Overclock.net › Forums › Intel › Intel - General › [Official] - Xeon X5660-X58 Review & Discussion [and Xeon L5639 benchmarks inside]
New Posts  All Forums:Forum Nav:

[Official] - Xeon X5660-X58 Review & Discussion [and Xeon L5639 benchmarks inside] - Page 700

Poll Results: Was my review helpful?

 
  • 95% (244)
    Yes
  • 4% (12)
    No
256 Total Votes  
post #6991 of 7945
OK, great news!

As far as the latencies go, you see, the highest I've dared to go so far with RAM voltage is 1.6 - 1.61V. I know that people run 1.65V 24/7 and then some other people bench at voltages as high as 1.75 - 1.8V, but I don't feel safe about either. At the moment I'm running 1600 @ 8-8-8-24 on Linux for perfect stability (I was getting crashes, and while those were with a 99% probability caused by Turbo Boost kicking in while it was turned off in the BIOS - Linux P.M. is weird - still I put everything back to stock having no time to debug the case properly at he moment), or 1600 7-7-8-21, both at 1.5V (1.53V with my motherboard).

Bottom line: do you think it's safe to use 1.6V to 1.65V on the Crucials with the X5650 and what should the RAM temps be like? Really, really don't want to accidentally degrade the kit. The idea behind running higher voltage is to get tighter latencies, if only for a few bench runs.
post #6992 of 7945
Quote:
Originally Posted by CrazyNightOwl View Post

Is that a golden chip you have there tbob22 or are the W's like that in contrast to the X's?

And a question for all the users in here. Are those read speeds good enough for 1600, 2000 and 2133 MHz RAM respectively, running on the X5650:
Warning: Spoiler! (Click to show)



How can I break the 30 GB/s barrier if at all?

I don't know about golden, a few others here have had a bit better chips (closer to 4.5ghz@1.3v) but certainly above average. A good W chip can clock just as well, but do usually get a bit warmer.

As far as hitting 30gb/s on read, you'll need to get your uncore to at least 3800mhz or so.
Main
(15 items)
 
Server
(8 items)
 
 
CPUMotherboardGraphicsRAM
E5-1660 @ 4.6ghz 1.275v Asus x79 Deluxe Gigabyte GTX 1080 Windforce 32gb Patriot Viper III @ 2400 11-12-11-24 1t 
Hard DriveHard DriveCoolingOS
4x4tb, 2x5tb, 2x8tb 7200rpm 850+840 Evo 500gb, BX100 500gb, 960 Evo 500gb Noctua NH-D15S Windows 10 Pro 
MonitorPower
2x Lenovo L220x, Hisense h8c 55" Corsair RM850i 
CPUMotherboardGraphicsRAM
2x Xeon x5670 ASUS Z8NA-D6C Windforce 7950 48gb NANYA DDR3 1333mhz 
Hard DriveCoolingPowerCase
2x Toshiba 2tb 7200rpm Raid 1 2x Raijintek Aidos 750w ToughPower Gold DIYPC D480-BK 
  hide details  
Reply
Main
(15 items)
 
Server
(8 items)
 
 
CPUMotherboardGraphicsRAM
E5-1660 @ 4.6ghz 1.275v Asus x79 Deluxe Gigabyte GTX 1080 Windforce 32gb Patriot Viper III @ 2400 11-12-11-24 1t 
Hard DriveHard DriveCoolingOS
4x4tb, 2x5tb, 2x8tb 7200rpm 850+840 Evo 500gb, BX100 500gb, 960 Evo 500gb Noctua NH-D15S Windows 10 Pro 
MonitorPower
2x Lenovo L220x, Hisense h8c 55" Corsair RM850i 
CPUMotherboardGraphicsRAM
2x Xeon x5670 ASUS Z8NA-D6C Windforce 7950 48gb NANYA DDR3 1333mhz 
Hard DriveCoolingPowerCase
2x Toshiba 2tb 7200rpm Raid 1 2x Raijintek Aidos 750w ToughPower Gold DIYPC D480-BK 
  hide details  
Reply
post #6993 of 7945
Quote:
Originally Posted by SmOgER View Post

With the corresponding 3.2 and 3.35Ghz uncore your read speeds are perfectly fine.
But since it's a trial version, it doesn't show the whole story unfortunately.

Copy speeds is where it likes to climb closer to 35Gb/s with good RAM.

But ignoring what it's not shown, your 2100@9-9-9-27|2T|_3.35Ghz setup scores almost the same like mine 1950@8-8-8-24|1T|_3.4Ghz. With the exception of latency being slightly higher.
So as far as the images go, I would say you are set.

Oh yeah.. those are much better. He was running some 60's and 70's in latency before. great improvement

You are right on par now!!

