Preliminary iMac duo-core Intel RESULTS.TXT

SO7783 · 2006-04-06, 00:27

Under WindowsXP on the iMac20" (using Apple's newly released
BOOTCAMP and the necessary April firmware and OS upgrade):

Genuine Intel(R) CPU T2500 @ 2.00GHz
CPU speed: 2000.16 MHz
CPU features: RDTSC, CMOV, Prefetch, MMX, SSE, SSE2
L1 cache size: 32 KB
L2 cache size: 2048 KB
L1 cache line size: 64 bytes
L2 cache line size: 64 bytes
TLBS: 128
Prime95 32-bit version 24.14, RdtscTiming=1
Best time for 512K FFT length: 31.856 ms.
Best time for 640K FFT length: 41.527 ms.
Best time for 768K FFT length: 51.003 ms.
Best time for 896K FFT length: 61.294 ms.
Best time for 1024K FFT length: 67.661 ms.
Best time for 1280K FFT length: 85.643 ms.
Best time for 1536K FFT length: 104.365 ms.
Best time for 1792K FFT length: 125.774 ms.
Best time for 2048K FFT length: 139.695 ms.
Best time for 2560K FFT length: 181.538 ms.
Best time for 3072K FFT length: 221.041 ms.
Best time for 3584K FFT length: 265.366 ms.
Best time for 4096K FFT length: 294.410 ms.
Best time for 58 bit trial factors: 6.631 ms.
Best time for 59 bit trial factors: 6.652 ms.
Best time for 60 bit trial factors: 6.517 ms.
Best time for 61 bit trial factors: 6.550 ms.
Best time for 62 bit trial factors: 9.294 ms.
Best time for 63 bit trial factors: 9.322 ms.
Best time for 64 bit trial factors: 15.936 ms.
Best time for 65 bit trial factors: 15.839 ms.
Best time for 66 bit trial factors: 15.855 ms.
Best time for 67 bit trial factors: 15.815 ms.

This is with v24.14 and affinity set to P-1 (no difference
whether choosing P-0 or "either")

Take with a grain of salt since there were a couple of corrupt
files on the XP installation disk and I haven't bothered with
any of the hardware drivers released for the first time today
from Apple -- will update this RESULTS.TXT with a "cleaner"
install in the next few days and report back if there is interest.

The *totally* quiet duo-core iMac didn't even make any fan
noise after four or five consecutive runs ... I don't know if
there is any MHz throttling going on at this point due to heat.

SO7783 · 2006-04-06, 00:54

FFT for *my* duo iMac @ 2001MHZ
versus *my* Athlon 64 @ 2009MHz:

(additional time for the Intel 2.0GHz to compute compared to AMD 3000+)

512K - 1.33
640K - 1.31
768K - 1.32
896K - 1.34
1024K - 1.34
1280K - 1.30
1536K - 1.31
1792K - 1.29
2048K - 1.29
2560K - 1.24
3072K - 1.23
3584K - 1.24
4096K - 1.21

58bit 1.09
59bit 1.09
60bit 1.07
61bit 1.08
62bit 0.85
63bit 0.86
64bit 1.14
65bit 1.14
66bit 1.14
67bit 1.14

Not that it should matter, but none of Apple's v1.00
drivers for Windows XP (04/05/06) have been loaded.

* running just 1 instance of Prime95 benchmark on iMac

dsouza123 · 2006-04-06, 06:59

That good news, Prime95 on a Intel Mac works well.

Which AMD 3000+ CPU is it ?

Sempron Socket A
Clock L2 FSB
2200 256 333

Sempron Socket 754
1800 128 400

Sempron64 Socket 754
1800 128 400

Sempron64 Socket 939
1800 128 400

Sempron64 Socket AM2
1600 256 667

Athlon64 Socket 754 Newcastle
2000 512 400

Athlon64 Socket 939 Winchester
1800 512 400

Athlon64 Socket 939 Venice
2000 512 400

Reread your second post,
Is the AMD 3000+, your Athlon64 at 2009 MHz ?
I am guessing so, but it could be from a benchmark page.
Even if so, which socket 754 or 939 ?

In any event, thanks for posting your benchmark results.

SO7783 · 2006-04-06, 20:57

>Which AMD 3000+ CPU is it ?

Good question, since it matches up 1:1 across-the-board (*) with the Sempr0n 3000+ (1.8GHz with 128K L2 cache) at http://www.mersenne.org/bench.htm !

