Coding 24/7... Limited forums/PMs time.
-Justice isn't blind, Justice is ashamed.
Many thanks to: Sue Wu, Yiwen Lin, Steven Kuo, Crystal Chen, Vivian Lien, Joe Chan, Sascha Krohn, Joe James, Dan Snyder, Amy Deng, Jack Peterson, Hank Peng, Mafalda Cogliani, Olivia Lee, Marta Piccoli, Mike Clements, Alex Ruedinger, Oliver Baltuch, Korinna Dieck, Steffen Eisentein, Francois Piednoel, Tanja Markovic, Cyril Pelupessy (R.I.P. ), Juan J. Guerrero
Same system.
8-7-6-23 timings
Coding 24/7... Limited forums/PMs time.
-Justice isn't blind, Justice is ashamed.
Many thanks to: Sue Wu, Yiwen Lin, Steven Kuo, Crystal Chen, Vivian Lien, Joe Chan, Sascha Krohn, Joe James, Dan Snyder, Amy Deng, Jack Peterson, Hank Peng, Mafalda Cogliani, Olivia Lee, Marta Piccoli, Mike Clements, Alex Ruedinger, Oliver Baltuch, Korinna Dieck, Steffen Eisentein, Francois Piednoel, Tanja Markovic, Cyril Pelupessy (R.I.P. ), Juan J. Guerrero
Very nice board. What do you think of its performance compared with other x58 boards?
I haven't tested it or any other X58 motherboard thoroughly enough to express my opinion yet.
It certainly has some potential.
It took my SuperTalent kit that maxxed out at DDR3-2001 8-7-6-21 in the past ( on various boards ) to DDR3-2102 with less voltage and tighter timings
And I haven't maxxed it out yet.
After today's and tomorrow's testing I shall be ready to start tweaking the system for maximum performance on stock clocks and start the measurements for the stock clocks comparison.
Coding 24/7... Limited forums/PMs time.
-Justice isn't blind, Justice is ashamed.
Many thanks to: Sue Wu, Yiwen Lin, Steven Kuo, Crystal Chen, Vivian Lien, Joe Chan, Sascha Krohn, Joe James, Dan Snyder, Amy Deng, Jack Peterson, Hank Peng, Mafalda Cogliani, Olivia Lee, Marta Piccoli, Mike Clements, Alex Ruedinger, Oliver Baltuch, Korinna Dieck, Steffen Eisentein, Francois Piednoel, Tanja Markovic, Cyril Pelupessy (R.I.P. ), Juan J. Guerrero
Coding 24/7... Limited forums/PMs time.
-Justice isn't blind, Justice is ashamed.
Many thanks to: Sue Wu, Yiwen Lin, Steven Kuo, Crystal Chen, Vivian Lien, Joe Chan, Sascha Krohn, Joe James, Dan Snyder, Amy Deng, Jack Peterson, Hank Peng, Mafalda Cogliani, Olivia Lee, Marta Piccoli, Mike Clements, Alex Ruedinger, Oliver Baltuch, Korinna Dieck, Steffen Eisentein, Francois Piednoel, Tanja Markovic, Cyril Pelupessy (R.I.P. ), Juan J. Guerrero
My last few benches before I start the "stock clocks" measurements for the review:
3D Mark06 GTX 280 SLI @ 702MHz/1512MHz/1215MHz
3D Mark06 GTX 280 @ 702MHz/1512MHz/1215MHz
Coding 24/7... Limited forums/PMs time.
-Justice isn't blind, Justice is ashamed.
Many thanks to: Sue Wu, Yiwen Lin, Steven Kuo, Crystal Chen, Vivian Lien, Joe Chan, Sascha Krohn, Joe James, Dan Snyder, Amy Deng, Jack Peterson, Hank Peng, Mafalda Cogliani, Olivia Lee, Marta Piccoli, Mike Clements, Alex Ruedinger, Oliver Baltuch, Korinna Dieck, Steffen Eisentein, Francois Piednoel, Tanja Markovic, Cyril Pelupessy (R.I.P. ), Juan J. Guerrero
wow 1.6 vcore, isn't that too high? also this is on air?
I mentioned in my previous posts that the Vcore shown in CPU-z is wrong
It's approximately 0.1V higher than the real voltage.
All these runs were done with a ThermalRight Ultra-120 ( the first version, not the Extreme ) and with a 68F/20C ambient room temp.
Coding 24/7... Limited forums/PMs time.
-Justice isn't blind, Justice is ashamed.
Many thanks to: Sue Wu, Yiwen Lin, Steven Kuo, Crystal Chen, Vivian Lien, Joe Chan, Sascha Krohn, Joe James, Dan Snyder, Amy Deng, Jack Peterson, Hank Peng, Mafalda Cogliani, Olivia Lee, Marta Piccoli, Mike Clements, Alex Ruedinger, Oliver Baltuch, Korinna Dieck, Steffen Eisentein, Francois Piednoel, Tanja Markovic, Cyril Pelupessy (R.I.P. ), Juan J. Guerrero
My remark regarding the voltages shown by CPU-z might be incorrect, because I think I've found a BIOS bug ( actually a simple error, somebody typed 2 instead of 1 in the CPU Target voltage indicator ) and if the measurements with a multimeter verify my findings I might have been running the CPU at the voltages shown in the previous screenshots!
