Not a problem with the card. decided to do one last test & it worked. Pulled drive out of caddy & connected direct to cable, as you can see...



..it worked (& all on C/S - Thanks Ozz). 2x optical drives, a.o.k, 2x HD's in caddies a.o.k. however optical drive with HD in caddy does not work - weird.

Ignore drives H20 (pun intended) as they are for VM's, on the card are drives G & Y.

So will need another card.

---

Query:

Since building this cruncher, it has completed 15 AP's, 14 validated with 1 waiting. I thought that once 10 or more had been completed, the estimates would stabilise?

Have received 6 AP's & all still show an estimated 326 hours to completion (all previous ones were completed in 14/15 hours).

---

Query:

Since building this cruncher, it has completed 15 AP's, 14 validated with 1 waiting. I thought that once 10 or more had been completed, the estimates would stabilise?

Have received 6 AP's & all still show an estimated 326 hours to completion (all previous ones were completed in 14/15 hours).

Your host details only showing 6 APs.
The rest probably had to much blanking.
<10% blanking required.

---

With each crime and every kindness we birth our future.

You've lost me there Mike. I take it that after completing 10 or more with <10% blanking, the estimates will stabilise?

---

Hmmm. AMD FX, GT630? 'This' cruncher will be

Application details for host 7260461

You may have 14 APs validated, but they appear to be split between different applications - some stock, some optimised.

The rule is that you have to have 11 tasks ('more than 10') "completed" for any one application, before the estimates for that application type are normalised.

And validation is a necessary, but not sufficient, condition for completion. In order to have a representative runtime, and hence be useful for correcting the estimates, the tasks have not to exit early (30 pulses, of either or both types), and not to suffer from a significant proportion of radar blanking.

Thanks Richard. I was sure I ran Lunatics as soon as I repaired the Windows hiccup, looks like I did not.

---

11 wu's completed & estimates for the 5 new wu's are 14h 06m which is about right.

---

11 wu's completed & estimates for the 5 new wu's are 14h 06m which is about right.

Damn, my old Q6600 will do an AP in about 10hrs and both my 2500K & 3750K will do them in an average 6.25hrs.

Cheers.

Ah but were they crunchers only or workhorses?

---

They're all workhorses.

Cheers.

Ah but I'm a dyed-in-the wool AMD fanboy :-)

---

Ah but I'm a dyed-in-the wool AMD fanboy :-)

8+ years ago nearly all my rigs were AMD too, but they have been going backwards since which is why they're now all Intel (even my last AMD Athlon II X630 would do an AP in 10hrs, it was on par with the Q6600 at that time).

Until AMD can get the CPU horsepower up (by a lot) and the power usage down (also by a lot) it looks like I'll be sticking with Intel for the foreseeable future. ;-)

Cheers.

Main reason for AMD, especially in today's economic climate...

Intel

AMD

---

Main reason for AMD, especially in today's economic climate...

Intel

AMD

Worth looking at the cost to run v work done.
AMD might be cheaper to buy, but Intel CPUs use a lot less power, and do a lot more work.

---

Grant
Darwin NT

True, but I have good reasons to use AMD...

1: Customer base (ask me no questions & I'll tell you no lies).
2: The links I provided are for consumers, I have a business account with that site & get slightly better deals :-)

---

Main reason for AMD, especially in today's economic climate...

Intel

AMD

Worth looking at the cost to run v work done.
AMD might be cheaper to buy, but Intel CPUs use a lot less power, and do a lot more work.

Being the "efficiency fanboy" that I am :-D, I look at it this way.

Sirius' FX-8320 is a 125W job and does about 13 AP's per day (if those times are based on using all 8 cores) and at the time I bought my 3570K was $60 cheaper at the time.

My 3570K on the other hand is a 77W job and does about 15 AP's per day (using all its 4 cores as it does).

So if I had gone with the cheaper option back then I'd been well behind on work done and that initial $60 saving would by now have been used up by the power bill and that's without finding out if the FX-8320 setup would be as responsive to use as the 3570K is under full load while feeding 2 GPU's 3 MB's each. ;-)

Cheers.

True,

I look at it this way

AMD are cheaper, run cooler but process work units slower than Intels on the CPU and require more percentage of a cpu for each GPU (meaning fewer work units per card)

Intels cost more, run hotter (needing better cooling units or liquid coolers), processor work units faster on the CPU and uses less percentage of a cpu per each GPU

So there are trade off to each type. In the end, the Intel will process more per day than the AMD but are going to run hotter requiring more cooling. If I had the money, then yes, all of my computers would have had Intel chips and Motherboards.


My 2 cents..

Zalster

AMD are cheaper, run cooler

Intels cost more, run hotter

You appear to be somewhat confused- AMD CPUs use much, much more power than Intel CPUs. AMD are the "hotter" running CPUs and have been ever since the introduction of the Core series.
Intel CPUs were well known for running hot, back in the days of the P4. That was a while ago now.

---

Grant
Darwin NT

AMD are cheaper, run cooler

Intels cost more, run hotter

You appear to be somewhat confused- AMD CPUs use much, much more power than Intel CPUs. AMD are the "hotter" running CPUs and have been ever since the introduction of the Core series.
Intel CPUs were well known for running hot, back in the days of the P4. That was a while ago now.

Not true.

AMDs using more power but are much cooler.

My 8350 is running at 48°C with air cooling.
Show me one Intel runnig this temps.

---

With each crime and every kindness we birth our future.

AMDs using more power but are much cooler.

That would be contrary to the laws of physics, thermodynamics in particular.

My 8350 is running at 48°C with air cooling.
Show me one Intel runnig this temps.

How? I don't own such a system.

But it's pretty straight forward that for a given sized area (in this case the CPU die), something that uses 125W of power will produce more heat than something that uses 84W.

---

Grant
Darwin NT