**Official** GIGABYTE Z87X-OC/OC Force Club

[sin0822]

Welcome to the Z87X-OC and OC Force Club!
Okay so it is that time of the year again where new boards are released and many new questions come up.
Get latest BIOS from here.

GIGABYTE product Page for:
Z87X-OC
Z87X-OC Force

Some FYI in General:
#1. To use full HD Mode one can use the iGPU, must use a GPU and monitor with pretty high resolution, I think 1080P. I also don't think a D-SUB/VGA cable can be supported.
#2. To change driver order in Windows mode you can select the drive and hold down ATL and then use the arrows.
#3. TAG profile is Profile #8 under save profiles, i like to setup TAG then load optimized defaults and just load TAG for OCing.
#4. If you run into issues with USB 3.0 without drivers in windows you can try disabling xHCI.

Some things for OC
#1. OC Touch for multiplier is capable of being increased even when trigger is engaged. So I can hit trigger and go to 8x and press + on multiplier and when i go back to full speed(when i disengage trigger) my multiplier will be one higher than before.
#2. Currently need to have EIST enabled for multiplier OC to work in Windows.
#3. GTL is coming out very soon.
#4. If your BCLK dividers don?t work at all then reflash the BIOS you might have corrupted ME firmware.
#5. BCLK on air 110+, add LN2 to CPU gain maybe 5 more MHz on top on air, LN2 on PCH gain maybe 5 mhz more as well.
#6. tRDRD =tRRSR on GBT BIOS
#7. For memory OC use 1.00x or 1.25x divider, 1.67x divider isn?t optimized for memory OC.
#8. Use Ring voltage for higher uncore frequency
#9. Use SA, CPU IO Analog and Digital for Memory OC. You can use +0.2 on SA and up to 0.15 on the others. The deltas here are important too.
#10. If you want to OC BCLK use the iGPU OR you can use the last PCi-E slot, the 4x PCI-E 2.0 slot's PCi-E frequency is locked at 100mhz always b/ it comes from the PCH.


Z87X-OC Force OC Features list:
#1. OC Touch (Multiplier, BCLK, Gear, Turbo, power, reset, clear CMOS)
#2. TAG
#3. OC Ignition
#4. Trigger
#5. Dual BIOS Selector
#6. Single BIOS Mode
#7. 15u gold plate on CPU pins, DIMM pins, and PCI-E pins
#8. 8-Layer 2oz Copper PCB
#9. 4-way SLI/CF at 8x/8x/8x/8x with PLX bypass
#10. 16 phase VRM using IR3563B and IR3550 (very high quality).

Z87X-OC OC Features:
#1. OC Touch(Multiplier, BCLK, Gear, Turbo, power, reset, clear CMOS)
#2. TAG
#3. OC Ignition
#4. Trigger
#5. Dual BIOS Selector
#6. Single BIOS Mode
#7. 15u gold plate on CPU pins, DIMM pins, and PCI-E pins
#8. GIGABYTE Mem SAFE
#9. Settings Lock
#10. Direct to BIOS
#11. CBAT Button
#12. 6-Layer 2oz PCB
#13. 8-Phase VRM with IR3563B and IR3553 and 60A Custom chokes.

Overall the Z87X-OC is perfect at its price, however if you want the best of the best the OC Force is the best high-end. Also there is a Z87X-UD7 TH which also has the same OC Features as the OC Force, but the board has gold highlights like the Z87X-UD5H.

Also GBT has really improved their memory OC for Z87, so enjoy.

[Solid Snake]

Hi guys , this is my first test on Z87X OC.

Air test...

Cinebench



Memory test with ADATA XPG V2





4900 MHZ for the moment

http://valid.canardpc.com/show_oc.php?id=2828274

3050 MHz on memory...

http://valid.canardpc.com/show_oc.php?id=2828728

[sin0822]

very nice job mate!

[Splave]

in? :P try some pi tonight, and hopefully some pie too

[ZenEffect]

im in

[ihog6hog]

Trident-X 2666C11 Hynix CFR @ 3300+

[stasio]

GA-Z87X-OC Force - F6f
GA-Z87X-OC - F5q

- Improve SATA RAID performance
- Enhance DDR performance
- 19.Jun 13

Dinos22 explains GIGABYTE’s OC Touch features on new Z87X-OC board:

http://www.youtube.com/watch?feature...&v=cnHbLqlwG18

[Solid Snake]

thanks for the info stasio I will try it with F5q

[sin0822]

i think F5q is what cookie used to do 195mhz BCLK on air, you can try it out!

