thestaton said:
Meh, hummer driver?
As I say my interest was more in seeing the chip functioning correctly and maybe seeing if there is indeed a consumption reduction. I'm interested for laptop reasons.
Got a tip for us?
Let us know
Become a MacRumors Supporter for $50/year with no ads, ability to filter front page stories, and private forums.
Intel iMac CPU Transplant Success
- Thread starter topgunn
- Start date
- Sort by reaction score
Meh, hummer driver?
As I say my interest was more in seeing the chip functioning correctly and maybe seeing if there is indeed a consumption reduction. I'm interested for laptop reasons.
yellow said:
Code:
Geekbench 2006 (build 190). Email geekbench@geekpatrol.ca with feedback.
System Information
Geekbench Version: Geekbench 2006 (build 190)
Geekbench Platform: Mac OS X x86
Geekbench Compiler: GCC 4.0.1 (Apple Computer, Inc. build 5341)
OS: Mac OS X 10.4.7 (Build 8J2135)
Model: iMac (Early 2006)
Motherboard: iMac4,1
Processor: Genuine Intel(R) CPU @ 2.16GHz
Processor ID: GenuineIntel Family 6 Model 15 Stepping 4
Logical Processor Count: 2
Physical Processor Count: 2
Processor Frequency: 2170 MHz
Bus Frequency: 664 MHz
Memory: 2048 MB
Integer Performance
Emulate 6502
single-threaded scalar 135.2 (rate: 1.0, result: 255.7 MHz)
multi-threaded scalar 266.8 (rate: 2.0, result: 504.1 MHz)
Blowfish
single-threaded scalar 189.5 (rate: 1.0, result: 78.2 MB/sec)
multi-threaded scalar 377.6 (rate: 2.0, result: 155.8 MB/sec)
bzip2 Compress
single-threaded scalar 181.0 (rate: 1.0, result: 28.2 MB/sec)
multi-threaded scalar 358.3 (rate: 2.0, result: 55.6 MB/sec)
bzip2 Decompress
single-threaded scalar 230.1 (rate: 1.0, result: 85.6 MB/sec)
multi-threaded scalar 470.1 (rate: 2.0, result: 169.3 MB/sec)
Floating Point Performance
Mandelbrot
single-threaded scalar 146.5 (rate: 1.0, result: 1.0 Gflops)
multi-threaded scalar 292.0 (rate: 2.0, result: 2.1 Gflops)
Dot Product
single-threaded scalar 294.2 (rate: 1.0, result: 468.1 Mflops)
multi-threaded scalar 550.9 (rate: 1.9, result: 908.3 Mflops)
single-threaded vector 122.3 (rate: 3.7, result: 1.7 Gflops)
multi-threaded vector 234.7 (rate: 7.2, result: 3.4 Gflops)
JPEG Compress
single-threaded scalar 131.2 (rate: 1.0, result: 12.2 Mpixels/sec)
multi-threaded scalar 262.8 (rate: 2.0, result: 24.3 Mpixels/sec)
JPEG Decompress
single-threaded scalar 122.9 (rate: 1.0, result: 20.4 Mpixels/sec)
multi-threaded scalar 240.7 (rate: 2.0, result: 39.9 Mpixels/sec)
Memory Performance
Read Sequential
single-threaded scalar 327.9 (rate: 1.0, result: 4.1 GB/sec)
multi-threaded scalar 284.5 (rate: 0.4, result: 1.7 GB/sec)
Write Sequential
single-threaded scalar 450.2 (rate: 1.0, result: 3.4 GB/sec)
multi-threaded scalar 273.4 (rate: 0.3, result: 1.0 GB/sec)
Stdlib Allocate
single-threaded scalar 224.4 (rate: 1.0, result: 7.9 Mallocs/sec)
multi-threaded scalar 104.8 (rate: 0.5, result: 3.7 Mallocs/sec)
Stdlib Write
single-threaded scalar 119.0 (rate: 1.0, result: 3.0 GB/sec)
multi-threaded scalar 136.1 (rate: 1.1, result: 3.2 GB/sec)
Stdlib Copy
single-threaded scalar 251.4 (rate: 1.0, result: 2.7 GB/sec)
multi-threaded scalar 294.3 (rate: 1.1, result: 3.0 GB/sec)
Stream Performance
Stream Copy
single-threaded scalar 161.5 (rate: 1.0, result: 2.0 GB/sec)
multi-threaded scalar 173.5 (rate: 1.1, result: 2.2 GB/sec)
single-threaded vector 157.1 (rate: 1.1, result: 2.1 GB/sec)
multi-threaded vector 160.0 (rate: 1.1, result: 2.2 GB/sec)
Stream Scale
single-threaded scalar 185.0 (rate: 1.0, result: 2.2 GB/sec)
multi-threaded scalar 184.6 (rate: 1.0, result: 2.2 GB/sec)
single-threaded vector 156.7 (rate: 1.0, result: 2.1 GB/sec)
multi-threaded vector 159.6 (rate: 1.0, result: 2.2 GB/sec)
Stream Add
single-threaded scalar 162.5 (rate: 1.0, result: 2.1 GB/sec)
multi-threaded scalar 191.5 (rate: 1.2, result: 2.6 GB/sec)
single-threaded vector 167.8 (rate: 1.1, result: 2.3 GB/sec)
multi-threaded vector 182.4 (rate: 1.2, result: 2.6 GB/sec)
Stream Triad
single-threaded scalar 160.8 (rate: 1.0, result: 2.1 GB/sec)
multi-threaded scalar 191.3 (rate: 1.2, result: 2.6 GB/sec)
single-threaded vector 135.9 (rate: 1.1, result: 2.3 GB/sec)
multi-threaded vector 146.4 (rate: 1.2, result: 2.6 GB/sec)
Overall Score: 221.6FWIW, someone else ran the test on their 2.0GHz iMac just before me and got 189.5 overall.
