tiborrr

I have used the wire from the 80-pin IDE cable (ATA66+), don't know what AWG gauge is that but it seems to be working without a hitch. Here you can see it connected to Vddr inductor: Thanks Roman for the mod! :banana:

It works on retail 5820K, tried both mods. For 2nd mod I did it without external power device. Just connect it to VDDR MOSFET or coil and your VDDR = you "uncore signal voltage". MSI X99S XPOWER AC Before mod: 3700MHz After Easy 4G mod: 4100MHz After 4G+ mod: 4500MHz (1.6Vddr = 1.6V "uncore signal voltage") This CPU is capped at 4500, test on X99-DELUXE, will have to try with my 5960X.

Thanks Stilt! Any performance figures?

Ha, my turn!

Thank you all! Unfortunately this system doesn't have much more in it, but a few more points are certainly possible

I'm just benching for fun and honor of Karl. We (EK) have organized some nice hardware prizes so get your benching spirit on! Top spots await!

Crazy!

Moonman trying to bypass GPU clock limit:

Bam!

Sub 20mins on A64 Venice, this is beyond crazy.

In my opinion, hwbot API would be perfect solution. Actually there's no validation required if you're able to save the result only when the checksum is correct. Thanks again, Matthias. You are making Karl proud!

@Genie: Excellent! This is what I wanted to hear!

Fixed in less than 20 seconds. @mat I may have a solution for this screenshot issue. The solution would be to save the result in form of a file. Once the calc is done the background info is encrypted and saved into a binary file, which is then uploaded to hwbot. This way the user would have to upload both valid file (encrypted) as well as screenshot. This way the photoshop overclockers would have more issues bypassing the results. This way the benchmark submission routine would consist of: - valid file - screenshot per hwbot rules You could add a "Save" button next to OK button to "Pi Calculation is done" splash window. What do you (Genie, mat, massbo) think?

Pushed Pitcairn below 1 minute for 1B: http://hwbot.org/submission/2696798_tiborrr_gpupi___1b_radeon_hd_7870_%28pitcairn_core%29_59sec_231ms?recalculate=true Also, there's an issue - you can open multiple instances of GPUPI, which is problematic: - you get end result whether the calculation is valid or not. it's easy to open another instance and run short 1M bench, hide it from screen and apply "Pi calculation is done" window to the invalid run. My recommendations would be: - locked main computing window while dialog window (Pi calculation is done) is opened. - only one instance of GPUPI can be run. - no result should be print to screen if there is an error in calculation. - if possible, an invalid run should halt the calculation right after it encounters rounding error or at least after each loop, like in original SuperPI (if possible by the algorithm). - Hwbot should add a valid checksum textbox, which should be mandatory for each submission. Regards, Niko

I fully concur with 1B and 32B. This way we're cool for couple of years (y) For the sake of nostalgia please: - use the original font - remove the cancel button on calculation start notification Also, here's the scaling of my HD7870 (Pitcairn):

Great benchmark, love it! I would vote for 2B and 32B to be accepted as 'retail' benchmarks once benchmark passes through the Hwbot validation: I would also lock down the benchmark window when splash screen "Pi Calculation is Done!" comes up. Just for the sake of nostalgia Here's my workstation machine with FirePro W9000: Wannabe W7000 (flashed 7870, slower DP):

Bravo!

Crazy core, excellent effort all around!

Incredible CPU! Congratulations on your effort!

1463/2000, like it says. This is a BIOS w/o Boost enabled.

zzolio, KPC sadly not working for me Just found this, thanks to Iliya: http://forums.evga.com/m/tm.aspx?m=2219559 Apparently they do all mods on PCB, not on the ON driver PCB. It's solder time!

I got lousy 30MHz in 3DMark03 with 1.35Vreal (measured through DMM). As expected voltage doesn't help shit when you have 125% power limit. Used 200k+100k trimer but it's way too much. You could go with 50k.

Voltmod in progress, testing now: IC: http://www.onsemi.com/pub_link/Collateral/NCP81174-D.PDF

@Tarakan: I'd say it's a tad better clock to clock compared to 2880SP GK100 with same memory speed. Given the fact GTX 980 runs such clockspeeds on water, which would require dry ice or single stage cooling on GK110, I believe the Haswell/GM204 platform is capable of breaking 200k. With 6.7GHz CPU of course