[DJMike]

check out this link i found.

http://www.barefeats.com/pb17.html

[Karim]

One relevant issue is that these tests were performed on the Macworld SF 17" Powerbook prototypes with a beta version of OSX 10.2.4

It is possible that on production hardware and with the 10.2.4 Release that the numbers will change. Hopefully showing more of a superiority of the MIA 17" powerbooks.

[Eug]

Originally posted by Karim:
Hopefully showing more of a superiority of the MIA 17" powerbooks.

I doubt it. The 17" in reality is very similar to the 15" 1 GHz. I'd be surprised to find a significant improvement in performance for most apps. The biggest differences would come with gaming, since they use different video cards, but it would depend on the game (and video drivers) as to which laptop would be faster.

By the way, the link was originally posted 2 months ago in this forum.

[ivi]

There are several benchmark posts on this forum now and they all have all sort of results, which are different from each other. Get real, why would you even consider something like this, I mean not only two 1Ghz ti will get different results, but the same test on the same machine today will give different results than it was yersteday. It's quite obvious that 12" will be noticably slower for some tasks and there won't be difference for others. 867 Ti is not really that much different than 1Ghz Ti etc.

One relevant issue is that these tests were performed on the Macworld SF 17" Powerbook prototypes with a beta version of OSX 10.2.4

So what?? You expect it to fly on non beta? Well hopefully 17" will outperform 15" in its final non beta version, will it make significant differenc1 that it did 0.0001 % better in tests? I honestly don't think that anyone should take these benches seriously, its not like you comparing iBook to Dual G4. Real thing here is that on 17" you got Firewire 800 (significant speed increase for some), integrated bluetooth (also very useful for some) or small size,invaluable for others or there is a bit of both in titanium enclosure. The rest is just priorities, if you need smaller machine more than extra power than do whats best to do.

[Karim]

In the past, there has been meaningful performance difference when comparing beta hardware/software with production versions.

The 12" shows a large difference in Xbench memory tests when compared to a 15" 1GHZ. I understand the G4 bus doesn't fire DDR ram on both sides of a clock pulse but their seems to be better performance out of the new memory subsystem in the Al books.

[Eug]

Originally posted by Karim:
The 12" shows a large difference in Xbench memory tests when compared to a 15" 1GHZ. I understand the G4 bus doesn't fire DDR ram on both sides of a clock pulse but their seems to be better performance out of the new memory subsystem in the Al books.

Yes. There will be a big increase in memory speed, IF the action is not CPU dependent. And there lies the rub... Real world apps are very different from synthetic memory bus benches, because real world apps usually aren't just shifting data around without the CPU.

If I remember correctly, there is a PC chipset out there that can support DDR on older CPUs, but it's analogous to the G4/DDR situation. Similarly, there is a negligible increase in overall speed for most apps with this setup.

On the Mac side there have been Xserve DDR vs. PowerMac SDR tests, and the results are almost identical.