找回密码
 新注册用户
搜索
查看: 9018|回复: 26

[已完成翻译] [Collatz Conjecture]cmdline参数请求翻译

[复制链接]
发表于 2009-10-3 20:47:36 | 显示全部楼层 |阅读模式
One can configure the app through some parameters in the supplied app_info.xml.
This is done by editing the line

<cmdline></cmdline>

There are several options one can put in there, the order does not matter. If the same argument
is given twice, the last one counts. Remember that for editing the app_info.xml one should take
the same precautions as for the installation of the application, i.e. one should stop BOINC
completely and restart it afterwards.

Options:

kernel frequency: f  (default 30)
The application determines the size of the work packages sent to the GPU in a dynamic manner. It
tries to get the number of executed packages per second above the value specified here by
splitting them to smaller ones. The OS is completely blocked from accessing the graphics card for
the duration of one package leading to a somewhat sluggish behaviour of the user interface.
Limiting the size and therefore reducing the execution time per package creates more opportunities
for the OS to slip in between two work packages and to react to user input. The default value of
30 Hz is tuned for usability of the system. Reducing it could increase the throughput of the app
slightly.

Example:
<cmdline>f10</cmdline>
The app will try to run 10 work packages per second (limits the execution time to 100ms or shorter).


Wait factor: w  (default 1.0)
The app tries to release the CPU during the the GPU computations. It does so by predicting
the runtime for the GPU calls and send the CPU thread to sleep for that time. One can manually
correct the prediction of the application with this factor. When raising it, the CPU thread
sleeps longer, decreasing the value will lead to a faster wakeup of the thread. After that time,
the polling of the GPU starts. Setting this value to 0.0 turns off the release of the CPU.
Default is of course a value of 1.0. If you see a low GPU load, you can try to decrease it (if
you have a load of 80%, set it to 0.8). If you see a high CPU load, a slight increase of this
value may help. Increase it too much and it will affect the crunching speed. You could use this
for throttling of the GPU in case the high GPU load leads to a very sluggish behaviour of the
user interface or even VPU recover events. Setting w1.1 could improve the situation (see also
the f, b and p options).

Example:
<cmdline>w1.3</cmdline>
This would increase the sleep time by 30% in relation to the prediction.


Maximum GPU RAM use: r  (default 33.4)
The application allocates memory on the graphics card to store the calculated steps. Increasing
the size can offer more parallel data to process and the f option more room to get the real
setting closer to itS specified value. But you should not increase the RAM usage, if more than
one WU are processed concurrently. The value is given in percent relative to the installed amount.
Too high or too low values can crash the application, so be careful! See also the explanation
about the graphics RAM usage at the end of the readme.

Example:
<cmdline>r40</cmdline>
This would increase the maximum RAM usage to 40%.


priority: p (default 1)
All BOINC applications normally run with the lowest possible priority to not disturb other
applications. This can lead to a low GPU load, as it may be not possible to fire up the next tasks
if all cores of the CPU are under load. Raising the priority may help here. BOINC recommends a
slightly increased priority for GPU applications. This setting is the default. Possible values:
p0: idle priority, used by CPU BOINC applications
p1: normal priority in idle priority class, this is the default and recommended by BOINC for GPUs
p2: normal priority in normal priority class
p3: normal priority in high priority class, use with care!
The priority will affect how much time it takes for the the app to get back control if it releases
its time slice (see also option b).

Example:
<cmdline>p2</cmdline>
This raises the priority to normal (the same as all normal user applications have as default).


polling behavior of the GPU within the Brook runtime: b (default 1)
See the option w for starters. If that time has elapsed, the GPU polling starts. This can be done
by continously checking if the task has finished (b-1), enabling the fastest runtimes, but potentially
creating a high CPU load (a bit dependent on driver version). Second possibility is to release the time
slice allotted by the OS, so other apps can run (b0). The catch is that there is some interaction with
the priority. The time slice is only released to other tasks of the same priority. So raising the priority
effectively disables the release and the behavior is virtually identical to setting this parameter
to -1. If a raised priority and a low CPU time is wanted, one should leave it at the default of 1. This
suspends the task for at least 1 millisecond, enabling also tasks of lower priority to use the CPU in the
meantime.
Possible values:
b-1: busy waiting
b0: release time slice to other tasks of same priority
b1: release time slice for at least 1 millisecond
See also options p and w.