[Phaseshift]

my cpuz says my bus is at 99.98, it's never 100mhz why is that?

[sin0822]

just the way it is, some BIOSes has it at 100 others at 99.8, GBT didn't want crap from reviewers about setting their BCLK too high.

[SonDa5]

Well I'm in.

Just ordered the Z87X-OC.


Choosing a Z87 MB was a hard choice for me because of the pricing and the many features and lack luster over clocking ability of HW with ambient temp cooling that I have seen so far.

I narrowed down my top 3 choices for my budget to the Asus ROG VI Hero, MSI Z87 Mpower Max and the Z87X-OC won.... My head is hurting and I'm not sure I made the right choice.

I not only enjoy ambient temp over clocking but I also enjoy gaming. I have an external headphone amp for audio and I prefer hard wired connection for NIC.


Haven't ordered HW cpu yet and I'm still not sure if I will get a 4670k or a 4770k. I plan on delidding.


Buying into this chipset and Haswell doesn't make alot of sense to me since I have a nice 5GHZ 3770k gaming RIG. Mainly did this for the challenge of over clocking HW and tweaking fun. Hope I made the right choice.


See you guys later.

[SonDa5]

Mine arrived today!

Some photos for the club.




VRM heat sink hole to hole measurement.





Cities where the OC Gods live.

[SonDa5]

That VRM is powered by IR 3553M which seems incredibly weak compared to the Force. From what I have researched online the IR 3553M is designed more for a server system not an over clocked gaming RIG and has max current rating of 40A which is way lower than the 60A rating that I have read a few reviews state it has. I think the reviewers got the amperage rating confused with the IR 3563B power unit which is rated for 60A from what I have researched.


With only 8 phases for cpu I think Gigabyte cut costs big time here. I got suckered into the "OC" hype of this MB and was sold on this IR power unit which from what I am learning is different than what many online reviews have stated. Considering the VRM is much weaker than the FORCE and only has half the numbers of phases for CPU.

The OC deserved at least 8 phases with top quality IR3550.

I already bought the chipset blocks as well.

[sin0822]

That VRM is powered by IR 3553M which seems incredibly weak compared to the Force. From what I have researched online the IR 3553M is designed more for a server system not an over clocked gaming RIG and has max current rating of 40A which is way lower than the 60A rating that I have read a few reviews state it has. I think the reviewers got the amperage rating confused with the IR 3563B power unit which is rated for 60A from what I have researched.


With only 8 phases for cpu I think Gigabyte cut costs big time here. I got suckered into the "OC" hype of this MB and was sold on this IR power unit which from what I am learning is different than what many online reviews have stated. Considering the VRM is much weaker than the FORCE and only has half the numbers of phases for CPU.

The OC deserved at least 8 phases with top quality IR3563B.

I already bought the chipset blocks as well.

No that isn't right, please list the reviews so I can out the reviewers who stated these numbers?

First off IR3563B is a PWM it controls the phases it doesn't limit output. Also 40A per phase is HUGE compared to what anyone else would use. It is also WAY more than needed, i don't even use the heatsink on those IR3553 as there is not much point.

Most of these LF-PAK and PowerPAKs might be able to pull off 15-25A per phase depending on the quality and amount used, NexFET that ASUS uses on ROG, and that ASRock is now using on their Extreme6 and on their OC Formula boards are also 25A continuous and 40A peak, basically same as IR3553. IR3550 is possible the best you can get, and so no there is no reason to use them on the OC board.
Okay let's put it this way, IR3550 are possibly THE MOST Expensive part aside from maybe DirectFETs that no one really uses anymore.

Let's do a quick calculation: Let's say 25A per phase(as that is 90% eff on the NexFET and the IR3553), that is 8 phases that can do 25A per phase so 8x25A= 200A times the VIN voltage which is 1.8v by default and upto 3v, so 1.8x200=360W or 600W for 3v and these IR3553 can do this on air, you really can't get that from other MOSFET designs, other than the NexFET. DrMOS doesn't even have that unless you use 50A DrMOS(which people only use the 35A varient on motherboards for OC). So think about it, how much power does your CPU pull? I woudl be VERY surprised if you could get it to use more than 120A let alone 200A hahahahahaha, and it is capable of bursts of 40A so that would make the OC VRM capable of 320A, then times by 3V and that is output its capble of when the oardis under Ln2 to cool the FET enough to allow it.

