Intel "LGA1700 Alder Lake" Bendgate, Thermalright has the solution.

[darknessblade]

Summary

 Thermalright launches Bending Corrector Frame for Alder Lake processors.

 

Quotes

Quote

Intel desktop Alder Lake processors are prone to bending or warping, an issue caused by a pressure applied by the latching system for LGA1700 socket. The bending problem causes the integrated heat spreader (IHS) to bend and lose the uniformity, which may lead to lower cooling efficiency by creating a gap between the cooler and the processor.

 

My thoughts

As I do not have a Intel Alder-lake CPU, I cannot really respond to issues regarding warping, or know how bad the issue actually is.

 

Sadly according to the Toms Hardware article you will void your "warranty" to fix a MANUFACTURER issue, trough a 3rd party part.

Which is quite ironic, as Intel does not come with a solution themselves, and downplays the actual issue.

 

It just shows how little Intel cares about consumers, that Thermalright/other brands, can do it with a bracket that cost less than 7$ [on Taobao] Which is in my opinion a better solution than a 3d Printed Bracket, which modders have made, or Intel saying that the warping issue is "near Non-existent", and does not require a fix.

 

 

Sources

https://videocardz.com/newz/chinese-company-came-up-with-anti-bending-frame-for-intel-alder-lake-s-processors

https://www.cowcotland.com/news/80999/thermalright-lga1700-bcf-un-accessoire-pour-empecher-les-processeurs-alder-lake-s-de-plier.html

Intel responds to warping/bending issues:

https://www.tomshardware.com/news/intel-comments-on-alder-lakes-warping-and-bending-issues-mods-void-warranty

https://www.rockpapershotgun.com/intel-play-down-alder-lake-cpu-bending-strongly-recommend-against-socket-mods

 

For those that do now want to wait for local stock.

you can order it the exact product now on aliexpress.

[Please note the link only serves as a example, I cannot guarantee that the store can be trusted]

http://aliexpress.com/item/1005004194894526.html?

[Moonzy]

19 minutes ago, darknessblade said:

Intel does not come with a solution themselves, and downplays the actual issue

Well, technically the CPU still runs within spec, so I agree with their statement

 

And I also treat their "warranty void" like overclocking, who would know if you don't tell em?

Average consumers aren't really affected, and overclockers that look for every ounce of performance already "voids" the warranty the moment they buy fast RAM and apply XMP, so modding the socket isn't too far off

[Blademaster91]

16 minutes ago, Moonzy said:

Well, technically the CPU still runs within spec, so I agree with their statement

So did their CPU's using toothpaste under the IHS, doesn't mean its something that should happen though.

[porina]

Is this a real problem? That's not to say it might not happen in rare cases but if it was really widespread you can be sure action would be taken. Feels like this is yet another case of clickbait articles and others trying to profit from it by offering solutions that are unnecessary outside of niche situations like extreme overclockers.

[Moonzy]

1 hour ago, Blademaster91 said:

So did their CPU's using toothpaste under the IHS, doesn't mean its something that should happen though.

Does the CPU run out of spec, though?

 

I mean, if you're not happy with it (I certainly am to a certain degree), u can always buy the competition, assuming they're doing better

or "void" the warranty -shrug-

[MageTank]

57 minutes ago, Blademaster91 said:

So did their CPU's using toothpaste under the IHS, doesn't mean its something that should happen though.

Gonna quote past MageTank here since he knows more about this than I do:

On 7/31/2017 at 3:41 PM, MageTank said:

@zMeul is right. First of all, the old linus video. That is exactly the result you see when you use a thermal paste that is terrible for thermal shock (especially under intense thermal cycles of bare die heat loads, both MasterGel and Noctua's stock paste are known for failing under this load) and secondly, he didn't spread the paste, and simply expected the IHS to do that for him. Your socket retention mechanism on the board absorbs the vast majority of the mounting pressure, so the IHS itself will not be able to evenly spread the paste across the die. Either way, linus did still see a per-core variance drop in his tests (meaning the cores max temperature differential was less after delidding) but he failed to mention it.

