet taking off when it spins up. Drives works fine, though.
They're using the new spiffy 640x480x(4bpp) framebuffer stuff in the new kernels which was put in mostly for this. As its the minimum that everyone will support unless you are using CGA, EGA, MDA, or some strange new video card that is not compatible with the old (correct me if I am wrong, but I believe this mode was introduced with the) IBM 8514's.
However, It does look prettier than the old text stuff. The Tetris at the end is a good idea.
Slashdot Mirror
I've got a Thinkpad A21p, and I regularily drag it around with me everywhere. It's got the 1600x1200, 15" screen. It's got some special shell over the screen, and the whole thing clamps together rather well. I'm, on the whole, incredibly happy with it. Though, I wish IBM would've omitted the floppy drive or at least, replaced the floppy drive with a zip drive. The screen hinges are tough, much tougher than the PowerBook hinges I've seen - also, the edge of the screen wraps down and catches with the side of the base, so it is less likely to stress the hinges with lateral movement (being jarred around.). I did put a piece of duct tape over the bottom dock-mate port. I felt that the flaps could be moved aside too easily, and didn't want anything getting in there. Oh, my thinkpad doesn't seem to have a fan that I know about, and if you use it for 30 hours straight, it gets warm on the bottom, but, I could easily leave it on my lap, not like some of the powerbooks.
Anyways, about wear and tear. The corners of the laptop are now showing some brassing. Otherwise, I just have a roll of foam in the bottom of my backpack's laptop pouch, to kushion the beast when I set my bag down.
I've had this thing since mid-summer, and it's still fine. I'm confident it could handle a reasonable drop if the lid was shut, but, otherwise, I'd feel safer with one of those iBooks with a polycarbonate shell. I take it everywhere - the law library reading room here at the university of michigan is a great place to study; I lug it there every day. Also, I tend to keep a few sheets of paper caught between the keyboard and the screen - and, no problems yet. The screen, again, is gorgeous. On the whole, I like the machine much better than the competing Dell models. It's definately sturdier and tougher, and, IMHO, a better engineered product.
Btw, I also like the Thinkpad T series, and was considering one - but, I got this A21p used, and was sold on the screen. Now, I'd be keen on a T22 or something with the 1400x1050 14.1", though. At the time, the best was a 1024x768, I believe.
With laptops, I've found that what matters most is the amount of ram, the size of the screen, and the battery life. I personally, couldn't care less if my laptop was 1ghz, 850 mhz, or 400 mhz. I speedstep my processor as slow as it can go. Hell, I use linux vga-console a lot.
The new iBook's tempt me for this reason - great battery life, OSX, decent screen, and industrial case.
I was not impressed with the Inspiron 3500's I worked with this summer. They were being used in the field by pipeline engineers, and were regularly coming back with the shit knocked out of them.I get the feeling my thinkpad would have fared better. Some of our execs were using 8000 series, and they were decent, but might as well have beeen desktop machines - they never left the office. The people with 5000's seemed to take them back and forth from home with enough regularity. (This might be that the people with the more expensiev laptops didn't do work at home, too.)
Someone mentioned that the 8000's had poor build quality - I agree. Though, I liked the 5000's - they were tight. I'd definately recommend T-series IBM replacements though, if I could.
Oh, yeah, I also had the pleasure of using a sony 505JS for a while, and loved it. The screen was small, but, the whole package was tiny. If owned one, I'd be tempted to mill out a box to keep it in, if I could keep it in something roughly the size of my calc text, I'd be happy and confident enough to throw it around recklessly. Granted, putting it in a thick metal box would defeat it's light weight, but I'd have the option between insane ruggedness and portability. If I ever replace my A21, I'd look at these.
In conclusion: Thinkpad's rule, so do iBooks.
What's wrong with that building? The 1st and 3rd floors had little redeeming value, but, the rest of it was good. In fact, I spent a lot of time in that building. It's a good building, a solid building - It's not falling down any time soon.
