I got a chance to do a seat-of-the pants comparison between my 8500 w/ G3@466MHz and an iMac 500MHz box. The 8500 was running 10.1 and Quartz compression while the iMac was running 10.0.4 stock. 8500 was using the onboard video which means no graphics acceleration exists.
The iMac was smoother than the 8500 in almost all respects. This smoothness and quickness was especially evident when I launched Apple System Profiler, selected the Extensions tab and paged up/down this list. The 8500 would wait a while before it displayed the next page while the iMac did the same thing without any hesitation. Bus speed may also contribute to the performance difference--the 8500's bus runs at 45MHz.
I thought that memory for older Macs like legacy PCI models were expensive because I looked at Crucial. OWC (where UUX can be downloaded) have very reasonable prices. Sure, it's not as inexpensive as PC-133 SDRAM; but it isn't as expensive as Crucial FPM DIMMs either. I believe 128MB costs around $40 dollars whereas it cost over $350 at Crucial.
There's a new version available. 2.0b4 at the same site. It's a nice piece of software, especially his L2CacheConfig utility that enables 3rd-party L2 caches for CPU upgrades.
I have run into overheating problems with a Quantum Viking 4GB 7200RPM SCSI in my Macintosh 8500 several years ago. The symptoms were clicking noises and frequent lockups of the system running Mac OS 9. This drive was, I guess, a 1st- or 2nd-generation 7200RPM because when I turned off the box and opened it up, the drive's surface was hot to the touch. If I left the cover off, the drive wouldn't fail. Closing up the machine, the drive failed within 24 hours while left on just idling.
I turned off the box for 24 hours and restarted with the cover off so I could save what data I had to another drive.
On the other hand, I have a 1st generation Seagate ST15150N 4GB SCSI 7200RPM in my HP workstation which gets very hot--too hot to touch with my bare hands--with an up time of 110 days so far without any problems. It's been in this system for around 4 months; hope it will last for many more.
It's interesting how Tandem Computers was started by ex-HP employees. Compaq purchased Tandem, and with HP purchasing Compaq, Tandem is folded back into HP. Kinda like Apple/NeXT.
Hailstorm is bad becausee it will allow full functionality only on Windows-based devices. This is against the original intention of the global Internet.
An effective way to reduce fan noise is to use an 80mm fan through the use of a 60mm-to-80mm adaptor. Some really slick adaptors can be purchased from www.plycon.com under fan adapters.
I have an AMD system and am relatively happy with it. But I wish AMD would include: a heat spreader for better dissipation of heat/prevent cracking the core when installing heatsinks; and thermal protection. One of these days, I'm going to come home and find my CPU has fried because my CPU fan has failed. My A7V's latest 1008 BIOS has fan/CPU temperature protection, but both features don't work right. Fan monitor thinks my 80mm fan is spinning too slow and makes the annoying alert sound; temp monitor is just broken and shuts down the box right after it is turned on. Consuming less wattage would be a nice feature, too. Can't wait for 0.13u.
The test on anandtech shows Intel's 845 chipset is intentionally crippled. Intel probably don't want people to say/think "Gee, your SDRAM chipset has latency as bad as DRDRAM and yet VIA's chipset has better latency; with performance within 5%, and a lot less $ investment using DDR SDRAM, I will buy the VIA product instead of yours."
if there's a branch, the pipeline has to be flushed
Your reply implies the following towards the end but wasn't clear. Pipelines aren't automatically flushed as you first imply. A CPU has to decide which fork to take when it loads instructions after the branch is read into the pipeline. Only if the code takes the branch that's not already in the pipeline does the CPU discard the pipeline's contents.
One of the things I really like about Apple's trackpad software is I can do a tap-and-a-half and have it function as a click-lock. I can then move my cursor to where ever I please and release with another tap. With Windows, I can do a tap-and-a-half but the software releases the click-lock as soon as I stop moving my mouse. This is just stupid. How can I configure the trackpad to work the way it's supposed to?
Spending more time on the net will not make a person more well-rounded. It will expose them to other people's experiences but reading about those experiences will not improve their practical experience. People who read about things are sometimes caught off-guard when their theory meet practice. Things like human interaction are hidden behind a facade when you use email or even instant messaging. People who surf a lot will have the opportunity to gain more knowledge, but it will not make them well-rounded in the traditional sense.
The reason PC emulation never took off on the Macintosh is because the emulation was slow compared to a real PC. I cannot recall the speed when Soft Windows came out, but it was orders of magnitudes slower.
Personally, I feel Palm devices still have an advantage in battery life.
Yes, it's true lots of people are exposed to the Windows interface. GNOME and some other desktop managers mimic the MS Windows metaphor but doesn't do it exactly and probably frustrates those who are already familiar with MS Windows. The fact that lots of people work on GNOME doesn't help with its inconsistencies.
I think that's the wrong attitude for them to take. Maybe they've overheard elitists say newbies are stupid or people have told them directly. When you hear people say this what do you say in response? If you don't say anything, it only reinforces how they feel.
