AidenShaw said:When a PowerPC is shifted to 64-bit mode, little changes except that the existing pointer registers are 64-bit instead of 32-bit. So no speedup. The potential slowdown is due to the increased amount of memory needed for pointers. It takes twice the memory bandwidth to move pointers to and from the main RAM, and only half as many pointers can fit into cache. (Due to alignment issues, there might also be additional wasted space in data structures - wasted space that's transferred from RAM to cache and back.)
- 32-bit Intel chips have supported 64 GiB of RAM for quite a few years, you don't need 64-bit to put more that 4 GiB of RAM on a system
- The big "more RAM" advantage of 64-bit is that a single program can easily use more than 4 GiB if it needs the space
- For PowerPC, 64-bit is no faster - and is most likely slightly slower - than 32-bit
- For Intel x64 chips, however, 64-bits is often quite a bit faster. 20% faster is an average speedup for 64-bit
When an x86 chip shifts to x64 mode, some major architectural changes occur. Most important, instead of having the 8 32-bit general purpose register of x86 - x64 has 16 64-bit registers. Since several registers are used for pre-defined purposes, this means that x64 mode has nearly 3 times as many available registers. Compilers can take advantage of this to reduce the traffic to cache and main memory.
x64 mode also has twice as many 128-bit SSE registers, so that SIMD and floating point can be much faster. (The SSE vector unit has a "vector length = 1" mode which is the recommended way of doing scalar floating point on x86/x64.)
Here's a bit from Bare Feats - if you look around you'll see many other benchmarks showing improved performance for 64-bit applications on x64.
Note the lines "PD 2.8 64/64", "PD 2.8 64/32", "PD 2.8 32/32". Those are three tests on the same 2.8 GHz Pentium D dual-core. "32/32" is 32-bit OS/32-bit app. "64/32" is 64-bit OS/32-bit app. "64/64" is 64-bit OS/64-bit app.
As you can see, the 64-bit app is 23% faster than the 32-bit app.
mark88 said:I agree, they've still got the Imac so I dont see why they couldn't keep iBook.
My initial claim, which you called wrong, was "The 64-bit product will be faster (like 20%) than the 32-bit".DKZ said:Yes and you'll find many more that shows no difference at all, and even a bit that shows a loss. But still this performance gain has nothing to do with the amount of bits, it has something to do with the AMD64 structure. As you yourself points out.
mikemodena said:I think since there's still the iMac it will be the iMac, the iBook, the MacBook Pro, and the Mac Pro... just losing the Power prefix. That's just my guess, and it's not based on anything really, except the fact that the iMac is still the iMac.
iBrow said:I Personally still think they should bring out a 12"/13.3" MacBook Pro. Because some people mite need a portable laptop and a good video card like a X1600 or a X1300 and a GMA950 wont cut it.
Digitalclips said:I just occurred to me why IBM dumped their PC manufacturing...
They had the inside info a long while back that Steve was dumping their chips and realized Mac with Intel was a win win for Apple and lose lose for PC makers.
mrichmon said:For what you describe (photoshop and design work -- I assume illustrator, indesign, quark, etc) then you are not doing anything that would benefit from a dedicated graphics card?
Current generation integrated graphics chips do accelerated 2D graphics very well and light to moderate 3D acceleration very well. Photoshop, inDesign, Illustrator, Quark, etc are use 2D graphics from the perspective of the graphics chip.
beatle888 said:the rumor claims that the ibook will be rebranded to "macbook".
this is what tells me that this isn't true. the iMac and the iBook are a consumer line. the only reason why apple changed to macbook pro was because they weren't using the PowerPC anymore and the Powerbook had the word "Power" in it. the iBook needs no such adjustment and since they didnt change the iMac i doubt they will change the iBook so....rumor looses credibility.
daschim said:What are the measurements of a 13"3 widescreen display actually? Why only launching ibooks with only one screensize? I can imagine that 13"3 laptops are ideal for people that need to travel a lot but not as a replacement for desktop pc(15" would be better for this). I really want a ibook but I think the 13"3 displays are to small. MacBookPro have 15" screens...but they are to expensive for me.
i dont understand you people who buy computers for christmas or graduation before you go to college, why not wait til the weeks before u go to school so you have the latest and greatest. Sure its nice to have the "gift" rigth there, but id sure as heck rather have the cash and save it for later and get a better computer.bill4588 said:this is great news. i hope they're released before my graduation (cuz im gettin one as my gift). I wonder how much they'll cost?
TheMasin9 said:i dont understand you people who buy computers for christmas or graduation before you go to college, why not wait til the weeks before u go to school so you have the latest and greatest. Sure its nice to have the "gift" rigth there, but id sure as heck rather have the cash and save it for later and get a better computer.
ImAlwaysRight said:According to Apple, there are currently 1,418 universal apps. So, quite far from "no" universal apps being available, as you stated.
As I said, the majority of the apps I use are already universal. I am in the market for a small laptop for presentations (using Keynote, which IS universal). Instead of listing what is universal, you are probably better off going through your apps and listing what isn't universal. There are only 4-5 apps I use that are not universal. And they all run under Rosetta. Sure, it may be 20% slower than a 1.33 iBook. But don't forget about overall system responsiveness, or the fact that the universal apps are going to be about 100% faster on a 1.67 core duo than a 1.33 G4.
Stella said:Hopefully, Apple will take the opportunity, along with the name change, from iBook -> MacBook ( which IMO, is a good change ), to make the MacBook less of a girlie looking machine.
I doubt it though!
AidenShaw said:My initial claim, which you called wrong, was "The 64-bit product will be faster (like 20%) than the 32-bit".
This claim is true - a product compiled for 64-bit will usually be significantly faster than a product compiled for 32-bit (I've seen quite a few benchmarks, in a range of a few percent faster to over 50% faster - 20% is a "typical" improvement).
The fact that it's due to the changes in the x64 ISA rather than the longer pointers doesn't negate the fact that 64-bit apps will usually be significantly faster than 32-bit apps on the same system. That's all I said, and I stand by it.
I also stand by the opinion that in the long term Apple (and Apple buyers) will regret the "9 months of Yonah". If Apple had waited for Merom, and had embraced true 64-bit for OSx64 (that is, there simply would be no 32-bit s/w whatsoever for OSX on Intel) - things would be simpler and cheaper for developers and users. As it is, there will be another big software transition - from the current 32-bit Intel to a 64-bit Intel, and "fat binaries" will have to become "even fatter binaries" to hold both 32-bit and 64-bit Intel code in addition to 32-bit PPC code.
AidenShaw said:I also stand by the opinion that in the long term Apple (and Apple buyers) will regret the "9 months of Yonah". If Apple had waited for Merom, and had embraced true 64-bit for OSx64 (that is, there simply would be no 32-bit s/w whatsoever for OSX on Intel) - things would be simpler and cheaper for developers and users. As it is, there will be another big software transition - from the current 32-bit Intel to a 64-bit Intel, and "fat binaries" will have to become "even fatter binaries" to hold both 32-bit and 64-bit Intel code in addition to 32-bit PPC code.