Trident X DDR3 2400 + Asrock Z77 Extreme6, Unable to use XMP

[Tradesman]

I think it's a problem with their BIOS and setting the advanced/secondary timings. I worked with them quite a while back on mt ASRock Z77 Extreme 4, the primary problem I found was they were limiting the tRFC to 255, they got a change in which helped, then released a version 1.6 which solved the problem. I've since upgraded to the Trident 32GB 2400 set and all is fine with 1.6 - if I stray from it, the 1.8 and 2.0 BIOS versions though all is lost and I have to manually set things again and they again limited the tRFC. Don't know why they are doing it...or why they pulled the 1.6 BIOS version for the EX 4, I'm nice cool and stable with 32GB at 2400 and my 3570K at 4.6 (when I go 4.7 the idle temps climb to 40-41 so I generally stay at 4.6 w/ 37-38 idle. Can send you the 1.6 if you want, I can email it to the address you registered w/

[StrikerX]

Yeah sure email me that bios please and also the settings you're using on your mobo, will appreciate it. I've also talked with Nick Shih about this and he has given me some beta bios's for Extreme 6 which I'll try on Monday, I think he agreed that its a bios related issue which he'll look into on Monday.

[Tradesman]

Will do, should go out in 5 min or so

[wopro]

new bios

asrock z77 extreme 4: bios 2.1
The first time now that the gskill memory works @ 2133 on that board
more later....

[Tradesman]

Great, they finally got back off the dime, will have to check it out....

[Draizer]

Hello!

Just so you know, I managed to post with no troubles with this memory kit @2400 with a close rig to the OP's (3770k instead of 3570k, ASRock Extreme6) when I updated to BIOS 2.10. The CPU is NOT OCd (so it is at its stock 3.50Ghz speed).

However, I'm getting troubles when trying to get truly stable. I have 2 kits of 2 sticks (both of which are Trident X 2400 2x4Gb kits), and always got errors 0x1a. I assumed either one of the kits had to be faulty or it was a memory mismatch (I know, I know, beginner's error), so I ran memtest on both kits separately @2400 and had plenty of errors on both, which is weird.

I didn't try it yet @1333 (which I intend to do tonight to be absolutely SURE it's faulty RAM sticks), but if it's not could it be a matter of voltages (VCCSA/VTT and all) and/or CPU speed (apparently the CPU gotta be OC'd too to follow the sticks or something)? I'm new to this, and while this thread helped me plenty I'd appreciate some more advice.

Thanks in advance to whoever is willing to answer!

[StrikerX]

See if this is of any help--> http://www.gskill.us/forum/showthread.php?t=11138

If the above doesn't work then I have a beta bios for extreme 6 if you're willing to test it out.

[Draizer]

Thanks, I'll try out what the thread says when I get back from work. Gotta admit I'm not too keen on getting my RAM above 1.65V though, but if that's what must be done...

I'd just like to know, because so far the answers I've found on Google were either useless or ambiguous: Can memtest86 find errors just because the RAM voltages and timings are too low for the speed I'm trying to run it at? Or is it necessarily because the sticks are bad?

By the way (again, assuming both my kits aren't faulty), if I find perfectly stable voltage configuration and timings for a 2x4Gb @2400, I suppose I'll have to tinker with them again if I wish to try it out with all 4x4Gb @2400?

[Tradesman]

Try DRAM 1.66, VTT 1.16 or so, VCCSA 1.2, and an OC of 4.3 with the vCore (CPU voltage at an offset of + 0.060 - see if that stabilizes.

[Draizer]

Are these settings for 2 or 4 sticks?

And while I'm willing to OC my CPU, I'm stuck with the default Intel fan for now, will it be able to handle the heat?

[Tradesman]

That would be for all 4 sticks, though can try it on just 2 and then if stable, lower them (starting with VCCSA, then VTT) a little at a time. With the stock fan, just watch the temps (can download Coretemp (free) which watches temps better than the mobo provided software

[Draizer]

Alright, I just followed your voltages and no troubles at boot, so far so good.

Temps at idle are perfectly okay I think (between 35-45 depending on core), but I'll run a stress test to see how it goes later.

Also I'll run a memtest before going to sleep to check if things went better thanks to those tweaks. If not, then I guess one of my kits is screwed.

Thanks again for your time, Tradesman and StrikerX!

[Tradesman]

We'll be waiting for your update with fingers crossed

[Draizer]

Welp, while I didn't notice any problem under Windows yet (usually it would have BSODed at least once during the time it was on, but perhaps I got lucky), memtest returned gajillions (well, more like a hundred of thousands as usual) of errors at pass 3.

Should I try out with sticks @1333 to make absolutely sure it doesn't come from them?

[Tradesman]

Try with the sticks individually to see if one is bad (don't need to run the whole test, 4-5 passes should suffice and go fairly fast)