1 |
I have: Atom(TM) CPU 330 @ 1.60GHz |
2 |
running it with: |
3 |
|
4 |
CFLAGS="-march=core2 -O2 -pipe" |
5 |
CXXFLAGS="${CFLAGS}" |
6 |
MAKEOPTS="-j5" |
7 |
CHOST="x86_64-pc-linux-gnu" |
8 |
|
9 |
Didn't run any test performance on it, but the only disappointing fact about this small box is that it will not reboot itself when the power goes down. |
10 |
It runs my asterisk and hylafax in remote location so I have to keep an eye on it :-/ |
11 |
|
12 |
-- |
13 |
Joseph |
14 |
|
15 |
|
16 |
On 12/11/12 18:36, Frank Steinmetzger wrote: |
17 |
>Hello list |
18 |
> |
19 |
>Long time no read... :) |
20 |
> |
21 |
>It follows a verbose preamble. For the actual questions see dashed line below. |
22 |
>TL;DR summary: it’s all about ricer-performance questions on a netbook. |
23 |
> |
24 |
> |
25 |
>I have the luck of having obtained a used netbook for free (Atom N450, single- |
26 |
>core with HT, 1 GB memory, 5400 rev HDD). During the last week I’ve been |
27 |
>experimenting with 32 and 64 bits on it and am still quite undecided which to |
28 |
>keep. My reasons: |
29 |
> |
30 |
>- They are not as far apart in CPU performance as is the Core2. |
31 |
> I posted a 32/64 comparison for Core2 a few months ago, which showed that |
32 |
> Lilypond speedup on 64 bit was 50%. On the Atom, it actually took 5% longer. |
33 |
> (Sadly, Blender doesn’t build on 32 bit right now). |
34 |
>- Startup times for hogs like Firefox and KDE are quite equal between the two |
35 |
> (that could be attributed in parts to the fact that the 64 bit partition |
36 |
> sits on the disk’s first sectors, while 32 bit sits at the other end, I |
37 |
> don’t know which end is faster). |
38 |
>- pro 64: it is very easy to use distcc, as opposed to 32 bits (see below). |
39 |
>- con 64: it uses about 50% more memory, 32 bit builds are a little faster. |
40 |
> |
41 |
>The RAM argument is the most convincing one right now, since more free RAM |
42 |
>means more cache, which means a faster system in the long run. Currently, KDE |
43 |
>after logon needs 150 MB on 32 bit, and 250 MB on 64 bit (without akonadi for |
44 |
>now). But awesome WM rocks on a netbook anyway. |
45 |
> |
46 |
> |
47 |
>----------[ Questions begin ]------------------------------------------------ |
48 |
> |
49 |
>So I’m interested in you opinion and own experience about the following |
50 |
>arising questions: |
51 |
> |
52 |
>* From my observations, the benefit of 64 bit over 32 is much smaller for an |
53 |
> Atom than it is for my Core2. Am I right to assume thus that the Atom |
54 |
> architecture doesn’t have much to offer to 64 bit (such as extra registers)? |
55 |
> I’m not talking about memory here, since it’s limited to 2 GB in any case. |
56 |
> |
57 |
>* The problem of distcc between different architectures: |
58 |
> The netbook already had an older 32 bit Gentoo installed. And since I have |
59 |
> a multilib host (march=core2), I though I could upgrade with distcc (using |
60 |
> march=atom on the netbook). But at some point more and more stuff stopped |
61 |
> working, eventually I got “invalid instruction” errors during emerge, hence |
62 |
> I figured that was a dead end. |
63 |
> |
64 |
> So is it possible to mix architectures in this way at all with distcc? |
65 |
> I also have crossdev for i686 installed, which even shares files with the |
66 |
> system’s normal multilib gcc. I find that odd. |
67 |
> I sped up the installation process for 32 bit by using a chroot on the big |
68 |
> machine, which worked nicely. But it’s not a long-term solution, b/c it |
69 |
> uses up too much disk space on the host. |
70 |
> |
71 |
>* I’m interested in the question of -O2 vs. -Os. |
72 |
> Some sources say -Os is bad, b/c it breaks debugging and is mainly untested. |
73 |
> I won’t do heavy developing on it anyway, and Atoms do have a puny cache. |
74 |
> So I wonder whether -Os would improve execution time and RAM usage |
75 |
> noticably. Diskspace itself is not an issue. |
76 |
> |
77 |
>* I’m also interested in comparing bin packages over self-compiled ones. |
78 |
> E.g. I did compile icedtea, even if it’s just for TV browser. :) |
79 |
> Can you name a Java benchmark to measure CPU performance? |
80 |
> |
81 |
>* The last thing I’m going to set up is filesystem encryption, at least for ~. |
82 |
> I already know/think that AES would be the best choice due to limited CPU |
83 |
> power, but what else is there to heed besides key size? |
84 |
> |
85 |
>* What other small benchmarks for CPU and memory can you recommend? So far I |
86 |
> tested with nbench and sysbench. The results are so-and-so. Some computation |
87 |
> stuff is much slower on 64 bit, some a bit faster. The applicability to |
88 |
> every-day use is of course a wibbly-wobbly argument. |
89 |
> I also tested the runtime of some application (packing and unpacking of |
90 |
> archives, throughput with dd, mencoder). If there is interest, I can post |
91 |
> the result of 21 runs on each platform, measured with GNU time. |
92 |
> |
93 |
>----------[ Questions end ]-------------------------------------------------- |
94 |
> |
95 |
> |
96 |
>PS.: I’m aware that benchmarks are always a bit subjective and none is |
97 |
>perfect. I also realise that most of the questions quite belong into the |
98 |
>ricer corner. But Netbooks are ricer devices, b/c they need to perform at |
99 |
>their limits all the time. :-D |
100 |
> |
101 |
>If you read until this point, thank you very much for your time. You get a |
102 |
>unicorn as a reward: `^nn~ |
103 |
>-- |
104 |
>Gruß | Greetings | Qapla’ |
105 |
>Please do not share anything from, with or about me with any Facebook service. |
106 |
> |
107 |
>“Time is money” said the waiter and put the date on the bill. |