Since the last Maintenance on the 30th of October all my computers (9 of them) have suffered a steep fall on host RAC's... Likewise the total credits per machine per day have also dropped significantly... Is there any other user(s) who have also suffered such a decrease? If so was this maintenance responsible for this? Any input will be most welcome...

---

I think it is generalised.
New BLC01 tasks pay (even) less than previous ones, which were some remnants of the old format.
In another thread this has been explained by the fact that new task format increase efficiency, which in turn causes decreased credit handout for the sillyness of CreditNew (long history here which I skip).

Same here. Iam down 10,000 on my RAC. it's coming up but very slow.

---

I can't say that my RAC has fallen much more than normal. But, it may be that since I swapped out my twin 750Ti for a single 1070 FE mine has climbed considerably since 17 October.

---

I don't buy computers, I build them!!

Thank you... At least I know I am not alone...

---

Also mine is not actually nosediving as in other occasions, it looks more like a gentle correction.
But if numbers are numbers. from the credit I am given for the new tasks, I think there is no other way than going down a bit.

As of this moment, mine are still falling down... Happy for you that yours are climbing up...

---

Glad for you...

---

Sorry for my next response's delay... Yesterday there was an unexpected power outage here which lasted around 8 hours, at the end of which I unfortunately lost one of my desktops... :( Today I checked all my other machines and I found little or no blc01 tasks, except those that were most recently uploaded and waiting in queue... Most of the tasks were 15,16,19,20,21,22,23,24,28,29oc's , and blc22 & 23's. Now if my sharp drop was due to these other task types, I shiver thinking about what will become of the RAC when blc01's will start running... I think as a voluntary tester you must have more access of the operations and developments inside the Seti@Home system... Can I ask you a favor of looking deeper into my problem, if it is not a big deal...? I really am extremely frustrated... I have been a seti member now slightly over 4 years, and I have even purchased laptops and desktops and put them into sole seti use (24/7)... I WILL APPRECIATE DEEPLY ANY LIGHT THAT SHINES OVER MY PROBLEMS... Thanks in advance if there is any new info, and also thanks if there isn't any... I'm hoping not to be a disturbance...

---

Volunteer tester just means the person concerned has at sometime joined SETI Beta, which anyone can do.

http://setiweb.ssl.berkeley.edu/beta/

So, in general, the RAC #'s for everyone is going down, even if the absolute # of tasks you are processing is holding steady or even increasing?

Or your speed of processing is increasing?

Tom

---

A proud member of the OFA (Old Farts Association).

As I haven't changed the anything recently on my four crunchers I can say that credit per hour has generally increased a bit over the last couple of weeks.

---

Bob Smith
Member of Seti PIPPS (Pluto is a Planet Protest Society)
Somewhere in the (un)known Universe?

As I haven't changed the anything recently on my four crunchers I can say that credit per hour has generally increased a bit over the last couple of weeks.

+1
RAC increased as we went from blc22 to blc01 and 04.

---

Humans may rule the world...but bacteria run it...

On Nov 1st my RAC graph started a nosedive worse than bitcoin, that is still in free fall. I was stable between 5,500 and 5,700 then BAM! the world dropped out. I'm at 3,700 and still falling at the same rate.
One machine started having CPU overheat and CTD issues and the other starts showing computing times that triple or quadruple the previous averages. I've uninstalled and reinstalled the BOINC on both machines and not seen any difference. This morning I uninstalled and reinstalled again but this time in a service mode with no GPU and I'm already seeing a marked performance improvement. So far no CTD and overheat issues.

Running SETI in service mode is not a good idea, and re-installing BOINC and rarely solves problems and can hinder in their resolution.
Sort out the overheating problem and you will stand a chance of getting back to near where you were - you have a shed load of error tasks (all "aborts" earlier today), bu that did clear out tasks from the internal GPU.

---

Bob Smith
Member of Seti PIPPS (Pluto is a Planet Protest Society)
Somewhere in the (un)known Universe?

Bob,

I think I had a couple things going on at the same time. I just happened to update both machines on the 1st from the previous version, I was actually a couple versions old for the install.
1) I traced the OH issue to the latest update and a mis-config that was self-induced. I had typically use the service install. I don't run a video card in most of my machines and did not want to task the Intel GPU (an oxymoron, that) with processing data at a rate and endurance it was not really designed to do. My laptop has a 550 card embedded but it tends to run hot so I don't task it unless I need it for something specific. Anyhow, I had inadvertently installed the update without spec'ing the service mode and both machines were using the GPU's
2) The re-base of the RAC's impacted my averages.

73,
Dave