 

I've delidded many CPU's too. More than I am able to count, and have used various kinds of paste, ranging from CLU to Gelid, and I can tell you, the difference is less pronounced when you use a solid silicone based paste and spread it properly, vs liquid metal alone. Granted, LM is still far superior, offering a 10-15C difference over Gelid, but it's still not the 20-25C difference seen when removing the glue in the process. Not only that, but I've also purchased a gigantic jar of Dow Corning paste (the exact same blend used on Intel CPU's from Haswell and above) and compared to Gelid, it was only 2-3C difference. Meaning it still dropped temperatures substantially compared to Intel using the exact same paste along with their gluing process.

 

If you need further evidence that the process itself was at fault, look no further than X299. Yes, not soldering that platform's SKU's certainly resulted in higher temperatures, and a bottleneck at the die->IHS heat transfer, the per-core variance on those chips are extremely low. Almost identical to what we see when delidding/using solder. This means that the paste is perfectly even this time around, and I blame this on the raised substrate in which the die is raised on an additional substrate layer, rather than be on the same layer as the glue itself. 

 

TL:DR? Liquid metal is still the best way to go for thermals when delidding, but stock paste wasn't the reason people needed to delid out of necessity. High per-core variance caused by the gluing application process caused higher than tolerable per-core temp variance, and people had to delid to fix that. The paste itself (Dow Corning TC-1996) while not as great as liquid metal, was still perfectly fine for even temps had the gluing process not impact the application of the paste.

 

The TL:DR is that Intel's stock paste at the time of Haswell/Skylake/Kaby Lake was Dow Corning TC-1996 which is really solid paste used in a ton of commercial applications due to its durability. The difference between some of the best and worst silicon based pastes is seldom ever greater than 2-3C. The thermal issues on Intel's older chips were a result of their adhesion process that caused small gaps in between the die and IHS to form while curing. We could see this when we removed the IHS as well as Intel's attempts to use more thermal paste as a means of making up for this mistake.

 

I've tested in great detail the thermal performance of TC-1996 by simply delidding a CPU that had it, cleaning it off, applying a new coat of it and attaching the IHS without any glue. Still saw a 10C drop on my old 6700K (not quite the 18C from liquid metal, but an improvement nonetheless).

 

I am all for memeing on Intel when they deserve it, but insulting arguably one of the most stable thermal pastes in our industry just seems silly to me, lol.

[LAwLz]

2 hours ago, darknessblade said:

I cannot really respond to issues regarding warping, or know how bad the issue actually is.

2 hours ago, darknessblade said:

Intel does not come with a solution themselves, and downplays the actual issue.

So let me get this straight.

You don't know how serious the issue is, yet you say that Intel downplays the issue? I feel like you, and a lot of reporters, are arguing in very bad faith here.

 

 

I don't know how big this issue is either, but after reading all of the sources thoroughly, except the one in French, I got the impression that it is not that big deal of a deal.

It may warp slightly, but even if it does it should only result in a couple of degrees rise in temperature. Of course Intel will say "modifying your socket will void the warranty". I mean, does anyone actually expect any different? I don't think that means "Intel doesn't care about their customers" or that "Intel are threatening people".

 

I can see how some people will be worried about their motherboards though, and I would say that if you are worried then getting this third party plate might be a good idea. Intel will say it voids your warranty, but it's not like they will be able to tell. 

[MageTank]

9 minutes ago, LAwLz said:

So let me get this straight.

You don't know how serious the issue is, yet you say that Intel downplays the issue? I feel like you, and a lot of reporters, are arguing in very bad faith here.

 

 

I don't know how big this issue is either, but after reading all of the sources thoroughly, except the one in French, I got the impression that it is not that big deal of a deal.

It may warp slightly, but even if it does it should only result in a couple of degrees rise in temperature. Of course Intel will say "modifying your socket will void the warranty". I mean, does anyone actually expect any different? I don't think that means "Intel doesn't care about their customers" or that "Intel are threatening people".

