Maximum OC Memory Clock Speed of Several Upcoming ASUS Z690 Leaked

[TheLostSwede]

While we wait for Intel's official launch of its Alder Lake platform, more leaks are making their way onto the internet and this time we get a sneak peek at the maximum memory speed at four different motherboards from ASUS. The model names of the boards leaked a couple of weeks ago, which makes this leak slightly more interesting, as we can get the full model names of the board this way, since the leaker was a bit sloppy here.

Two of the boards, the TUF Gaming Z690-Plus D4 and what we presume is the ROG Strix Z690-A Gaming D4, but the Wi-Fi version, as there are no Prime gaming boards, top out at 5,333 MHz. However, this shouldn't come as a surprise, as both boards rely on DDR4 and the highest memory speed is on par with ASUS' current Z590 boards.



Next we have the Prime Z690-A, which seems to be a mid-range non ROG board and it looks like it'll top out at 6,000 MHz. The last model appears to be a board that hasn't leaked as yet, but which we presume is the ROG Strix Z690-F Gaming, as there's a DDR4 version of this board. Here memory speeds are slightly higher at 6,400 MHz, but this still seems fairly low, as there have been leaks of a board from Gigabyte with memory clocks of 8,000 MHz, so either ASUS is playing it conservatively on its lower-end boards, or these are not the final overclockable memory speeds for these boards.

View at TechPowerUp Main Site

 

[bonehead123]

hummm... leaky sneeky peeky squeeky....this really won't be all that interesting UNTIL we see some actual boards being tested & reviewed, then the fun will start, trying to figure out what boards will run what memory at what speeds and how much headroom there is for overclocking, if any at all, AND how many arms, legs, kidneys, beloved pets, and 1st born children all this is gonna cost us

a snoozerfester until then......

 

[Axaion]

those are generally much lower end than the gigabyte tachyon, which should be a 2dimm board on top.
like.. low/mid vs extreme high end for memory oc

best asus for memory should be Apex, but.. didnt see a release for it

 

[TheLostSwede]

those are generally much lower end than the gigabyte tachyon, which should be a 2dimm board on top.
like.. low/mid vs extreme high end for memory oc

best asus for memory should be Apex, but.. didnt see a release for it

I guess it could also be because nothing faster than 6400MHz is expected for now, so why bother tuning the UEFI for crazy high speeds.
But yes, I did forget the Tachyon was HiCookie's special design, although at the same time, at every new chipset/CPU launch, we tend to get a list of more or less unattainable memory clock options in the UEFI, which doesn't seem to be the case from Asus this time around.

 

[gasolina]

the price of ddr5 16gb 4800 without heatsink is 200$ and 85$ for 8gb
12900k es currently 4500¥ which is 740$ the mobo cost around 350$ at base model up to 700$+

 

[docnorth]

It seems rog strix series will also (according to this leak) support DDR5 from the beginning, few days ago they were listed only as DDR4 boards. We shall know soon.

 

[AlwaysHope]

Interesting, but like what has been commented on before, its only words atm.
Personally, I'm bored with AM4 atm, & are in 2 minds about jumping onto rocket lake or hanging out for alder lake. I have fond memories of early adopters trap with new tech & I really don't wanna have to go there again.
So wait n' see eh?

 

[TheLostSwede]

Interesting, but like what has been commented on before, its only words atm.
Personally, I'm bored with AM4 atm, & are in 2 minds about jumping onto rocket lake or hanging out for alder lake. I have fond memories of early adopters trap with new tech & I really don't wanna have to go there again.
So wait n' see eh?

The trap this time is going to be cost...

 

[AlwaysHope]

The trap this time is going to be cost...

Very true, also buggy bioses & drivers.

 

[Chaitanya]

Very true, also buggy bioses & drivers.

Most importantly lack of software support.

 

[Chrispy_]

DDR5-4800 40-40-40-77 at $$$$ with flaky launch-generation support, numerous patches, BIOS flashes, and rapid obsolescence

or

DDR4-4400 19-19-19-39 which is dirt cheap (e.g.Viper Steel) and close to double the performance in all metrics except raw bandwidth, where it still has over 90%....

 

[AusWolf]

I'm sure these DDR5 speeds will be reached with extremely high latencies at first. I'm also sure modules supporting these speeds will be extremely expensive. I think for people running 9-10-11th gen Intel, or Ryzen 3000 or 5000 series, it'll be worth to wait a couple more generations until these speeds are supported as JEDEC standard. Early adopters pay a disproportionately high price, as usual.

 

[Chrispy_]

I'm sure these DDR5 speeds will be reached with extremely high latencies at first. I'm also sure modules supporting these speeds will be extremely expensive. I think for people running 9-10-11th gen Intel, or Ryzen 3000 or 5000 series, it'll be worth to wait a couple more generations until these speeds are supported as JEDEC standard. Early adopters pay a disproportionately high price, as usual.

More of the industry-standard "paying beta testers" mentality.

Intel's last platform launch was a buggy, hurried mess at great expense to early adopters. Intel has deleted the trust they had, and have to re-earn it. I'm sure some fanboys will buy without hesitation anyway, but I think the majority of people have learned that in this two-horse race, Intel aren't as careful and motherboard vendors aren't given anywhere near as much time as they used to get back when Intel was the only real option.

