Welcome to MilkyWay@home

Posts by Thamir Ghaslan

1) Message boards : Number crunching : Can we have some more work, please? (Message 40237)
Posted 6 Jun 2010 by Thamir Ghaslan
Thats the main reason I've set my cache to 10 days.

Does'nt make a difference with MW, but with collatz and DNET, it does make me survive if all 3 projects go down for a few days!
2) Message boards : Number crunching : 5970 + 4870x2 = downclocked 4870x2 (Message 39923)
Posted 25 May 2010 by Thamir Ghaslan
Sounds like the cards are "stuck" in low power mode. Is BOINC crunching happily on all GPU cores - ie does BOINC see 4 GPUs?

I'd also close CCC and use MSI Afterburner.

I have a 5970 and a single 4870 crunching happily away and only see them downclock when they run out of work.

Thanks for the neat after burner utility. Great graphs! :)

I've discovered the problem from CCC, right click, 4870, 3d setting, force maximum performance clocks. it now locked the 4870 to 750 mhz.

3) Message boards : Number crunching : 5970 + 4870x2 = downclocked 4870x2 (Message 39915)
Posted 24 May 2010 by Thamir Ghaslan
Any ideas or similar experience for having hybrid cards?

At catalyst control center, the 5970 is fully throttled in regards to core or mem MHZ. slider 750 mhz core, 500 mhz mem, correctly reflected under current values.

The 4870x2 however, has been locked down to 500 MHZ core, and no amount of overdrive tweaking can increase that value!
4) Message boards : Number crunching : Bittersweet Milestone (Message 39782)
Posted 17 May 2010 by Thamir Ghaslan

Still.. even 'sadder'... all that time lost.. all the energy spent over 17 years... completely blown away in less than a week from one, single, solitary GPU.


Question is, where do we go from here? Whats the next step in the future?

Quantum computing in your home after all the algorithms and production costs gets ironed out?

Biological computing, having your own brain in a box and feed it nutrients instead of electricity? :P

No doubt GPU is king for cheap, green, fast computing, wonder whats next!
5) Message boards : Number crunching : Errors on new 5970 (Message 39732)
Posted 15 May 2010 by Thamir Ghaslan

My mobo is an ASUS P5Q3 Deluxe/WiFi-AP@n motherboard. Quote from a review is "Another remarkable thing is three PCI Express x16 slots. However, don’t expect any wonders here: if you use two graphics cards, the first two slots will share 16 PCI Express 2.0 lanes. The third PCI Express x16 slot supports PCI Express x4 logical interface and is connected to the chipset South Bridge.

Boinc benchmarked the 5970 4.6 tera flops peak and 4870x2 1.6 tera flop peak.

~2.3 tflops vs ~800 gflops per gpu, ~2.9 multiple.

Still suspect your mobo. :/
6) Message boards : Number crunching : Errors on new 5970 (Message 39727)
Posted 15 May 2010 by Thamir Ghaslan
I was under the impression that a 5970 would be doulbe the speed of a 4870x2, turns out that one GPU in a 5970 is roughly 3 times as fast as a 4870 GPU, so overall that should give out a rough 6x increase.

Based on my results I'm seeing each OC'd GPU on the 5970 doing just over twice the throughput of my 4870. See here.


Doing almost 100 second for 5800s, 300 seconds for 4800s. Same ratio hold in other projects!


I'm suspecious of your PCI slot setups, maybe it would help if your device 0 and 1 were the 5800s?

Running a gigabyte motherboard, GA-EP45-DQ6, not sure if pci-e slot speed plays a role here, if yours are both full 16x or not?

Or it might be somethign else all together? :/

the 4800 and 5800 cards are both saphire, no OC.
7) Message boards : Number crunching : Errors on new 5970 (Message 39718)
Posted 14 May 2010 by Thamir Ghaslan
Power usage with 5970 @ 876/1000MHz & 4870 @ 801/900MHz on MW with 1 WU per GPU and Q9450 @ 2.66GHz running Aqua = 608W to 625W.

Idle = 245W with the 5970 fan at 40%.

At full load before the card has warmed up, so the fan is still at 40%, the power draw was 580W.

As the card temps leveled out at 67°C & 76°C, my fan profile has the fan running at 56% and the power draw is back to 608 to 625W. Glad my PSU is 850W.

Wow, the power bill is going to be BIG, BIG!

I have a 4870x2 and a 4 day old 5970 on the same board. antec 850 watts, on q6600.