 

I can see how some people will be worried about their motherboards though, and I would say that if you are worried then getting this third party plate might be a good idea. Intel will say it voids your warranty, but it's not like they will be able to tell. 

I've investigated this quite thoroughly in our lab and found that boards that primarily use the Foxconn ILM (Most ASUS boards) saw almost zero difference by adding 1mm washers to the ILM retention mechanism. The LOTES design however, did see a decent 5C drop in temps on our 12900K non-delidded sample with this mod alone. We primarily find this design on MSI and Gigabyte boards, as well as our ASUS Prime series samples. For overclockers, 5C is a pretty significant chunk of thermal overhead, but I don't see this making or breaking performance at stock. You'll see far more significant gains by fixing Intel's aggressive LLC settings on these boards. RobertoSampaio wrote a fantastic guide on this for those of you curious about taming Alder Lake thermals: https://rog.asus.com/forum/showthread.php?127565-ASUS-MAXIMUS-Z690-EXTREME-amp-i9-12900K-GUIDE-Load-Lines-VF-Curves-Adaptive-Voltage. TL:DR? You can use LLC and AC/DC inputs to force an adaptive undervolt on your processor that is dependent on load. On our 12900KF sample, I dropped 20C and 50W off my testing while pushing our sample 100mhz higher at 5ghz all-core vs the 4.9 all-core boost. 800 point MC boost in R23 as well.

 

Do I think Intel needs to address this socket warping "issue"? Not really. This is more of a tolerance issue with the designs between the ILM manufacturers and their slight deviations based on Intel's specs. It's not significant enough to induce thermal throttling for most people under normal operating conditions. If you are overclocking, you'll hit thermal barriers due to the poor solder and aggressive voltage curves long before this IHS warping becomes an issue. Besides, people buying 12900K's are likely buying water blocks, which are naturally bowed by design to meet the mating surfaces of these CPU's. Flat air coolers would suffer more from a bowed socket than a convex cooler would when attempting to mate to a concave processor. Ask the people that lapped their Heatkiller Pro IV coolers and they'll tell you using a flat cooler on a non-flat processor is just an awful idea. Just don't lap these Intel CPU's without doing the socket mod, as they'll likely continue to warp a little beyond your initial lap, so you have to wait until the shape of the IHS no longer changes before lapping.

[Lightwreather]

2 hours ago, darknessblade said:

For those that do now want to wait.

I mean by the time it ships and ends up in your hands, I doubt Alder Lake would even be relevant /s

[darknessblade]

1 hour ago, J-from-Nucleon said:

I mean by the time it ships and ends up in your hands, I doubt Alder Lake would even be relevant /s

Aliexpress is at most 60 Days delivery time, depending on the seller. {at least that is in the EU}

[grg994]

Isn't this all the same problem that overclockers solved with the "washer mod", now months ago? (detail on that:

 

[darknessblade]

16 minutes ago, grg994 said:

Isn't this all the same problem that overclockers solved with the "washer mod", now months ago? (detail on that:

 

It is indeed similar to the Washer mod, but this is a more professional looking mechanism.

 

Which in my opinion looks much better, and might work even better than a few "plastic" washers

[YamiYukiSenpai]

I wonder if someone will test this out and report back 6 months later, or somehow artificially simulate what it'll be like half a year from now.

[darknessblade]

50 minutes ago, YamiYukiSenpai said:

I wonder if someone will test this out and report back 6 months later, or somehow artificially simulate what it'll be like half a year from now.

If that is the case someone would need 3 identical setups.

 

Where

1 is the normal one

1 is the washer mod

1 is the Thermalright solution

 

Then you can really know what the best solution is, in the long run.

[Quackers101]

so my CPU can go full kamikaze on me? delid itself and hurt itself in it's confusion? should I just pour liquid metal all around it and hope for the best?

*metal pool of hot silicon babes*

[Falkentyne]

So where can you buy this in the USA?

[darknessblade]

5 hours ago, Falkentyne said:

So where can you buy this in the USA?

Not in actual stores. but you can buy a different "unbranded" model trough Aliexpress.