Example:
<cmdline>b-1</cmdline>
Enable busy waiting (no release of the time slice during polling).


The options can be combined in arbitrary order. Therefore the following argument list is valid:
<cmdline>w0.9 f20 r50 p2 b1</cmdline>
(wait factor 0.9, frequency 20Hz, 50% RAM used at max, normal priority, and CPU friendly polling)
Deprecated options (use with extreme care, will likely be removed in future versions)

Number of concurrent WUs: n  (default 3)
It is possible to define the maximum number of concurrently running WUs per GPU. Default value is
three. If the BOINC client schedules more WUs to run, they are put to a wait state and not crunched
until a GPU finishes another WU. Running more than a single WU can raise the efficiency of the GPU
as the remaining CPU calculations for one WU can be hidden during the wait for the GPU of another
WU. More than 2 WUs don't help anymore and running too many WUs concurrently first slow down the
calculation and finally crash if there is not enough memory. But as mentioned, more than 2 won't
help either way. If the high GPU load leads to a very sluggish behaviour of the user interface or
even VPU recover events, limiting the app to one WU (n1) could improve the situation (look also
into the w and f options). Do not use it with an ATI aware BOINC client! It should be ignored in
that case. If you run more than one GPU project (Collatz and Milkyway), this will be a shared
setting and should be the same for both in all cases!


Example:
<cmdline>n2</cmdline>
The app will run a maximum of two WUs per GPU.

Exclusion of GPUs: x  (default none)
If a certain GPU should not be used by the app, one can exclude it. The GPU number has to be in
the range 0..31. The first GPU in the system has the number 0. This option can be used multiple
times. Per default, all supported ATI GPUs are used. Do not use it with an ATI aware BOINC client!
It should be ignored in that case. But there may be some unwanted side effects.

Example:
<cmdline>x1 x2</cmdline>
The app will not use the second and the third GPU in the system.

如果能全部翻译完,一定会帮助所有GPU加速的同学,请各位有时间的帮下忙

[ 本帖最后由 ledled 于 2009-10-3 20:48 编辑 ]
回复

使用道具 举报

发表于 2009-10-3 21:08:28 | 显示全部楼层
厄,理论上来讲,虽然英语水平有限,不过灵格斯词霸+google+我的智商,应该能翻译下来。

不过好长的说。
回复

使用道具 举报

 楼主| 发表于 2009-10-3 21:15:08 | 显示全部楼层

回复 #2 cuihao 的帖子

第二个引用里面的内容可以不用翻,反正我打算让他们直接上6.10.X客户端

如果你能将它全部翻译完毕并且上了维基页,我会让饼干赏你几块拼图

[ 本帖最后由 ledled 于 2009-10-3 21:18 编辑 ]
回复

使用道具 举报

发表于 2009-10-3 21:20:07 | 显示全部楼层
我提供Options前面的部分:

通过编辑app_info.xml中的一些参数,可以配置计算程序的运行。
编辑类似于以下形式的内容即可:
<cmdline></cmdline>

这里有多个参数,顺序并不重要。如果同样的参数出现两次,则以最后一个为准。记住:在编辑app_info.xml之前,需要完全关闭BOINC,编辑后再打开。

下面的有点麻烦....好专业的说。
回复

使用道具 举报

发表于 2009-10-3 21:24:41 | 显示全部楼层

回复 #3 ledled 的帖子

拼圖有甚麼用?

ps:把文章分細一點吧~不用一下了譯一篇,很費神的~

[ 本帖最后由 universebreaker 于 2009-10-3 21:26 编辑 ]
回复

使用道具 举报

发表于 2009-10-3 21:44:08 | 显示全部楼层
再翻译一个(水平有限 ):
此处的work packages我翻译成了任务包,我的理解是程序把一个WU分成若干个小包再计算。

内核频率: f  (默认 30)

The application determines the size of the work packages sent to the GPU in a dynamic manner.
计算程序动态地决定送交GPU任务包的大小。

It tries to get the number of executed packages per second above the value specified here by splitting them to smaller ones.
计算程序从这里获得每秒钟运行任务包数量的具体值,以便把一个任务分割更小。