Hey, I don't know if the Aki pictures are legitimate or not. I do recall when the pictures of the Macintosh G4 Cube were leaked several days prior to their introduction and this 2D graphics expert itemized something like 5-6 things wrong with the photos. They sounded very reasonable but the truth is the photographs were legitimate despite the well reasoned speculations. Unless someone on the inside is actually posting, we'll probably never know.
No inside scoop, but if you do a strings on many utilities such as update_prebinding, you'll see that there are indications of portable code with architectures ranging from the 68k through PPC7400, with some PA-RISC and SPARC sprinkled in.
Most likely a lot of it is from the codebase's NeXTSTEP lineage.
Nothing stops another company from introducing another service. But if the masses are dependent upon Microsoft applications, these companies will have a more difficult time getting these MS apps to work seamlessly with their services instead of Microsoft's. Microsoft thus leverages their desktop monopoly to gain a foothold and perhaps dominance/monopoly in another market.
Compiling under FreeBSD 4.3, with g++ 2.95.3, just invoking g++ gives me a file size of 5228 bytes; stripping it gives me 3500 bytes.
The executable requires the libstdc++, libm, and libc shared libraries.
At work, our binaries display a stack trace if it gets a critical signal which would prevent function names from being displayed if these binaries were stripped. So we leave them in for supportability.
A non-technical person will not infer there are any different types of "open source." The question asked was Linux and open source. Nowhere does GPL, GNU, BSD get mentioned. Without Ballmer specifically qualifying his statements, he is incorrectly generalizing.
FreeBSD's port system provides a way for you to compile a program along with all its dependent programs/libraries. i.e. if you want to make Mozilla, just
Taking anti-inflammatories like Ibuprofin and Aspirin often will have the effect of suppressing your body's natual ability to cause inflammation--the act of healing injuries. It's much better to fix the cause, whether it be excessive computer use, not taking enough breaks and stretching, or lack of exercise.
RSI is just the first stage of repetitive stress. You don't want to get RSD, which is basically permanent nerve damage. I highly recommend the book Repetitive Strain Injury by Pascarelli. Here's a link at Amazon. I'm not affiliated with Amazon.com in any way. Fatbrain also has this book, here; but it is more expensive.
I have had RSI since the mid '90s. Whatever treament is used, one shouldn't over do it. While I was on short term disability, I had to type for 20-30 mins, rest 5 mins for an 8-hour period for a week before my doctor would send me back to work. Trying to do this the first time with ice packs during the breaks put me into a relapse that was much more severe than it was before. My RSI is caused by overwork and lack of good circulation. Although the cold slowed inflammation, it prevented good circulation which ultimately is required to heal the damaged soft tissues.
For me, the only things that work are exercise and knowing what parts of my body cause various symptoms. The former part works well generally while the latter part gives me a way to deal with acute symptoms. Having RSI has increased my awareness of my body's signals to my brain whenever I overdo a physical activity--i.e. typing, mousing (cannot mouse anymore, must use a track ball), and even laying on the couch for extended amounts of time (static load on the neck).
About exercise, I'm not talking about going out for a walk. One needs to do 30 minutes/day at 80% of maximum heart rate 3 times a week. Discontiguous workouts per day can be combined so that they total 30 minutes, but since you need to warm up/down for 5 minutes each--to prevent injury--it would be better to do it once.
When you use DVD Genie with a DVD-ROM that's been modified to RPC-1, you can change the region code as many times as you like and be able to play legacy and RCE discs.
RPC-1 firmware is for getting around the drive's region code change limit, you can change it more than 4-5 times; DVD Genie is for getting around the player's region code limit. Both the DVD-ROM and the software's region must match or else your discs won't play.
Slashdot Mirror
The iMac was smoother than the 8500 in almost all respects. This smoothness and quickness was especially evident when I launched Apple System Profiler, selected the Extensions tab and paged up/down this list. The 8500 would wait a while before it displayed the next page while the iMac did the same thing without any hesitation. Bus speed may also contribute to the performance difference--the 8500's bus runs at 45MHz.
I thought that memory for older Macs like legacy PCI models were expensive because I looked at Crucial. OWC (where UUX can be downloaded) have very reasonable prices. Sure, it's not as inexpensive as PC-133 SDRAM; but it isn't as expensive as Crucial FPM DIMMs either. I believe 128MB costs around $40 dollars whereas it cost over $350 at Crucial.
There's a new version available. 2.0b4 at the same site. It's a nice piece of software, especially his L2CacheConfig utility that enables 3rd-party L2 caches for CPU upgrades.
I turned off the box for 24 hours and restarted with the cover off so I could save what data I had to another drive.
On the other hand, I have a 1st generation Seagate ST15150N 4GB SCSI 7200RPM in my HP workstation which gets very hot--too hot to touch with my bare hands--with an up time of 110 days so far without any problems. It's been in this system for around 4 months; hope it will last for many more.
It's interesting how Tandem Computers was started by ex-HP employees. Compaq purchased Tandem, and with HP purchasing Compaq, Tandem is folded back into HP. Kinda like Apple/NeXT.