Edit: I think that was OCmember
Edited by DRKreiger - 6/15/16 at 7:20am
ROGBeast RIP
(16 items)
 
- X58 Xeon Club -
(22 photos)
 
CPUMotherboardGraphicsRAM
Xeon W3680 asus ramage iii extreme Galaxy GTX 780 Hall of Fame Crucial Ballistix Tactical LP 
Hard DriveHard DriveOptical DriveCooling
2x Samsung 850 evo raid 0 raptor 10k blueray something EK Supremecy Evo 
CoolingMonitorMonitorKeyboard
EK 780-FC HOF asus pg279 Asus pg279 Corsair Strafe RGB 
PowerCaseMouseAudio
Cooler Master V1200 Platinum  Corsair Carbide 760T Corsair Scimitar Sound Blaster 3D recon 
  hide details  
Reply
ROGBeast RIP
(16 items)
 
- X58 Xeon Club -
(22 photos)
 
CPUMotherboardGraphicsRAM
Xeon W3680 asus ramage iii extreme Galaxy GTX 780 Hall of Fame Crucial Ballistix Tactical LP 
Hard DriveHard DriveOptical DriveCooling
2x Samsung 850 evo raid 0 raptor 10k blueray something EK Supremecy Evo 
CoolingMonitorMonitorKeyboard
EK 780-FC HOF asus pg279 Asus pg279 Corsair Strafe RGB 
PowerCaseMouseAudio
Cooler Master V1200 Platinum  Corsair Carbide 760T Corsair Scimitar Sound Blaster 3D recon 
  hide details  
Reply
post #6994 of 7945
Judging by the screens I've posted here, used to have 60's - 70's latencies as well. Does it make a difference in everyday usage though? The bench results have increased for sure, but other than that, the only area where I think I might feel a difference is BF gaming, and it's been hard to estimate since after a BF3 reinstall, terrible stutter emerged. BF4 was smooth though without a hiccup in the performance overlay graph with ~60 FPS @ 1920x1080 on all lows (and no NVIDIA CP tweaking had been made!), but I only played it after installing the Crucials, through the Origin Game Time limited 1-week trial.

However, the 6 vs 12 threads IBT difference still is there. E..g on my last test run, for 3.5 GHz I think, it was 58 vs 64 GFLOPS.
post #6995 of 7945
Quote:
Originally Posted by CrazyNightOwl View Post

Judging by the screens I've posted here, used to have 60's - 70's latencies as well. Does it make a difference in everyday usage though? The bench results have increased for sure, but other than that, the only area where I think I might feel a difference is BF gaming, and it's been hard to estimate since after a BF3 reinstall, terrible stutter emerged. BF4 was smooth though without a hiccup in the performance overlay graph with ~60 FPS @ 1920x1080 on all lows (and no NVIDIA CP tweaking had been made!), but I only played it after installing the Crucials, through the Origin Game Time limited 1-week trial.

However, the 6 vs 12 threads IBT difference still is there. E..g on my last test run, for 3.5 GHz I think, it was 58 vs 64 GFLOPS.

I don't think it'll make much difference. If you are getting 20gb/s+ you're unlikely to see any difference in anything but the most memory intensive applications (not games).

That IBT seems a little low, but maybe that's about right for 3.5. How are your cinebench scores?
Main
(15 items)
 
Server
(8 items)
 
 
CPUMotherboardGraphicsRAM
E5-1660 @ 4.6ghz 1.275v Asus x79 Deluxe Gigabyte GTX 1080 Windforce 32gb Patriot Viper III @ 2400 11-12-11-24 1t 
Hard DriveHard DriveCoolingOS
4x4tb, 2x5tb, 2x8tb 7200rpm 850+840 Evo 500gb, BX100 500gb, 960 Evo 500gb Noctua NH-D15S Windows 10 Pro 
MonitorPower
2x Lenovo L220x, Hisense h8c 55" Corsair RM850i 
CPUMotherboardGraphicsRAM
2x Xeon x5670 ASUS Z8NA-D6C Windforce 7950 48gb NANYA DDR3 1333mhz 
Hard DriveCoolingPowerCase
2x Toshiba 2tb 7200rpm Raid 1 2x Raijintek Aidos 750w ToughPower Gold DIYPC D480-BK 
  hide details  
Reply
Main
(15 items)
 
Server
(8 items)
 
 
CPUMotherboardGraphicsRAM
E5-1660 @ 4.6ghz 1.275v Asus x79 Deluxe Gigabyte GTX 1080 Windforce 32gb Patriot Viper III @ 2400 11-12-11-24 1t 
Hard DriveHard DriveCoolingOS
4x4tb, 2x5tb, 2x8tb 7200rpm 850+840 Evo 500gb, BX100 500gb, 960 Evo 500gb Noctua NH-D15S Windows 10 Pro 
MonitorPower
2x Lenovo L220x, Hisense h8c 55" Corsair RM850i 
CPUMotherboardGraphicsRAM
2x Xeon x5670 ASUS Z8NA-D6C Windforce 7950 48gb NANYA DDR3 1333mhz 
Hard DriveCoolingPowerCase
2x Toshiba 2tb 7200rpm Raid 1 2x Raijintek Aidos 750w ToughPower Gold DIYPC D480-BK 
  hide details  
Reply
post #6996 of 7945
Well, I've been getting less freezes in BF3 on 64 player Metro with the new RAM, so it's either the doubled amount or the increased "speed" of it, or maybe both.