Never had any issues on any projects, and temps are below 85c for all cards and card noise are ok, card fan speed below 40% thanks to a well ventilated thermal take beast of a case, front, top, bottom, rear fans for good circulation. :)

Can go quiter and way cooler with lower card mem speed.

I was under the impression that a 5970 would be doulbe the speed of a 4870x2, turns out that one GPU in a 5970 is roughly 3 times as fast as a 4870 GPU, so overall that should give out a rough 6x increase.
8) Message boards : Number crunching : Sudden mass of WU's finishing with Computation Error (Message 34297)
Posted 6 Dec 2009 by Thamir Ghaslan

With that in mind, I experimented a bit yesterday. I dropped the boxes down to running 1 WU at a time (n1)and set w1.01 (w1.15 on the old P4). Hey presto, they have all been running happy and smooth.

Yes, seems running multiple tasks will dry up the GPU ram!

300 MB per task on the longer WU.
9) Message boards : Number crunching : Sudden mass of WU's finishing with Computation Error (Message 34296)
Posted 6 Dec 2009 by Thamir Ghaslan
Ok, if these tasks requires roughly 300 MB then I suggest to run 1 task at a time depending on GPU ram rather than running multiple tasks.

I think most used to run 2 to 3 tasks at a time in the past?

My system is set to run one tasks at a time, I have 1 GB shared on the 4870x2, so thats roughly 600 MB for two tasks, one task per GPU.

10) Message boards : Number crunching : Sudden mass of WU's finishing with Computation Error (Message 34267)
Posted 5 Dec 2009 by Thamir Ghaslan
Bad news, it's not about insufficient system memory. I ran four long 4x MW WUs with 8GB of system RAM, only one GPU installed and all unnecessary services and processes disabled on my Win2003 Server Enterprise Edition with PAE enabled and it still finished with "fitness: 1.#QNAN000000000000000". I ran them with network disabled so don't be surprised that they have not been reported, yet.

I have a screen shot of the memory usage here.

My next idea was to try an optimized app but there is no opt app for CUDA. Why? Is the stock CUDA app so well optimized that it's not necessary?

From what I've been following it seems the majority of Nvidia users are erroring out.

Again, the problems cooincided with the increased WU size. ATIs seem to handle it while Nvidias don't.

I have'nt seen any feed back from any of the ATI or CUDA (inhouse?) coders so their insight would be highly valuable.
11) Message boards : Number crunching : Sudden mass of WU's finishing with Computation Error (Message 34214)
Posted 4 Dec 2009 by Thamir Ghaslan
I doubt this is a hardware issue.

And I have'nt heard from Gipsel in a while. </hint> :P

Notice how these errors only crept up with the 4x increase in WU size.

Time for another GPU software update to correctly handle the 4x increase?
12) Message boards : Number crunching : Sudden mass of WU's finishing with Computation Error (Message 34155)
Posted 3 Dec 2009 by Thamir Ghaslan
MW is still my backup project, collatz is my primary.

I have to say I'm pleased with the 4x increase in wokrunit size, and no problems what so ever on a 4870x2.

13) Message boards : Number crunching : ATI Radeon HD5970 performance on boinc (Message 33481)
Posted 21 Nov 2009 by Thamir Ghaslan
I'm planning on getting a bigger case but I'm not sure of the power requirements if I mix my current 4870x2 with a 5970. the PSU is an ANTEC 850 watts.

If you believe this calcuator:


Then I'd need 860 watts to run 4870x2 and a 5970!
14) Message boards : Number crunching : credit table 2.0 (Message 32828)
Posted 27 Oct 2009 by Thamir Ghaslan
i'm a forum observer, an ex project participant, as i've found collatz much more inviting, from active project admins and never ending work feeds.

road runner, the fastest computer on earth, is primarily made up of cell, the stuff that runs play station 3, and as far as i'm concerned, its a GPU!


i know my contribution is pointless to the admins here, but just food for thought!

15) Message boards : Number crunching : How many choose to stick with CPU over GPU? (Message 32014)
Posted 6 Oct 2009 by Thamir Ghaslan

Any stats on mortality rates between vid cards and cpu's? Pure speculation here but I would think vid cards would fail more frequently since gpu's run warmer while cpu heatsinks tend to be beefier. Or maybe gpu's are simply designed to run hot for sustained periods without adverse effects?

Don't have exact stats, but from personal experience I've burned a gtx 280 after 5 months of contious use last year. Its been 10 months since I've replaced it with a 4870x2 that has also been running 24x7 without failure.

