MacBidouille posts PPC 970 benchmarks
From http://www.macbidouille.com/niouzcon...date=2003-05-5 . They offer their own English translation which I've transcribed below. Also, there are a few speed comparison charts at the site I did not post below (to save on loading times); just click the link above to see them.
Mods: If this is a duplicate post, feel free to delete/lock.
Quote:
By reading these benchmarks you'll understand that we couldn't publish them before.
Now we know that PM G4 sells are stuck at a very low level, the following test results won't have much incidence. It will however make the ones switching to PC wait for the next generation of Power Macs.
The first benchmarks were done during March 2003 on a preview model running at 1.4 GHz. OS was an alpha version 7B5 and 7B8 of Panther, optimised for 64 bits processor, but the applications tested were only using 32 bits.
Photoshop : PPC 970 mono 1.4 is 87% faster than a Dual 1.42 GHz Final Cut Pro : PPC 970 mono 1.4 is 112% faster than a Dual 1.42 GHz Alias|Wavefront Maya Render : PPC 970 mono 1.4 is 254% faster than a Dual 1.42 GHz.
The second series of benchmarks were done on the same computers that will be sold. There is however a doubt on the presence of the up-market dual 2.0 GHz as the availability of these chips isn't sure. It seems Apple will surely be able to sell Mono 1.4 GHz, Dual 1.6 and Dual 1.8.
The result is that the G4 compared to the PPC 970 is now a secretary computer.
A few explanations to the results:
- The Altivec shows a 80% increase of performances with the 970. This is not due to the chip itself, but to the high speed access between processor and central memory. The Mach 64 motherboard is highly optimised for the use of DDR-SDRAM.
- There is no performance loss when the PPC 970 executes some 32 bits apps.
- The motherboard optimization almost allows dual processors to reach double performance. In fact it's about 90% efficiency gained with the second processor, compared to 50% for the G4.
Looking closely to these results we understand why Apple didn't need to wait for a 64 bits OS to launch the PPC 970. We'll take advantage of a 50% gain of performance between he up-market Pro G4 and the first PPC 970. We can imagine the difference with the top level PPC 970. It will be the best evolution ever between two Mac generations.
Mac fans, our wait will be rewarded. The fight is over and Apple will soon rule the world!
By reading these benchmarks you'll understand that we couldn't publish them before.
Now we know that PM G4 sells are stuck at a very low level, the following test results won't have much incidence. It will however make the ones switching to PC wait for the next generation of Power Macs.
The first benchmarks were done during March 2003 on a preview model running at 1.4 GHz. OS was an alpha version 7B5 and 7B8 of Panther, optimised for 64 bits processor, but the applications tested were only using 32 bits.
Photoshop : PPC 970 mono 1.4 is 87% faster than a Dual 1.42 GHz Final Cut Pro : PPC 970 mono 1.4 is 112% faster than a Dual 1.42 GHz Alias|Wavefront Maya Render : PPC 970 mono 1.4 is 254% faster than a Dual 1.42 GHz.
The second series of benchmarks were done on the same computers that will be sold. There is however a doubt on the presence of the up-market dual 2.0 GHz as the availability of these chips isn't sure. It seems Apple will surely be able to sell Mono 1.4 GHz, Dual 1.6 and Dual 1.8.
The result is that the G4 compared to the PPC 970 is now a secretary computer.
A few explanations to the results:
- The Altivec shows a 80% increase of performances with the 970. This is not due to the chip itself, but to the high speed access between processor and central memory. The Mach 64 motherboard is highly optimised for the use of DDR-SDRAM.
- There is no performance loss when the PPC 970 executes some 32 bits apps.
- The motherboard optimization almost allows dual processors to reach double performance. In fact it's about 90% efficiency gained with the second processor, compared to 50% for the G4.
Looking closely to these results we understand why Apple didn't need to wait for a 64 bits OS to launch the PPC 970. We'll take advantage of a 50% gain of performance between he up-market Pro G4 and the first PPC 970. We can imagine the difference with the top level PPC 970. It will be the best evolution ever between two Mac generations.