The OS is completely blocked from accessing the graphics card for the duration of one package leading to a somewhat sluggish behaviour of the user interface.
操作系统在一个任务包运行期间会完全阻断显卡的使用,以至于用户界面可能会略有迟钝。

Limiting the size and therefore reducing the execution time per package creates more opportunities for the OS to slip in between two work packages and to react to user input.
限制包的大小并因此缩短每个包占用的时间,会给操作系统更多时间响应用户输入。

The default value of 30 Hz is tuned for usability of the system. Reducing it could increase the throughput of the app slightly.
默认值30Hz(即每秒发出30个包)对于系统是比较合理的,减少这个值会稍微增加程序运行效率。

Example:
<cmdline>f10</cmdline>
The app will try to run 10 work packages per second (limits the execution time to 100ms or shorter).

例子:
<cmdline>f10</cmdline>
程序每秒会运行10个任务包(限制运行时间在100毫秒以内)。

[ 本帖最后由 cuihao 于 2009-10-3 21:46 编辑 ]

评分

参与人数 3基本分 +12 维基拼图 +10 收起 理由
霊烏路 空 + 10
cicikml + 6 辛苦了!
ledled + 6 不错,辛苦了 ^_^

查看全部评分

回复

使用道具 举报

发表于 2009-10-3 22:00:43 | 显示全部楼层

回复 #6 ledled 的评分

真的吗?
说实话,我都没整太明白 ,总结一下个人理解的大意:
计算程序把WU分割成更小的WU计算,因为进行计算时GPU是完全独占的,所以需要给OS一些GPU资源,WU分割的越小(即f越大),包计算的间隔就会增多,OS就会更顺畅,但计算效率会降低。

剩下的明天再看吧,睡觉了。

话说,论坛英语比我好的人,快来。
回复

使用道具 举报

发表于 2009-10-3 23:32:26 | 显示全部楼层
突然不想睡了,再翻译一个。


Wait factor: w  (default 1.0)
等待因数:w(默认 1.0)

The app tries to release the CPU during the the GPU computations.
程序在进行GPU计算时,会尝试释放CPU资源。

It does so by predicting the runtime for the GPU calls and send the CPU thread to sleep for that time.
计算程序会预计GPU调用(即CPU向GPU发出运算命令)所需要的时间,并告知CPU线程在GPU运算期间休息。

One can manually correct the prediction of the application with this factor.
通过这个因数,可以人为校正程序的预测时间。

When raising it, the CPU thread sleeps longer, decreasing the value will lead to a faster wakeup of the thread. After that time, the polling of the GPU starts.
若提升此参数,CPU线程会闲置更长时间(GPU运算减少),若减少,会导致CPU线程更快被唤醒(GPU运算增多)。

Setting this value to 0.0 turns off the release of the CPU.
设置此数值为0.0将会关闭CPU资源释放(一直进行计算)。

Default is of course a value of 1.0. If you see a low GPU load, you can try to decrease it (if you have a load of 80%, set it to 0.8).
默认值自然是1.0(100%),如果你发现GPU使用率低,可以尝试减少这个值(如果使用率为80%,设置为0.8)。

If you see a high CPU load, a slight increase of this value may help.
若CPU使用率过高,增加这个数值或许有所帮助。

Increase it too much and it will affect the crunching speed. You could use this for throttling of the GPU in case the high GPU load leads to a very sluggish behaviour of the user interface or even VPU recover events. Setting w1.1 could improve the situation (see also the f, b and p options).
增加该值过多会减慢GPU运算效率(但响应用户会加快)。万一过高的GPU使用率使你的计算机相应太慢甚至VPU重置,设置成w1.1可以减少GPU调用,缓解这种情况(另见f,b,p参数)。

Example:
<cmdline>w1.3</cmdline>
This would increase the sleep time by 30% in relation to the prediction.
例子:
<cmdline>w1.3</cmdline>
这样会增加30%的休息时间。

[ 本帖最后由 cuihao 于 2009-10-4 07:39 编辑 ]

评分

参与人数 1维基拼图 +15 收起 理由
霊烏路 空 + 15

查看全部评分

回复

使用道具 举报

发表于 2009-10-4 07:56:32 | 显示全部楼层
谁能提供一个关于GPU计算过程的资料,不然有些内容实在不懂

Maximum GPU RAM use: r  (default 33.4)
最大显存使用率:r (默认 33.4)