Aight, sorry - I do agree with you, and have a stack of IBM DDYS disks to show for it - BUT, you said some things that were wrong:
Just about all PCI ide controllers can use DMA, which cuts the cpu intensiveness down to almost the scsi-level.
Further, Not all drives have to have their own individual interrupt - this depends on the ide chip and how they are arranged on the pci daughterboard, or on the motherboard. (interrupt sharing, etc.) Promise chips will use one interrupt for two interfaces.
SCSI does offer a whole slew of advantages, disconnect, command queueing, etc. These are advantages in a RAID setup. IDE does suck, but not because it's cpu intensive or gobbles interrupts.
Instead of listening to vague stories about someone's hairdresser's uncle.. why don't you learn something about the technology? 802.11b uses frequency hopping and direct sequence spread spectrum transmission techniques to avoid interference. The sequence correlation in the DSSS will make your transmissions distinct from those of the cordless phone. Even large bursts of RF, like microwave ovens, will not significantly impair the 802.11b functionality.
First off, I suggest you get a college education - it's an amazing asset. Try taking a signals course, even a simple one that covers filters, z-transforms, and DFT's. You'll hit a topic called sliding correlation, which is the basis of transmission in most 2.4 ghz equipment. We hit this my freshman year.
Do your homework and make your own decisions - don't let others make them for you.
(I'm not trying to troll, but - this person is asking for it! They're not bothering to investigate the issue and complaining about it. Laziness is one thing, but to complain about being lazy is disgusting.)
Somehow, another U of M, the University of Michigan, manages to do this for about 40K students. I don't know what your operating budget is, but, It's certainly not outside the scope of a university. Oh, our IFS space is backed up regularily and you can submit requests for files from tape; so, it's not just some mirrored FC cabinet.
>> When you dig down into the SBP-2 layer of IEEE-1394 you will find that it is SCSI. SCSI commands and responses are used for the mass storage device on IEEE-1394. The only thing that is different is the physical and low level signal transmission. So, at the software level (once you get above the lowest level packet sender/receiver) there is no difference from scsi.
:-) A 12% cost savings will win in the end.
Hey - funny that.. you know, ATAPI, FibreChannel, and USB all use SCSI commands too?
>> At the physical level you get to trade a 50 or 68 pin connector and cable for a 6 or 4 pin connector and cable. The controller chips probably cost about the same in volume, maybe a couple of bucks different. A good SCSI cable (and don't mess with bad ones) is $50. A good firewire cable is $7.
You're also trading 100, 160, or 200 MegaBytes/s (ata-100,ultra-160,2 gig FC) for 50 MegaBytes/s.. And if minimizing the number of conductors is your goal, FC over copper only uses two LVPECL pairs..
>> There is your reason. A $300 disk is $350 with SCSI and $308 with Firewire. (I added a dollar for the $0.50 license fee on the ports at each end of the cable.
Since u160 scsi is almost three times as fast, with firewire, you get almost a 70% saving in usable bandwidth!
>> Non-tangibles such as easy configuration, the ability to pile a dump truck load of disks on a single interface, and not becoming ensnared in a wriggling mass of cables are just nice bonuses.
What a bonus! Not only is your bus markedly slower, it supports more devices, so you can take advantage of that absent bandwidth!
Honestly - I'm not trolling. (honest) - I'm trying to add perspective.. '1394 is nifty, and certainly better than USB, but, it is by no means a replacement for ATA, SCSI, or FC. Sure, It's cheaper, sure it uses cheaper cable.. but, you get a cheaper technology. Like dad always said, 'You get what you pay for.' Me? I'll pay the $42 extra for my U160 drives.
>For the KVM switcher that needed 9 volts, I tried using the 7 volt level, which almost worked, but not well enough, so I took the 12VDC line, dropped a variable resistor on in a voltage adjustment (not rheostat) config, measured with the DVM for the proper voltage (a little over 9VDC), then tried it - it worked fine. I let it run for a while, then felt it for heat - not too big a deal, ran cool. I could have also dropped the voltage to 9 volts over a few diodes as well, or if I wanted to be really cool - use a variable regulator circuit. But this particular device didn't need it.
