G.Skill Memory Helps Achieve DDR5-12000 Memory OC on Core Ultra 9 285K

[btarunr]

G.SKILL International Enterprise Co., Ltd., the world's leading brand of performance overclock memory and PC components, is excited to announce that four extreme overclockers - BenchMarc from US, OGS from Greece, Dreadzone from Australia, and CENS from Germany have all successfully shattered the DDR5-12000 overclocking barrier using G.SKILL DDR5 memory. The four extreme overclockers have achieved DDR5-12066, DDR5-12046, DDR5-12046, and DDR5-12042 respectively with their liquid nitrogen (LN2) cooling setup, all based on the latest Intel Core Ultra 9 285K desktop processor and ASUS ROG Maximus Z890 Apex motherboard.

DDR5-12000 is a tough milestone to reach for the memory overclocking community, but thanks to the incredible memory overclock performance of the latest Intel Core Ultra 200 K-series desktop processors, ASUS ROG Maximus Z890 Apex motherboard, and the latest G.SKILL DDR5 memory, we hope to see more overclockers joining the DDR5-12000+ club. For more details about the 4 memory overclock scores, please refer to the HWBOT links below:

• BenchMarc @ DDR5-12066
• OGS @ DDR5-12046
• Dreadzone @ DDR5-12046
• CENS @ DDR5-12042

Congratulations to extreme overclockers BenchMarc, OGS, Dreadzone, and CENS for this historical achievement. These extreme overclock records required precision tuning, optimum thermal management, and unparalleled expertise, and we are looking forward to more overclocking achievements from them in the future.

"We are thrilled to see G.SKILL DDR5 memory hit the DDR5-12000 mark," said Tequila Huang, Vice President at G.SKILL. "This achievement reflects our commitment to deliver cutting-edge memory products that are designed for overclocking enthusiasts seeking the highest performance levels. It also shows the incredible potential of DDR5 technology as we continue to push memory speed to the limits."

View at TechPowerUp Main Site

 

[RaceT3ch]

DAMN DUDE

 

[Hyderz]

It wasn’t so long ago they announced 9000kits for the new intel platform… I wonder how high ddr5 can go

 

[StimpsonJCat]

AMD, don't you find this embarrassing when there are people with ultra expensive x870E motherboards unable to reach 8000? Your IO die's memory controller and IF is crap and should have been refreshed with Zen 5.

 

[_roman_]

I'm not impressived.

It's just logic. The rising and falling flank of a clock signal is being used. Processors were long term at 6 GHz.

Impressive would be DDR5-35000 at 283 Kelvin.

 

[Durhamranger]

AMD, don't you find this embarrassing when there are people with ultra expensive x870E motherboards unable to reach 8000? Your IO die's memory controller and IF is crap and should have been refreshed with Zen 5.

Nope..... I`d be more embarrassed at releasing Sparrow Puke

 

[qcmadness]

AMD, don't you find this embarrassing when there are people with ultra expensive x870E motherboards unable to reach 8000? Your IO die's memory controller and IF is crap and should have been refreshed with Zen 5.

More embarrassing is Arrowlake at DDR5-10000 could not defeat Zen 5 at DDR5-8000.

 

[Wirko]

I'm not impressived.

It's just logic. The rising and falling flank of a clock signal is being used. Processors were long term at 6 GHz.

But internally, at least some parts of the memory controller must run at 12 GHz to process the incoming and outgoing data. That's impressive. PCIe 5.0 is even more impressive of course.

 

[StimpsonJCat]

More embarrassing is Arrowlake at DDR5-10000 could not defeat Zen 5 at DDR5-8000.

Not because of its memory controller though, was it!

 

[AnarchoPrimitiv]

AMD, don't you find this embarrassing when there are people with ultra expensive x870E motherboards unable to reach 8000? Your IO die's memory controller and IF is crap and should have been refreshed with Zen 5.

Doesn't Intel find it embarrassing that AMD boards only use 6000MT/s memory and STILL beat Intel at gaming even when the Intel board is using a memory kit that costs more than the motherboard?

In fact, now that I think of it, Intel has literally every single imaginable advantage over AMD and still can't win....3x the R&D budget, node advantage now that they use TSMC (cant remember how many times I've heard "AMD has a node advantage" in the past several years), more employees, relationships with OEMs that exclude AMD products from the top laptop models, top of mind awareness (your average Best Buy customerprobably doesnt evn know AMD exists), etc.....and AMD still holds the gaming crown (and enterprise)

 

[Durhamranger]

Not because of its memory controller though, was it!