The application allocates memory on the graphics card to store the calculated steps.
计算程序使用显存来记录计算的内容。

Increasing the size can offer more parallel data to process and the f option more room to get the real setting closer to itS specified value.
增加该值可以为进程提供更多的并行数据,并会有更多的空间使f参数接近指定的值(不大理解)。

But you should not increase the RAM usage, if more than one WU are processed concurrently.
如果超过一个WU同时处理,则不应该增加显存使用。

The value is given in percent relative to the installed amount.
此数值是占用实际显存的百分率。

Too high or too low values can crash the application, so be careful!
注意,太高或太低的数值可能导致程序崩溃。

See also the explanation about the graphics RAM usage at the end of the readme.
另见readme最后关于显存用量的说明。

Example:
<cmdline>r40</cmdline>
This would increase the maximum RAM usage to 40%.
例子:
<cmdline>r40</cmdline>
将增加最大显存使用率到40%。

评分

参与人数 3基本分 +12 维基拼图 +12 收起 理由
霊烏路 空 + 12
cicikml + 6 感谢你的翻译!
ledled + 6 ( ̄▽ ̄)翻译还不错嘛,自信

查看全部评分

回复

使用道具 举报

 楼主| 发表于 2009-10-4 08:34:00 | 显示全部楼层
不懂的话请看之前f参数的说明:计算程序动态地决定送交GPU任务包的大小。
他的意思大概是增加该值会有更多的显存运行任务包

Default graphics RAM usage:

As a rule of thumb Collatz adapts to the amount of installed memory on the graphics card and
doesn't use more than a third of it per WU. It has a maximum usage of 528MB per WU (only on cards
with more than 1.5GB). The used memory is always (16 + a power of 2) MB, i.e. 17MB, 18MB, 20MB,
24MB, 32MB, 48MB, 80MB, 144MB, 272MB or 528MB. That means it will use 32MB per WU on a 128MB card,
80MB per WU on a 256MB card, 144MB per WU on a 512MB card, and 272MB per WU on a card with 1GB
or more. But this doesn't mean a card with more memory will be faster just because it is
allowed to use more RAM. Only real fast cards (think next generation cards) need the higher
amount of parallel tasks offered by using more RAM. For a HD4870 it doesn't make any noticeable
difference if you run the 512MB or the 1GB model.
I you use the r option the behaviour will be similar, just with another percentage.

关于显存用量的说明,最好还是翻译一下免得被那些不懂的新手调成r100直接崩溃
回复

使用道具 举报

发表于 2009-10-4 09:34:28 | 显示全部楼层

回复 #10 ledled 的帖子

o(╯□╰)o,你会错意了。
意思是计算程序把一个WU分成若干个小的包给GPU计算,这样GPU可以干一会儿休息一会儿。
增加这个值,分的就会更多,如同屏幕刷新一样,计算间隔可以响应用户。

我在润色一下语言( 话说我的语文作文水平很烂)。

[ 本帖最后由 cuihao 于 2009-10-4 09:38 编辑 ]
回复

使用道具 举报

发表于 2009-10-5 08:29:26 | 显示全部楼层
还没有高手过问吗?那我就继续翻译

priority: p (default 1)
优先级:p (默认 1)

All BOINC applications normally run with the lowest possible priority to not disturb other applications.
正常情况下,所有的BOINC计算程序为了不影响其他程序,都会以最低优先级运行。

This can lead to a low GPU load, as it may be not possible to fire up the next tasks if all cores of the CPU are under load.
当所有CPU核心在被使用时,程序暂时可能无法启动下一次计算(没有空闲的CPU资源来调用GPU)。这样做可以减少GPU使用率.

Raising the priority may help here.
提高优先级可能对计算效率会有一些帮助。

BOINC recommends a slightly increased priority for GPU applications.
对于GPU计算程序,BOINC提供了一个略高的优先级。

This setting is the default. Possible values:
有以下可行的设置:

p0: idle priority, used by CPU BOINC applications
p0:低优先级,BOINC的CPU计算程序使用。

p1: normal priority in idle priority class, this is the default and recommended by BOINC for GPUs
p1:较低优先级,GPU计算程序推荐。

p2: normal priority in normal priority class
p2:普通优先级。

p3: normal priority in high priority class, use with care!
p3:高优先级,谨慎使用。

The priority will affect how much time it takes for the the app to get back control if it releases its time slice (see also option b).
优先级会影响到计算程序获得多少的控制时间(另见b参数)。