Edit: In BootCamp, my Merom iMac scored 268.2
Nice job man! I bet she flies like the....like the........really fast!
How long do you have the Eng. Sample for? Or do you get to keep it!?
How long do you have the Eng. Sample for? Or do you get to keep it!?
topgunn said:
From the screenshot we can see the following:
1. The operating system doesn't recognise the exact type of the chip. It recognises that it is an unknown kind of Intel chip (just wondering how many different types of existing Intel chips the system would recognise, since all Macs built so far used Core Solo, Core Duo or P4 (old development samples)).
2. The operating system apparently doesn't know how to find out the size of the L2 cache. Shouldn't be a problem; L2 cache will work whether you know its size or you don't. Might be a small problem for high performance code that adapts itself to the size of the L2 cache.
netdog said:
I think what she/he said was uncalled for, but it was only said once and we should probably let it go... at least, the one person calling troll (fatsoforgotso) seems to be the only trolling going on in the thread.
To the OP, thank you again for providing all of this information.
It seems essentially consistent with what others are reporting, suggesting that overall this chip may present a modest improvement. (I think it's sad it doesn't really appear to be substantially cooler, but... that's neither here nor there).
How about same game benchestopgunn said:
Doom,Quake,Far Cry....... This new Intel stuff is just pretty awesome in my view putting everyone in the dust far far back, AMD,G5,G4,,,,,The only FPS games I have are Half-Life 2 and UT 2004. HL2 wouldn't really be a good test since it is already GPU bound. UT may work but I have never tried to bench it. Anyone have any tips for doing this?Dont Hurt Me said:How about same game benches
I did run SuperPi to 1 million digits and it came in under 23 seconds.
Where do you get SuperPi from? I've searched, and either i cant find it, or i dont know what im looking for.topgunn said:
gnasher729 said:
That's weird. I've once written a small utility to get some system information for troubleshooting problems reported to us by our users. Among other things, it gets the cpu name.
For the Core Duo, this name contains the model number (T2400 for my iMac). So why wouldn't the Core 2 Duo's name also include the model number?
Processor Name
Yes, maybe your utility does, but Apple System profiler does not. My system only says "Intel Core Duo" even though it is a T2600. And with an engineering sample (or even a new processor that did not exist publicly when the hardware was made) this is entirely understandable.
I am also curious about the power consumption, i.e. for a MacBook Pro...
reflex said:
Yes, maybe your utility does, but Apple System profiler does not. My system only says "Intel Core Duo" even though it is a T2600. And with an engineering sample (or even a new processor that did not exist publicly when the hardware was made) this is entirely understandable.
I am also curious about the power consumption, i.e. for a MacBook Pro...
I checked the power consumption with a kill-a-watt and the consumption levels at load, idle and while asleep were identical with the Merom as they were with the Yonah. This is supported by the recent article over at AnandTech comparing the T7600 to the T2700.darthmullet said:
I was wondering if this was possible, thanks Topgunn.
But from hearing how difficult you found it, I think I won´t ever try this at home
Do you think anyone can do this or just the tech savy people?
But from hearing how difficult you found it, I think I won´t ever try this at home
Do you think anyone can do this or just the tech savy people?
GOOD LORD! That is a nightmare.
I got my kid the 2ghz core 2 duo 17" imac a while back and all he does is surf and play games. He is only 4 yrs old so he doesn't do much. I figured I could do some hardware swapping of his imac with the 1.66ghz mac mini I have seeing as how he doesn't need the power I figured I would grab his superdrive and his cpu and let him have mine. Well after a few hrs of starring at the superdrive in the imac I was finally able to get it out and swap it with my mac mini combo drive but that imac is setup is suicide. Worse than the powermac 8500 which was the most difficult mac to upgrade. I didn't even attempt it. He gets to keep the 2ghz core 2 duo chip by I got that superdrive.
I think I will wait and buy the chip later when the prices come down.
I got my kid the 2ghz core 2 duo 17" imac a while back and all he does is surf and play games. He is only 4 yrs old so he doesn't do much. I figured I could do some hardware swapping of his imac with the 1.66ghz mac mini I have seeing as how he doesn't need the power I figured I would grab his superdrive and his cpu and let him have mine. Well after a few hrs of starring at the superdrive in the imac I was finally able to get it out and swap it with my mac mini combo drive but that imac is setup is suicide. Worse than the powermac 8500 which was the most difficult mac to upgrade. I didn't even attempt it. He gets to keep the 2ghz core 2 duo chip by I got that superdrive.
I think I will wait and buy the chip later when the prices come down.
Did the design of the iMac change so much between the G5 and the Intel models that you now go in through the front rather than the back?
The designed changed between the iMac G5 and the iMac G5 with iSight. I can only think of two reasons why this might have been done. A) Apple didn't want people to easily fool around with the guts of their computers and B) the change was required due to the thinning of the enclosure that happened at the same time as the iMac getting the iSight.