Cinebench 15 is about 925-930 for 200 BCLK, 4.0 GHz, 3.6 GHz QPI, 3.2 GHz uncore, and good RAM settings. OpenGl test in CB15 is slightly above 100 FPS.

Cinebench 11.5 awarded me with a score between 10 and 11 points for 4 GHz, can't recall precisely straight off my head, since I have two runs, one < 10.5 and one > 10.5, recorded.
post #6997 of 7945
Quote:
Originally Posted by Ferro10 View Post

Hello
I am new to the forum and would like to ask a help on coming overclock a X5660.

my system is as follows:
p6t deluxe V2
ZOTAC GTX 970 AMP! Omega Core Edition
i7 920 2.6ghz@ 4.2ghz (ex cpu Now mount the X5660)
Thermalright venus x
6gb G.skill Trident 1600mhz 6-7-6-18 t1


I managed to be stable at 4.4ghz with these settings:
210x21=4.4ghz
Vcore= 1.39
qpi/dram core vol= 1.34
dram bus Voltage 1.64
cpu pll= 1.84

temp max 77°


a 1.39 Vcore you can keep it for daily use?

no one to help me?
post #6998 of 7945
Quote:
Originally Posted by Ferro10 View Post

no one to help me?

No, i wouldn't go for more than 1.36v for 24/7 and the official guideline is 1.35v.Just drop to 205x21 and put cpu vcore to 1.328-1.350v and start stress testing.You'll be golden.
post #6999 of 7945
I'd say the VTT (also called QPI) voltage is a bit too high, what's the exact reason to run it like that? Especially seeing how the RAM is at a mere 1600 MHz, that QPI voltage appears unjustified to me. Regarding safety, it's under the 1.35V limit and is thus safe by the specifications, but from what I understand, it's on the very edge.

77C is high enough on its own and too high for 1.39V vCore in my opinion, since it's not just the voltage that might degrade the CPU, but rather a certain amount of voltage being used under high temps for a prolonged period of time. 1.39V for 4.4.GHZ is alright, there are better results out there, but for my X5650 wants about 1.44V - 1.45V at that frequency, for example.

Also, it may be possible that at a lower temperature, say 60 - 65C, you may be able to run a lower vCore for the same frequency. I'd recommend investing into a stronger air cooler or getting a decent AIO, since the latter have become more affordable than ever now.
post #7000 of 7945
^^^ for sure. If you could get 4.2 @ 1.35v's and keep the temps in the 60's, you would be quite good.
ROGBeast RIP
(16 items)
 
- X58 Xeon Club -
(22 photos)
 
CPUMotherboardGraphicsRAM
Xeon W3680 asus ramage iii extreme Galaxy GTX 780 Hall of Fame Crucial Ballistix Tactical LP 
Hard DriveHard DriveOptical DriveCooling
2x Samsung 850 evo raid 0 raptor 10k blueray something EK Supremecy Evo 
CoolingMonitorMonitorKeyboard
EK 780-FC HOF asus pg279 Asus pg279 Corsair Strafe RGB 
PowerCaseMouseAudio
Cooler Master V1200 Platinum  Corsair Carbide 760T Corsair Scimitar Sound Blaster 3D recon 
  hide details  
Reply
ROGBeast RIP
(16 items)
 
- X58 Xeon Club -
(22 photos)
 
CPUMotherboardGraphicsRAM
Xeon W3680 asus ramage iii extreme Galaxy GTX 780 Hall of Fame Crucial Ballistix Tactical LP 
Hard DriveHard DriveOptical DriveCooling
2x Samsung 850 evo raid 0 raptor 10k blueray something EK Supremecy Evo 
CoolingMonitorMonitorKeyboard
EK 780-FC HOF asus pg279 Asus pg279 Corsair Strafe RGB 
PowerCaseMouseAudio
Cooler Master V1200 Platinum  Corsair Carbide 760T Corsair Scimitar Sound Blaster 3D recon 
  hide details  
Reply
New Posts  All Forums:Forum Nav:
  Return Home
  Back to Forum: Intel - General
Overclock.net › Forums › Intel › Intel - General › [Official] - Xeon X5660-X58 Review & Discussion [and Xeon L5639 benchmarks inside]