Intel X299 Platform Called a "VRM Disaster" by Overclocker der8auer

[RejZoR]

I agree. It's also sad that now AMD has very competitive products, lets face it, Ryzen CPU's are amazing, they still aren't making good enough traction because of initial quite honestly, flawed reviews and shit taken by the Intel fanboys who still swear to vastly inferior but highly clocked quad cores (like 7700K) just because they posted better scores in games that were essentially specifically written for them. But there is no freaking way in this world that 500-900MHz higher clock can offset a lack of 4 physical cores (and 8 freaking threads!). That's like missing a whole quad core CPU! Ryzen has a tiny clock penalty with single threaded or lower threaded apps and games, but even as things stand now, Ryzen is not far behind. And if you buy CPU for long term and lets face it, we usually do, Ryzen will fare far better for far longer with more moderately clocked octa core than Intel will with higher clocked quad core. Games are already going past 4 cores in usage and it'll just get even worse as time goes on.

 

[erixx]

All AMD's fault, for turning the markets into a PR deathmatch!



PD: and for delivering amazing stuff and staying undead.

 

[dj-electric]

@buildzoid This is where i get concerned. Would you like a picture of the X299-A board i have?

 

[Nihilus]

Tomorrow wins!

True.

 

[buildzoid]

Board makers didn't fuck up (at least not everything). Intel (and AMD) have to be VERY clear and strict about minimum specs for the VRM. Board makers have to strictly obey that. They can go crazy and overdo the VRM segment of their boards, but they are NOT allowed to go below what Intel or AMD specify for their processors. Only fuck up by the board makers is the fact that Intel has now given them single board that takes anything from crappy quad core up to insane 10 cores. This means VRM needs to be bonkers on all boards even if they'll be primarily bought by budget people for their "high end" quad cores. If board makers try to save things on their own here, that's another thing. It's also possible Intel did fuck up. They can have loads of $ to throw at anything, but rushing things can lead to mistakes even if you have billions to throw around.

Intel's spec sheet for the X series says boards need to handle 73A continuous current draw on VCCIN. If the mobo makers are building for that it's no surprise that the boards can't handle overclocks. The VRMs on X299 should be built to handle around 150-200A 2V continuous(basically they need better cooling) if you plan to max out a 10 core.(imagine what an 18 core will do once OCed)

 

[atomicus]

So eager, they released the fastest consumer CPU ever made, on the best HEDT platform ever, with better single core performance than almost every other CPU (except for their own i7-7700K), and did it at a much lower price than the last generation. You call that a trainwreck? I call it brilliant! Granted, they rushed the release by a couple months, and this thinned out the number of motherboards that were available at launch. But are you really heartbroken about that? Were you planning a knee-jerk upgrade from your present X99 rig? I'd wait for the better motherboards, myself, let the platform mature a few months. Are you upset because they run so much faster than X99 CPUs (10-15%), or because they cost so much less (and you feel that you paid too much for your X99 system)? I'm not trying to be a smartass, I really am curious as to why a few reasonable, intelligent people are having such a negative reaction to X299, which seems all out of proportion to any perceived shortcomings. Also, nobody is forcing anyone to buy this platform. I just don't understand your position, except for seeing that it pushed some kind of emotional button in you, not a considered rational response. Okay, it's not perfect - please tell me a better alternative (past, present, or known future).

There is so much wrong with everything you've said I don't even know where to begin. Fortunately I don't have to as Tomorrow has done a damn good job of ripping it apart already.

Seriously, you come across as some sort of paid for Intel shill.

 

[buggalugs]

Everyone is jumping to conclusions based on one guy on the internet nobody has heard of.

I doubt there is a problem with the VRM design, or the heatsinks. Its most likely a very raw, unoptimized bios. These boards were rushed out, motherboard makers didnt have time to optimize the bios and the voltages.

My guess is an updated bios will solve this issue.

 

[Caring1]

Everyone is jumping to conclusions based on one guy on the internet nobody has heard of.

I doubt there is a problem with the VRM design, or the heatsinks. Its most likely a very raw, unoptimized bios. These boards were rushed out, motherboard makers didnt have time to optimize the bios and the voltages.

My guess is an updated bios will solve this issue.

Hahahaha, so you've never heard of him, too bad, so sad.

 

[dj-electric]

Everyone is jumping to conclusions based on one guy on the internet nobody has heard of.

No. Its one guy you didn't hear about.
He is an overclocking expert, and many world records holder.
The guy probably knows more about hardware than the people who make it

 

[R0H1T]

Everyone is jumping to conclusions based on one guy on the internet nobody has heard of.

I doubt there is a problem with the VRM design, or the heatsinks. Its most likely a very raw, unoptimized bios. These boards were rushed out, motherboard makers didnt have time to optimize the bios and the voltages.

My guess is an updated bios will solve this issue.

Yeah when was the last time you saw an updated BIOS solve a leakage issue as bad as this ~ https://www.computerbase.de/2017-06/intel-core-i9-7900x-test

It;s almost as bad as the FX 9570/90 except that was on 32nm. The BIOS won;t fix the leakage or high temps, that's dead certain.

 

[Tomorrow]

Everyone is jumping to conclusions based on one guy on the internet nobody has heard of.

I doubt there is a problem with the VRM design, or the heatsinks. Its most likely a very raw, unoptimized bios. These boards were rushed out, motherboard makers didnt have time to optimize the bios and the voltages.

My guess is an updated bios will solve this issue.

