Re: Are Sunblade 1000s slow?
On Fri, 03 Oct 2003 08:32:12 -0500
Kent West <westk@acu.edu> wrote:
> Desktop usage/tinkering. I wanted Solitaire to have the cards fly by
> faster than a speeding bullet; I wanted Mozilla to pop up faster than IE
> on their dorm computers; I wanted zippiness.
The card motion in the solitaire game is probably moved at a fixed
rate using timers. There's probably an option in the preferences
to turn off the fancy card motion or make it faster.
But anyways, using an ATI graphics card in a sb1000 is like putting
tricycle tires on a Ferrari. Stick a creator3d in there if you can.
> It's apparently just a matter of unrealistic expectations. I haven't had
> a lot of experience with Sun; a couple of years ago when I started
> working with Sun/Solaris/the SB100s, I was expecting great machines;
> after all, this is Sun!
It's got a 600mhz cpu in it, what in the world do you expect?
For ~$300.00 USD it's very easy to put together a P4 multi-GHZ little
desktop machine that's going to blow your sb1000 away. And you're
going to be able to use a bazillion more apps than on your Sparc.
That's the facts of the market today, everthing is commoditized
to the tilt. Sun's still asleep at the wheel, and while McNealy
continues to complain about how Dell and others are "parts suppliers"
his company's market share and technology slowly sinks into a black
hole.
> "Like wow! Why didn't we just purchase much cheaper Athlon boxes
> and get more for our money?"
Like wow, why didn't you. I'd never waste my own hard earned money on
Sun hardware, the only reason I have so many boxes is that I've gotten
them all for free during my years of development.
Look, do the math, the cheapest lowest end Athlons you can even buy on
the market today are in the 1.2GHZ range (and these are for laptops).
This means that even if your 600Mhz UltraSPARC-III could execute twice
as many instructions per cycle as the Athlon (btw it can't), they
would perform about the same.
I only hack Sparc64 for two reasons:
1) I find it aestetically pleasing to hack it in assembly.
2) Continuing to maintain the sparc64 kernel helps quickly find
portability problems in changes that are made to Linux in general.
My 2-processor sb1000 is fast enough for my work, all I do is edit
code and crank out kernel builds all day long.
Reply to: