I get errors on my ATI 6450 when running MB tasks.
Any idea whats wrong, or isn't card supported?
http://setiweb.ssl.berkeley.edu/beta/results.php?hostid=55958&offset=0&show_names=0&state=6&appid=

from the 25'th and forth.

I get errors on my ATI 6450 when running MB tasks.
Any idea whats wrong, or isn't card supported?
http://setiweb.ssl.berkeley.edu/beta/results.php?hostid=55958&offset=0&show_names=0&state=6&appid=

from the 25'th and forth.

<message>
Maximum elapsed time exceeded
</message>

It's BOINC own issue.

I get errors on my ATI 6450 when running MB tasks.
Any idea whats wrong, or isn't card supported?
http://setiweb.ssl.berkeley.edu/beta/results.php?hostid=55958&offset=0&show_names=0&state=6&appid=

from the 25'th and forth.

You Basically can't run GPUs that are too disimiliar together, if the slower GPU is more than 10 times slower all it's tasks will end in Maximum Time Exceeded,

OpenCL Platform Name: AMD Accelerated Parallel Processing
Number of devices: 2
Max compute units: 2
Max work group size: 256
Max clock frequency: 625Mhz
Max memory allocation: 199753728
Cache type: None
Cache line size: 0
Cache size: 0
Global memory size: 536870912
Constant buffer size: 65536
Max number of constant args: 8
Local memory type: Scratchpad
Local memory size: 32768
Queue properties:
Out-of-Order: No
Name: Caicos
Vendor: Advanced Micro Devices, Inc.
Driver version: CAL 1.4.1703 (VM)
Version: OpenCL 1.2 AMD-APP (923.1)
Extensions: cl_khr_global_int32_base_atomics cl_khr_global_int32_extended_atomics cl_khr_local_int32_base_atomics cl_khr_local_int32_extended_atomics cl_khr_3d_image_writes cl_khr_byte_addressable_store cl_khr_gl_sharing cl_ext_atomic_counters_32 cl_amd_device_attribute_query cl_amd_vec3 cl_amd_printf cl_amd_media_ops cl_amd_popcnt cl_khr_d3d10_sharing
Max compute units: 18
Max work group size: 256
Max clock frequency: 750Mhz
Max memory allocation: 536870912
Cache type: None
Cache line size: 0
Cache size: 0
Global memory size: 1073741824
Constant buffer size: 65536
Max number of constant args: 8
Local memory type: Scratchpad
Local memory size: 32768
Queue properties:
Out-of-Order: No
Name: Cypress
Vendor: Advanced Micro Devices, Inc.
Driver version: CAL 1.4.1703 (VM)
Version: OpenCL 1.2 AMD-APP (923.1)
Extensions: cl_khr_fp64 cl_amd_fp64 cl_khr_global_int32_base_atomics cl_khr_global_int32_extended_atomics cl_khr_local_int32_base_atomics cl_khr_local_int32_extended_atomics cl_khr_3d_image_writes cl_khr_byte_addressable_store cl_khr_gl_sharing cl_ext_atomic_counters_32 cl_amd_device_attribute_query cl_amd_vec3 cl_amd_printf cl_amd_media_ops cl_amd_popcnt cl_khr_d3d10_sharing

Claggy

I get errors on my ATI 6450 when running MB tasks.
Any idea whats wrong, or isn't card supported?
http://setiweb.ssl.berkeley.edu/beta/results.php?hostid=55958&offset=0&show_names=0&state=6&appid=

from the 25'th and forth.

You Basically can't run GPUs that are too disimiliar together, if the slower GPU is more than 10 times slower all it's tasks will end in Maximum Time Exceeded,

OpenCL Platform Name: AMD Accelerated Parallel Processing
Number of devices: 2
Max compute units: 2
Max work group size: 256
Max clock frequency: 625Mhz
Max memory allocation: 199753728
Cache type: None
Cache line size: 0
Cache size: 0
Global memory size: 536870912
Constant buffer size: 65536
Max number of constant args: 8
Local memory type: Scratchpad
Local memory size: 32768
Queue properties:
Out-of-Order: No
Name: Caicos
Vendor: Advanced Micro Devices, Inc.
Driver version: CAL 1.4.1703 (VM)
Version: OpenCL 1.2 AMD-APP (923.1)
Extensions: cl_khr_global_int32_base_atomics cl_khr_global_int32_extended_atomics cl_khr_local_int32_base_atomics cl_khr_local_int32_extended_atomics cl_khr_3d_image_writes cl_khr_byte_addressable_store cl_khr_gl_sharing cl_ext_atomic_counters_32 cl_amd_device_attribute_query cl_amd_vec3 cl_amd_printf cl_amd_media_ops cl_amd_popcnt cl_khr_d3d10_sharing
Max compute units: 18
Max work group size: 256
Max clock frequency: 750Mhz
Max memory allocation: 536870912
Cache type: None
Cache line size: 0
Cache size: 0
Global memory size: 1073741824
Constant buffer size: 65536
Max number of constant args: 8
Local memory type: Scratchpad
Local memory size: 32768
Queue properties:
Out-of-Order: No
Name: Cypress
Vendor: Advanced Micro Devices, Inc.
Driver version: CAL 1.4.1703 (VM)
Version: OpenCL 1.2 AMD-APP (923.1)
Extensions: cl_khr_fp64 cl_amd_fp64 cl_khr_global_int32_base_atomics cl_khr_global_int32_extended_atomics cl_khr_local_int32_base_atomics cl_khr_local_int32_extended_atomics cl_khr_3d_image_writes cl_khr_byte_addressable_store cl_khr_gl_sharing cl_ext_atomic_counters_32 cl_amd_device_attribute_query cl_amd_vec3 cl_amd_printf cl_amd_media_ops cl_amd_popcnt cl_khr_d3d10_sharing