You mean der8auer is a guy no one has ever heard of? Talk about suprising statements. Granted not everyone knows him but he's still fairly known and respected by people who keep tabs on overcloking or just general hardware news. He know's what he's talking about. I mean the guy works for (or with) ASUS for god's sake. http://hwbot.org/news/12661_overclocker_focus_roman_hartung_der8auer_interview

Also i'd like to see how they fix bad VRM-s with a BIOS update. The only way to do that is to disable all voltage controls that could affect VRM temperature or maybe even undervolt (tho that would affect stability even at stock speeds). Essentially killing overcloking ability. Then again Intel might fly in and say that people should not overclock their unlocked processors right?

 

[RejZoR]

Everyone is jumping to conclusions based on one guy on the internet nobody has heard of.

I doubt there is a problem with the VRM design, or the heatsinks. Its most likely a very raw, unoptimized bios. These boards were rushed out, motherboard makers didnt have time to optimize the bios and the voltages.

My guess is an updated bios will solve this issue.

"Jumping to conclusions based on one guy on the internet nobody has heard of..."

I suggest you watch this video taken from ASUS RealBench tool where you'll see one of the worlds most prominent and respected extreme overclockers and extreme modders... You may spot a familiar face and name through the video...

 

[dorsetknob]

Everyone is jumping to conclusions based on one guy on the internet nobody has heard of.

Seriously you have never heard of one of the Best/world record holder Overclocker

I doubt there is a problem with the VRM design, or the heatsinks. Its most likely a very raw, unoptimized bios. These boards were rushed out, motherboard makers didnt have time to optimize the bios and the voltages.

Solved by expensive Board redesign ( No FREE RMA Owners will just have to suck it up) Or new T&C restricting what you can install

My guess is an updated bios will solve this issue

I very much doubt any Bios upgrade will solve anything but the most Trivial faults

Conclusion
Perhaps you should Delete your Ill Informed Opinionated Post ( my Opinion )

 

[Rahmat Sofyan]

"We do not recommend running outside the processor specifications, such as by exceeding processor frequency or voltage specifications, or removing of the integrated heat spreader (sometimes called ‘de-lidding’). These actions will void the processor warranty.”

Intel already told us guys ...

 

[atomicus]

"We do not recommend running outside the processor specifications, such as by exceeding processor frequency or voltage specifications, or removing of the integrated heat spreader (sometimes called ‘de-lidding’). These actions will void the processor warranty.”

Intel already told us guys ...

Yes, correct, Intel told everyone who bought a CPU marketed for overclocking not to overclock it.

 

[Rahmat Sofyan]

Yes, just for fun fact and cheers up guys...

 

[Duality92]

They need to make functional heatsinks again, like the ones used on the ASUS X99M WS

 

[Hood]

Tomorrow wins!

If you call negativity a win - all my positive points are valid - depends on your viewpoint, I guess. I'll never understand how people get all bitter and angry because some product is better or worse than some other product, even things they don't own and never will. EVERYTHING is better or worse than something else in some way, so I guess you could be bitter and angry about every aspect of life, if that's how you want to be. I prefer to enjoy life...

 

[Solidstate89]

Looks like this is the year of rushed MOBO/platform launches. Can't wait to see what happens with Threadripper and Coffee Lake at this rate.

 

[Frick]

If you call negativity a win - all my positive points are valid - depends on your viewpoint, I guess. I'll never understand how people get all bitter and angry because some product is better or worse than some other product, even things they don't own and never will. EVERYTHING is better or worse than something else in some way, so I guess you could be bitter and angry about every aspect of life, if that's how you want to be. I prefer to enjoy life...

You have clearly said X299 is the best HEDT platform ever though, on multiple occasions iirc, and you define why it should be regarded as such.

 

[EarthDog]

I have the prime x299 deluxe... the vrms were warm on top running p95 at stock... didnt check the bottom. I also ran at 4.5ghz, vrms were toasty, but nothing to worry about. This was on an open air bench without direct airflow.

They need to make functional heatsinks again, like the ones used on the ASUS X99M WS

LOL...

 

[FR@NK]

CPU segmentiting forcing you to buy atleast a thousand dollar CPU to get the full 44 PCI-E lanes (wich is still 20 short of Threadripper)? Check

Dude you've drank too much of AMD's cool-aid!

7900x paired with x299 will have more PCIe lanes then threadripper.

Threadripper CPU has 48(24x2) PCIe lanes split between two separate dies. The x399 chipset adds another 16 lanes which brings the platform total to 64 lanes. But you also lose 4(or maybe 8, need to do more research) lanes because they are used to connect the cpu to the chipset; so now you are at 60(or 56) usable lanes. Because of how the lanes are spread out between the dies; you can only get 2x full 16x PCIe slots.

The 7900x has 48 lanes minus 4 lanes for the DMI link to the chipset so 44 lanes. Then add the 24 lanes from the chipset and you get 68 total usable lanes.

source

 

[buggalugs]

I dont care if hes a great overclocker, I lost interest in those guys 10 years ago. Im not really interested in unattainable overclocking records that are completely useless other than as marketing for mainboard manufacturers. I've only ever been interested in 24/7 overclocking.

We'll see how it pans out. I doubt all the manufacturers have the same problems with VRM at the same time. Motherboard makers usually choose VRM on their own boards, and for highend boards, they are usually overkill in specs. I dont believe there is a hardware problem, but we'll see.

 

[_JP_]

They need to make functional heatsinks again, like the ones used on the ASUS X99M WS

No, but seriously:

A top-down cooler should work fine, but who wants to use one of those these days.....besides me...

 

[Duality92]

No, but seriously:

A top-down cooler should work fine, but who wants to use one of those these days.....besides me...

I have a few, like the small Phanteks one, the Raijintek Pallas, but they're not the easiest coolers to mount. They also are more performance limited than most tower coolers.