Intel is far from the only manufacturer of good chips. DEC, Sun and HP all have better chips in most senses... Sun needs to release their UIII at 1GHz and that will go a long way towards impressing the worlds.
I am dying to try this out but the Mozilla developers have made it all but impossible to get Mozilla compiled without Sun's commercial C compiler (GNU tools do not work, spent an afternoon hacking on makefiles and gave up).
Do you agree that the Amiga and OS/2 had an extreme advantage over PC systems in EVERY way in 1988-1991? It was FUD press releases like those of Ed Muth's that got Microsoft into position stepping over its superior rivals - maybe you just weren't on the receiving end of that like I was. Go get 'em ESR!
Even Sun will tell you privately that it was way off on its Sparc 4/5 packages and it was dying painfully with its SuperSPARC architecture. UltraSPARC is a whole new way to rock and roll, though - it should not be compared to the old architecture.
One mistake not to make is to compare compile times between SPARC and Intel boxes. I've had a few people compare compile times as a way of comparing the two but the SPARC systems are RISC-like and the compiler does a lot more work than the CISC-like compiler needs to do for Intel.
Given good optimum code, the UltraSPARC systems are much faster, and server duty under multiple I/O streams, hundreds of TCP/IP connections, etc. is exemplary. We finally are taking a machine out of service and moving it to a more modern one. This Ultra 140 with 256MB memory ran a heavily-used 200MB database, a heavily used web server, e-mail services including POP3 and IMAP4 for 9,000 accounts, a radius server, a heavily-used name server, interactive services like Pine, syslogging, minor NFS, SNMP services/probes, and more - and maintained a 1.3 load under its heaviest usage. It only had 4 disk packs for I/O, too, and a lot of those were the older 5800 RPM drives on a SCSI-II bus!
Sun already has boxes that approach those speeds and shortly will be beyond that. And you can shoot a bullet through an e10000 as long as you don't hit anything vitally important (like going through all the power supplies for instance) and it will still run. Not so very different from this 'discovery'.
...no offense buddy, but wearable computers and a barcode reader on the trashcan? You need to get out more, my friend. Send me your address and I'll buy you a cab ride to the local park.:)
The doorbell cam has applications though. I'd like one of those.
I won't mind because Microsoft's cruft is so sub-standard even they hate it (and thus Windows 2000 which they will work years on trying to scale back down IMHO)
Slashdot Mirror
Intel is far from the only manufacturer of good chips. DEC, Sun and HP all have better chips in most senses... Sun needs to release their UIII at 1GHz and that will go a long way towards impressing the worlds.
I am dying to try this out but the Mozilla developers have made it all but impossible to get Mozilla compiled without Sun's commercial C compiler (GNU tools do not work, spent an afternoon hacking on makefiles and gave up).
...a Solaris binary. Damned if I can get Mozilla to compile on a Solaris system with GNU tools.
So Rob... code update? Once a year, maybe?
Clanlib is not multi-platform! Do NOT tie things to Linux only! All *IX should benefit... GGI works perfectly so far on both my Sun and Linux systems.
I appreciate that they got so far but it is not open software!
I wonder how long before their full-blown web hosting business starts, designed to undercut everyone else on the planet.
Do you agree that the Amiga and OS/2 had an extreme advantage over PC systems in EVERY way in 1988-1991? It was FUD press releases like those of Ed Muth's that got Microsoft into position stepping over its superior rivals - maybe you just weren't on the receiving end of that like I was. Go get 'em ESR!
I explicitly turned off the ten Star Wars trailers a day, Rob. What's going on?
I'll second that.
Even Sun will tell you privately that it was way off on its Sparc 4/5 packages and it was dying painfully with its SuperSPARC architecture. UltraSPARC is a whole new way to rock and roll, though - it should not be compared to the old architecture.
One mistake not to make is to compare compile times between SPARC and Intel boxes. I've had a few people compare compile times as a way of comparing the two but the SPARC systems are RISC-like and the compiler does a lot more work than the CISC-like compiler needs to do for Intel.
Given good optimum code, the UltraSPARC systems are much faster, and server duty under multiple I/O streams, hundreds of TCP/IP connections, etc. is exemplary. We finally are taking a machine out of service and moving it to a more modern one. This Ultra 140 with 256MB memory ran a heavily-used 200MB database, a heavily used web server, e-mail services including POP3 and IMAP4 for 9,000 accounts, a radius server, a heavily-used name server, interactive services like Pine, syslogging, minor NFS, SNMP services/probes, and more - and maintained a 1.3 load under its heaviest usage. It only had 4 disk packs for I/O, too, and a lot of those were the older 5800 RPM drives on a SCSI-II bus!
{typing from his Ultra 5}
Sun already has boxes that approach those speeds and shortly will be beyond that. And you can shoot a bullet through an e10000 as long as you don't hit anything vitally important (like going through all the power supplies for instance) and it will still run. Not so very different from this 'discovery'.
Still, interesting.
Cuz he ain't getting much dating done, I'll bet.
...no offense buddy, but wearable computers and a barcode reader on the trashcan? You need to get out more, my friend. Send me your address and I'll buy you a cab ride to the local park. :)
The doorbell cam has applications though. I'd like one of those.
Something my UltraSPARC systems can crunch on... my Ultra 5 managed 4.4m keys/sec even while I work on it and play MP3s in the background, et al.
:(
So frustrating that the RC5 is inefficient on SPARC systems
We get excellent speeds out of our MAXen. 6.0KB/s is the norm for good solid 56K connects. ISDN connections are a solid 14KB/s.
I won't mind because Microsoft's cruft is so sub-standard even they hate it (and thus Windows 2000 which they will work years on trying to scale back down IMHO)