The Official Memory Frequency WR List

[Splave]

http://valid.canardpc.com/tyaddx
1671.5MHz @ 5-31-31-63 | 1x4096MB Avexir MFR | Core i7 4790K @ Gigabyte Z97 LN2| Splave

http://valid.canardpc.com/zn771i
2100MHz @ 6-31-31-63 | 1x4096MB Avexir MFR | Core i7 4790K @ Gigabyte Z97 LN2| Splave

[Charles Wirth]

Back in the day, we wired the 3.3v rail directly to the DDR with a pot adjust to crank it higher

[Splave]

[Don_Dan]

2100MHz CL6?!

[TaPaKaH]

I'm not entirely sure CL5 and CL6 are real.
Even super good MFR barely have enough potential to suicide at 1000-1050 CL6 on air (with 2V+).
I wouldn't expect them to scale on LN2 so well that they do twice the freq.

[Don_Dan]

No idea, but I'm sceptical as well. After all it's a lot more than Nick Shih did on ASRock Z87 (#72).

[Don_Dan]

First DDR4 scores added to the first post.

[Splave]

What's to be skeptical about? Set cas6 and boot up. How exactly would one trick cpuz

[Don_Dan]

What's to be skeptical about? Set cas6 and boot up. How exactly would one trick cpuz

I'm skeptical, because whenever something looks too good to be true, it's usually not true.
Those clocks are way higher than anyone else has ever achieved on MFR at CL5 and CL6 or on CL5 and CL6 in general.

Tricking CPU-Z is easy, here's a validation of DDR3 running CL4, although DDR3 doesn't even support CL4. Something similar might be happening here as well.
More info about the CL4 validation(s) can be found in the HWBot OC Challenge October 2011 thread.

[Splave]

So I need to post it closer to the current record? That's strange logic for pushing hardware. Lol glad I wasted my time I guess

[Don_Dan]

Well, I've added your scores nonetheless, I'm not here to decide if they're valid or not. I'm just saying I find them hard to believe, but it might be due to the fact that no one else has tried MFR on LN2 for CL5 and CL6 before.

[websmile]

Everyone who can read will see these are bugged - show me one reason why these wonder ics that opt out at 1050 on air while psc do 1300+ now shall go much higher on c6 than they can do at c7, 8 and even 11,12. Gigabyte bugged runs ftw - only comment I will make on this^^ - only a waste of time

[Don_Dan]

You know I care, Sam, I just didn't want to waste my time arguing about this.
Thanks for proving what I thought in the first place.

[Splave]

Well, I've added your scores nonetheless, I'm not here to decide if they're valid or not. I'm just saying I find them hard to believe, but it might be due to the fact that no one else has tried MFR on LN2 for CL5 and CL6 before.

Different ics act different isn't that obvious? How can you even compare air to ln2. These were attained the same way 2200mhz runs were so better take those down too. So how can one prove validity?

I have cas 8 9 10 11 and 12 done as well just have been away on vacation

[Splave]

http://valid.canardpc.com/sdv4d0
2085.2 7-31-31-63

http://valid.canardpc.com/vdk71i
2072.7 8-31-31-63

http://valid.canardpc.com/8sdwy8
2085.7 9-31-31-63


http://valid.canardpc.com/sk0j5d
2085.7 10-31-31-63


http://valid.canardpc.com/fnk11d
2114.8 11-31-31-63


http://valid.canardpc.com/5s38uh
2114.3 12-31-31-63



All done at the same time, decdided to try because the all out mhz of the chip was not that great. CL11 thru CL14 same max

[I.nfraR.ed]

So at CAS6 you went higher than 7, 8, 9 and 10?

[TaPaKaH]

what were the subtiming values for your runs?

[Don_Dan]

http://valid.canardpc.com/tyaddx
1671.5MHz @ 5-31-31-63 | 1x4096MB Avexir MFR | Core i7 4790K @ Gigabyte Z97 LN2| Splave

http://valid.canardpc.com/zn771i
2100MHz @ 6-31-31-63 | 1x4096MB Avexir MFR | Core i7 4790K @ Gigabyte Z97 LN2| Splave

http://valid.canardpc.com/sdv4d0
2085.2 7-31-31-63

http://valid.canardpc.com/vdk71i
2072.7 8-31-31-63

http://valid.canardpc.com/8sdwy8
2085.7 9-31-31-63


http://valid.canardpc.com/sk0j5d
2085.7 10-31-31-63


http://valid.canardpc.com/fnk11d
2114.8 11-31-31-63


http://valid.canardpc.com/5s38uh
2114.3 12-31-31-63



All done at the same time, decdided to try because the all out mhz of the chip was not that great. CL11 thru CL14 same max

Different ics act different isn't that obvious? How can you even compare air to ln2. These were attained the same way 2200mhz runs were so better take those down too. So how can one prove validity?

