_mat_

In the last few weeks I've been working on a benchmark to calculate pi on graphics cards. It's an idea that's in my head for some years now, but I couldn't put it to life until now. Somehow the death of our beloved Turrican inspired me - I wanted to do something good, something that matters. So I tried hard and it finally worked out. I want to dedicate this benchmark to Turrican, like a virtual monument to never forget his legend. A benchmark in his honour. Together with the community of overclockers.at I've tested the benchmark for the last week and put out some new versions, wrote a technical description and added some frequently asked questions. It's still a beta version, but it should be pretty stable with most of the graphics cards and processors around. But be sure to install the latest GPU/OpenCL drivers for the target device, old ones can be very bugged. Here are a few examples for calculating the billionth digit of pi: GPUPI 1B: AMD Radeon R9 290, NVIDIA GeForce GTX 980 and Intel Core i7-4960X@4 GHz I think that there's a lot of potential in a benchmark like this, it's just fun to crunch pi on a gpu. If you like it, I would consider implementing multi gpu support and hwbot direct submission. Bugs and feature suggestions can be made here or in the international support thread at our forums. Download, technical details & FAQ: GPUPI

Great score. Congrats!

Before any doubts of hardware sharing come up, we've binned 11 modules of Hynix MFR from ADATA and Corsair. Our best module died from high voltage but 2nd and 3rd stick could do 1930 mhz (splmann's result) and 1901 mhz (my result). Pictures and a back story (in German) can be found on our site: http://www.overclockers.at/articles/the-show-must-go-on/page_3

It's a relative link (/submission/...), but it should be an absolute one (http://hwbot.org/submission). Just click the link, go to the address bar an remove the "forum." part.

We have binned a 3770K that easily does 6 GHz benchstable and maybe a bit more with 1.4 volts under cold. But guess what ... it does not scale with voltage, so 6.2 GHz is about the maximum the cpu can do with 4 cores. And that was one piece out of 55 cpus we tested.

We used the ASUS Engineering BIOS, so the voltage was locked at 1,21 volts. We had no throttling and slowdowns after that. Our Titan was also pretty easy to work with, no shutdowns until -140 °C (measured on the pot but near gpu). But cold did not help that much, without vmods the card is limited as hell. But I guess it could get some pretty decent scores, if it's modded. Congrats to everybody who made the cut. Have a nice time in Moscow, you earned it!

We poured ln2 on the card, but that was already risky. No volt mods allowed though, so we couldn't do too much damage either way.

Everyone is doing zombie mods now? Overclocking has certainly evolved! oanvoanc and me just started with our first results. We have a good cpu, but only a borrowed Titan - and we're not allowed to hurt it (whatever that means! ). Well, we're doing our best anyways, because that's what it's all about, right? Here are some pictures from our first days: We are trying to get the Austrians involved in our progress, so you can find more pictures in our (german) article: http://www.overclockers.at/articles/asus-open-overclocking-cup-2013-wir-sind-dabei

That's the way it is. And doing it live in front of an audience doesn't help either! So congratulations on your huge score, Mike!

Great work, keep pushing!

GPU on LN2 in PCMark05? Great work!

You sure know how to have fun.

Very nice result for 5.8 ghz! Keep pushing it, mate!

Nice, 1425 mhz on gpu with water cooling is huuuge! Did you overvolt the card by firmware?

I have some ATTO scores with more than 4 GB/s read and write with 8 SSDs. I will post them if you like. But depends very much on stripe size and virtual drive properties.

No, my friend. WE nailed it! :celebration: Regarding the storage pictures, I couldn't find ones now but we have a few from our preparations: It shows everything we used, but I guess it's not easy to see: => Four SSDs in Raid-0 connected to the raid controller (in front of the radiator, red SAS to SATA cable) => The SSD2go in every possible color. => Two SSDs in Raid-0 connected to the Intel Z77 raid controller (behind the radiator) => One system disk in the light blue drive bay Edit: Photos with a better overview of the system can be found here and here.

Reserved. Will post further information, some images of the event and a little how-to on PCMark05 here soon!

PCIe slot 1 and 2 are 16x on this board. This is the 8x bug that seems to happen without any obvious reasons on some M5E boards. Andre, did you have a look into this? Sometimes it can be fixed with forcing PCIE1-16x in BIOS, sometimes you need to set GEN2 for the slot or try the GEN3-Presets. A strange one, this is! Congrats anyways, great work!

We wanted to use a raid controller, but performance was not scaling well and we didn't have enough time to fix it before the event. So we went for 1x Angelbird Crest 6 Master as system disk and 2x Angelbird SSD in Raid-0 on Intel Z77 controller. Write back cache was enabled! For the XP Bootup we used a USB 3.0 device called Angelbird SSD2go, without turbo mode or UASP. We will show more images and a video very soon!

More infos about the event can be found at overclockers.at. This is just the first result, we will host another livestream next Friday where we will try to push the score higher! After this we will release all tweaks that we have used for our scores to the public.

Interesting. I was asking because I am currently testing a LSI MegaRAID 9271 with 8 ssds and I have the worst performance. Like 200 mb/s General Usage. So I thought that this has to be a reason for people using the software. Could not gain much with different controller or vd settings. Any hints on this? Very much appreciated!

Thanks! It's not stated in the rules, so I am going to ask: Is LSI's CacheCade and Fastpath software allowed? I guess it is, because it's not very different to Intel RST and AMD RAIDXpert, but I just wanted to be sure!

Very nice score, congrats! You didn't specify your disk configuration. Virus Scan is very very very high!

2,3 volts on this little FX-8120 (video inside) ... we thought it died, but actually just the mainboard did. So what I actually want to say: Christian, keep pushing it!

Maybe they scale only with cold?