I don't think Solaris is going away anytime soon - after all, there was a big push to get Sun to release Solaris 9 (SunOS 5.9) for the x86.
I've run Linux and Solaris on x86 boxes - Linux has a lot of neat bells and whistles, but Solaris seems to be a bit better thought out (have yet to try a BSD on x86). My biggest complaints with Solaris are the lack of an "smbmount" facility. and limited HW support My biggest complaint with Linux is that the desktops lack the refinement of CDE - CDE was designed, KDE and Gnome evolved.
Remember that Linux is just the kernel - a good portion of what you find in a typical distro can easily be ported to Solaris.
Prof Bean (taught Cal history at UCB) was quite the believer in the authenticity of the plate.
UFS does support logging
on
The Faded Sun
·
· Score: 1
For the record a form of journaling is available by specifying the "logging" option in/etc/vfstab. This has been in Solaris well before ext3 was available for Linux.
since gnome2 uses A LOT more ram and cpu cycles than good old cde
Not surprising. HP's VUE, the "father" of CDE was designed to run on 25 MHz 68040's with 8 MB of RAM. It also was designed to be workable with smaller screen sizes - 11 years ago, an HP monitor capable of 1280 by 1024 would set you back 6 grand!
The thing is, Oregon cannot require California-registered cars (or cars registered to any of the 49 other states) to have their tracking devices
Not to mention Mexican and Canadian cars and trucks.
If the problem is road damage, then the real solution is to require these devices for trucks (meaning semi's, not pickups). In addition, base the fees on weights recorded at weigh stations.
Oregon would be better off simply raising the fuel taxes to cover their revenue needs. This would also serve to encourage people to buy more fuel efficient vehicles.
This proposal would make the most sense for alternate fueled vehicles, e.g. electric, natural gas, etc.
IIRC, the Hiroshima bomb had about 50 kg of U-235. Fissioning 1 kg or U-235 or Pu-239 yields about 20 kilotons. The yield quoted for Hiroshima was 12.5 kT, which implies an efficiency (usually reffered to as burnup) of a bit over 1%. !00% burnup would yield about 1 MT.
The Nagasaki (Fat Man) bomb had a higher yield and IIRC also had less fissile material, I'd guess the burnup was 5 to 10%. The best reported burnups are on the order of 25%.
As other posters have pointed out, there wasn't enough fissile material in the US to make a bomb in 1944.
The other beauty of Opteron is the ease of building multiway systems up to eight-way...as opposed to HYPErthreading. Personally, I prefer multiple real processors if I'm going to pay additional license fees...
IMHO, that is the where AMD has the strongest edge over Intel and that it does > 32 bit addressing. Micron has demo'ed 1 Gb DDR's which can be used to make a 4 GB DIMM.
Probably the closest competitor is the UltraSPARC IIIi, which is also really easy to do a 2 and 4 way system.
Gamers [toastyfrog.com] have been begging Trip Hawkins to stop wasting time with games for years. I guess Hawkins' prof was just ahead of his time.
Make that way ahead of his time - Trip started Harvard in Fall '72 which would have limited gaming to Spacewar or maybe Adventure. Besides he wasn't that much of a techie at the time.
He found out about Apple at the first West Coast Computer Faire in '77, hired on as head of marketing shortly thereafter. He left Apple in early '82 (wasn't too happy about the Lisa) and started EA not too long after that.
And I find that it's rather preposterous to say that Sun started charging for StarOffice because people thought its $0 price tag made it inferior to Microsoft Office.
Apparently one German Bank told Sun that while StarOffice fitted their needs better than M$ Office, they decided to go the M$ route. Reason being is that they were afraid that Sun would not have the same interest maintaining StarOffice if it was free.
There is some non-Sun code in Solaris and Sun does owe royalties for distributing copies of Solaris. One example is the PostScript code licensed from Adobe. I've heard there is a bit of code copyrighted by M$ itself (from the Xenix days).
