Discussion in 'TechPowerUp Club Forum' started by cadaveca, Jun 9, 2013.
I talk to that guy doesn't seem anyone from msi on there fourm does any kind of zub zero cooling
Kinda curious where people are getting beta BIOS.
But good news: Z87-Plus BIOS update! Can anyone shed more light on item 4? (Z87-A also got this BIOS update).
Z87-PLUS BIOS 1204
1. Revise help string for the German / Simplified Chinese / Korean Language.
2. Updated over voltage warning range.
3. Enhance PCIE 3.0 compatibility.
4. Enhance CPU performance when all CPU power saving functions are disabled in BIOS.
5. Enhance SATA Port Remaning function to remember SATA port name after BIOS update.
6. Enhance DRAM compatibility.
7. Enhance USB HID device compatibility on USB 3.0 ports.
8. Add new option "Memory Scrambler" under Advanced\system agent configuration\memory configuration for more DRAM tuning flexibility.
P.S. My 4.8GHz overclock started crashing in games, I'm guessing its no coincidence that it started when this heat wave hit. Good thing UPS should be dropping off my h220 tomorrow. Also, dropping my cache speed to 3.9GHz helped with stability when I used Per-Core overclocking. Anyone know the difference between 48x Sync All Cores and Per-Core with all cores at 48x? Why would Per-Core be less stable?
I just saw a simple and well written Haswell OC guide at overclockers. A month old already, so please excuse me if it was posted earlier, I missed it somehow until now.
Nothing groundbreaking there, but I thought it could come handy for people who are about to start OC-ing their first Haswell CPU.
That's just a shortened version of the ASUS guide.
I did not want to discredit the original source, nor to post something useless. Just thought some people here might find it handy, that's all. I can delete the post if you want, or change the url if you could give me the original link. Thanks and sorry.
Nah, what I meant was that it was more geared towards ASUS products only, not that there was any issues with it.
click on MSI BETA & MOD BIOS Section
there on there fourm for all there mbs just have to look for it
im having hell trying to get SLI to work on xp
No luck finding something like this for Asus boards. I guess people used to get them from a FTP, but Asus shut that down.
kingpincooling has betas for ASUS ROG boards.
I have many BETA BIOSes for many boards, but I find sometimes that BETAs may not be that stable right at launch, as some are specifically geared towards clock-record attempts or memory overclocking.
If you guys want a BETA I have, I can share, just be aware that if stuff screws up, that's on your head, not mine. Just send me a PM, and I'll help if I can.
At least most if not all the Z87 boards have dual EEPROMs so there's probably a much lesser chance of bricking the board.
Anyway, I'm testing 4.4GHz/4.15GHz 1.2v/1.17v at 125.7MHz BCLK and it has been rock solid for 27h. My Fractal Design XL R2 arrived today so I'm going to go back to the TT Water 2.0 Performer and see how it likes being mounted in the front so it isn't bombarded by 7970 heat. I think that's why my temps are so high even with my TPC-812 because my video card runs at 100% load 24/7 with the CPU and the case ambient is in the mid-high 30s Celsius even with 7 noisy fans.
I think my CPU's cache is only good for under 4.2GHz as I have tried 4.2/4.2 at 1.16v for both and the machine would still crash randomly. I think my current overclock proves that to be true, so I'll keep the cache/ring as close to 4.1GHz as I can.
No wait sorry, i think cache ring bus is only 0.977v, can it be?
Edit: I played with bclk a bit @ 106mhz,
but there is a strange issue with memory speeds @ Aida64, I loose at least 6gb read and 4gb/s copy... Its like mobo puts to loose adv timings or something, weird. Happens with normal bclk 100mhz as well (2200mhz @ XMP)
voltage...yes, you probably has "fully Manual" enabled, which sets 1.05 V or so as the lowest possible. updating your results in the ranking to reflect he lower voltage.
AIDA issues...some dividers do this. Try new BIOS, or other divider. Not sure if stability issue...or what...still investigating this myself. BCLK increases also automatically enable different divider if needed, so that might be the thing here.
Just started overclocking yesterday, since I am noob in this thing I started to raise my multi to 45 and raised the voltage to 1.25 V. It did boot and passed every stress test that I threw on it.
Lowered the voltage to 1.15 V and it still kicked every stress test that I threw but I experienced lock up at this voltage. I am now at 46 multi and 1.2 voltage, hopefully it will stable for 24/7 at this clock and voltage
Everyone seems to choose 4770k over 4670k here
And my stock core and cache voltage are 0.986 V and 1.15 V respectively.
Cadaveca, thank you for the Z87-GD65 review. You made me confident to buy this board
Seems this is the max so far ran out of ln2
Did you just "slam" into that 6ghz wall? Have you had a chance to try that same cpu on a different mobo?
Asrock OCF-M ram tests
Yes that was the max I could boot at tried 125 bclk but couldn't get 6.1 to boot to the bios but didn't try OS app oc or the bclk buttons
And no other mb to try
Dom you're crazy but i love the effort.
Got the Swiftech H220 installed yesterday. I'm a bit underwhelmed with the 6C improvement at stock ratio/auto voltage (71C vs 77C). And one of the fans is making that "ddddd" noise like it needs lube. Unfortunately, my other Zalman CNPS10x data points were from before this damn heat wave, so I'm not seeing any more than a couple degrees improvement right now at higher voltages. Is this TIM-Mate 2 stuff any good, or should I use Noctua nt-h1?
Also annoying, I'm getting freezes all the time on known-good settings. Like 39x with Auto voltage. I've that nasty feeling like something is breaking, anyone else seeing this?
The SECOND you notice something weird, clear CMOS.
In other words, yes, this can happen, sometimes it is the BIOS being a bit funny, sometimes it's the chip starting to give up its magic smoke.
Maybe you're doing something wrong and do you have the fans hooked to the headers on the mobo or on the pumps header?
I have mine hooked up to the pumps header, even though they are pwm' control, I turned up the speed on my mobo and that made a huge difference in temps. I dropped down another 9-12 degrees now idling anywhere from 29-33 instead of 38-45
I did update the BIOS, which is kind of like clearing the CMOS. Saw the freezing before and after the update though. I'll clear it and load optimized defaults, it's still possible I changed some seemingly minor setting. I just hope the hour or two (total) I've stressed it at high voltage offsets (1.45v peak load voltage) hasn't degraded it already. I figured as long as it doesn't sit at 90-100C constantly, it'll be fine.
I'm using the little header thing for the pump and both fans. The pump is plugged into channel 1 and the fans take two adjacent spots. The fans automatically ramp up as temperatures increase, and the motherboard has some options for fan speed. Idle temps seem good: 28-31C, which is 1-3C better than the Zalman.
I'm going to remount the water block so I can see how the TIM looks. I'll try some nt-h1 this time, which will make for a better comparison with the Zalman anyways.
Even with temps considered, I think 1.3V might be a good safe "limit". Anything over...yeah, interesting. I wonder if VCC-in boosted itself, perhaps, if CPU is damaged.
Anyway, CPU replacement is easy if you have the Intel Tuning Plan Warranty...it's there specifically to cover these sorts of situations where we might push a bit too far.
I'm going by that 1.3 V limit myself, for vCPU, vCache, vccSA, vDIO, and vAIO. Max 1.9V for VCC-in.
what kind of volts are you pushing on the samsungs?
I wanna say 2.05 V? Those are some pretty tight timings. Dumo never holds back.
Howszit going with your clocking? Did you get any tips to break 6 GHz?
Separate names with a comma.