PowerBook To Get Dual-Core G4?

[gekko513]

Given that the P3 beat the P4 at similar clockspeeds for most uses, i thought the Pentium M's changes were purely for power to insure stablity at the higher clock speeds.

The Pentium M has among other things a deeper pipeline, improved branch prediction, and enlarged instruction window compared to the PIII. It also introduces macro-ops (fused micro-ops) and a stack execution unit. (source is arstechnica)

 

[MacsRgr8]

Hmm.. rumored Dual Core 970 , now Dual Core G4....

All boils down to: "can't get the MHz or GHz up".
😛

Still, gr8 news if true.

 

[jhu]

Given that the P3 beat the P4 at similar clockspeeds for most uses, i thought the Pentium M's changes were purely for power to insure stablity at the higher clock speeds.

The addition to stages doesn't really matter because Intel would have to do similar tasks.

the pentium m has more pipeline stages than the p3, but it also has a higher ipc than the p3. now that's some snazzy engineering.

 

[jhu]

Hmm.. rumored Dual Core 970 , now Dual Core G4....

All boils down to: "can't get the MHz or GHz up".
😛

Still, gr8 news if true.

why not just graft an altivec unit to the power4 chip?

 

[Corozive]

I hope not for ages

I hope we dont see these for a while yet 🙁

i have just bought a new 17" PB

want to have a while with it being the top one available

 

[AliensAreFuzzy]

Well, whatever chip is in the Powerbook in a year is probably the one I'm getting. I'm hoping for a G5 for all of the G5 optimized programs run to their full potential.

 

[Sun Baked]

why not just graft an altivec unit to the power4 chip?

All those extra units/transistors not needed on the desktop version that make it more expensive (ie, extra real estate) to produce -- Fabric Controller, Inline L3 cache controller, Massive L2 Caches, etc.

Just adding Altivec units to the Power4 would basically make the CPU more expensive than the Power4, because suddenly there are fewer CPUs per wafer.

 

[jhu]

Thanks for commenting on my logical reasoning unit and sentence creation shareware. I read that the fpu on the G4 is per unit weaker than that of the P4 from www.arstechnica.com the link is:
http://arstechnica.com/cpu/01q4/p4andg4e2/p4andg4e2-3.html

and i quote "The P4's FPU hardware is a little beefier than that of the G4e, but a little less beefy than that of the PIII or the Athlon. The P4's designers slimmed down the FPU and weakened it a bit, banking on increasing clock rates and the widespread replacement of x87 code for SSE2-optimized code to take up the resulting performance slack. A closer look at the P4's FPU will show just how it has been weakened."

i read that article too. i just can't help but feel that on a clock-for-clock basis, the ppc 74xx's fpu is better than the p4's. perhaps it's the shorter pipeline that gives it that advantage.

 

[itsa]

Like notebook sales are not hurtting enough!
Even more reason to NOT buy one. Man... I am soooooo glad I waited.

 

[Muzukun]

What's this mean for gaming?

I remember reading that if you just got rid of sound in games for a mac you'd get a major performance boost. Considering this could act as a dual CPU system would it be possible that companies (if they cared enough to do so) could write patches for games enabling them to leave sound processing to one core and the rest to another? IE would we be seeing much better performance on mac gaming with all this?


any chance of this coming to the ibook? (yeah worth asking that as practically no info has been released from apple and it's all happy guessing and all at this point in time -.-)

 

[drsuse]

Well, whatever chip is in the Powerbook in a year is probably the one I'm getting. I'm hoping for a G5 for all of the G5 optimized programs run to their full potential.

similar, i'm getting whatever's shipping in 6-10 months depending mainly on cash flow. originally i was hoping for the g5, but i'm actually more keen on the dual core g4 now.

 

[adamcoop]

New iMac

And no one has even considered that this, instead of a G5 processor, will be going into the next iMac? I mean, Apple didn't actually say that the iMac will be a G5. And if the components are to be mounted behind the screen (as per previous rumours), you'd think some sort of low heat solution would be needed.

 

[drsuse]

And no one has even considered that this, instead of a G5 processor, will be going into the next iMac? I mean, Apple didn't actually say that the iMac will be a G5. And if the components are to be mounted behind the screen (as per previous rumours), you'd think some sort of low heat solution would be needed.

I understood that the iMac was likely coming out at the end of august, if that's true, and freescale isn't releasing the dual core g4 until october at the very earliest, i don't think the imac will be dcg4.

