New SMP clients 6.29

[orion456]

Windows client 6.29 (drop in binary near the bottom)

http://folding.stanford.edu/English/DownloadWinOther


Linux client 6.29

http://folding.stanford.edu/English/Download

I had no problem stopping the client and then dropping in the windows binary or extracting the Linux client and saving it over the current client. Restarting simply took off from where I was before.

Remember you have to sign up for passkey to get bonus points when they become available.

 

[Norcalsteve]

So, the question still is... if you are on i7 with HT'ing: "8 Cores," and folding on VM linux client with Bigadv flag, there is no need to upgrade to 6.29? or is this the upgrade for ALL of us "power users?"

 

[orion456]

So, the question still is... if you are on i7 with HT'ing: "8 Cores," and folding on VM linux client with Bigadv flag, there is no need to upgrade to 6.29? or is this the upgrade for ALL of us "power users?"

You definitely need the new client. In your case, the linux version. They said
the -bigadv boys will be the first to see the new A3 core.

 

[deadlysyn]

See here:

http://www.overclockers.com/forums/showthread.php?t=632417

 

[AmbientFiction]

All my clients are on -oneunit runs right now... Gonna back up the drop-in I've been running in case I don't like my new numbers. My main rig should be fine seeing that its running an AM2 but my others are 939 rigs so might need to go back to the one I've been running.
AM2 turns over in 17hrs
939 rigs 2 - 3 days

 

[ChasR]

Dual cores aren't going to like the new client.

 

[AmbientFiction]

Dual cores aren't going to like the new client.

YAY!!!! Great so I guess I'm hosed!

 

[ChasR]

Afraid so. I'm afraid this points scheme is going to obsolete the Q6600 as well. Bonuses for i7 class machines are going to be so high that all the work thats been done before now will be devalued.

 

[orion456]

Afraid so. I'm afraid this points scheme is going to obsolete the Q6600 as well. Bonuses for i7 class machines are going to be so high that all the work thats been done before now will be devalued.

I have all my dual cores folding with two GPUs. Looks like that's going to be the only way duals will get decent points.

 

[Lancelot]

Dual cores aren't going to like the new client.

Why not? I can confirm the new client is more optimized for Windows. I had an A1 WU running taking 17-18 minutes per step. The new client resumed the same WU at 13-14 minutes per step! (on a E8500 dual-core @stock)

 

[Bobnova]

YAY!!!! Great so I guess I'm hosed!

Me too. Rosetta time


Actually i'm going to keep running as long as there is a vague chance of me winning something, but beyond that stanford has made me grumpy at them.

 

[Norcalsteve]

I had no problem stopping the client and then dropping in the windows binary or extracting the Linux client and saving it over the current client. Restarting simply took off from where I was before.

confirmed... however, i still recommend, ctrl+c, -oneunit, restart and let finish before you drop in. to each his own

 

[Bobnova]

I just dropped it in as well.
Worked fine, and got an A3 unit as soon as the A2 finished.

 

[Norcalsteve]

I still keep the -bigadv flag with the 6.29?

 

[PeddlerOfFlesh]

Why not? I can confirm the new client is more optimized for Windows. I had an A1 WU running taking 17-18 minutes per step. The new client resumed the same WU at 13-14 minutes per step! (on a E8500 dual-core @stock)

I just did the same thing on my Q8200 and it dropped 60 ppd.

 

[ChasR]

Changing from 6.24 to 6.29 will not make any difference in the speed of an A1 WU. If it changed, something else made the difference.

 

[Shelnutt2]

Of course this happens as I get an dumb a1 wu.

 

[orion456]

Of course this happens as I get an dumb a1 wu.

I only have 3 w3's and the rest are w1 and w2's. Those 3's are assigned very low points. It's a bit weird having to wait to see what bonus points are assigned.

 

[ChasR]

HFM tell you what the credit will be.