To set the record straight - they bought it from Seattle Computer Products. Tim Patterson was an employee of SCP when he wrote QDOS and 86-DOS. Tim was then hired by M$ to continue working on DOS (and then went on to start Falcon). The DOS numbering schemes continue from 86-DOS, the last release of 86-DOS was v1.14 (the basis for PC-DOS 1.0) and the first MS-DOS release was v1.25 (PC-DOS 1.1).
QDOS/86-DOS was designed to make it easy to translate CP/M programs written in asembler and have them run with minor tweaking. This extended to using pretty much the same API for the file control blocks, pretty much the same numbers for the function calls, pretty much the same layout for the first 256 bytes of the transient program area.
Where 86-DOS differed from CP/M, it tended to be more UNIX like, e.g. copy source destination rather than PIP destination source . More functions were included with the command interperter and the batch files were a bit nicer to use than CP/M's submit files.
'Course you've got to remember that CP/M was designed to run in 32K of memory.
The incident with DOS wasn't the only time that SCP got the shaft from M$. SCP was the outfit that designed the Z-80 card for use on the Apple II.
The next version of SunOS WILL be designated 5.10 (this has been stated on comp.unix.solaris by several people from Sun). One side effect is that simple minded version checking schemes will break - There's a '1' after the '.', so it must be 5.1...
From what the articles said, I'd reckon that the antenna should be good for about 20 dB gain over a dipole. Figure that with the inverse square law, you'd be getting 10 dB in range (a factor of 10). 4 miles in free space sounds like a reasonable figure IF the laptop on the other end is in a reasonably quiet area.
The big advantage of this system is to be able to generate multiple beams - something that is a bit difficult to do with parabolic reflector antennas.
If you don't mind a little bit of board design, Intersil makes a nice little 4 channel receiver chip that can take the outputs of 65 MSamples/s ADC's. Effectively these chips have a DDS and quadrature mixer for each channel followed by a programmable CIC filter. Even with the decimation, the data rate causes some problems with a 33 MHz / 32 bit PCI bus. Hard to beat dedicated hardware.
Having said that, I'm curious what you would recommend for the rf tuner for the sound card application. My preference (based on lots of experience) would be a quadrature direct conversion front end.
Further questions along thos elines would be trade-offs with analog pre-filtering, whether to do it before or after the conversion.
So it was more like "we got a moron for a CEO who was totally in love with Bill Gates".
Who was the same moron who trashed HP-UX on the desktop. Back in the realy 90's, HP had persuaded Lotus to port 123 and Ami Pro to HP-UX, VUE kicked Windblows 3.1 ass all over creation.
If you are getting jammed, the worst that can happen is that you can't call 911 when you have to, then your family sues the state government for damages after you die.
When my wife and I go out on a movie date, we carry the cell phone into the theater with the phone turned on. This is in case the babysiiter needs to get in touch with us about a medical problem. If the theater was intentionally jamming the cellphones and that messed up an emergency call to us, I would go after the theater owner and operators with both civil and criminal complaints.
Interfering with emergency communications is a good way of getting yourself into some really deep shit!
After seeing the I&T article, I was suddenly struck as to where Disney got the inspriation for the original Tomorrowland. The Eisner version just doesn't cur it for me (plus they can't even get the new rides right - i.e. the "Rocket Rods").
There have been a couple of references about the recently deceased Ward Kimball (had the pleasure of meeting him 10 years ago) helping with the Collier's series on space travel. It would seem that the flow went both ways.
However, the locomotives made to move, and are not as efficient as a fixed diesel generator of the same size.
Actually the reduction in efficiency isn't as much as you imply. Modern locomotive engines have a thermal efficiency on the order of 45% - the best gas turbines are 44% (stationary NOT aero derived and running at 100% output). The best steam plants are 40%.
The best stationary plants are combined cycle units with thermal efficiencies on the order of 60%.
FWIW, the efficiency of the electric transmission is on the order of 85% for a typical U.S. locomotive. It is comparable to what can be achieved with a mechanical transmission.