CUDIMM

 

[azrael]

AMD, don't you find this embarrassing when there are people with ultra expensive x870E motherboards unable to reach 8000? Your IO die's memory controller and IF is crap and should have been refreshed with Zen 5.

So you're taking an overclocking scenario and try to apply it to the real world?

 

[Steevo]

Not because of its memory controller though, was it!

What great performance does this new high speed record at sloppy timings provide good sir? Surely Intel just knew that People love to tinker and they would find that missing performance under the stone of faster speed memory..... astounding how no reviewer could fathom this undeniable truth. But your superior intellect has deduced the obvious. Hurry and procure the items required to defend thy good name of Intel, prove to the mortals they alone deserve to produce products for our consumption, all others are of inferior quality and you will light the beacon of truth alight with your results!!!!


Huzzah!!!!

 

[Oberon]

*still somehow manages to lose to Zen 5 in gaming*

 

[evernessince]

AMD, don't you find this embarrassing when there are people with ultra expensive x870E motherboards unable to reach 8000? Your IO die's memory controller and IF is crap and should have been refreshed with Zen 5.

The irony of this statement is two-fold, that Intel cannot even beat Zen 5 with such high memory speed and that the memory latency on Arrow lake is now worse than Zen 4/5 even with AMD holding it's IO die improvements only to enterprise this generation. Wendel from Level1Techs showed that he was originally getting 100ms memory latency on Arrow Lake and 80ms after a patch, both of which are worse than Zen's 70ms.

At the end of the day this is bad for consumers, AMD has been holding back adding more cores to it's consumer CPUs and now holding back the IO die improvements to focus on enterprise all because Intel isn't competitive enough in the consumer CPU space. Saying that AMD's memory speed is low and IF is bad is missing the forest through the trees.

 

[SOAREVERSOR]

Doesn't Intel find it embarrassing that AMD boards only use 6000MT/s memory and STILL beat Intel at gaming even when the Intel board is using a memory kit that costs more than the motherboard?

In fact, now that I think of it, Intel has literally every single imaginable advantage over AMD and still can't win....3x the R&D budget, node advantage now that they use TSMC (cant remember how many times I've heard "AMD has a node advantage" in the past several years), more employees, relationships with OEMs that exclude AMD products from the top laptop models, top of mind awareness (your average Best Buy customerprobably doesnt evn know AMD exists), etc.....and AMD still holds the gaming crown (and enterprise)

Not everyone is focused on gaming. I would never buy an X3D chip because it's worse for actual work and power user tasks. I have as much use for it and what it does as the plastic tricycle my five year old nephew rides around my brothers house.

 

[damric]

Did it help?

 

[tpuuser256]

Jesus christ, I'm not sure the 8700G could reach that

 

[Oberon]

Not everyone is focused on gaming. I would never buy an X3D chip because it's worse for actual work and power user tasks. I have as much use for it and what it does as the plastic tricycle my five year old nephew rides around my brothers house.

You can't make a blanket statement like that re: "actual work." There are a number of non-gaming workloads where vCache chips hold a significant performance advantage (but you'd probably have to be someone who actually does that kind of work to know that.)

 

[mechtech]

It's so shiny it's like a mirror, and didn't even see it at first.

I was trying to figure out why the dimm sticker was mirrored..............

ahhhhhh that's why..............

 

[StimpsonJCat]

So you're taking an overclocking scenario and try to apply it to the real world?

I forgot that nobody overclocks in the real world. Sorry.

But AMD does state that their products can do 8000 via the performance destroying 1:2 memory divider, and many people have found that they cannot even boot at that speed. Ask Actually Hardcore Overclocking...

More embarrassing is Arrowlake at DDR5-10000 could not defeat Zen 5 at DDR5-8000.

Because the CPU itself is shit. AMD is let down by its memory controller and high latency. Why else does the x3D version exist?

 

[Dr. Dro]

AMD, don't you find this embarrassing when there are people with ultra expensive x870E motherboards unable to reach 8000? Your IO die's memory controller and IF is crap and should have been refreshed with Zen 5.

Apples to oranges, Ryzen doesn't benefit from this because of internal bandwidth limitations anyway. If you need more memory oomph, the X3D chip is the way to go

 

[_roman_]

What great performance does this new high speed record at sloppy timings provide good sir?

That's the main point why I wrote I'm not impressed.