But the 3000+ CPU I compared to in yesterday's posting, was a socket 754 Athlon64 (2009MHz) with Corsair DDR400 and aggressive memory timings on a decent nForce3 motherboard.

*Of course that is just half of its processor capabilities, and not too bad for a 2.0GHz mobile chip versus a "desktop" CPU. The whole thing doesn't draw 100W and I can read a book at night from the 20" backlight while the system didn't begin to drain my UPS after twenty minutes during a full load test.

SO7783 · 2006-04-06, 21:09

Quote:

Originally Posted by dsouza123

That good news, Prime95 on a Intel Mac works well.

In any event, thanks for posting your benchmark results.

I had no idea last night how accurate those figures might be, since the whole system "feels" faster than the timings indicated (and I touch multiple-CPU systems every day).

Turns out that the T2500 on the Mac falls right between the already-charted P95 numbers for the T2400 and T2600 on PC laptops. Without having loaded Apple's drivers, I wasn't sure whether some "green" throttling function -- heat or otherwise -- might have been in effect. [I've seen power_conscious laptops which drop in speed 20% or more under certain circumstances like not running on AC, etc.]

The only XP-on-iMac dissapointment so far was the bootup time ... which in practice I could care less about ... was twice the usual 10 seconds of a fresh Windows install. Also, MSINFO32 was slow as it "tried" pulling the usual information, but failed. If this happens to others, I'd bet it has to do with it trying to pull nonexistant BIOS information with Apple's EFI.

bayanne · 2006-04-13, 08:35

Anyone tried using Parallels yet ......

bayanne · 2006-04-15, 14:38

OK, to answer my own question ....

Code:

Genuine Intel(R) CPU            1600  @ 2.16GHz
CPU speed: 2335.28 MHz
CPU features: RDTSC, CMOV, Prefetch, MMX, SSE, SSE2
L1 cache size: 32 KB
L2 cache size: 2048 KB
L1 cache line size: 64 bytes
L2 cache line size: 64 bytes
TLBS: 128
Prime95 32-bit version 24.14, RdtscTiming=1
Best time for 512K FFT length: 25.411 ms.
Best time for 640K FFT length: 37.884 ms.
Best time for 768K FFT length: 44.309 ms.
Best time for 896K FFT length: 51.622 ms.
Best time for 1024K FFT length: 59.983 ms.
Best time for 1280K FFT length: 75.410 ms.
Best time for 1536K FFT length: 96.943 ms.
Best time for 1792K FFT length: 120.301 ms.
Best time for 2048K FFT length: 131.405 ms.
Best time for 2560K FFT length: 170.659 ms.
Best time for 3072K FFT length: 211.014 ms.
Best time for 3584K FFT length: 255.166 ms.
Best time for 4096K FFT length: 284.042 ms.
Best time for 58 bit trial factors: 3.823 ms.
Best time for 59 bit trial factors: 6.199 ms.
Best time for 60 bit trial factors: 4.653 ms.
Best time for 61 bit trial factors: 3.651 ms.
Best time for 62 bit trial factors: 8.554 ms.
Best time for 63 bit trial factors: 7.088 ms.
Best time for 64 bit trial factors: 10.592 ms.
Best time for 65 bit trial factors: 10.778 ms.
Best time for 66 bit trial factors: 11.590 ms.
Best time for 67 bit trial factors: 13.203 ms.

2006-04-06, 00:27	#1
SO7783 Nov 2003 77377/77477 (either way) 2²·3 Posts	Preliminary iMac duo-core Intel RESULTS.TXT Under WindowsXP on the iMac20" (using Apple's newly released BOOTCAMP and the necessary April firmware and OS upgrade): Genuine Intel(R) CPU T2500 @ 2.00GHz CPU speed: 2000.16 MHz CPU features: RDTSC, CMOV, Prefetch, MMX, SSE, SSE2 L1 cache size: 32 KB L2 cache size: 2048 KB L1 cache line size: 64 bytes L2 cache line size: 64 bytes TLBS: 128 Prime95 32-bit version 24.14, RdtscTiming=1 Best time for 512K FFT length: 31.856 ms. Best time for 640K FFT length: 41.527 ms. Best time for 768K FFT length: 51.003 ms. Best time for 896K FFT length: 61.294 ms. Best time for 1024K FFT length: 67.661 ms. Best time for 1280K FFT length: 85.643 ms. Best time for 1536K FFT length: 104.365 ms. Best time for 1792K FFT length: 125.774 ms. Best time for 2048K FFT length: 139.695 ms. Best time for 2560K FFT length: 181.538 ms. Best time for 3072K FFT length: 221.041 ms. Best time for 3584K FFT length: 265.366 ms. Best time for 4096K FFT length: 294.410 ms. Best time for 58 bit trial factors: 6.631 ms. Best time for 59 bit trial factors: 6.652 ms. Best time for 60 bit trial factors: 6.517 ms. Best time for 61 bit trial factors: 6.550 ms. Best time for 62 bit trial factors: 9.294 ms. Best time for 63 bit trial factors: 9.322 ms. Best time for 64 bit trial factors: 15.936 ms. Best time for 65 bit trial factors: 15.839 ms. Best time for 66 bit trial factors: 15.855 ms. Best time for 67 bit trial factors: 15.815 ms. This is with v24.14 and affinity set to P-1 (no difference whether choosing P-0 or "either") Take with a grain of salt since there were a couple of corrupt files on the XP installation disk and I haven't bothered with any of the hardware drivers released for the first time today from Apple -- will update this RESULTS.TXT with a "cleaner" install in the next few days and report back if there is interest. The totally quiet duo-core iMac didn't even make any fan noise after four or five consecutive runs ... I don't know if there is any MHz throttling going on at this point due to heat.

