I wasn't referring CPU performance.

The impact of CPU performance when the iGPU is already bottlenecking the RAM bandwidth is not very important.

The overall performance loss caused by the iGPU not getting max. RAM bandwidth is the key. The CPU cores anyway hardly contribute much to the total output, unless it is a very potent CPU coupled with a rather slow iGPU (although that can depend on the project).

The potential output gain from running CPU cores is typically more than offset but the significantly reduced iGPU performance under these conditions.

I am not how you would come to the conclusion that the CPU is not contributing as much as the iGPU.

0 CPU + 1 iGPU MB
~1hr for iGPU = 24 iGPU tasks a day.
Total 24 tasks a day

4 CPU MB + 1 iGPU MB
~2hr for CPU & ~1hr for iGPU = 48 CPU tasks & 24 iGPU tasks a day.
Total 72 tasks a day

4 CPU MB + 0 iGPU
~1hr for CPU = 96 CPU tasks day.
Total 96 tasks a day

I think maybe you misunderstand use of iGPU in this case? iGPU is being used to describe Intel HD Graphics 4600 GPU built into the CPU & not an ATI or NVIDIA GPU.

---

SETI@home classic workunits: 93,865 CPU time: 863,447 hours
Join the [url=http://tinyurl.com/8y46zvu]BP6/VP6 User Group[

Ah, okay...

Running a similar setup (albeit with a rather slow mobile Core i3) myself since just a few days, I also found out the performance ratio on SETI comparing CPU and iGPUs is unexpectedly close.

So in this case, you're right as the iGPU doesn't contribute so much to the overall score when bundled with fast CPU cores.

---

Ah, okay...

Running a similar setup (albeit with a rather slow mobile Core i3) myself since just a few days, I also found out the performance ratio on SETI comparing CPU and iGPUs is unexpectedly close.

So in this case, you're right as the iGPU doesn't contribute so much to the overall score when bundled with fast CPU cores.

My Bay Trail system the iGPU I think adds to overall output. I am still early into test, but so far the data looks like this.

4 CPU AP + 1 iGPU MB
~23.3hr CPU time + ~5.5hr iGPU time

4 CPU AP + 0 iGPU
~21.5hr CPU time

Using 650 average AP credit & 90 average MB credit
4 CPU AP + 1 iGPU MB
~3070 estimated RAC
4 CPU AP + 0 iGPU
~2900 estimated RAC

When running SETI@home for both CPU & iGPU it looks like overall output is a gain. Most notably CPU is not slowed to nearly double run time when using iGPU. Some CPU slowdown must be expected as iGPU does share resources with CPU. So far CPU is slowed only 8% it would seem.

---

SETI@home classic workunits: 93,865 CPU time: 863,447 hours
Join the [url=http://tinyurl.com/8y46zvu]BP6/VP6 User Group[

HAL, I'm a bit confused here.

From your first post:

Configuration: 4 CPU MB + 1 iGPU MB
CPU times ~2 hr, iGPU times ~1 hr

Configuration: 4 CPU MB, No iGPU running
CPU times ~2 hr

And from one of your last posts:

4 CPU MB + 1 iGPU MB
~2hr for CPU & ~1hr for iGPU = 48 CPU tasks & 24 iGPU tasks a day.
Total 72 tasks a day

4 CPU MB + 0 iGPU
~1hr for CPU = 96 CPU tasks day.
Total 96 tasks a day

So does the iGPU really slow down the CPU that much or not?


*edit* Nevermind, I already found the answer:

Further testing on my Haswell systems has shown that I was mistaken on the MB CPU times when not running the iGPU. It could be that the MB CPU tasks were also VLARs and I just didn't notice at the time.

So I will revise my data from before.
Configuration: 4 CPU MB + 1 iGPU MB
CPU times ~2 hr, iGPU times ~1 hr

Configuration: 4 CPU MB, No iGPU running
CPU times ~1 hr

Wow, so the iGPU is really slowing down the CPU to 50% ?? That's interresting!

---

Wow, so the iGPU is really slowing down the CPU to 50% ?? That's interesting!

For my Haswell, i5-4670K, systems.
Yes, For SETI@home project running any combination of MB or AP tasks with CPU & iGPU has the effect of about double CPU run times. I had hoped it would be something simple like "you are doing this wrong". However, I think what Joe mentioned about the cache is accurate.
Probably good to mix non cache heavy project on either CPU or iGPU. Then SETI@home on other processing device.

For my Silvermont/Bay Trail-D system.
CPU tasks are slowed about 8% when running iGPU. At least for AP. I must still do more test to see what the % is for MB tasks, but it may be the same.
It appears that iGPU produces enough output to more than make up for slower CPU times. Much like when running Hyper-Threading on a CPU. The processing times are slowed 15-30%, but output increases. Because of greater number of tasks completed.

---

SETI@home classic workunits: 93,865 CPU time: 863,447 hours
Join the [url=http://tinyurl.com/8y46zvu]BP6/VP6 User Group[