 

[AusWolf]

More of the industry-standard "paying beta testers" mentality.

Intel's last platform launch was a buggy, hurried mess at great expense to early adopters. Intel has deleted the trust they had, and have to re-earn it. I'm sure some fanboys will buy without hesitation anyway, but I think the majority of people have learned that in this two-horse race, Intel aren't as careful and motherboard vendors aren't given anywhere near as much time as they used to get back when Intel was the only real option.

Well, I have a Core i7-11700 with a B560 motherboard, and I haven't experienced any of the "buggy mess" every media outlet was writing about at launch. I think it all comes down to 1. motherboard choice and quality, and 2. the media writing about every small thing like it was the end of the world just to gain readers/viewers.

It's easy to hate on the latest Intel products without trying them first hand. I've had several Ryzen 3000 and 5000 series CPUs, as well as my 11700 right now, and I'm actually happier with Intel. It is easier to cool in a SFF case with limited airflow, its BIOS options are much clearer and easier to understand, it doesn't need any stupid chipset software to run properly like the Ryzen 3000 series does, and it behaves much better during idle by just adapting to Windows power settings.

Edit: About the "paying beta tester" mentality: I agree that it's at play here, and it's wrong. People tend to advocate choosing a fresh platform with a fresh socket for long-term compatibility. As for me, I tend to go with the last generation of a specific platform to make sure I get something that's well-tested by the public and all issues have been ironed out.

 

[Noreng]

those are generally much lower end than the gigabyte tachyon, which should be a 2dimm board on top.
like.. low/mid vs extreme high end for memory oc

best asus for memory should be Apex, but.. didnt see a release for it

The Tachyon was a single DIMM at extremely loose timings. It's entirely possible that DDR5-7200 or 7533 with proper subtimings will be the limit at launch

 

[TheLostSwede]

Well, I have a Core i7-11700 with a B560 motherboard, and I haven't experienced any of the "buggy mess" every media outlet was writing about at launch. I think it all comes down to 1. motherboard choice and quality, and 2. the media writing about every small thing like it was the end of the world just to gain readers/viewers.

It's easy to hate on the latest Intel products without trying them first hand. I've had several Ryzen 3000 and 5000 series CPUs, as well as my 11700 right now, and I'm actually happier with Intel. It is easier to cool in a SFF case with limited airflow, its BIOS options are much clearer and easier to understand, it doesn't need any stupid chipset software to run properly like the Ryzen 3000 series does, and it behaves much better during idle by just adapting to Windows power settings.

Edit: About the "paying beta tester" mentality: I agree that it's at play here, and it's wrong. People tend to advocate choosing a fresh platform with a fresh socket for long-term compatibility. As for me, I tend to go with the last generation of a specific platform to make sure I get something that's well-tested by the public and all issues have been ironed out.

I think you forget about one thing, precedence. Intel has always been known as the company that delivered really stable platforms at launch and it's pretty unheard of that Intel launches a really buggy platform. Have there been buggy boards from its partners in the past? Hell yes, there have been some truly horrible boards, but it was never due to Intel.

AMD on the other hand has a history of buggy product launches, some worse than others, so it's less of a surprise when it happens again and again.

What you're saying about the UEFI, well, some of it comes down to the fact that Intel hides a lot of knobs, so you can't turn them, most likely because there's no reason for you to turn them, whereas AMD left every knob and dial user accessible, for better or worse. You do in fact not need to touch most of it imho, so it's not a huge issue. Some of it also comes down to the motherboard vendor, as I can't stand the current mess of a UEFI UI/UX that MSI has created. It's just impossible to navigate and the layout makes zero sense.

As for paying beta testers, it's gotten a lot worse over the past, maybe five years, to the extend that almost every company ships half finished products from a software perspective and in best case it gets fixed six months down the line or so, worst case, never. In fairness to both Intel and AMD here, they do a lot better job than some companies.

 

[AusWolf]

I think you forget about one thing, precedence. Intel has always been known as the company that delivered really stable platforms at launch and it's pretty unheard of that Intel launches a really buggy platform. Have there been buggy boards from its partners in the past? Hell yes, there have been some truly horrible boards, but it was never due to Intel.

Still, I haven't experienced any of the bugginess everyone was so on about. My experience of 11th gen has been stable as usual. The only new thing for me is when I change some settings in the BIOS and save it, I get a hard restart. If things like this are issues for some people, they shouldn't buy a PC, really.

AMD on the other hand has a history of buggy product launches, some worse than others, so it's less of a surprise when it happens again and again.

It's true that they've improved a lot lately. I remember memory support on Athlon 64 was horrendous, and the way whole motherboards (chipset + VRM) overheated on FX was unheard of. Though I also can't forget the disaster that Pentium 4 was.

What you're saying about the UEFI, well, some of it comes down to the fact that Intel hides a lot of knobs, so you can't turn them, most likely because there's no reason for you to turn them, whereas AMD left every knob and dial user accessible, for better or worse. You do in fact not need to touch most of it imho, so it's not a huge issue. Some of it also comes down to the motherboard vendor, as I can't stand the current mess of a UEFI UI/UX that MSI has created. It's just impossible to navigate and the layout makes zero sense.

