[pdovinh]

Congratulations Gordio!!!

Initial thoughts:

Rush out and buy the Mac mini 1.66 Intel Core Duo RIGHT NOW!!!
This new baby is NICE and SMOOTH! It's a winner!!!

I'm too excited to say much more but below are some Xbench data.

Xbench for 1.42 OC'ed to 1.58:

Results 43.36
System Info
Xbench Version 1.2
System Version 10.4.5 (8H14)
Physical RAM 1024 MB
Model PowerMac10,1
Processor PowerPC G4 @ 1.58 GHz
L1 Cache 32K (instruction), 32K (data)
L2 Cache 512K @ 750 MHz
Bus Frequency 167 MHz
Video Card ATY,RV280
Drive Type ST9100823A
CPU Test 59.50
GCD Loop 147.20 7.76 Mops/sec
Floating Point Basic 33.74 801.69 Mflop/sec
vecLib FFT 92.81 3.06 Gflop/sec
Floating Point Library 49.96 8.70 Mops/sec
Thread Test 70.86
Computation 68.31 1.38 Mops/sec, 4 threads
Lock Contention 73.61 3.17 Mlocks/sec, 4 threads
Memory Test 24.71
System 26.40
Allocate 132.03 484.85 Kalloc/sec
Fill 24.17 1175.06 MB/sec
Copy 15.46 319.29 MB/sec
Stream 23.23
Copy 22.47 464.19 MB/sec [altivec]
Scale 22.52 465.21 MB/sec [altivec]
Add 23.95 510.19 MB/sec [altivec]
Triad 24.09 515.33 MB/sec [altivec]
Quartz Graphics Test 59.33
Line 43.61 2.90 Klines/sec [50% alpha]
Rectangle 52.02 15.53 Krects/sec [50% alpha]
Circle 52.09 4.25 Kcircles/sec [50% alpha]
Bezier 73.61 1.86 Kbeziers/sec [50% alpha]
Text 107.14 6.70 Kchars/sec
OpenGL Graphics Test 73.84
Spinning Squares 73.84 93.67 frames/sec
User Interface Test 35.30
Elements 35.30 162.02 refresh/sec
Disk Test 31.92
Sequential 55.65
Uncached Write 32.65 20.05 MB/sec [4K blocks]
Uncached Write 64.18 36.31 MB/sec [256K blocks]
Uncached Read 86.01 25.17 MB/sec [4K blocks]
Uncached Read 71.21 35.79 MB/sec [256K blocks]
Random 22.38
Uncached Write 7.22 0.76 MB/sec [4K blocks]
Uncached Write 66.86 21.40 MB/sec [256K blocks]
Uncached Read 68.05 0.48 MB/sec [4K blocks]
Uncached Read 94.93 17.62 MB/sec [256K blocks]

Xbench for 1.66 Core Duo:
Results 52.86
System Info
Xbench Version 1.2
System Version 10.4.5 (8H1619)
Physical RAM 512 MB
Model Macmini1,1
Drive Type ST98823AS
CPU Test 63.86
GCD Loop 216.72 11.42 Mops/sec
Floating Point Basic 75.72 1.80 Gflop/sec
vecLib FFT 36.85 1.22 Gflop/sec
Floating Point Library 56.57 9.85 Mops/sec
Thread Test 166.22
Computation 150.29 3.04 Mops/sec, 4 threads
Lock Contention 185.93 8.00 Mlocks/sec, 4 threads
Memory Test 95.47
System 100.50
Allocate 108.21 397.37 Kalloc/sec
Fill 94.27 4583.51 MB/sec
Copy 99.98 2065.12 MB/sec
Stream 90.91
Copy 77.54 1601.61 MB/sec
Scale 76.20 1574.35 MB/sec
Add 112.01 2386.04 MB/sec
Triad 110.49 2363.60 MB/sec
Quartz Graphics Test 58.05
Line 58.91 3.92 Klines/sec [50% alpha]
Rectangle 53.40 15.94 Krects/sec [50% alpha]
Circle 53.20 4.34 Kcircles/sec [50% alpha]
Bezier 75.57 1.91 Kbeziers/sec [50% alpha]
Text 54.33 3.40 Kchars/sec
OpenGL Graphics Test 183.74
Spinning Squares 183.74 233.08 frames/sec
User Interface Test 25.08
Elements 25.08 115.12 refresh/sec
Disk Test 26.50
Sequential 49.56
Uncached Write 48.53 29.80 MB/sec [4K blocks]
Uncached Write 49.69 28.12 MB/sec [256K blocks]
Uncached Read 46.21 13.52 MB/sec [4K blocks]
Uncached Read 54.53 27.41 MB/sec [256K blocks]
Random 18.09
Uncached Write 7.36 0.78 MB/sec [4K blocks]
Uncached Write 56.18 17.99 MB/sec [256K blocks]
Uncached Read 41.24 0.29 MB/sec [4K blocks]
Uncached Read 23.12 4.29 MB/sec [256K blocks]

More thoughts:

1) Very quick, responsive, and smooth with Safari, iTunes, and iPhoto (as compared to G5 2.0 DP w/ 2 STRIPED 74G Raptors)!!! WOW!

2) Very quiet. Yes, more heat escapes from the back as compared to 1.42 and 1.25 minis, but the fan doens't rev up.

3) Front Row and Remote works flawlessly!

http://forums.macnn.com/images/smilies/laughing.gif

Will add 2 gigs tonight and post a new Xbench after that.