Mac fans, our wait will be rewarded. The fight is over and Apple will soon rule the world!
Mods: If this is a duplicate post, feel free to delete/lock.
Comments
It's a good thing its good material!
*runs with glee whilst tearing hair from head*
Lemon Bon Bon
I wonder what the DVD MPEG-2 encoding speed will be on the new machines (I haven't checked mb.com so if it's there, I apologize) ...
-J
Lemon Bon Bon
Originally posted by Fat Freddy
AFAIK, Bryce 5 doesn't support Multiprocessing... F A K E \
Exactly what I spotted too!
The numbers for the G4 and P4 is the exact same as barefeats got, which isn't too bad for the G4 IMHO (but more suspicious, since I doubt macbidouille benchmarks the same way as barefeats). Either I'm overlooking something here, or the DP1.8-number is a clear fake. Which of course makes all of the other benchies just as fake (why fake numbers when you don't need to?)
Anyways, I just want to say that these benchmark results are about what I'd expect from a PPC970-mac, I believe that's how they got their "benchmarks results". Find out what kind of operations the apps use, and then estimate by using the again estimated Spec-numbers.
Well, I'll smile happily when I know these benchies are real for sure (when the 970's are officially out, and being tested for real).
Originally posted by Mount_my_floppy
Why does it make it fake because bryce does not support SMP? It also does not support Altivec but that does not mean it does not run it.. And that also does not mean a machine thats 2X as fast as the dual 1.42 would not be when in bryce.. The dual 1.42 does not get to use altivec or SMP in bryce either..
The machine still has a xxx xxxx 900MHz bus, a better processor architecture and yadi yadi I don`t understand why it cant be 2x as fast in bryce also..
Good point, at least as far as I can see.
Not the first time I jumped too soon.
Anyway, what could explain the big difference in the benchmarks (assuming they're real) if bryce is not SMP capable. Yeah, I know, bus speed +++, but that can hardly explain the difference between the 1.4 and 1.8 dp entirely. [wild speculation mode] Can panther explain some of the difference, making two processors appear as one to certain non-smp avare apps. [/wild speculation mode] Or is there any other possible explanation?
"If it's too good to believe it's probably false"
There's know way we go from a large deficit in processor speed to and large advantage. This is fake .
If these benchmarks are true.....THEN.....Rev B 970.....come to me baby
If those are true---IBM wont even sell a PPC 970 that's as slow as the Pentium IV, at least for a little while. And Apple is using them as DUALs, with real DDR support. I've always wondered how much the G4's bus held back altivec--now we know. Motorola suffocates the G4 with it's pitiful bus--only a truly stupid design would saddle such a brilliant technology as Altivec with a bus designed to not supply the right amount of data I/O.
It looks like SPEC benchmarks don't tell the whole story of this monster--fire up altivec and it reams Pentiums! In hindsight the G4 will look like it's purpose was to prepare developers for Altivec support on the almightly PPC 970.
Not even Apple can fsck this one up...unless the numbers are fake. Probably are, but I'm hoping they're close to the truth or even a little conservative.
Originally posted by hmurchison
Fake.
"If it's too good to believe it's probably false"
There's know way we go from a large deficit in processor speed to and large advantage. This is fake .
While I share your doubts, would an 80% increase it AltiVec provide an explanation? How Alti-enabled is Cinema 4D or Maya?
Purely an interested layman, here, btw.
Originally posted by NETROMac
D@mn, I should have posted a new thread in the first place. These multiple thread-same discussions are making me dizzy
Apologies for the disorientation.
I posted a new thread since I didn't see this news mentioned in the two or three other active 970 threads...at least when I looked at the posts dated May 5th earlier today.
It's a shame that wide dissemination of these tests is going to completely destroy PM sales (and hence Apple's cashflow/profitability) until those 970 boxes are out of the door.
And someone's going to get their backside slapped: I don't even want to think what the NDA on that kit is like...