Example:
<cmdline>p2</cmdline>
This raises the priority to normal (the same as all normal user applications have as default).
例子:
<cmdline>p2</cmdline>
提升优先级到普通优先级(与大多数应用程序相同)。

[ 本帖最后由 cuihao 于 2009-10-5 08:31 编辑 ]

评分

参与人数 1维基拼图 +10 收起 理由
霊烏路 空 + 10

查看全部评分

回复

使用道具 举报

发表于 2009-10-5 11:02:30 | 显示全部楼层
我也来掺活~~~小修小补~~~
内核频率: f  (默认 30)

The application determines the size of the work packages sent to the GPU in a dynamic manner.
计算程序动态地决定送交GPU任务包的大小。

It tries to get the number of executed packages per second above the value specified here by splitting them to smaller ones.
计算程序从这里获得每秒钟运行任务包数量的具体值,以便把一个任务分割更小。
程序会试图输送高于这里设定数字的每秒计算任务数量,这时程序会把任务包分割成较小的部分。

The OS is completely blocked from accessing the graphics card for the duration of one package leading to a somewhat sluggish behaviour of the user interface.
操作系统在一个任务包运行期间会完全阻断显卡的使用,以至于用户界面可能会略有迟钝。
在一个任务包的运行时期间,操作系统对显卡的访问将被阻断,这会导致用户界面反应迟钝现象的发生。

Limiting the size and therefore reducing the execution time per package creates more opportunities for the OS to slip in between two work packages and to react to user input.
限制包的大小并因此缩短每个包占用的时间,会给操作系统更多时间响应用户输入。

The default value of 30 Hz is tuned for usability of the system. Reducing it could increase the throughput of the app slightly.
默认值30Hz(即每秒发出30个包)对于系统是比较合理的,减少这个值会稍微增加程序的吞吐量

Example:
<cmdline>f10</cmdline>
The app will try to run 10 work packages per second (limits the execution time to 100ms or shorter).

例子:
<cmdline>f10</cmdline>
程序每秒会运行10个任务包(这将把运行时间限制在100毫秒以内)。
Wait factor: w  (default 1.0)
等待系数:w(默认 1.0)

The app tries to release the CPU during the the GPU computations.
程序在进行GPU计算时,会尝试释放CPU资源。

It does so by predicting the runtime for the GPU calls and send the CPU thread to sleep for that time.
计算程序会预计GPU调用(即CPU向GPU发出运算命令)所需要的时间,并告知CPU线程在GPU运算期间休息。

One can manually correct the prediction of the application with this factor.
通过这个因数,可以人为校正程序的预测时间。

When raising it, the CPU thread sleeps longer, decreasing the value will lead to a faster wakeup of the thread. After that time, the polling of the GPU starts.
若提升此参数,CPU线程会闲置更长时间,若减少,会导致CPU线程更快被唤醒。这一时间过后,对GPU的轮询会重新开始。

Setting this value to 0.0 turns off the release of the CPU.
设置此数值为0.0将会关闭CPU资源释放。

Default is of course a value of 1.0. If you see a low GPU load, you can try to decrease it (if you have a load of 80%, set it to 0.8).
默认值自然是1.0,如果你发现GPU使用率低,可以尝试减少这个值(如果使用率为80%,设置为0.8)。

If you see a high CPU load, a slight increase of this value may help.
若CPU使用率过高,增加这个数值或许有所帮助。

Increase it too much and it will affect the crunching speed. You could use this for throttling of the GPU in case the high GPU load leads to a very sluggish behaviour of the user interface or even VPU recover events. Setting w1.1 could improve the situation (see also the f, b and p options).
增加该值过多会减慢GPU运算效率(但响应用户会加快)。万一过高的GPU使用率使你的计算机相应太慢甚至VPU重置,设置成w1.1可以减少GPU调用,缓解这种情况(另见f,b,p参数)。

Example:
<cmdline>w1.3</cmdline>
This would increase the sleep time by 30% in relation to the prediction.
例子:
<cmdline>w1.3</cmdline>
这样会增加30%的休息时间。
Maximum GPU RAM use: r  (default 33.4)
最大显存使用率:r (默认 33.4)

