• Some users have recently had their accounts hijacked. It seems that the now defunct EVGA forums might have compromised your password there and seems many are using the same PW here. We would suggest you UPDATE YOUR PASSWORD and TURN ON 2FA for your account here to further secure it. None of the compromised accounts had 2FA turned on.
    Once you have enabled 2FA, your account will be updated soon to show a badge, letting other members know that you use 2FA to protect your account. This should be beneficial for everyone that uses FSFT.

SMP + GPU2 + Console super Q6600 build

S

smileyscout

Gawd
Joined
Mar 29, 2005
Messages
631
Xilikon said:
At what speed your Q6600 is running ?

If the speed is high enough (like 3.0-3.2 GHz), you can assign the SMP client to core 0 and 1 only, assign the GPU2 client to core 2 then install a console client and assign it to core 3 only. You would be looking at about 4500-4700 ppd if thing is configured right.

Assigning cores is essential because if a client is hopping around the cores, it lose efficiency. Locking to a single core (or 2 for the SMP) will minimize this and it will run at peak efficiency.

Click to expand...

Xilikon has inspired me to make my build for a dedicated folding machine more [H]ard.

https://secure.newegg.com/NewVersion/WishList/MySavedWishDetail.asp?ID=6218585

  • MSI P35 Neo2-FR
  • SAPPHIRE Radeon HD 3870 512MB
  • Intel Core 2 Quad Q6600
  • G.SKILL 2GB (2 x 1GB) DDR2 800
  • XIGMATEK HDT-S1283 120mm Rifle CPU Cooler

for $616

I'm also thinking about ditching the retail Q6600 and getting the G0 SLACR Q6600 OEM from Clubit for $223.99

My power supply will be the old one from my main rig AeroCool ZERODBA 620 Watt

I'll use my OEM copy of Windows XP Home since I'm a Linux noob. This program SMP Seesaw Pro which Tigerbiten recommended should balance it all out.

 
I hate to say this but if its a pure folding box then I would not run the GPU2 client on it.
Ditch the vid card and just run 2x SMP clients on it.

If you are going to run the GPU2 client then you will need to run the SMP client inside a VM so it stays locked to 2 CPU cores all the time.

But one thing I was thinking about a GPU + SMP setup.
The SMP client set to run at 75-80%inside a VM, with 2x VM's locked to the same three CPU cores.
Then the GPU2 client locked to the remaining CPU core.
Would you over-saturate the memory bus ?
If not would it pull in +1,000 PpD over 2x plain VM's ?
Would the VM's still make the preferred deadlines if you got a chewy protien ?

Luck ............... :D
 
Tigerbiten said:
But one thing I was thinking about a GPU + SMP setup.
The SMP client set to run at 75-80%inside a VM, with 2x VM's locked to the same three CPU cores.
Then the GPU2 client locked to the remaining CPU core.
Would you over-saturate the memory bus ?
If not would it pull in +1,000 PpD over 2x plain VM's ?
Would the VM's still make the preferred deadlines if you got a chewy protien ?

Luck ............... :D
Click to expand...

I tried that and SMP would often (depending on WU) miss preferd deadline:(

GPU2 client is IMO almost equal to 1 SMP in VM (3870 @ stock is ~1800 PPD and E6600 3.15 is ~2000 PPD). If someone want to buy GPU folding rig I would wait for HD4xxx (should be available in June) - it was already confirmed that they will work with GPU2 client.


 
Rebel44 said:
GPU2 client is IMO almost equal to 1 SMP in VM (3870 @ stock is ~1800 PPD and E6600 3.15 is ~2000 PPD). If someone want to buy GPU folding rig I would wait for HD4xxx (should be available in June) - it was already confirmed that they will work with GPU2 client.
Click to expand...

GPU2 is equal or better depending on the WU you get now. The new 47xx WU get me just over 2500 PpD. Combine that with 1 SMP client using 3 cores (2700 PpD) on my Quad and its pulling down over 5200 PpD. The SMP client when it gets a chewy protein goes down to about 2400. I'd like to see any quad running 2 SMPs pull anywhere near 4900PpD when they are running any of those 306x proteins.
 
Also there is a huge difference in Points per Watt. Your best combo right now is a quad with dual SMPs you will get anywhere from 3600-4500 for a well overclocked quad.



 
Sunin said:
Also there is a huge difference in Points per Watt. Your best combo right now is a quad with dual SMPs you will get anywhere from 3600-4500 for a well overclocked quad.



Click to expand...

Up to 5300 ppd in my case (Q9450 at 3.52GHz, 2 SMP VM + 1 console).

 
I have a smp client, the newest beta, and running it on core 0,1 and only showing 25%cpu usage although I told it to use 100%, so whats going on?

and how do I run another instance of the Smp client to use core 2,3?
 
Rebel44 said:
I tried that and SMP would often (depending on WU) miss preferd deadline:(
Click to expand...

I'm going to take a wild guess at this and say you didn't miss any preferred deadlines with a quad at 3.3 GHz. You were probably running the older version of FAHMon that was rounding down preferred deadlines that weren't whole numbers. For example, some of the new work units have a preferred deadline of 1.8 days. FAHMon used to round that down to 1 day.





 
dwilson041781 said:
I have a smp client, the newest beta, and running it on core 0,1 and only showing 25%cpu usage although I told it to use 100%, so whats going on?
Click to expand...

Please take a screenshot of Task Manager > Processes and show us the problem.

and how do I run another instance of the Smp client to use core 2,3?
Click to expand...

Let's get the first instance running right before trying a second one.


 
aldamon said:
Please take a screenshot of Task Manager > Processes and show us the problem.



Let's get the first instance running right before trying a second one.


Click to expand...

yp538.jpg


just to make sure Im setting the affinity correct, Im setting it on FahCore_78.exe and not Folding@home.exe.

and I noticed the Priority is set to low, should that be changed to normal?
 
aldamon said:
Just as I suspected, that's not the SMP client. Get the SMP version.




Click to expand...

palm:forehead!

Thanks man.

Questions about advanced settings. are these Ok, and Optimal?

Use Internet Explorer Settings (no/yes) [no]? yes
Allow receipt of work assignments greater than 5MB in size (no/yes) [no]?yes
Change advanced options (yes/no) [no]? yes
Core Priority (idle/low) [idle]? low
CPU usage requested (5-100) [100]? 100
Disable highly optimized assembly code (no/yes) [no]? no
Pause if battery power is being used (useful for laptops) (no/yes) [no]? no
Interval, in minutes, between checkpoints (3-30) [15]? 30
Memory, in MB, to indicate (2046 available) [2046]? 2046
Request work units without deadlines (no/yes) [no]? no
Set -advmethods flag always, requesting new advanced
scientific cores and/or work units if available (no/yes) [no]? yes
Ignore any deadline information (mainly useful if
system clock frequently has errors) (no/yes) [no]? yes
Machine ID (1-16) [1]? 1
 
dwilson041781 said:
regarding changing that, how do I go about reconfiguring that?
Click to expand...
  1. While SMP is running, back up the entire SMP folder. SMP can be unstable after stopping and restarting it.
  2. Hit Ctrl > C to shut down SMP.
  3. Run the SMP shortcut with the switch -configonly to re-configure SMP.
  4. Run install.bat again and re-enter your password twice as usual.
  5. Delete the -configonly switch and run SMP.



 
You must log in or register to reply here.
Back
Top