 

[Guest]

Intel processors warping seems way too trivial in the first place - just use more toothpaste! thermal paste!

I do not understand the whole point of this entire product other than, IDK, extreme overclockers.

[Dogzilla07]

Der8auer is doing a similar, 3x more expensive, but more advanced bracket, with alternative-mode for easy lapping via the Thermal Grizzly brand

 

 

[darknessblade]

22 hours ago, Dogzilla07 said:

Der8auer is doing a similar, 3x more expensive, but more advanced bracket, with alternative-mode for easy lapping via the Thermal Grizzly brand

 

If even Thermal-Grizzly is getting into the market, then the problem might be more widespread than we thought.

 

Though mostly only the users who build custom PC's are the ones getting it, the tech-N00Bs who buy a prebuild, will never think of getting this or even know what its purpose is, Those users often do not care about a 5 degrees Celsius difference. as they are not prone to fully loading their CPU for a prolonged time.

[porina]

20 minutes ago, darknessblade said:

If even Thermal-Grizzly is getting into the market, then the problem might be more widespread than we thought.

I'd still question if this can even be considered a real problem. Der8auer had sold de-lid tools before. Do you consider those CPUs to be faulty? I see this as no more than offering hard core overclockers a possible option to squeeze out a bit more performance regardless of the cost.

[RejZoR]

Like Der8auer said, this is for extreme overclockers where even 50Mhz more can mean a difference between world record or second place. For most casual users, all the deviations in surface flatness are well within acceptable levels and shouldn't affect day to day use or experience.

 

In 99% of cases, using capable enough AiO and best thermal paste and it's fine. Hell, sometimes you make compromises for the sake of convenience. I know I'm avoiding liquid metal because I had nothing but problems with it and would rather have slightly higher temperatures and have none of idiocies that come with liquid metal. Like idiotic staining, metal diffusion and just hassle of cleaning it. I don't care how much better is with temperatures, I'm not using that crap ever again.

 

Also as side note, people need to stop applying thermal paste with those stupid X and pea size dots in the middle patterns. You're not applying thermal paste every 3 days, don't be lazy and spread it fully across entire surface of IHS. Any excess will be pushed out and that's perfectly fine. It will guarantee to fill in all inperfections and flatness deviations without exceptions. That's the point of thermal paste.

 

[Falkentyne]

56 minutes ago, RejZoR said:

Like Der8auer said, this is for extreme overclockers where even 50Mhz more can mean a difference between world record or second place. For most casual users, all the deviations in surface flatness are well within acceptable levels and shouldn't affect day to day use or experience.

 

In 99% of cases, using capable enough AiO and best thermal paste and it's fine. Hell, sometimes you make compromises for the sake of convenience. I know I'm avoiding liquid metal because I had nothing but problems with it and would rather have slightly higher temperatures and have none of idiocies that come with liquid metal. Like idiotic staining, metal diffusion and just hassle of cleaning it. I don't care how much better is with temperatures, I'm not using that crap ever again.

 

Also as side note, people need to stop applying thermal paste with those stupid X and pea size dots in the middle patterns. You're not applying thermal paste every 3 days, don't be lazy and spread it fully across entire surface of IHS. Any excess will be pushed out and that's perfectly fine. It will guarantee to fill in all inperfections and flatness deviations without exceptions. That's the point of thermal paste.

 

Some people are getting overheating at stock turbo boost settings because of this and are not trying to get world records.

[Moonzy]

48 minutes ago, Falkentyne said:

because of this

Any example?

How's the rest of their setup like?

[Falkentyne]

1 hour ago, Moonzy said:

Any example?

How's the rest of their setup like?

I don't have the links and I can't be bothered taking all that time to search for them, buried in threads like that, I apologize but I have work to do.

But I can tell you It was on the Asus and 12900K forum sections with several users on OCN (overclock.net), and with one user having problems with a waterblock and getting proper IHS contact (very bad thermal paste spread and contact with pictures) until he exchanged his Apex for another board (I don't know if it was a MSI or Gigabyte) but as soon as he exchanged it, there were no temp problems at all.