 

[MikeBike]

G4 vs. G5 XBench

The G4 at 1.5 vs. the G5 at 1.6:
The G4 130 / G5 145.
The G4 having a 15 point disadvantage, but also a 100mhz slower clock.
Had the G4 shipped with a 7200 rpm hard drive it's disk scores would be much better, possibily passing the G5.

So, a Dual core G4, with memory controller looks great.

Also, the new GCC optimizations coming in 3.4 will benefit both the G4 and G5 processors.

With a dual core, some apps would see a 100% performance improvement.
Folding@home for one.

Apple's Java virtual machine would also benefit.
Aside from ALL java applications and development tools:
JBuilder/JDeveloper/JEdit/and Oracle database and mgmt tools.

A truly AWESOME tool for java developers.

Note: I didn't take a knife to all non-essential services,
so, your scores might be better. Also, test was run with Energy setting:
Highest Possible Performance.

The only disadvantage the G4 currently has is there would be no 64bit Registers. It would be great to see Apple's JVM with true 64 bit longs.


My XBench scores for a 1.5ghz Powerbook,
128meg video memory, 1Gig ram:
Results 130.47
System Info
Xbench Version 1.1.3
System Version 10.3.5 (7M34)
Physical RAM 1024 MB
Model PowerBook5,5
Processor PowerPC G4 @ 1.50 GHz
L1 Cache 32K (instruction), 32K (data)
L2 Cache 512K @ 1.50 GHz
Bus Frequency 167 MHz
Video Card ATY,RV360M11
Drive Type TOSHIBA MK8026GAX
CPU Test 167.56
GCD Loop 173.87 6.79 Mops/sec
Floating Point Basic 133.02 481.04 Mflop/sec
AltiVec Basic 183.36 5.33 Gflop/sec
vecLib FFT 181.83 2.82 Gflop/sec
Floating Point Library 177.99 7.12 Mops/sec
Thread Test 129.13
Computation 92.26 1.25 Mops/sec, 4 threads
Lock Contention 215.08 2.70 Mlocks/sec, 4 threads
Memory Test 132.96
System 145.84
Allocate 913.07 595.60 Kalloc/sec
Fill 148.32 1180.59 MB/sec
Copy 78.53 392.67 MB/sec
Stream 122.18
Copy 121.07 885.06 MB/sec [altivec]
Scale 125.92 929.31 MB/sec [altivec]
Add 124.25 795.22 MB/sec [altivec]
Triad 117.78 719.63 MB/sec [altivec]
Quartz Graphics Test 180.47
Line 149.27 3.80 Klines/sec [50% alpha]
Rectangle 155.11 10.91 Krects/sec [50% alpha]
Circle 166.68 3.84 Kcircles/sec [50% alpha]
Bezier 175.52 1.91 Kbeziers/sec [50% alpha]
Text 349.42 5.70 Kchars/sec
OpenGL Graphics Test 111.61
Spinning Squares 111.61 78.10 frames/sec
User Interface Test 219.88
Elements 219.88 70.72 refresh/sec
Disk Test 74.80
Sequential 92.01
Uncached Write 81.66 34.04 MB/sec [4K blocks]
Uncached Write 73.46 30.08 MB/sec [256K blocks]
Uncached Read 198.45 31.41 MB/sec [4K blocks]
Uncached Read 79.51 32.12 MB/sec [256K blocks]
Random 63.01
Uncached Write 51.68 0.78 MB/sec [4K blocks]
Uncached Write 62.56 14.11 MB/sec [256K blocks]
Uncached Read 67.25 0.44 MB/sec [4K blocks]
Uncached Read 75.31 15.50 MB/sec [256K blocks]


😉

 

[stingerman]

i read that article too. i just can't help but feel that on a clock-for-clock basis, the ppc 74xx's fpu is better than the p4's. perhaps it's the shorter pipeline that gives it that advantage.

The P-M's macro-Ops are just a fused load or store operation with a math op. The P-M still is the same old narrow processor compared with the G4e.

 

[stingerman]

The G4 at 1.5 vs. the G5 at 1.6:
So, a Dual core G4, with memory controller looks great.

Also, the new GCC optimizations coming in 3.4 will benefit both the G4 and G5 processors.

With a dual core, some apps would see a 100% performance improvement.
Folding@home for one.)