Typical practice since the mid 1960's has been using an alternator with retctifiers instead of a generator - above 2500 HP, the generators were getting just too big - the 8500/10,000 HP gas turbine locomotives built for the UP in the early 60's used a gearbox to distribute the power to four generators.
Recent practice is to replace the DC traction motors with AC induction motors - main advantage is low speed torque, resistance to overheating and improved adhesion.
So when is Red Hat coming out with a kernel capable of supporting 128 processors? The licensing fee for Solaris 9 goes up rapidly with the number of processors, the 400 grand fee is probably aimed at the Fujitsu customers. Sun makes the money needed to update Solaris by selling hardware - if people by used Sun equipment, that means less money to sun for development.
Sun makes a big distinction between systems bought from Sun or an authorized reseller versus EBay, etc. This is probably done to keep the resellers happy.
There has been a lot of wailing and gnashing of teeth on comp.unix.solaris, primarily from people having old 4 processor servers lying around (which are worth less than the license). The license for Solaris 8 was really nice, free for machines that could hold 8 or fewer processors. BTW, that license is still in effect for people with media in hand (although it applies just for their organization).
Sun's hurting themselves more by not getting the Jalapeño systems out - keep up the pressure on the low end. Rumor was that the Jalapeño machines were to be cost competitive with the intel boxes.
ISTR that the ARRL website had an article about a 900 MHz netwrok being shut down in the Dallas area. Part 15 states that devices operating under these rules cannot interfere with licensed services.
Slashdot Mirror
What!? No TROFF?
Was nice to see UNIX Text Processing available as a download.
Thanks.
RUN AWAY, RUN AWAY!!!!
and I even like Fortran!
I've run Linux and Solaris on x86 boxes - Linux has a lot of neat bells and whistles, but Solaris seems to be a bit better thought out (have yet to try a BSD on x86). My biggest complaints with Solaris are the lack of an "smbmount" facility. and limited HW support My biggest complaint with Linux is that the desktops lack the refinement of CDE - CDE was designed, KDE and Gnome evolved.
Remember that Linux is just the kernel - a good portion of what you find in a typical distro can easily be ported to Solaris.
Prof Bean (taught Cal history at UCB) was quite the believer in the authenticity of the plate.
For the record a form of journaling is available by specifying the "logging" option in /etc/vfstab. This has been in Solaris well before ext3 was available for Linux.
WWII started out in Manchuria in 1931 - the European was started in 1939 when Poland was invaded by Germany and the Soviet Union.
Not surprising. HP's VUE, the "father" of CDE was designed to run on 25 MHz 68040's with 8 MB of RAM. It also was designed to be workable with smaller screen sizes - 11 years ago, an HP monitor capable of 1280 by 1024 would set you back 6 grand!
Not to mention Mexican and Canadian cars and trucks.
If the problem is road damage, then the real solution is to require these devices for trucks (meaning semi's, not pickups). In addition, base the fees on weights recorded at weigh stations.
Oregon would be better off simply raising the fuel taxes to cover their revenue needs. This would also serve to encourage people to buy more fuel efficient vehicles.
This proposal would make the most sense for alternate fueled vehicles, e.g. electric, natural gas, etc.
The Nagasaki (Fat Man) bomb had a higher yield and IIRC also had less fissile material, I'd guess the burnup was 5 to 10%. The best reported burnups are on the order of 25%.
As other posters have pointed out, there wasn't enough fissile material in the US to make a bomb in 1944.
IMHO, that is the where AMD has the strongest edge over Intel and that it does > 32 bit addressing. Micron has demo'ed 1 Gb DDR's which can be used to make a 4 GB DIMM.
Probably the closest competitor is the UltraSPARC IIIi, which is also really easy to do a 2 and 4 way system.
Make that way ahead of his time - Trip started Harvard in Fall '72 which would have limited gaming to Spacewar or maybe Adventure. Besides he wasn't that much of a techie at the time.
He found out about Apple at the first West Coast Computer Faire in '77, hired on as head of marketing shortly thereafter. He left Apple in early '82 (wasn't too happy about the Lisa) and started EA not too long after that.
