Muldersoft/LameXP
Muldersoft/LameXP
1
0
Fork 0
Code Releases 2 Activity

Slightly tweaked formula to selected the number of instances based on CPU count. Maximum value is now 16.

Browse Source
This commit is contained in:
LoRd_MuldeR 2011-11-27 02:06:03 +01:00
parent 3a74f55c5f
commit c1b88a7d07
4 changed files with 18 additions and 17 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

1
doc/Changelog.html
View File

@ -21,6 +21,7 @@ a:visited { color: #0000EE; }
<li>Updated LAME encoder to v3.99.2 Final (2011-11-18), compiled with ICL 12.1.7 and MSVC 10.0 (<a href="http://lame.cvs.sourceforge.net/viewvc/lame/lame/doc/html/history.html?revision=1.134" target="_blank">details</a>)
<li>Updated MediaInfo to v0.7.51+ (2011-11-19), compiled with ICL 12.1.6 and MSVC 10.0
<li>Implemented coalescing of update signals in order to reduce the CPU usage of the LameXP process
<li>Run more than four instances in parallel on systems with more than four CPU cores (<a href="FAQ.html#89cbd3d0" target="_blank">details</a>)
</ul><br>
<a name="4.03"></a>Changes between v4.02 and v4.03:<br><ul>

12
doc/FAQ.html
View File

@ -479,21 +479,21 @@ mapping! Instead a custom-made 'cubic spline' function is used to map the number
number of instances. This function has the following properties: On systems with at most four CPU cores, the<br>
maximum number of parallel instances will be identical to the number of CPU cores. On systems with eight CPU<br>
cores, the maximum number of parallel instances is six. On systems with 16 cores, the maximum number of<br>
parallel instances is eight. On systems with 32 cores, the maximum number of parallel instances is ten. And<br>
on systems with 64 cores, the maximum number of parallel instances is twelve. Twelve is the upper limit.<br>
parallel instances is eight. On systems with 32 cores, the maximum number of parallel instances is eleven.<br>
And on systems with 64 cores, the maximum number of parallel instances is 16. 16 is the upper limit.<br>
<br>
<a href="http://img265.imageshack.us/img265/9200/cpucoresmapping.png" target="_blank"><img src="http://img265.imageshack.us/img265/9200/cpucoresmapping.th.png" border="0" alt="thumb"></a><br>
<a href="http://img685.imageshack.us/img685/9453/cpucoresmappingnew.png" target="_blank"><img src="http://img685.imageshack.us/img685/9453/cpucoresmappingnew.th.png" border="0" alt="thumb"></a><br>
<br>
You may wonder why LameXP doesn't always create one instance for each CPU core. In theory, the more instances<br>
we create, the more CPU cores can be utilized. In reality, however, there are more "shared" resources on the<br>
computer. And, the more instances we run in parallel, the more processes will be competing for the shared<br>
we create, the more CPU cores can be utilized. In reality, however, there are some "shared" resources on the<br>
computer. And, the more instances we run in parallel, the more processes will be competing for these shared<br>
resources! More specifically, the amount of main memory (RAM) is limited. Creating a huge number of instances<br>
in parallel can easily use up all RAM, which will cause the operating system to make heavy use of the page<br>
file. This can result in HDD thrasing and may significantly hurt the overall performance! But even when there<br>
is enough RAM available on the system, each encoder or decoder instance needs to access the HDD, e.g. for<br>
reading the input file and for writing the output file. Thus too many instances will cause an I/O bottleneck!<br>
If, however, you think that LameXP's choice of the number of parallel instances is too conservative, you may<br>
overwrite the number of parallel instances on the "Advanced Options" tab. The upper limit is 16.<br>
overwrite the number of parallel instances on the "Advanced Options" tab. The upper limit is 16 for now.<br>
<br>
Finally, note that LameXP only controls the number of instances that will run in parallel, but it does NOT<br>
control how many threads an individual instance will create! Some encoders use "built-in" multi-threading and<br>

2
src/Config.h
View File

@ -30,7 +30,7 @@
#define VER_LAMEXP_MINOR_LO 4
#define VER_LAMEXP_TYPE Alpha
#define VER_LAMEXP_PATCH 4
#define VER_LAMEXP_BUILD 792
#define VER_LAMEXP_BUILD 793
///////////////////////////////////////////////////////////////////////////////
// Tool versions (minimum expected versions!)

20
src/Dialog_Processing.cpp
View File

@ -987,18 +987,18 @@ bool ProcessingDialog::shutdownComputer(void)
static int cores2instances(int cores)
{
//This function was approximated as a "cubic spline" with sampling points at:
//(1,1); (2,2); (4,4); (8,6); (16,8); (32,10); (64,12)
//This function is a "cubic spline" with sampling points at:
//(1,1); (2,2); (4,4); (8,6); (16,8); (32,11); (64,16)
static const double LUT[8][5] =
{
{ 1.0, 0.01440972, -0.04322917, 1.02881944, 0.00000000},
{ 2.0, -0.02881944, 0.21614583, 0.51006944, 0.34583333},
{ 4.0, 0.01017795, -0.25182292, 2.38194444, -2.15000000},
{ 8.0, 0.00005425, -0.00885417, 0.43819444, 3.03333333},
{16.0, 0.00011122, -0.01158854, 0.48194444, 2.80000000},
{32.0, 0.00000949, -0.00182292, 0.16944444, 6.13333333},
{64.0, 0.00000000, 0.00000000, 0.00000000, 12.00000000},
{DBL_MAX}
{ 1.0, 0.014353554, -0.043060662, 1.028707108, 0.000000000},
{ 2.0, -0.028707108, 0.215303309, 0.511979167, 0.344485294},
{ 4.0, 0.010016468, -0.249379596, 2.370710784, -2.133823529},
{ 8.0, 0.000282437, -0.015762868, 0.501776961, 2.850000000},
{16.0, 0.000033270, -0.003802849, 0.310416667, 3.870588235},
{32.0, 0.000006343, -0.001217831, 0.227696078, 4.752941176},
{64.0, 0.000000000, 0.000000000, 0.000000000, 16.000000000},
{DBL_MAX, 0.0, 0.0, 0.0, 0.0}
};
double x = abs(static_cast<double>(cores)), y = 1.0;