*** Official Clevo X170SM-G/Sager NP9670M Owner's Lounge ***

I don't think they're preparing a mxm, they'll barely be producing mobile bga 6800s

 

Technically the 5700M in the Alienware 51m R2 was on a modular, proprietary Dell board. There is also a modular 5500XT/5700 in this eGPU I think (have not seen any disassemblies to be sure though).

My hope is that this generation Apple will not be selling anything with AMD mobile GPU's so hopefully AMD can start producing more for PC laptops instead of just the handful of models as it has been in the past.

 

Here is a good one. I have 5 laptops the size and weight of the X170sm-g and 4 x 49 inch curved wide screen monitors. Based on weight requirements, who makes combination laptop / monitor arms that will support that? I am thinking 1 arm per 2 laptops and 2 monitors.

So I would need 2 or 3 arms total. based on 5 laptops and 4 screens. I need them off my desk.

 

You look like the perfect guy for this:
https://www.acmeportable.com/products/notepac-iii-prov

 

This thing looks awesome! Unfortunately it'll probably also cost a pretty penny.

 

That is really cool, but it does not solve my problem...lol

 

That's down to Nvidia who could do with some more competition.

 

They haven't had any competition in the mobile space for more than a decade, and I am not optimistic that AMD is going to bring any. They're just barely doing so on the desktop graphics and we know their efforts on mobile will be half-hearted because everything almost everyone seems to do with respect to laptops is half-hearted, half-assed, or both.

 

Is there any need for added precaution in case the liquid metal "spills" out of the cpu die? ( That is if such a thing could happen if one is careful enough not to use tons of it )

 

Yes. Many threads on this issue such as this one: http://forum.notebookreview.com/thr...and-liquid-metal.806840/page-49#post-11088800

 

I was hoping for something a bit more direct than a 50 page thread, however I do appreciate it.

 

Quick pointers: coat surrounding area with nail polish or electric tape, and use a dam (foam is best, many use another, viscuous non-LM paste)

Also good to watch: https://www.reddit.com/r/hardware/c...aste/?utm_medium=android_app&utm_source=share

 

Keeping in mind less is more is also useful.

 

You mean they don't bother with so few models that can handle 200W graphics? Yeah, AW, Asus and Msi haven't models anymore that can handle 200w TGP. Game over. All of them offer only thin Jokes nowadays. Clevo alone is probably to small asking for 200W vBios for Ampere graphics.

 

If you apply an excessive amount or your laptop is subject to moderately rough handling, then there is a risk of leakage. Not really from under the IHS... that is possible but very unlikely... but from where the heat sink makes contact with the IHS and/or GPU die.

 

AMD GPU’s right now are the definition of “Participation Trophy”. It’s funny to me that people still keep thinking they’re gonna come out with something that’s gonna slaughter NVIDIA. With the track record they’ve had I’m surprised they still try.

 

For mobile, AMD GPUs would be very good since they have a better perf per watt than nVidia. Unfortunately, they no one seems interested in doing MXM modules with AMD cards

 

How about a intel mxm inside this clevo, they have a intel partnership since forever anyway..

 

If performance per watt is important I guess it could be good. Lack of features, modest overall performance, resistance to overclocking/tuning, inferior driver support and inconsistent reliability make them generally undesirable.

 

Exactly just like their cpus. That's why we need a intel mxm, atleast they ain't treating enthusiasts like some kind of alien.

 

It's a shame but that's what I guess.

 

I think it is pretty obvious that at Nvidia there is currently no desire to produce a really good notebook GPU and it possibly makes business sense but it is not the honorable thing to do.

With Nvidia being such a big outfit that is probably swimming in money right now they should strive to offer maximum performance (within reason) for those who desire it. To reduce mobile card TGP compared to the previous generation when in the same generation TGP is going up in desktop cards certainly does NOT qualify as a serious effort.

 

What are they competing against though? They can already sell every chip too.

 

The problem is the lowest common denominators are the group defining "performance" as the term applies to laptops. In their sick, twisted definition that is defined by lack of size and weight and how long it can run between charges. Actual performance and horsepower aren't even on the radar because they are incompatible with their dystopian pseudo-smartphone aspirations for laptops

 

Exactly - this is why I said it would be the honorable thing to do. You know, produce something great that you can be proud of instead of mobile chips that fall further behind their desktop counterparts for the second generation in a row.

 

Oh I’m sure they’re proud of these mobile cards. “Look how much we can charge for these things! And it’s blatant false advertising!”

 

They were always going to be further behind regardless but yeah, 200W + options would have been nice.