IR3553 is the size down variant. Did you check what the ASUS ROG boards use? They use D87350D http://pdf1.alldatasheet.com/datashe...D87350Q5D.html

Which is also upto 40A. ASUS uses one of these per phase, with 60A choke as well.

IR3563B is a PWM, it doesn't output current, what review did you read this crap in?

Just read my reviews for VRm analysis please, or look at THE VRM LIST I have been updating:
http://www.sinhardware.com/images/vrmlist.png

60A per phase with 60A power stage and 60A inductors were unheard of for motherboard VRM. You are talking AMPs here you aren't talking watts.

Z87X-OC VRM ~ Maximus Vi Extreme/Formula/Hero/Impact/Gene VRM, with Z77 GBT has strongest VRM, this 16 phase Z87X-OC Force VRM has about 30A-35A inductors, still outputs way more than any other board out there. It is total overkill.

The Z87X-OC VRM is perfect for its use, not using IR3550 is how they are able to keep the cost down, as using IR3550 would have increased cost and the board might have cost $250 for no reason, you wouldn't see an improvment in OC.

Limiting factor in VRm design for output is first the input(the PCH and connector), then the high-side FET, then the inductor. So the inductor is the last place it can bottleneck, Z87X-OC has 60A chokes, they wont bottleneck, but right now the Z87X-OC Force only uses 35A chokes so the IR3550 are just for show to justify charging $410 for the board. No way they can even go to 35A as the CPU could never pull that much from each phase on a 16 phase board.

[SonDa5]

No that isn't right, please list the reviews so I can out the reviewers who stated these numbers?


Limiting factor in VRm design for output is first the input(the PCH and connector), then the high-side FET, then the inductor. So the inductor is the last place it can bottleneck, Z87X-OC has 60A chokes, they wont bottleneck, but right now the Z87X-OC Force only uses 35A chokes so the IR3550 are just for show to justify charging $410 for the board. No way they can even go to 35A as the CPU could never pull that much from each phase on a 16 phase board.

60AMP chokes should have been paired with IR3550 parts. Based on what I am seeing it looks like the Z87X-OC is not good design and will end up failing before the Force will. Force has double the number of higher amperage rated true "OC" (per IR manufacturer guidlines) 3550 part. With the low rated higher spec choke the power has better digital distribution of cleaner power for pushing cpu. The Force is a much more cleaner power design and the "OC" with 3553M is has a much higher potential to put out low quality power.

The Z87X-OC is marketed as being budget no thrill get to the point over clocking mother board yet uses the same CPU power IR3553M part found in the much cheaper priced Z87 being used in the UD3H. I could have spent less money and got a UD4H with double the amount of IR3553 power(16) for CPU. This IR3553M part is recommended to be used in server motherboard not over clocked gaming mother boards.

Sorry but I coudn't find some of the reviews that lead me to believe the CPU VRM IR were IR3550. Vortez still shows some nice art work for the "OC" series that show IR3550.

http://www.vortez.net/articles_pages..._review,5.html


I really thought the "OC" series featured this IR 3550 VRM for the CPU power.

No theory here. Just sharing what I have found. To me it seems Gigabyte was trying to offer a "OC" series MB with the Z87X-OC but cheapened out on parts to make the most money possible from hyping it up and marketing.


To the enthusiast it makes no sense to buy this MB because its obvious the main VRM power output desing was cheapened out. I know this now but when I ordered it I didn't realize the IR's being used were cheapened out.



I'm not seeing other results with high CPU over clocks for ambient temp cooling with the Z87X-OC either. My delidded bare die mounted 4770k requires 1.4v withis mother board to clear a pass of the pre configured tests of the new Linpack with AVX2 and temps are in the low to mid 90C so I really don't have much head room left.

In that VOrtez review link that I posted they had a much better quality 4770k that required much less voltage and wasn't delided and they could only get to 4.7GHZ.

If I had cleaner cooler power from VRM I think my CPU would do better.