The application allocates memory on the graphics card to store the calculated steps.
计算程序在显存中开辟空间以记录计算内容。

Increasing the size can offer more parallel data to process and the f option more room to get the real setting closer to itS specified value.
增加该值可以为进程提供更多的并行数据,并会有更多的空间使f参数接近指定的值(不大理解)。

But you should not increase the RAM usage, if more than one WU are processed concurrently.
如果超过一个WU同时处理,则不应该增加显存使用。

The value is given in percent relative to the installed amount.
此数值是占用实际显存的百分率。

Too high or too low values can crash the application, so be careful!
注意,太高或太低的数值可能导致程序崩溃。

See also the explanation about the graphics RAM usage at the end of the readme.
另见readme最后关于显存用量的说明。

Example:
<cmdline>r40</cmdline>
This would increase the maximum RAM usage to 40%.
例子:
<cmdline>r40</cmdline>
将增加最大显存使用率到40%。

priority: p (default 1)
优先级:p (默认 1)

All BOINC applications normally run with the lowest possible priority to not disturb other applications.
正常情况下,所有的BOINC计算程序为了不影响其他程序,都会以最低优先级运行。

This can lead to a low GPU load, as it may be not possible to fire up the next tasks if all cores of the CPU are under load.
当所有CPU核心在被使用时,程序暂时可能无法启动下一次计算(没有空闲的CPU资源来调用GPU),这样做可能会导致GPU使用率过低.

Raising the priority may help here.
提高优先级可能对计算效率会有一些帮助。

BOINC recommends a slightly increased priority for GPU applications.
对于GPU计算程序,BOINC提供了一个略高的优先级。

This setting is the default. Possible values:
有以下可行的设置:

p0: idle priority, used by CPU BOINC applications
p0:低优先级,BOINC的CPU计算程序使用。

p1: normal priority in idle priority class, this is the default and recommended by BOINC for GPUs
p1:较低优先级,GPU计算程序推荐。

p2: normal priority in normal priority class
p2:普通优先级。

p3: normal priority in high priority class, use with care!
p3:高优先级,谨慎使用。

The priority will affect how much time it takes for the the app to get back control if it releases its time slice (see also option b).
优先级会影响到计算程序在释放CPU时间后多久可以重新获得控制(另见b参数)。

Example:
<cmdline>p2</cmdline>
This raises the priority to normal (the same as all normal user applications have as default).
例子:
<cmdline>p2</cmdline>
提升优先级到普通优先级(与大多数应用程序相同)。

评分

参与人数 4基本分 +207 维基拼图 +70 收起 理由
BiscuiT + 200 + 60
霊烏路 空 + 10
cuihao + 1 多谢多谢,学习学习,大家共
ledled + 6 Thanks ^_^

查看全部评分

回复

使用道具 举报

发表于 2009-10-6 13:50:46 | 显示全部楼层
越来越难了,这个参数名称谁翻译一下。
polling behavior of the GPU within the Brook runtime: b (default 1)

See the option w for starters.
首先请看w参数的介绍。

If that time has elapsed, the GPU polling starts.
在等待时间过去后,GPU轮询开始。

This can be done by continously checking if the task has finished (b-1), enabling the fastest runtimes, but potentially creating a high CPU load (a bit dependent on driver version).
GPU轮询可以通过持续不断地进行检查运算是否完成来实现(b-1),这样用的时间最少,但是可能导致很高的CPU使用率(部分依赖于硬件驱动版本)。

Second possibility is to release the time slice allotted by the OS, so other apps can run (b0).
第二种选择是由操作系统分配处理时间,这样别的程序可以正常运行(b0)。

The catch is that there is some interaction with the priority.
需要注意的是,优先级在这里会有一些影响。

The time slice is only released to other tasks of the same priority.
处理时间只会分配给相同优先级的其它进程。

So raising the priority effectively disables the release and the behavior is virtually identical to setting this parameter to -1.
所以提升优先级,禁止释放计算资源,实质上和设置此参数为-1作用相同。

If a raised priority and a low CPU time is wanted, one should leave it at the default of 1.
如果你提升优先级并且希望所用的CPU时间尽量少,你应该设置此参数为默认值1。