Uhh. Your voltage will vary with your load - it suits you to use a voltage regulator. Use an LM7809 and attach it to your 12Volt line and ground. You should be able to get these from jameco or even radio shack.
They really aren't criticizing .NET languages, software, or architecture. Microsoft is positioning it's passport system to collect phenomenal amounts of information about people without their knowledge. Hence, the attack on privacy.
Honestly - the open source operating system community stands to benefit from Itanium more than Hammer. Kernel extensions for itanium and hammer are already underway, with gcc targets already developed. There are few 64-bit architectures that are being actively developed anymore. Alpha and MIPS are basically EOL, only UltraSparc survives. If Intel has a fast, efficient 64 bit processor, we can expect Linux to immediately benefit, recompile everything and you're gold. Hammer includes a set of 64 bit extensions to ia-32. IA-64, by all reports, is much better designed. With the ability to recompile all code for native ia-64 bytecode, people don't need to bother with ia-32 emulation. x86 should've died more than a decade ago; unfortunately, it seems that in general the worst technology seems to be most fundamentally embraced.
I think that your needs are best solved with the ucsimm. Its relatively cheap ($210), however, it already runs uCLinux, has 8 mb ram, 2 mb flash, and all the development tools set out. That means more of your time will be spent developing your application, rather than hacking up TI hardware. ;) Check out the uclinux website.
Well. 6.1 megapixels through software-interpolation. The CCD is not really 6.1 megapixels. Also, the camera is based off of the Nikon N60 body. IMHO, an unfortunate choice. The N60 doesn't support many of the really novel features of the newer lenses coming out, like AFS and VR. Its' autofocus also leaves much room for improvement. A better choice to have built this camera on would have been the Nikon F100, N70, or N90s. It is, though, still cheaper than the Nikon D1 or a Kodak Hybrid.
The Nikon D1.
It is beautiful. Its a professional quality Digital Camera, takes Nikon's F-Mount Autofocus lenses. This isn't a camera with a cheap, small CCD. Its not a full 35mm size ccd, but, its still a really really good. And, Only $6,000. Btw, you'd have to spend a good $2000 on lenses, too, like the 17-35f2.8, 50f1.8, and 80-200f2.8 AFS.
Nikon Website's Propaganda
Or, You could get a Kodak DCS 620, which has a full-frame 35mm ccd. Its ungodly sharp and crisp. Only, what, $20,000? (This too, uses Nikon Lenses.)
Nikon lenses are quality. Yeah, They're not Zeiss lenses, but, They're not shabby either, better than that Tamron/Canon/Fuji crap.
The reason I point out these two cameras, is because often times I see people buying just by the numbers. They want to maximize the amount of pixels, and minimize the cost. Now, In my humble estimation, 1024x768 sharp color-balanced pixels is better than 16000x12000 pixels that have been software-interpolated from some crappy ccd.. I really want a D1, and may purchase on this summer. It'll use all of my nikon lenses, and is a really really nice camera. Though, I'm really tempted to wait until they come out with a Digital camera which has the same frame size as a my film cameras. So, until then, I shall stay with the best digital photography solution: Nikon film camera, Fujichrome (velvia, provia, astia, whatever.), and a film-scanner. The Fujichrome has a lot more resolution than a ccd. However, The CCD's they put in slide-scanners take minutes to go over the whole slide, thus, they do a much better job than the CCD's in a regular digital camera.
Moral: Check out Nikon Cameras.
In regards to silent operation.. I have a Fujitsu M2934QAU that sounds much like an unlubricated jet taking off when it spins up. Drives works fine, though.
They're using the new spiffy 640x480x(4bpp) framebuffer stuff in the new kernels which was put in mostly for this. As its the minimum that everyone will support unless you are using CGA, EGA, MDA, or some strange new video card that is not compatible with the old (correct me if I am wrong, but I believe this mode was introduced with the) IBM 8514's.
However, It does look prettier than the old text stuff. The Tetris at the end is a good idea.