Hailstorm is bad becausee it will allow full functionality only on Windows-based devices. This is against the original intention of the global Internet.
An effective way to reduce fan noise is to use an 80mm fan through the use of a 60mm-to-80mm adaptor. Some really slick adaptors can be purchased from www.plycon.com under fan adapters.
I have an AMD system and am relatively happy with it. But I wish AMD would include: a heat spreader for better dissipation of heat/prevent cracking the core when installing heatsinks; and thermal protection. One of these days, I'm going to come home and find my CPU has fried because my CPU fan has failed. My A7V's latest 1008 BIOS has fan/CPU temperature protection, but both features don't work right. Fan monitor thinks my 80mm fan is spinning too slow and makes the annoying alert sound; temp monitor is just broken and shuts down the box right after it is turned on. Consuming less wattage would be a nice feature, too. Can't wait for 0.13u.
The test on anandtech shows Intel's 845 chipset is intentionally crippled. Intel probably don't want people to say/think "Gee, your SDRAM chipset has latency as bad as DRDRAM and yet VIA's chipset has better latency; with performance within 5%, and a lot less $ investment using DDR SDRAM, I will buy the VIA product instead of yours."
Easier to write vaporware.
Your reply implies the following towards the end but wasn't clear. Pipelines aren't automatically flushed as you first imply. A CPU has to decide which fork to take when it loads instructions after the branch is read into the pipeline. Only if the code takes the branch that's not already in the pipeline does the CPU discard the pipeline's contents.
One of the things I really like about Apple's trackpad software is I can do a tap-and-a-half and have it function as a click-lock. I can then move my cursor to where ever I please and release with another tap. With Windows, I can do a tap-and-a-half but the software releases the click-lock as soon as I stop moving my mouse. This is just stupid. How can I configure the trackpad to work the way it's supposed to?
Spending more time on the net will not make a person more well-rounded. It will expose them to other people's experiences but reading about those experiences will not improve their practical experience. People who read about things are sometimes caught off-guard when their theory meet practice. Things like human interaction are hidden behind a facade when you use email or even instant messaging. People who surf a lot will have the opportunity to gain more knowledge, but it will not make them well-rounded in the traditional sense.
Personally, I feel Palm devices still have an advantage in battery life.
Yes, it's true lots of people are exposed to the Windows interface. GNOME and some other desktop managers mimic the MS Windows metaphor but doesn't do it exactly and probably frustrates those who are already familiar with MS Windows. The fact that lots of people work on GNOME doesn't help with its inconsistencies.
I think that's the wrong attitude for them to take. Maybe they've overheard elitists say newbies are stupid or people have told them directly. When you hear people say this what do you say in response? If you don't say anything, it only reinforces how they feel.
Hey, I don't know if the Aki pictures are legitimate or not. I do recall when the pictures of the Macintosh G4 Cube were leaked several days prior to their introduction and this 2D graphics expert itemized something like 5-6 things wrong with the photos. They sounded very reasonable but the truth is the photographs were legitimate despite the well reasoned speculations. Unless someone on the inside is actually posting, we'll probably never know.
Most likely a lot of it is from the codebase's NeXTSTEP lineage.
Nothing stops another company from introducing another service. But if the masses are dependent upon Microsoft applications, these companies will have a more difficult time getting these MS apps to work seamlessly with their services instead of Microsoft's. Microsoft thus leverages their desktop monopoly to gain a foothold and perhaps dominance/monopoly in another market.
The executable requires the libstdc++, libm, and libc shared libraries.
At work, our binaries display a stack trace if it gets a critical signal which would prevent function names from being displayed if these binaries were stripped. So we leave them in for supportability.
A non-technical person will not infer there are any different types of "open source." The question asked was Linux and open source. Nowhere does GPL, GNU, BSD get mentioned. Without Ballmer specifically qualifying his statements, he is incorrectly generalizing.
cd /usr/ports/www/mozilla
make
RSI is just the first stage of repetitive stress. You don't want to get RSD, which is basically permanent nerve damage. I highly recommend the book Repetitive Strain Injury by Pascarelli. Here's a link at Amazon. I'm not affiliated with Amazon.com in any way. Fatbrain also has this book, here; but it is more expensive.
For me, the only things that work are exercise and knowing what parts of my body cause various symptoms. The former part works well generally while the latter part gives me a way to deal with acute symptoms. Having RSI has increased my awareness of my body's signals to my brain whenever I overdo a physical activity--i.e. typing, mousing (cannot mouse anymore, must use a track ball), and even laying on the couch for extended amounts of time (static load on the neck).
About exercise, I'm not talking about going out for a walk. One needs to do 30 minutes/day at 80% of maximum heart rate 3 times a week. Discontiguous workouts per day can be combined so that they total 30 minutes, but since you need to warm up/down for 5 minutes each--to prevent injury--it would be better to do it once.
RPC-1 firmware is for getting around the drive's region code change limit, you can change it more than 4-5 times; DVD Genie is for getting around the player's region code limit. Both the DVD-ROM and the software's region must match or else your discs won't play.
www.google.com is your friend.