I have cas 8 9 10 11 and 12 done as well just have been away on vacation

Thanks for posting the rest of the scores!

I don't know how to prove validity. Of course, different ICs act differently and results on air can't be directly related to results on LN2, but I just find chips that show 100% scaling and reach almost the same clocks on CL6 to CL14 rather peculiar. That's all, it isn't a personal attack, I'm just wondering about this.


1671.5MHz @ 5-31-31-63 | 1x4096MB Avexir MFR | Core i7 4790K @ Gigabyte Z97 | Splave
2100.8MHz @ 6-31-31-63 | 1x4096MB Avexir MFR | Core i7 4790K @ Gigabyte Z97 | Splave
2085.2MHz @ 7-31-31-63 | 1x4096MB Avexir MFR | Core i7 4790K @ Gigabyte Z97 | Splave
2072.7MHz @ 8-31-31-63 | 1x4096MB Avexir MFR | Core i7 4790K @ Gigabyte Z97 | Splave
2085.7MHz @ 9-31-31-63 | 1x4096MB Avexir MFR | Core i7 4790K @ Gigabyte Z97 | Splave
2085.7MHz @ 10-31-31-63 | 1x4096MB Avexir MFR | Core i7 4790K @ Gigabyte Z97 | Splave
2114.8MHz @ 11-31-31-63 | 1x4096MB Avexir MFR | Core i7 4790K @ Gigabyte Z97 | Splave
2114.3MHz @ 12-31-31-63 | 1x4096MB Avexir MFR | Core i7 4790K @ Gigabyte Z97 | Splave

[Splave]

I think its mostly matter what tWCL you are using not just CL

Some Dumps corrupted, as well as I changed some sub timings a tad bit near the end when trying Cas7 and Cas6. I stared at C14 and went down just changing Cas but then changed other subs for c7 and c6 and c5 but I think the c5 could be better but I ran out of ln2 it was like -90c lol full pot for everything else.

@ Sam any values in particular you would like or just secondaries?

[TaPaKaH]

Could there a superloose subtiming or special option for raw freqs in the BIOS, which renders the CL value you set in by hand irrelevant?

[Splave]

Im not smart enough to find it if there is man there are IMC sub timings page but I leave them auto

[bullant]

Cl4 lol is cl4 a bug? seem real in bios,hit a wall at cl4 then when I bumped it to cl5- it pushed along further

http://valid.canardpc.com/8l8vfs

[Don_Dan]

Cl4 lol is cl4 a bug? seem real in bios,hit a wall at cl4 then when I bumped it to cl5- it pushed along further

http://valid.canardpc.com/8l8vfs

Please take a look at what I wrote in the thread about HWBOT's OC Challenge October 2011:

I already sent a PM regarding stage 2 to Massman three days ago, but I'll repost it in public.
I had a hard time believing all the CL4 scores that have been posted, so I checked some datasheets from both Elpida and Micron:

Micron D9GTR Datasheet, page 111:
CAS Latency (CL):
The CL is defined by MR0[6:4], as shown in Figure 53 on page 109. CAS latency is the
delay, in clock cycles, between the internal READ command and the availability of the
first bit of output data. The CL can be set to 5, 6, 7, 8, 9, or 10. DDR3 SDRAM do not
support half-clock latencies.

Also, right at the beginning it says:

• CAS (READ) latency (CL): 5, 6, 7, 8, 9, 10, or 11
• POSTED CAS ADDITIVE latency (AL): 0, CL - 1, CL - 2
• CAS (WRITE) latency (CWL): 5, 6, 7, 8, based on tCK

Then I checked Elpida's datasheets, I think the correct MGH-E datasheet is not available, so I had a look at the one for EDJ1108BASE in general. Right at the beginning it says:

• /CAS Latency (CL): 5, 6, 7, 8, 9, 10, 11
• /CAS Write Latency (CWL): 5, 6, 7, 8

Seems like CL4 is not supported by DDR3 at all and the boards are running at some other CAS Latency. I know some guys think those clocks are real, but please show me CL5 at those clocks, it should be easy, right?

Also looking at another datasheet (Hynix MFR):

• Programmable CAS latency 5, 6, 7, 8, 9, 10 and 11, 13 supported
• Programmable additive latency 0, CL-1, and CL-2 supported
• Programmable CAS Write latency (CWL) = 5, 6, 7, 8

Summing up, I don't think CL4 is real.

[Don_Dan]

New DDR frequency record by kotori! (Validation, HWBOT)

[varand]

3200mhz cl 8-12-11-18-1T with tight secondary & third timing