I decided to give the Ultra-120 some more CFM to help her withstand the heavy load
Gonna run some 3D Mark Vantage now and maybe some wPrime v1.55 1024m.
And then I'm off to start the default clocks measurements for the review.
Coding 24/7... Limited forums/PMs time.
-Justice isn't blind, Justice is ashamed.
Many thanks to: Sue Wu, Yiwen Lin, Steven Kuo, Crystal Chen, Vivian Lien, Joe Chan, Sascha Krohn, Joe James, Dan Snyder, Amy Deng, Jack Peterson, Hank Peng, Mafalda Cogliani, Olivia Lee, Marta Piccoli, Mike Clements, Alex Ruedinger, Oliver Baltuch, Korinna Dieck, Steffen Eisentein, Francois Piednoel, Tanja Markovic, Cyril Pelupessy (R.I.P. ), Juan J. Guerrero
Time for some BIOS shots:
Coding 24/7... Limited forums/PMs time.
-Justice isn't blind, Justice is ashamed.
Many thanks to: Sue Wu, Yiwen Lin, Steven Kuo, Crystal Chen, Vivian Lien, Joe Chan, Sascha Krohn, Joe James, Dan Snyder, Amy Deng, Jack Peterson, Hank Peng, Mafalda Cogliani, Olivia Lee, Marta Piccoli, Mike Clements, Alex Ruedinger, Oliver Baltuch, Korinna Dieck, Steffen Eisentein, Francois Piednoel, Tanja Markovic, Cyril Pelupessy (R.I.P. ), Juan J. Guerrero
Test GSkill 3x1gb F3 16000 kit on BloodRage:
cpu core 1.475
cpu vtt +200mv
1.8v pll 1.87
cpu clk amp ctrl .800
Dram 1.93
QPI multi x18
uncore multi x18
mem bus multi x8
It's the best I could get so far. I thought this kit would run a higher multi but I don't know to get there.
tripgood
Big hardware pile
tripgood, is that much voltage necessary ?
No. I was focusing on getting the mem multi higher with no success and just left it there. H2o cooling low 50's loading 32mPI. Diablo's are next up.
tripgood
Big hardware pile
Set the UnCore multiplier to 20x, and the memory multiplier to 10x.
It should work if your memories can handle it and you're feeding the UnCore with the required voltage.
In relatively high overclocks, to get the UnCore stable at anything over 4GHz you'll more than likely have to raise the memory voltage as well ( apart from the rest voltages ) [ and make sure you keep the Vram - Vunc => 0.5V ]
Coding 24/7... Limited forums/PMs time.
-Justice isn't blind, Justice is ashamed.
Many thanks to: Sue Wu, Yiwen Lin, Steven Kuo, Crystal Chen, Vivian Lien, Joe Chan, Sascha Krohn, Joe James, Dan Snyder, Amy Deng, Jack Peterson, Hank Peng, Mafalda Cogliani, Olivia Lee, Marta Piccoli, Mike Clements, Alex Ruedinger, Oliver Baltuch, Korinna Dieck, Steffen Eisentein, Francois Piednoel, Tanja Markovic, Cyril Pelupessy (R.I.P. ), Juan J. Guerrero
Thanks. I will try that later today.
tripgood
Big hardware pile
Bench, in your opinion, what's the final word in terms of VDIMM and Uncore Voltage safe max's?
Core i7 920 D0 @ 4833.55Mhz~EVGA "Classified" E760~XFX HD5870~6GB G.Skill Perfect Storm DDR3 2000~Corsair HX1000~2x WD Velociraptor 150GB RAID 0~WD Cavier Black 1TB~Custom Danger Den Water Box Plus~Windows 7 x64 RTM~Dell 2407WFP
H20! CPU Loop: MCP-655~Swiftech MCR-220~HEATKILLER® 3.0 LT GPU Loop: MCP-655~XSPC RS240~EK FC5870 Plexi+Nickel
I've been feeding the CPU with UnCore voltages up to 1.72V ( +560mV setting ), Vdimm up to 2.25V, and Vcore up to... hold your breath.... 1.62V on air ( you can laugh now ) and it's still clocking fine, like day 1
I'd say 1.65V UnCore is safe for 24/7 and Vdimm up to 2.1V for 24/7 ( you need to have the UnCore at 1.6V or a bit higher to keep the Vdimm at 2.1V for 24/7 without risking your CPU's IMC ).
Coding 24/7... Limited forums/PMs time.
-Justice isn't blind, Justice is ashamed.