Not trying to start a fanboy war, and don't consider myself one, but I have yet to come across one ATI owner who burned his card, most of the numerous burns I've seen in other forums are Nvidia related unless some one can correct me.
16) Message boards : Number crunching : Problem with tiny cache in MW (Message 31814)
Posted 1 Oct 2009 by Thamir Ghaslan

Another question, do the admins participate in the forum or are we just talking to ourselves?

Talking to ourselves.

17) Message boards : Number crunching : Asymmetric GPU Installations (Message 31799)
Posted 1 Oct 2009 by Thamir Ghaslan
I have a 4850 and a 4870 in the same box and will be staying that way for a while unless I win the lottery....

But I will never purchase cards in batch lots so I will always have an asymmetric installation as I upgrade one card etc.....

Then best complain so I am not the only voice ... because UCB is going to take silence as consent ... and will never consider reality because I don't know of anyone that can always afford to upgrade in full sets.

Even if they can, like on my next upgrade cycle I have two GTX260s and if they are both running and I get a MB with three PCI-e slots ... will I be able to find GTX260s? By then the fuller line of GTX300 cards may be filling the niches and I am supposed to junk two perfectly good GPUs?

Anyway, silence is consent ...

Add my vote, I usually sell a card and upgrade, and if I can't sell I'll upgrade anyway and keep the second card to work with the new asymmetrically.
18) Message boards : Number crunching : HD5870 (Message 31798)
Posted 1 Oct 2009 by Thamir Ghaslan
@ Ironwork16

Actually there is no driver for the 58x0 and WinXp from the ATI/AMD website. But with my Asus 5870 i got a cd with a driver for WinXp. This driver works but i believe it is less stable than the driver for Vista/Win7. Crunching unter Win7 is fun with the 5870...

Early game benchmarks and even one guy caught crunching in collatz seem dissapointing!

One 5870 is equal to or slightly faster than a 2 4870 or a 4870x2.

If we can expect a 10% performance improvement with every month driver release which is the historical case for both card vendors, then I'd expect card maturity within 10 months.
19) Message boards : Number crunching : CUDA Application Updated (Message 31751)
Posted 30 Sep 2009 by Thamir Ghaslan
That's why other projects are "whining" about the overpay at MW. For most of them it's just impossible to utilize the hardware to this extent. It's not just about being lazy programmers - the algorithm and the problem itself don't allow it. They could never achieve the same Flop/s even if they made an ATI app.


That is very true.
The MW algorithm is really perfect for a GPU. Take vast amounts of parallelism (millions of threads), no branching (except you want to call a loop with a counter checked each iteration a branch), a very compute intense algorithm with only a few memory accesses, minimal communication between the threads (the values are just added in the end), and what you get is virtually the peak performance of a given GPU for the instruction mix of the algorithm. It's not all about multiply-adds, so you won't get exactly peak performance. But the v0.20 has cut all the overhead down to a minimum so you really arrive within 10% of what is theoretically possible with the algorithm's instruction mix. That's better than any current CPU achieves, even relative to its peak performance.

And that will continue to scale, the new ATI HD5870 should easily double the performance of a HD4890 at Milkyway. And when the next nvidia generation arrives, I'm quite sure it will do much more than to double the DP performance of a GTX285 ;)

Far be it that I'd like to start a flame war.

Maybe this analogy is appropriate, maybe not. But lets suppose I'm running Crysis and get 10 FPS (Credit or WU), and I'm running another game and get 100 FPS. For the sake of argument lets suppose in both scenarios the CPU is 0% and the GPU is at 100%.

My point? Not sure exactly, just making an analogy.

I can imagine a boinc future where GPU users will cry foul for quantum computer owners running quantuamilly optimzed apps. :P

20) Message boards : Number crunching : CUDA Application Updated (Message 31707)
Posted 29 Sep 2009 by Thamir Ghaslan
You're right, I was talking about a 4870 and it's maximum dp performance is indeed 240 GFlops at 750 MHz. Mine runs at 800 MHz (256 GFlops) and achieves ~190 GFlops at MW. That's a really really good optimization done by CP, so even achieving something close to these numbers is challenging.

Which makes me disappointed when other projects whine overpay in milky way!

I know its already done, credit lowering and all, optimization will overcome or surpass lowering and all, but Boinc and its projects are decentralized, so MW admin should not bow under pressure from other projects!

CUDA is not equal to Brooks. ATI is not equal to Nnvidia. Intel is not equal to AMD.

Next 20

©2023 Astroinformatics Group