2006-04-06, 06:59	#3
dsouza123 Sep 2002 296₁₆ Posts	That good news, Prime95 on a Intel Mac works well. Which AMD 3000+ CPU is it ? Sempron Socket A Clock L2 FSB 2200 256 333 Sempron Socket 754 1800 128 400 Sempron64 Socket 754 1800 128 400 Sempron64 Socket 939 1800 128 400 Sempron64 Socket AM2 1600 256 667 Athlon64 Socket 754 Newcastle 2000 512 400 Athlon64 Socket 939 Winchester 1800 512 400 Athlon64 Socket 939 Venice 2000 512 400 Reread your second post, Is the AMD 3000+, your Athlon64 at 2009 MHz ? I am guessing so, but it could be from a benchmark page. Even if so, which socket 754 or 939 ? In any event, thanks for posting your benchmark results. Last fiddled with by dsouza123 on 2006-04-06 at 07:02

Similar Threads
Thread	Thread Starter	Forum	Replies	Last Post
48-core chip - Intel	hj47	Hardware	6	2009-12-20 13:59
Intel Quad Core optimization	joblack	Hardware	1	2009-06-11 02:42
Optimising work for Intel Core 2 Duo or Quad Core	S485122	Software	0	2007-05-13 09:15
Intel's 80 core beast...	TehPenguin	Hardware	10	2006-11-27 11:13
Intel Core Duo processor	drew	Hardware	5	2006-05-29 07:00

2006-04-06, 00:54	#2
SO7783 Nov 2003 77377/77477 (either way) 2²·3 Posts	FFT for my duo iMac @ 2001MHZ versus my Athlon 64 @ 2009MHz: (additional time for the Intel 2.0GHz to compute compared to AMD 3000+) 512K - 1.33 640K - 1.31 768K - 1.32 896K - 1.34 1024K - 1.34 1280K - 1.30 1536K - 1.31 1792K - 1.29 2048K - 1.29 2560K - 1.24 3072K - 1.23 3584K - 1.24 4096K - 1.21 58bit 1.09 59bit 1.09 60bit 1.07 61bit 1.08 62bit 0.85 63bit 0.86 64bit 1.14 65bit 1.14 66bit 1.14 67bit 1.14 Not that it should matter, but none of Apple's v1.00 drivers for Windows XP (04/05/06) have been loaded. * running just 1 instance of Prime95 benchmark on iMac

2006-04-06, 20:57	#4
SO7783 Nov 2003 77377/77477 (either way) 14₈ Posts	>Which AMD 3000+ CPU is it ? Good question, since it matches up 1:1 across-the-board () with the Sempr0n 3000+ (1.8GHz with 128K L2 cache) at http://www.mersenne.org/bench.htm ! But the 3000+ CPU I compared to in yesterday's posting, was a socket 754 Athlon64 (2009MHz) with Corsair DDR400 and aggressive memory timings on a decent nForce3 motherboard. Of course that is just half of its processor capabilities, and not too bad for a 2.0GHz mobile chip versus a "desktop" CPU. The whole thing doesn't draw 100W and I can read a book at night from the 20" backlight while the system didn't begin to drain my UPS after twenty minutes during a full load test.

2006-04-13, 08:35	#6
bayanne "Tony Gott" Aug 2002 Yell, Shetland, UK 332₁₀ Posts	Anyone tried using Parallels yet ......