Claggy

Ok.
Thank you for the answers.
I use exclude gpu 5850 in cc_config.xml to only run on the 6450
And I use it on POEM and that works.
I have reconfigured my cc_config.xml to only run POEM on my 6450 for now.

I found out that buying that 6450 probebly was a mistake.
I bought it with the intention to have it as a second card and use it for Movies.
I found out again that it's still not possible to "clone" desktops between 2 graphic cards. Lucky for me that card was second hand and pretty cheap.

I just wanted to test if it would work with the stock ATI app. It probebly would if it was alone in one computor. I have the intention to set up a small web server some day. There the card is perfect(low power card).

Now I try to run the MB ATI tasks on my 5850.......GAHHHH
I use exclude in the cc_config.xml file to only run on 5850.
It doesn't work to well with the other card in the computor...

I will remove the 6450 soon though. It computes "freeky" with this config on some project combinations and it's not worth the effort to figure out what will work and what will.

with amd releasing the new accelerated streaming sdk 2.8

will we be seeing new compile's of the clients soon?

also when I try to add seti beta to boinc it say's that i have already joined

but does not show as project?

I have done beta work in the past but no longer know how to access it help.

---

also when I try to add seti beta to boinc it say's that i have already joined

but does not show as project?

I have done beta work in the past but no longer know how to access it help.

Well, reattach your host to Seti Beta using the Seti Beta Website url, your email address and your password then.

Claggy

thought i did that will try again

are you running eyefinity under catalyst control center?

must have missed something the first time works.

---

are these wu the x4 ones?

---

are these wu the x4 ones?

No, they are standard v7 Wu's.

Claggy

sorry about aborting a bunch astropulse units yesterday then loosing the mb when i reset

project.

as it turns out cuda jobs will only run about 2.68% on an ati card before

crashing the driver.

basically I did not ketch the client before I checked the preference's and they

were set wrong.

evreything is fine now will be more careful from here on.

LOL

---

as it turns out cuda jobs will only run about 2.68% on an ati card before

crashing the driver.

basically I did not ketch the client before I checked the preference's and they

were set wrong.

evreything is fine now will be more careful from here on.

I was half expecting you to have trouble with the 6.99 ATI app and the Cat 12.11 driver and to have driver restarts,

A bit of background regarding the ATI 6.99 app, when first run the app compiles some kernels and wisdom files using whatever driver is installed, yours were done on this Wu:

Task 11648887

Up until Cat 12.8 this worked right, with Cat 12.10 the compilation caused a faultly kernel file (or enabled a latent app Bug), causing the 6.99 ATI app to find the wrong number of pulses,
My HD7770's compilations were done on Cat 12.8, so i've been producing valid work even when running Cat 12.10, (unlike a lot of Cat 12.10 running hosts around here)
i've recently upgraded to Cat 12.11 Beta 8, now on my offline runs the app produces kernels that produce valid work,
i did remove my Cat 12.8 produced kernel and wisdom files yesterday, and let the ATI app produce new ones, but i got a driver restart soon after it was done,
it corrupted the display enough that i had to reboot, then i had another driver restart soon afterwards, reverted back to the Cat 12.8 produced files and everything is running sweet again,
Raistmer knows about the Cat 12.10 problem, and at some point will look into it,

Claggy

first astropulse finished ok but afterburner and tthrottle show vary low gpu utilization.

went to look at xml in boinc no xml how do I ajust client?

---

that is weird ran primgrid to check gpu's went back to setibeta gpu's dropped to 1/4.

the first time gpu #1 was 40% gpu #2 was 10% at least there now balenced.

---

at this point only running beta astropulse gpu utilization all over the place but averaging low?

---

first astropulse finished ok but afterburner and tthrottle show vary low gpu utilization.

Make sure you free a CPU core, there is a low GPU usage Bug/feature with the ATI/AMD and Nvidia OpenCL apps, if there isn't enough free CPU then GPU utilisation suffers.

Claggy

first I am running 2 ati hd6870's.

second I am only running gpu tasks just to check that.

so I have two gpu tasks that are bouncing between 10% and 25%

with nothing else running.

after upgrading to boinc 7.31 and downgrading back to 7.28

there is no xml file to adjust.

should i go back farther to 6.81 maybe?

---

did a clean install of 12.11 beta11 no differance.

did a clean istall of 12.8

still have problem but graph smother will wait and see.