1 |
Hi Janusz, |
2 |
|
3 |
On Sat, Sep 5, 2009 at 11:31 PM, <jsyrytczyk@×××××××××.pl> wrote: |
4 |
> Hi, I got Atom 330 with this motherboard sice six months or so. |
5 |
> |
6 |
> http://www.intel.com/products/desktop/motherboards/D945GCLF2-D945GCLF2D/D945GCLF2-D945GCLF2D-overview.htm |
7 |
> |
8 |
> The machine runs all right, but the compilation times are similar to *very* |
9 |
> outdated Celeron 1.7 GHz (which is one core only). The Atom goes very snappy |
10 |
> on every other task it performs (software mirror, backup, rsync, remote NX |
11 |
> station, email, iscsi, puppet, pulseaudio server etc.), |
12 |
> |
13 |
> but |
14 |
> |
15 |
> compilation times *are* slow. |
16 |
|
17 |
Your reply has been the most relevent to date, having been the only |
18 |
person who has an Atom 330 device for reference. |
19 |
|
20 |
After reading about your compilation speeds, I did a bit of googling |
21 |
for "intel core 2 duo vs intel atom 330", and found [1] which reveals |
22 |
quite a lot. It would seem that all of the Atom 330 "slowdowns" are |
23 |
caused by memory latency. |
24 |
|
25 |
It seems, that the Atom (both the 230 and 330) were not designed to |
26 |
use the blazingly fast FSB frequencies that all other modern Intel |
27 |
processors use, which is likely the primary reason for (sub-par) |
28 |
performance, and probably also the reason for their low power |
29 |
consumption. Most of the 45nm Core-2 processors support FSB |
30 |
frequencies around 1 or 1.3 GHz. On the other hand, the Atom 330 only |
31 |
supports FSB frequencies of 533 MHz. In terms of silicon / FETs, high |
32 |
clock speeds == high power leakage. |
33 |
|
34 |
So essentially, if the clock speeds of a Duo and Atom core were the |
35 |
same, then the Atom would require twice as much time as the Duo for |
36 |
the the same amount of "work" (i.e. memory reads / writes). Assuming |
37 |
that the Atom in-use power is about 1/2 of the duo, then both systems |
38 |
consume the same amount of power for a "task", but the Atom takes |
39 |
twice as long. |
40 |
|
41 |
In reality, the Atom consumes over half of the Duo in-use power. |
42 |
Therefore, power-efficiency ironically favours the Intel Core-2 Duo |
43 |
rather than the Atom for computation-intensive applications. For |
44 |
multimedia, I would say that the Atom is slightly more |
45 |
power-efficient. |
46 |
|
47 |
In my estimation, the lower in-use power of the Atom would be lost if |
48 |
it used a 1.3 GHz FSB controller. Does anyone disagree? |
49 |
|
50 |
Conclusion: |
51 |
|
52 |
The main bottleneck on the Atom 330 is not the CPU frequency, but |
53 |
rather the FSB frequency. Therefore, for a dedicated HTPC and / or NAS |
54 |
device, the an Intel Atom 330 device is a good choice. For a |
55 |
dedicated, low-power build machine, the Atom 330 is a bad choice for |
56 |
performance, but a good choice if only moderate performance is |
57 |
required. |
58 |
|
59 |
For a box that is intended to be used for HTPC / NAS and also a |
60 |
dedicated low-power build machine, an Atom 330 device is still a |
61 |
decent choice, because at least it performs efficiently for 2 out of 3 |
62 |
functions, and it's unlikely (physically impossible?) that one will |
63 |
find a comparable dual-core, low-power, fanless (and not liquid |
64 |
cooled) device with a 1.3 GHz FSB. |
65 |
|
66 |
So ... yea, I think I'll probably grab one of these ZOTAC boards |
67 |
anyway, at least for having an HTPC. Using it as a dedicated build |
68 |
machine would still be useful, even if the performance isn't |
69 |
particularly great. In any event, it won't be building packages |
70 |
constantly for my purposes, but only periodically. |
71 |
|
72 |
|
73 |
|
74 |
Cheers, |
75 |
|
76 |
Chris |
77 |
|
78 |
|
79 |
[1] http://www.tomshardware.com/reviews/dual-core-atom-330,2141-6.html |
80 |
[2] http://www.intel.com/products/processor/core2duo/specifications.htm |
81 |
[3] http://processorfinder.intel.com/details.aspx?sSpec=SLG9Y |