 

Remember when Epic showcased Unreal Engine 5 on the PS5?

The CEO of Nvidia commented that the PS5 is “only” as powerful as a laptop RTX 2080.

The comment was intended to insult consoles, but it actually insults us, too.

It goes to show how Nvidia regards laptops.

 

This hurts and confuses my brain on so many levels that I don't even know how to respond to that. . .

 

Jenson has always been an interesting guy...

 

I think that was very helpful and it shows that they are very much aware how much they crippled their mobile cards by now.

 

Eh, it's more the TDP difference that's doing that.

 

Welp, Amazon were having a sale on the 10850k.

Would be rude not to. So I did.

Thinking of getting it delidded after it arrives and i've checked it. Never had a delidded and liquid metal cooled CPU before, so is there anything I need to be careful of? Are the copper IHS's still a problem with this machine?

 

This topic of copper ihs has been discussed around 3 trillion times in this thread before. Better use the search feature.

 

There are indeed multiple ways these are limited. Still I am pretty sure that the mobile 3080 performance would scale very well up to 200W TGP.

 

It would be interesting to see that scaling and it should be an option.

 

I already did. I found YOU saying "it's bad, don't do it" and someone else saying "I had it done and it's fine."

And that's why I asked since some posts are old and some aren't.

EDIT: I think I might skip the copper and stick with the normal IHS.

Anything I need to do/ask for when having it delidded? As mentioned, i've never had a delidded CPU before and don't want to let the magic smoke out. Service i'm looking at using apparently coats it in clear nail polish to prevent shorts and then re-seals with silicone. Will that be enough?

If this goes well, i'll get the 10700K done as well but i'll have that replaced with copper since that'll be going in a desktop.

 

The normal IHS is still copper (nickel plated) by the way. It's more about height and shape differences.

 

That's honestly what concerned me. Shape/height as i've seen some folks say it causes issue in the laptop, and others say it's fine.

So I think i'll stick with the original.

 

With mine I have about the tallest you could get but it works well in my machine.

 

And your machine haven't Unified heatsink If the unified heatsink fits well on both the Cpu and Gpu and you add etc 1mm taller custom IHS on the Cpu it will disrupt the Gpu cooling. Same also for thermal pads.

 

Thankfully my unified heatsink already had a 1mm gap between heatsink and stock IHS, so a 1mm taller BartX IHS was perfect.

 

Indeed. I remember the custom IHS you had made by Rockit, which was pretty awesome after Fitz polish!

 

Just had an ideea. what if in the non vc p8 heatsink we drill the plate where the mounting holes for the Gpu Pcb are + little bit longer longer screws... let the pcb screws hold in place the gpu and the heatsink iteself (would need to remove the xbracket of the rtx 30 series to not interfere with the heatsink). might be also doable with the vc if we can somehow fix it in place tight enough (without drilling holes in it of course)

 

I don't think that would be possible because half the mounting holes for the GPU PCB are outside of the GPU coldplate itself.

 

You would need some sort of adaptor plate to do it really well I think.

 

Hi. I've got a X170SM-G from 3XE in the UK with the i9-10900K. It updated the Windows to 20H2 just after initial build. It has v1.07.03 BIOS firmware.

I have the CCC power util set to "Performance" but the CPU runs very slowly. It seems capped at x41 all core even for single core loads. I run Cinebench R23 and on a single core task the CPU sits there at x41 (4.1GHz) when it should go over 5GHz on Core #0.
On a multicore run, it slows to between x41 and x36 but the CPU is cool (77C all cores) and the fan spins to about 80%. There are no thermal or power limit flags set in HWInfo and the package power is only 110W peak - it's not using the PL1 or PL2 limits at all - and 109W steady run.

Is this some problem with CCC (and the Intel XTU behind it) or the Intel DPTF software being overly protective?

I had the machine delivered and it developed a keyboard fault. Sent it back for repair and they changed the whole chassis. The old one ran much faster than this one but the old one didn't update to Win 20H2. It also had older BIOS on it v1.07.02

Anyone else seeing this?

I've noticed the Clevo web site has a newer BIOS just released (April 9th v1.07.04) anyone tried this out?

 

I do, and have run it for weeks now, tbh, I prefer xmg's , but to be fair they both pale in comparison to prema's

 

Use Throttlestop

 

Throttlestop can help, but there is much more that needs to be done to tame these beasts without sacrificing performance. On my 9900K, I tweaked TS as much as I could. But in the end, there was simply too much heat being exhausted to the chassis.