My Asus Tuf B560M-Plus doesn't hide any of the knobs. It just sorts the things that 99% of users don't need into menus like "advanced CPU settings" and has them on auto. The Asus Tuf A520M-Plus and B550M-Plus that I've tried on the other hand, have CPU settings in the overclocking menu, advanced menu, and there's things like AMD CBS and whatnot that aren't explained anywhere, and they're mostly just duplicate options that nobody needs to be there. All three BIOSes are stable and fairly logical, but the B560M-Plus wins for me because it doesn't have duplicate options in several different menus.

Also, Intel's power options are simpler and more logical. You also don't have to play with them to suit your cooling (at least on 65 W CPUs), which is a huge plus.

As for paying beta testers, it's gotten a lot worse over the past, maybe five years, to the extend that almost every company ships half finished products from a software perspective and in best case it gets fixed six months down the line or so, worst case, never. In fairness to both Intel and AMD here, they do a lot better job than some companies.

That is unfortunately true. That's why I never adopt a platform too early. You can also find better deals on platforms that are slowly being phased out.

 

[Noreng]

What you're saying about the UEFI, well, some of it comes down to the fact that Intel hides a lot of knobs, so you can't turn them, most likely because there's no reason for you to turn them, whereas AMD left every knob and dial user accessible, for better or worse.

This is just wrong, AMD doesn't expose nearly as many settings to the user as Intel does. However, Intel also allows motherboard manufacturers to tweak and adjust a lot more than AMD.

AGESA locks down a lot with AGESA, the most obvious one being tREFI, but control over stuff like Precision Boost is also very lacking. On Intel you can adjust Turbo Boost up or down, AMD allows you to raise the PBO frequency by 200 MHz (with minimal control over the V/F curve with curve optimizer)

 

[TheLostSwede]

Asus Tuf B560M-Plus doesn't hide any of the knobs. It just sorts the things that 99% of users don't need into menus like "advanced CPU settings" and has them on auto.

How do you know this when they're hidden from all users? Intel BIOSes and UEFIes have always had features hidden from the end user that we've not been able to access. You may think there are things that aren't hidden, but there are plenty of knobs and dials Intel doesn't want consumers to mess with.

 

[zo0lykas]

So now we get every day a new post about alder Lake?
Price
Benchmark
Packed
Ram speed
What next?

 

[TheLostSwede]

This is just wrong, AMD doesn't expose nearly as many settings to the user as Intel does. However, Intel also allows motherboard manufacturers to tweak and adjust a lot more than AMD.

Yes, Intel allows the board makers to change settings that the end users don't have access to, AMD not so much, although board makers can clearly hide further options as we've seen.

AGESA locks down a lot with AGESA, the most obvious one being tREFI, but control over stuff like Precision Boost is also very lacking. On Intel you can adjust Turbo Boost up or down, AMD allows you to raise the PBO frequency by 200 MHz (with minimal control over the V/F curve with curve optimizer)

You sure all this is actually implemented in a way that it could be user accessible? The AGESA isn't exactly a great piece of software from what I understand.

So now we get every day a new post about alder Lake?
Price
Benchmark
Packed
Ram speed
What next?

Didn't bother with all the pictures...
If you're not interested, don't read it, problem solved.

 

[AusWolf]

How do you know this when they're hidden from all users? Intel BIOSes and UEFIes have always had features hidden from the end user that we've not been able to access. You may think there are things that aren't hidden, but there are plenty of knobs and dials Intel doesn't want consumers to mess with.

Point taken. Let's just say that there's no setting that I miss from my motherboard's BIOS.

 

[docnorth]

@TheLostSwede wrote"...as there have been leaks of a board from Gigabyte with memory clocks of 8,000 MHz, so either ASUS is playing it conservatively on its lower-end boards, or these are not the final overclockable memory speeds for these boards."
BTW those 8000 MT/s are controlled by Gear 2, if I'm not wrong. Gear 2 or 4 and latency are going to be a hot topic until (or even after) launch. Anyway it's very good to already see 6000-6400 RAM speed with more or less the standard timings and 8000 with primary timing 50.

 

[TheLostSwede]

Point taken. Let's just say that there's no setting that I miss from my motherboard's BIOS.

Which is fair enough and I think that applies to most, if not all of us.

 

[Noreng]

@TheLostSwede wrote"...as there have been leaks of a board from Gigabyte with memory clocks of 8,000 MHz, so either ASUS is playing it conservatively on its lower-end boards, or these are not the final overclockable memory speeds for these boards."
BTW those 8000 MT/s are controlled by Gear 2, if I'm not wrong. Gear 2 or 4 and latency are going to be a hot topic until (or even after) launch. Anyway it's very good to already see 6000-6400 RAM speed with more or less the standard timings and 8000 with primary timing 50.

Note that 8000 was achieved with a single DIMM, som it's hardly representative of what we can expect.

I'll be surprised if we can push more than 7500 with 2 DIMMs