Many thanks to: Sue Wu, Yiwen Lin, Steven Kuo, Crystal Chen, Vivian Lien, Joe Chan, Sascha Krohn, Joe James, Dan Snyder, Amy Deng, Jack Peterson, Hank Peng, Mafalda Cogliani, Olivia Lee, Marta Piccoli, Mike Clements, Alex Ruedinger, Oliver Baltuch, Korinna Dieck, Steffen Eisentein, Francois Piednoel, Tanja Markovic, Cyril Pelupessy (R.I.P. ), Juan J. Guerrero
Core i7 920 D0 @ 4833.55Mhz~EVGA "Classified" E760~XFX HD5870~6GB G.Skill Perfect Storm DDR3 2000~Corsair HX1000~2x WD Velociraptor 150GB RAID 0~WD Cavier Black 1TB~Custom Danger Den Water Box Plus~Windows 7 x64 RTM~Dell 2407WFP
H20! CPU Loop: MCP-655~Swiftech MCR-220~HEATKILLER® 3.0 LT GPU Loop: MCP-655~XSPC RS240~EK FC5870 Plexi+Nickel
Default cpu vtt (uncore) is 1.16v?
tripgood
Big hardware pile
Coding 24/7... Limited forums/PMs time.
-Justice isn't blind, Justice is ashamed.
Many thanks to: Sue Wu, Yiwen Lin, Steven Kuo, Crystal Chen, Vivian Lien, Joe Chan, Sascha Krohn, Joe James, Dan Snyder, Amy Deng, Jack Peterson, Hank Peng, Mafalda Cogliani, Olivia Lee, Marta Piccoli, Mike Clements, Alex Ruedinger, Oliver Baltuch, Korinna Dieck, Steffen Eisentein, Francois Piednoel, Tanja Markovic, Cyril Pelupessy (R.I.P. ), Juan J. Guerrero
Things are getting a little better with this GSkill kit.
tripgood
Big hardware pile
So why the whole obsession now with making 1.65v DDR3 kits. I mean, if they'r okay up to 2.1V as long as the Vtt is kept at under .5 from Vdimm, then 1.65v is not the limit right?
Unless 1.65v are value kits and someone is planning on making triple channel 2.1v kits, but I don't know of anyone. Those seems like they'd be the ultimate overclockers.
Like the ocz gold vx series back in DDR1 days.
Computer: i7 2600k @ 4.7Ghz | Asus P8P67 Evo | Corsair LP 16gb 1600CL9 | Silver Arrow | Essence STX | Crucial m4 128gb | Silverstone Raven 3|
Video: 2x Sapphire 6950 Toxic 2gb @ 6970 Switch @ 880 / 1350 | Asus VG248QE |
Audio: ODAC+O2 => JH|13 Pro | STX => ATH-AD700X / Audioengine A5
Problem is in keeping voltage of VTT that high. Difference must not be bigger than .5V from Vdimm.. 1.65 isnt limit, I guess that 1.75V maybe.
There are triple kits with 2V and more, made by A-Data.. But I guess anyone who likes to keep his CPU working would not try that..
Core i7 920 D0 @ 4833.55Mhz~EVGA "Classified" E760~XFX HD5870~6GB G.Skill Perfect Storm DDR3 2000~Corsair HX1000~2x WD Velociraptor 150GB RAID 0~WD Cavier Black 1TB~Custom Danger Den Water Box Plus~Windows 7 x64 RTM~Dell 2407WFP
H20! CPU Loop: MCP-655~Swiftech MCR-220~HEATKILLER® 3.0 LT GPU Loop: MCP-655~XSPC RS240~EK FC5870 Plexi+Nickel
Ok I have the same g-skills sticks as you but Im having wierd results with oc'ing tmy 920 on the bloodrage. I went with a 190 Bclock, 20x multi on the core, 18x multi on QPI, 20x multi on Uncore, and a 10x multi with the ram (around 2300mhz). I did go light on the vcore volts (1.35v), and the uncore I had at +220mV with ram at 1.85v (this is per XMP profile enabled). With those settings my target core speed is suppose to be 3.8ghz but on post screen it shows me booting at like 3.2ghz (160 Bclock) with the ram at the default it seems 1067. I can boot into windows fine but cpu-z reports my clock speed at 3.5ghz?? and my ram at 1330mhz with 8 8 8 19 timings ??? Huh??? My ram is 1600mhz default with timings of 9 9 9 24.
I disabled xmp profile and dropped the ram volts to 1.65v, and dropped the multi on the ram to like a 6 or 8x multi giving me around 1504mhz. I kept Uncore at 20x multi. Rebooted. Same thing happened. Post is reporting 3.3ghz, and 1067mhz ram. CPU-Z reports 3.5ghz, and 1330mhz ram with the same 8 8 8 19. I even went back into bios and manually changed the timings but that didnt work.
Im running the g13 bios.
Help!!!
Bookmarks