Apparently one German Bank told Sun that while StarOffice fitted their needs better than M$ Office, they decided to go the M$ route. Reason being is that they were afraid that Sun would not have the same interest maintaining StarOffice if it was free.
There is some non-Sun code in Solaris and Sun does owe royalties for distributing copies of Solaris. One example is the PostScript code licensed from Adobe. I've heard there is a bit of code copyrighted by M$ itself (from the Xenix days).
QDOS/86-DOS was designed to make it easy to translate CP/M programs written in asembler and have them run with minor tweaking. This extended to using pretty much the same API for the file control blocks, pretty much the same numbers for the function calls, pretty much the same layout for the first 256 bytes of the transient program area.
Where 86-DOS differed from CP/M, it tended to be more UNIX like, e.g. copy source destination rather than PIP destination source . More functions were included with the command interperter and the batch files were a bit nicer to use than CP/M's submit files.
'Course you've got to remember that CP/M was designed to run in 32K of memory.
The incident with DOS wasn't the only time that SCP got the shaft from M$. SCP was the outfit that designed the Z-80 card for use on the Apple II.
The next version of SunOS WILL be designated 5.10 (this has been stated on comp.unix.solaris by several people from Sun). One side effect is that simple minded version checking schemes will break - There's a '1' after the '.', so it must be 5.1...
The big advantage of this system is to be able to generate multiple beams - something that is a bit difficult to do with parabolic reflector antennas.
Having said that, I'm curious what you would recommend for the rf tuner for the sound card application. My preference (based on lots of experience) would be a quadrature direct conversion front end.
Further questions along thos elines would be trade-offs with analog pre-filtering, whether to do it before or after the conversion.
Who was the same moron who trashed HP-UX on the desktop. Back in the realy 90's, HP had persuaded Lotus to port 123 and Ami Pro to HP-UX, VUE kicked Windblows 3.1 ass all over creation.
When my wife and I go out on a movie date, we carry the cell phone into the theater with the phone turned on. This is in case the babysiiter needs to get in touch with us about a medical problem. If the theater was intentionally jamming the cellphones and that messed up an emergency call to us, I would go after the theater owner and operators with both civil and criminal complaints.
Interfering with emergency communications is a good way of getting yourself into some really deep shit!
There have been a couple of references about the recently deceased Ward Kimball (had the pleasure of meeting him 10 years ago) helping with the Collier's series on space travel. It would seem that the flow went both ways.
So does dtpad under CDE (formerly vuepad under VUE). dtpad is much nicer for all around test editing than notepad.
Having said that, NEdit is much nicer to use for prgramming than dtpad.
Actually the reduction in efficiency isn't as much as you imply. Modern locomotive engines have a thermal efficiency on the order of 45% - the best gas turbines are 44% (stationary NOT aero derived and running at 100% output). The best steam plants are 40%.
The best stationary plants are combined cycle units with thermal efficiencies on the order of 60%.
Typical practice since the mid 1960's has been using an alternator with retctifiers instead of a generator - above 2500 HP, the generators were getting just too big - the 8500/10,000 HP gas turbine locomotives built for the UP in the early 60's used a gearbox to distribute the power to four generators.
Recent practice is to replace the DC traction motors with AC induction motors - main advantage is low speed torque, resistance to overheating and improved adhesion.
Sun makes a big distinction between systems bought from Sun or an authorized reseller versus EBay, etc. This is probably done to keep the resellers happy.
There has been a lot of wailing and gnashing of teeth on comp.unix.solaris, primarily from people having old 4 processor servers lying around (which are worth less than the license). The license for Solaris 8 was really nice, free for machines that could hold 8 or fewer processors. BTW, that license is still in effect for people with media in hand (although it applies just for their organization).
Sun's hurting themselves more by not getting the Jalapeño systems out - keep up the pressure on the low end. Rumor was that the Jalapeño machines were to be cost competitive with the intel boxes.
ISTR that the ARRL website had an article about a 900 MHz netwrok being shut down in the Dallas area. Part 15 states that devices operating under these rules cannot interfere with licensed services.