Actually Apple is jumping right to GCC 3.5 (in Tiger, maybe a point release for Panther too.) It is significantly improved with a completely new PRE-SSA, auto-vec and auto-parallezation, inner loop optimizer, etc. It will allow for another major performance jump for all Macs, but the G5's should fly.

 

[dongmin]

MacOSRumors.com has already reported on this stuff. No new news here. But it's nice to see more than one publication say it. My only gripe is the terrible track record of the Register. Then again, MOSR isn't particuarly accurate, either.... 🙂

Normally I wouldn't give a rat's ass about Register rumors, but it's confirmed: Freescale will present the dual-core e600s at the upcoming Microprocessor Forum in October.

 

[dizastor]

I hope we dont see these for a while yet 🙁
i have just bought a new 17" PB want to have a while with it being the top one available

If this were the case, there would never be an upgrade to any product.
Admittedly it feels crappy to buy something and see the upgrade product that is x-amount faster, for the same price... but if it does what you need what does it really matter.

 

[aswitcher]

With Apple's G5 supply problems for desktops it might just be that G4s get a new lease on life with what could be a very powerful chip...but I dont see production until next year...

 

[dizastor]

All boils down to: "can't get the MHz or GHz up".

They make a pill for that don't they?

 

[mikeyredk]

look at the xbench emac somone over-clocked on this forum.

over-clocked to 1.6 ghz from 1 ghz. if it had a better gpu it would have scored so much better. but think about over-clocking the newer 1.25 emacs to 1.6+ with there improved internals.

 

[dongmin]

And no one has even considered that this, instead of a G5 processor, will be going into the next iMac? I mean, Apple didn't actually say that the iMac will be a G5. And if the components are to be mounted behind the screen (as per previous rumours), you'd think some sort of low heat solution would be needed.

It's possible but a recent discovery has indicated that the imac will have a G5. Why? Because there's a bit of code in 10.3.5 that works with the 970fx's PowerTune feature.

<dict>
<key>PowerBook7,1</key>
<string>SMU_Neo2_PlatformPlugin</string>

<key>PowerBook7,2</key>
<string>SMU_Neo2_PlatformPlugin</string>

<key>PowerMac8,1</key>
<string>SMU_Neo2_PlatformPlugin</string>

<key>PowerMac9,1</key>
<string>SMU_Neo2_PlatformPlugin</string>

The PowerMac8,1 refers to a consumer-grade G5 Power Mac, i.e. the new iMac. Using this same logic, the next PowerBooks will probably be G5 based.

 

[szark]

I mean, Apple didn't actually say that the iMac will be a G5.

Yes, they did.

I think a dual-core G4-based Powerbook would be a screamin' machine, as long as they increase the system bus speed or use an on-board memory controller.

 

[shawnce]

And no one has even considered that this, instead of a G5 processor, will be going into the next iMac? I mean, Apple didn't actually say that the iMac will be a G5. And if the components are to be mounted behind the screen (as per previous rumours), you'd think some sort of low heat solution would be needed.

Actually Apple explicitly stated that the new iMac will be G5 based during last quarters earning meeting.

"We normally don't talk about unannounced products, but we feel you need to know about the current situation. The iMac is based on the G5 processor. We could not secure the necessary supply of G5 processors to launch our new iMac on schedule, and as we indicated a few weeks ago, we now to plan to announce and ship it in September... We believe that IBM has placed enormous resources on improving the situation, and based on what they have told us, we expect the supply problems to be behind us by the beginning of Apple's fiscal Q1 05."

 

[AidenShaw]

Monopolies are quite legal

If Apple dies, Microsoft is suddenly a COMPLETE monopoly, the only competition being *free* Linux not owned by any company.

Since MS will be the only OS company, and control the market, they'd get antitrusted to death by the US courts.

There is nothing illegal about being a monopoly - the charges against Microsoft were that Microsoft used its position as a monopoly in illegal ways to leverage and pressure.

Note that nothing that Microsoft did would have been illegal for a smaller company, in fact Apple does many of the same things.

The antitrust laws try to level the field by requiring monopolies to "play nice" - in effect they can fight, but with one hand tied back.

So, if Apple dies on its own (say due to chip supply problems), there would be no change in the status for Microsoft. A monopoly while Apple was alive, a monopoly after Apple's demise.

For Microsoft to kill Apple by stealing all the PPC970s would be legally very, very bad for Microsoft - of course.