Actually the biggest difference between the IIe or IIi procs and the standard II proc is that all multiprocessing capability is removed from the proc. Since it only comes on systems without the capability of adding a second proc, it really isn't a loss at all.
The IIi and IIe are meant as embedded processors and/or lower-cost alternatives to the full-bore II. As such, their caches are smaller and they use electronics for lower-cost solutions such as RAM type. This affects their interaction with external devices, I/O and graphics. There's also no UPA bus in the IIi/IIe systems.
I wouldn't expect a IIe or IIi to be any slower than a standard II, assuming the cache is the same size (which it often isn't).
It all depends on what you're using the CPU/box for. As a workstation, the IIi and IIe aren't bad at all, but as a server, they'd be lacking. You're talking about a 256/512K cache as compared to 2, 4 and even 8MB. They also suffer worse floating point power:
440MHz UltraSPARC IIi delivers SPECint95 of 18.7 and SPECfp95 of 21.1
450MHz UltraSPARC II delivers SPECint95 of 17.4 and SPECfp95 of 25.7
500MHz UltraSPARC IIe delivers SPECint95 of 20.0 and SPECfp95 of 21.0
Of course, there's also the debatability of investing in a UltraSPARC II derived system, when the UltraSPARC III is out, and it's not backward compatible. Then again, on a workstation like these you probably won't upgrade the CPU as readily as you might in a server - but that's one of the nice things about Sun hardware...
This is not the same CPU that's in the Ultra 5/10. The Ultra5/10 uses the UltraSPARC IIi processor in 270Mhz-440MHz speeds, and varying cache sizes from 256k to 2MB.
Correct - good catch!
I'd wager it's still a lot slower than the equivalent UltraSPARC II/III, which was my original point.:(
What also sucks is that you can't use existing UltraSPARC II CPU modules in the Blade 1000. You *must* upgrade to new CPUs with the box. That's gonna hurt...
Some things to point out about this system, as opposed to a slightly higher-end model:
500MHz UltraSPARC IIE processor - this is the same type used in the Ultra 5/10 machines. Not nearly as fast as a "real" UltraSPARC II (or III) processor
IDE subsystem - again, the slightly higher-end machines use SCSI featuring hot-swap capability and quick-loading drive sleds
PGX64 or Expert3D-Lite graphics options are weak. Next step up is the Creator3D, followed by the Elite3D cards which are much faster. With the built-in card, 1280x1024 at 24-bits is a little weak. The Creator3D can do 1600x1280 at 24-bits.
Plus side? It's got USB and Firewire ports... But you can outfit the higher-end Ultra 30/60/80 workstations with a $39 USB card from CompUSA...
I've already got an Ultra 60 with dual 450MHz processors, 1GB of RAM and two internal 9GB drives, with an attached Multipack containing six 18GB drives in a RAID 0+1 configuration... Sporting two Creator 3D cards and 21" monitors...
It's getting to the point where they almost have to give it away for free to continue "competing" with these other OSes...
But judging by their past practices, they'll probably snub that idea along the lines of "we own the market anyway and have computer resellers by the balls anyway."
If they think I'm going to hand over code that I developed for use in their projects without compensation to me, regardless of license - then forget it. Because - if I can't do the reverse, that's hardly an equitable arrangement.
They would pay me to write code for them, and them only. Well, why shouldn't I have the same policy and keep them both separate?
Granted, this doesn't take advantage of re-usable code and modularity - but it beats giving up the rights to MY code.
I had this same situation once, and there was no way I was signing off on any of my code.
I didn't advise them of any of my prior works. I chose to keep it separate - private work is private work and business coding is just that - business. They have no right to my code, and I don't have any right to there.
My code is released under a license. If they choose to accept that license - so be it. Otherwise, hands off!
In 1986, I was a security policeman (SP) in the Air Force, station at MacDill AFB in Florida (in Tampa).
I was out on my balcony, getting ready for work and I was patiently watching the skies for the launch.
The time came and I saw the plume rise to the sky, when the explosion occured. I, like many people figured "Hmm, something doesn't look right" and ran inside to put on the news...
Sure enough, there had indeed been a "major malfunction.":(
While others at neighboring bases were sent to guard the beach and obvious wreckage, I luckily didn't get this grisly duty.
My heart is certainly out for the crew of Challenger, and it's good to know that 15 years later, they're still remembered, and the ideal of a "teaching mission" was not in vain.