This suspends the task for at least 1 millisecond, enabling also tasks of lower priority to use the CPU in the meantime.
b参数设为1,每次调用GPU计算中间至少会有1毫秒间隔,使得低优先级的任务在此期间可以使用CPU资源。

Possible values:
b-1: busy waiting
b0: release time slice to other tasks of same priority
b1: release time slice for at least 1 millisecond
See also options p and w.
可行的设置:
b-1:忙等待
b0:向同一优先级程序释放CPU处理时间
b1:至少释放1毫秒处理时间
另见p、w参数

Example:
<cmdline>b-1</cmdline>
Enable busy waiting (no release of the time slice during polling).
例子:
<cmdline>b-1</cmdline>
允许忙等待(GPU轮询期间不释放CPU资源)


发现漏了一小段:

The options can be combined in arbitrary order. Therefore the following argument list is valid:
<cmdline>w0.9 f20 r50 p2 b1</cmdline>
(wait factor 0.9, frequency 20Hz, 50% RAM used at max, normal priority, and CPU friendly polling)

各个参数可以以任意顺序写在同一行,因此这条参数是允许的:
<cmdline>w0.9 f20 r50 p2 b1</cmdline>
(等待系数0.9,内核频率20,最大显存使用率50%,标准优先级,CPU优先的轮询方式)

[ 本帖最后由 cuihao 于 2009-10-7 08:51 编辑 ]

评分

参与人数 2基本分 +8 维基拼图 +10 收起 理由
霊烏路 空 + 10
ledled + 8 谢谢,我明天整理一下……

查看全部评分

回复

使用道具 举报

 楼主| 发表于 2009-10-7 08:18:31 | 显示全部楼层
Note on Catalyst driver versions:

ATI/AMD renamed the CAL libraries with their later driver releases. You have to run the right
version which fits to the names of the CAL libraries of the driver release you have installed.
Catalyst 8.12 and 9.1 use the amd prefixed names, later ones use ati prefixed names (CAL 1.4).

If you choose the right version, there shouldn't be any need to meddle with the dll files. But
if you want to run the version designed for amd prefixed dll names to run with later drivers,
duplicate the following files in your Windows\system32 folder (the suffix "64" applies only to
64Bit systems):

aticalrt.dll or aticalrt64.dll
aticalcl.dll or aticalcl64.dll
aticaldd.dll or aticaldd64.dll

Now Rename the copies to

amdcalrt.dll or amdcalrt64.dll
amdcalcl.dll or amdcalcl64.dll
amdcaldd.dll or amdcaldd64.dll

The simplest way to do that is using the console (command prompt). Start -> execute, type "cmd",
press enter ;). Type in the following two commands:

cd %systemroot%\system32
copy atical*.dll amdcal*.dll

That's all. You should have now six dll files in your Windows\system32 folder, three named
amdcal*.dll and three with the name atical*.dll. A simple rename of atical*.dll to amdcal*.dll
is not sufficient!

Default graphics RAM usage:

As a rule of thumb Collatz adapts to the amount of installed memory on the graphics card and
doesn't use more than a third of it per WU. It has a maximum usage of 528MB per WU (only on cards
with more than 1.5GB). The used memory is always (16 + a power of 2) MB, i.e. 17MB, 18MB, 20MB,
24MB, 32MB, 48MB, 80MB, 144MB, 272MB or 528MB. That means it will use 32MB per WU on a 128MB card,
80MB per WU on a 256MB card, 144MB per WU on a 512MB card, and 272MB per WU on a card with 1GB
or more. But this doesn't mean a card with more memory will be faster just because it is
allowed to use more RAM. Only real fast cards (think next generation cards) need the higher
amount of parallel tasks offered by using more RAM. For a HD4870 it doesn't make any noticeable
difference if you run the 512MB or the 1GB model.
I you use the r option the behaviour will be similar, just with another percentage.

还有这两段未翻译,请继续

[ 本帖最后由 ledled 于 2009-10-7 09:00 编辑 ]
回复

使用道具 举报

您需要登录后才可以回帖 登录 | 新注册用户

本版积分规则

论坛官方淘宝店开业啦~

Archiver|手机版|小黑屋|中国分布式计算总站 ( 沪ICP备05042587号 )

GMT+8, 2024-3-29 14:30

Powered by Discuz! X3.5

© 2001-2024 Discuz! Team.

快速回复 返回顶部 返回列表