Couldn't a bios update fix the power? It's all digital so I think there is room for BIOS improvement but right now things don't seem right for my over clocks on ambient temps and I am not seeing many people results with ambient temp cooling and most people seem more interested in RAM speed. I am more interested in MAX CPU over clocks.

[Dumo]

That VRM is powered by IR 3553M which seems incredibly weak compared to the Force. From what I have researched online the IR 3553M is designed more for a server system not an over clocked gaming RIG and has max current rating of 40A which is way lower than the 60A rating that I have read a few reviews state it has. I think the reviewers got the amperage rating confused with the IR 3563B power unit which is rated for 60A from what I have researched.


With only 8 phases for cpu I think Gigabyte cut costs big time here. I got suckered into the "OC" hype of this MB and was sold on this IR power unit

The OC deserved at least 8 phases with top quality IR3563B.

Did you really clock this board air/cold and see in fact the difference with any board using different vrromm?

[SonDa5]

Did you really clock this board air/cold and see in fact the difference with any board using different vrromm?

Don't need to. I know the VRM parts used on the Z87X-OC are rated for server mother boards not over clocked gaming mother boards. I read this on the spec sheets for the parts. The VRM design on the Z87X-OC is much cheaper than that found on the Z87X-Force and only has half the number of CPU VRMS phases. Also i know that my max over clocks are not great.

[Dumo]

^^Yes, you have to go through all the steps of overclocking this mobo air/h20/cold to prove that its using unacceptable vroomm

Just reading the spec sheets won't get you anywhere, overclock it!

[Boulard83]

GREATTTT ! I jump in the club

Rig spec
Gigabyte Z87X-OC
I7 4770k + PH-TC14PE
2x8gb ballistix 1600mhz CAS 8
Zotac GTX 780 + Accelero Xtrem3
Corsair HX750
2x Agility 4 256gb
Tecnofront HWD BenchTable
3x Asus VE278Q 5760x1080







My rig thread on my forum.
http://infodupat.com/forum/viewtopic.php?f=10&t=330

Addon, video of the Xtrem3 installation
http://infodupat.com/forum/viewtopic.php?f=37&t=397

[Fosco]

Got my Z87X-OC today. Fun board to play with..

DDR3 3100 MHz :



Uploaded with ImageShack.us

[acebmxer]

Some things I have noticed with my board may cross over to there boards but here are some voltage read outs I have seen while overclocking. These settings were done with LLC set to Extreme

                      Bios                    Dmm full load

Vcore                1.2v-1.35v          0.01v-0.15v Higher
                      1.35v+              0.02v+ Higher
                      1.330v                    1.344v

vRing                 1.170v                   1.189v

This is what I have tested/seen so far. But look at vcore the closer you get to 1.4v the more the board over volts. Cpuz 1.64.0 and higher seem to show correct vcore at full load compared to a dmm but only 1.64.0 for me show correct idle voltage.

[sin0822]

Some things I have noticed with my board may cross over to there boards but here are some voltage read outs I have seen while overclocking. These settings were done with LLC set to Extreme

                      Bios                    Dmm full load

Vcore                1.2v-1.35v          0.01v-0.15v Higher
                      1.35v+              0.02v+ Higher
                      1.330v                    1.344v

vRing                 1.170v                   1.189v

This is what I have tested/seen so far. But look at vcore the closer you get to 1.4v the more the board over volts. Cpuz 1.64.0 and higher seem to show correct vcore at full load compared to a dmm but only 1.64.0 for me show correct idle voltage.

if you test with like wprime opposed to a bench like IBT/AVX or LinX with AVX does the vcore go lower? AVX instruction set will add 0.05 to 0.15v to the CPU VCore to compensate for its strain on the CPU, there is nothing anyone can do to stop that, it is built into the power control unit.

[bullant]

I have one of these boards too,im in! Havnt really spent a lot of time on this board yet.Started testing my PSC,single Dim testing to see how they fair on Haswell
Memory on air and tWCL6



8M stable Memory on air 1.9v twcl6

[sin0822]

I have one of these boards too,im in! Havnt really spent a lot of time on this board yet.Started testing my PSC,single Dim testing to see how they fair on Haswell
Memory on air and tWCL6



8M stable Memory on air 1.9v twcl6

Question PSC master, why don't you go in the SuperPI competition?