Linux also works very well with the Palm and Handspring handhelds. There's a good article on this subject, entitled "Using the PalmPilot Under Linux" on this subject.
This is the correct URL. Jeez. I gotta lay off the caffeine...
...to base your product on a limited supply of resources, no?
I think it would have been better to reverse engineer or have more of the SID chips made first!
Slashdot Mirror
The IIi and IIe are meant as embedded processors and/or lower-cost alternatives to the full-bore II. As such, their caches are smaller and they use electronics for lower-cost solutions such as RAM type. This affects their interaction with external devices, I/O and graphics. There's also no UPA bus in the IIi/IIe systems.
I wouldn't expect a IIe or IIi to be any slower than a standard II, assuming the cache is the same size (which it often isn't).
It all depends on what you're using the CPU/box for. As a workstation, the IIi and IIe aren't bad at all, but as a server, they'd be lacking. You're talking about a 256/512K cache as compared to 2, 4 and even 8MB. They also suffer worse floating point power:
Of course, there's also the debatability of investing in a UltraSPARC II derived system, when the UltraSPARC III is out, and it's not backward compatible. Then again, on a workstation like these you probably won't upgrade the CPU as readily as you might in a server - but that's one of the nice things about Sun hardware...
Correct - good catch!
I'd wager it's still a lot slower than the equivalent UltraSPARC II/III, which was my original point. :(
What also sucks is that you can't use existing UltraSPARC II CPU modules in the Blade 1000. You *must* upgrade to new CPUs with the box. That's gonna hurt...
Better off with Linux than Solaris x86 - way better device support.
Plus side? It's got USB and Firewire ports... But you can outfit the higher-end Ultra 30/60/80 workstations with a $39 USB card from CompUSA...
Can be found on the Sun Blade 100 Workstation page.
Heheh. :)
Check out the newly opened Everything Solaris site - for unique Solaris related content in the same vain as Everything Linux!
A little blunt, if you ask me - but an option.
But judging by their past practices, they'll probably snub that idea along the lines of "we own the market anyway and have computer resellers by the balls anyway."
Bastards.
So long, Katz! ;>
It's not the creation of Earth, but rather the command to compile XFree86... ;>
If they think I'm going to hand over code that I developed for use in their projects without compensation to me, regardless of license - then forget it. Because - if I can't do the reverse, that's hardly an equitable arrangement.
They would pay me to write code for them, and them only. Well, why shouldn't I have the same policy and keep them both separate?
Granted, this doesn't take advantage of re-usable code and modularity - but it beats giving up the rights to MY code.
I didn't advise them of any of my prior works. I chose to keep it separate - private work is private work and business coding is just that - business. They have no right to my code, and I don't have any right to there.
My code is released under a license. If they choose to accept that license - so be it. Otherwise, hands off!
Mine came first, and I'm not giving up nuthin...
I was out on my balcony, getting ready for work and I was patiently watching the skies for the launch.
The time came and I saw the plume rise to the sky, when the explosion occured. I, like many people figured "Hmm, something doesn't look right" and ran inside to put on the news...
Sure enough, there had indeed been a "major malfunction." :(
While others at neighboring bases were sent to guard the beach and obvious wreckage, I luckily didn't get this grisly duty.
My heart is certainly out for the crew of Challenger, and it's good to know that 15 years later, they're still remembered, and the ideal of a "teaching mission" was not in vain.
This is the correct URL. Jeez. I gotta lay off the caffeine...
That's "...should bear some interesting fruit..."
Simply blow up the other plane before they blow you up!
Without the hair, maybe.
I think you meant here - http://www.freebsd.org
I already have a bunch of Gore-Tex lined gear, and it's great stuff. Not stain resistant, but hell, nothing a can of ScotchGard can't fix.
I do wonder though, is the only reason this got mentioned on Slapmeat because of the word "nano" in the name? ;>
"Don't throw stones in Glassco... Err, houses."
Bad monkey!
At least they sent you an Email letting you know what they were doing, and how to correct it, should you wish to change it.
Granted, setting it all to "yes" is a little bold - but for most users, they probably want it this way. "Oooh, more info - sure, send it!"
If they hadn't told you about it, I'd be a little more worried. No harm done - just go and set your preferences back.
...to base your product on a limited supply of resources, no? I think it would have been better to reverse engineer or have more of the SID chips made first!
Sure, 2.4 still isn't here yet. But Linux developers have their wish lists for the 2.5 kernel