[JoshuaZ]

I actually like the redesign of the mini. My only hardware concern is with the video card, which in all reality doesn't have to be good on the ultra low end computer. My other concern is the $100 base price mark up. It would be nice if they'd kept it at $500.

[mavherzog]

Originally Posted by JoshuaZ

I actually like the redesign of the mini. My only hardware concern is with the video card, which in all reality doesn't have to be good on the ultra low end computer. My other concern is the $100 base price mark up. It would be nice if they'd kept it at $500.

From another thread:

Originally Posted by mduell

The single core model with Airport, Bluetooth, 80GB HDD, and 512MB RAM is the same price as the old single core model with Airport, Bluetooth, 80GB HDD, and 512MB RAM. Apple eliminated the $499 price point with less features/capacity.

[Grrr]

I don't mean to be a party pooper.. But how is that 4X faster? Maybe i missed something, but it doesn't look anywhere near 4X faster to me. I know this is being compared to an over clocked original Mini, but still.. From where im sitting, its slower at many of the tests.

[JKT]

Firstly, XBench is a very imperfect benchmarking tool. Secondly, 4x or whatever faster is the optimal speed increase in a synthetic benchmark. In real world use, it will never achieve that, but in some cases it will get close.

[Grrr]

It doesn't look anywhere near close to me.
As for Xbench.. The only imperfect thing about it, is that it doesn't prompt the user to match and standardize various system settings between tests. But for those people that are less than retarded, this really isn't a problem.
Xbench simply times how long it takes for the Mac to do various simple tasks. The faster it does it, the better the test result. It doesn't need to be any more sophisticated than that.
I don't think there is so much wrong with it myself..

[Stratus Fear]

Originally Posted by Grrr

It doesn't look anywhere near close to me.
As for Xbench.. The only imperfect thing about it, is that it doesn't prompt the user to match and standardize various system settings between tests. But for those people that are less than retarded, this really isn't a problem.
Xbench simply times how long it takes for the Mac to do various simple tasks. The faster it does it, the better the test result. It doesn't need to be any more sophisticated than that.
I don't think there is so much wrong with it myself..

Xbench isn't very reliable because it can (and often does) vary, sometimes wildly, between test runs on the same machine. The other problem comparing old PPC systems to new Intel ones with Xbench is the fact that Xbench, as compiled for the PPC systems, doesn't force beam synchronization (compatibility with <10.4). The Intel systems don't have an older version of the OS to support, so beam sync is enabled with the Intel Xbench build. If you force beam sync off (w/ Quartz Debug), you're far more likely to get a result possibly a couple times higher than the old mini. But yeah, that's pretty much why I wouldn't use Xbench as a method of comparison. It's not that accurate.

[funkboy]

Hold on though... the hard drive is *slower* on the intel mini? That is depressing... unless you swapped the HD on your PowerPC one...

[Eug Wanker]

Originally Posted by JKT

Firstly, XBench is a very imperfect benchmarking tool. Secondly, 4x or whatever faster is the optimal speed increase in a synthetic benchmark. In real world use, it will never achieve that, but in some cases it will get close.

Cinebench's CPU rendering is 3X as fast on the Core Duo 1.66 as it is on the G4 1.5.
The Core Solo 1.5 is about 1.4X as fast as it is on the G4 1.5.

I suspect that Handbrake is around 3-4X as fast on the Core Duo 1.66 as compared to the G4 1.5 as well, and that's an app I use regularly. (It's pretty fast on my G5 iMac, but it's very slow on my G4 iBook.)

[pdovinh]

Hi guys,

You guys are a bunch of cool people and I have learned much from the threads that you've posted.

I ran the Xbench just to get some sort of comparisons for discussion purposes only. Conlusions should be made by hands-on experiences or those from reliable sources or veteran users/tester.

The Xbench tests posted included 1.42 OC'ed to 1.58, 1G RAM, ATA 100 G HD VS. the new mini 1.66, 512K RAM, SATA 80 G. Therefore no claims have been made.

With the mini 1.66, I found user experience of iTunes, Safari, and iPhoto improved. They ran faster, smoother, and much more responsive than the other minis.

Subjectively, I felt it was almost as responsive as my G5 DP 2.0 w/ two 74 G Raptors STRIPED! Now that means something. I was impressed with the results.

Besides all the posted specifications of the new mini, I believe that the price increase of this version is justified.

There are many improvements made internally:

1) Airport and Bluetooth transmitters are now secured on top of spring-loaded posts, with all cables nicely tucked away behind built-in tabs. Cables are tidied up and no longer taped in place like before. They are easily accessible for removal. Groundings, noise dampenings, ventilation, and vibration-reduction measures were taken and implemented.

2) The cooling assembly and fan combination has been redesigned and seemed to be more efficient. More heat are being thrown to the back without the fan reving up.

3) The DVD could be removed easily as before with four screws from the sides and two from the back, holding the ATA PCB.

4) The plastic housing holding the HD could be removed and flipped toward the back for easy access to the main mother board by taking off 1 screw at each corners.

5) The two DDRAMs slots are located on the main board and easily accessible. Maximum 2Gs is great for this mini.

6) Remote control and Front Row worked flawlessly. The sensor is located at one end of the DVD disk slot. Well hidden and cleverly done.

7) Faster SATA drive is a definite plus!

8) Optical IN and OUT for sound is something that I really appreciate. I hooked the optical out to a decent Musical Fidelity DAC and I have achieved very nice audios. I love it.

9) Extra USBs are so nice to have.

10) It's just so nice to see Activity Monitor showing both processors humming along in this little guy.

This version of the mini is a definite improvement and I highly recommend it.

Phil Dovinh.