[jclarkv]

MacBidoouille has posted (http://translate.google.com/translat...language_tools) a performance comparison of 10.2.7 vs 10.3 on a G5/1.8 with XBench 1.1.1.

A total score of 139 under 10.2.7 becomes 174 under 10.3. Wow.

[daydreamer]

interesting,

Check the floating point test woaah

[Horsepoo!!!]

That's a 35 point boost by simply upgrading to 10.3. It's too bad the G5s didn't originally ship with 10.3.

[SMacTech]

10.3 will make your older hardware faster too. I use it on my 400 G3 pismo PB with 192mb RAM as a low-end test setup for application testing. It makes the pismo feel like a lot faster.

[simifilm]

I don't want to spoil the fun, but be careful with XBench results, they're not very reliable. XBench is known for delivering greatly varying benchmarks on identical configurations. Even on the same machine you can get quite different results just by running it several times.

[Eug]

Originally posted by simifilm:
I don't want to spoil the fun, but be careful with XBench results, they're not very reliable. XBench is known for delivering greatly varying benchmarks on identical configurations. Even on the same machine you can get quite different results just by running it several times.

Xbench isn't a great test no, but usually the fluctuations are in individual tests, not across the board. In this case, most of the important tests are improved, not just one here or there.

Furthermore, usually the overall score is nearly identical. For instance, on my TiBook, the fluctuation in the score is no more than a couple of points.

Also, the test result would agree with the general impression of X.3.

It looks like Panther is going to be a huge upgrade.

[devmage]

I happen to know a developer who recently got his Dual G5, and have run both 10.2.7 and 10.3 on it. He tells me in real world practical use of the system 10.3 is noticably faster than 10.2.7. I asked him as soon as I saw those benchmarks because I never put much stake in them.

Devmage

[KidRed]

It's as I've been saying on a few mac boards, people shouldn't be whining about a G5 with less then a good amount of ram running a hacked 10.2 OS. This is what I was expecting to see with Panther.

[Hydra]

Originally posted by KidRed:
It's as I've been saying on a few mac boards, people shouldn't be whining about a G5 with less then a good amount of ram running a hacked 10.2 OS. This is what I was expecting to see with Panther.

I totally agree. The flip side is that I hope Apple can find it in it's little heart to give us G5 early adopters a free copy of 10.3 to really make our new boxes shine. I'm not complaining, I'm a dual G5 waiter and I'll be buying Panther ASAP if I have to and I buy my G5 full well knowing that I may have to pay for Panther. Ah, the price of being on the bleeding edge...or is is the price of stupidity I can't tell the difference sometimes.

-Jerry C.

[alligator]

Originally posted by Hydra:
I totally agree. The flip side is that I hope Apple can find it in it's little heart to give us G5 early adopters a free copy of 10.3 to really make our new boxes shine. I'm not complaining, I'm a dual G5 waiter and I'll be buying Panther ASAP if I have to and I buy my G5 full well knowing that I may have to pay for Panther. Ah, the price of being on the bleeding edge...or is is the price of stupidity I can't tell the difference sometimes.

-Jerry C.

I got some "coupons" in my G5 box, what are they for? Hopefully for a free upgrade to 10.3????

[scottiB]

The coupons are for upgrading the OS if you've purchased the Mac after the OS is shipping, yet your system doesn't have it installed--because it was sitting in a warehouse or stockroom or FedEx truck or something.

[skyman]

Originally posted by jclarkv:
MacBidoouille has posted (http://translate.google.com/translat...language_tools) a performance comparison of 10.2.7 vs 10.3 on a G5/1.8 with XBench 1.1.1.

A total score of 139 under 10.2.7 becomes 174 under 10.3. Wow.

Xbench sucks... Give us REAL world performance stats.