electrosoft
-
Posts
2,599 -
Joined
-
Last visited
-
Days Won
110
Content Type
Profiles
Forums
Events
Everything posted by electrosoft
-
I haven't found someone who tested it in both standard and water cooled version yet. They are priced cheaply on Aliexpress now though so I've been mulling over ordering one to see. https://www.aliexpress.com/item/3256802893647618.html?pdp_npi=2%40dis!USD!US %2436.00!US %2428.80!!!!!%4021031a5516542938709864495efa22!12000025851981212!sh&spm=a2g0o.store_pc_home.productList_2002555959075.pic_0 Ah, ok. I use Phobya too but I think I'm going to pop the LTX back in since it is already delidded with the Nickel plated BartX IHS and run it a few days with Phobya again and then a few days with LM and a foam perimeter to note the differences. My heatsink was already previously used in a LM mount from Ztec so it is well soaked as is. I just ordered a more precise, lower range torque screw driver too recommended by a few fellow Pi robots buddies I collab with who use it for all their precision assemblies for a better price (rebrand of many other models): https://smile.amazon.com/gp/product/B09YXYHVFS/ref=ppx_yo_dt_b_asin_title_o00_s00?ie=UTF8&psc=1#customerReviews to really dial it in. I might just stick with the LTX depending on results and sell the SL since the SL is in pristine condition (never delidded) with original box, SL paperwork, etc. I only want to hold onto one of them and the LTX was close enough. Basically, whichever one sells I'll keep the other.
-
True, I found this out early on. Basically eating your own chow. The blessing and the curse of using your own screws is you don't need shims or taller IHS's. The curse is of course dialing in the optimal amount of torque. Get a manual (or powered) screw driver with torque settings. Dial in the optimal settings for your unit for maximum pairing pressure: https://www.amazon.com/CDI-61NSM-Torque-Micro-Adjustable-centinewton/dp/B000JXPF12/ref=cm_cr_arp_d_product_top?ie=UTF8&th=1 @Clamibot are you running standard TIM or LM on your CPU?
-
I like they gave the CPU a full 4 post mounting but even looking at their past 15.6" designs for desktop CPUs for both Intel and AMD I have no idea how this was their final boss heatsink especially looking at that early prototype. Here's hoping Aliexpress has a good aftermarket design like those that have shown up for the X170, P870 and P77X series. -
BGA issues aside.... My issue is looking at the prototype for this at the trade show versus the final product is a slap in the face in regards to the heatsink. The original heatsink shown was much beefier for the CPU. Whoever decided to go from this: To this: Needs to be evaluated.....unless it is a bean counter then we know why. Who in their right mind evaluated the 12th gen performance and heat generation and decided this was an acceptable solution is mind boggling.
-
VMD is a bane when trying to clean install new BGA laptops from Acer too. Their pre-installs come with it enabled but as soon as you nuke the drive it disappears except in the BIOS. You have to disable it there too.
-
*Official Benchmark Thread* - Post it here or it didn't happen :D
electrosoft replied to Mr. Fox's topic in Desktop Hardware
So are the bulk of you happy or sad about the AVX512 being disabled? I'm indifferent as it didn't make or break my leisure or workflow 🤷♀️ -
Very nice bro! Definitely post all the juicy stuff! 😁
-
*Official Benchmark Thread* - Post it here or it didn't happen :D
electrosoft replied to Mr. Fox's topic in Desktop Hardware
AMD looks like it is coming in guns ablazing and while Intel will retain the single core performance crown, AMD will win at everything else. What is the logic behind fusing off AVX512? -
2. Yep. 4. Glad the info helped out!
-
Price: *** SOLD *** Condition: Used Warranty: None Reason for sale: Too many 10900k's, Many other toys to buy Payment: PayPal (non-cc) , F2F, trade Item location: Southern NJ Shipping: USPS International shipping: To known former NBR users Handling time: 3 days Feedback: https://www.ebay.com/fdbk/feedback_profile/electrosoft Specification: LTX Golden Sample (1 of 200) that has been delidded and re-lidded and ready to rock. Asus rates this as an SP96. There is one discolored contact pin on the bottom that does not affect performance. Relidded w/ liquid metal on 7/15/2022 and a 10 min CB23 stress run pictured 7/18/2022 showing those sweet, sweet temps under that pull. Tested on an MSI Z590 w/ EVGA 360mm CLC and Gelid Extreme. Price includes USPS Priority shipping CONUS Proof of ownership:
-
Or a cherry picked unit was sent his way.
-
MSI's version of the 12950HX hitting 24140 @ ~160w / 95c -50uv in MSI BIOS out of the box in CB23:
-
So how goes the comparison with the NH55 vs the Scar? Leaning one way or the other yet? I suspect the Scar cost much more but the one stand out feature I do like is the 17.3" display. I wish they offered a version with the full fat HX but 3060 at a much reduced cost. Any side by side pics?
-
Welcome! 1. It has definitely been fixed and improved. You have to exert a little bit of force to get them to come out now. 2.780w will work but some of the old reports on the old forums said it had a not so great design and was a bit on the dirty side. 3. Problems with running 4 sticks at 3200 is a common problem. Locked down BIOS is another. 4. 12th gen Clevo is a 15.6" model for desktop CPU supported laptop here: Apparently the 17.3" X270 is BGA across the board. 5. Prema was supposed to be in the works but it looks like it is DOA due to many issues with the system implementing it. Prema is available on select X170SM-G models.
-
*Official Benchmark Thread* - Post it here or it didn't happen :D
electrosoft replied to Mr. Fox's topic in Desktop Hardware
To add to this, here is what it looks like when you pull the heatsink off after a K5 application. This is before clean up and with a LTX Golden 10900k with a BartX nickel plated IHS installed. It is a wash what sticks to what either the heatsink or the system but like I said before clean up is pretty quick with a TIM spreader. -
*Official Benchmark Thread* - Post it here or it didn't happen :D
electrosoft replied to Mr. Fox's topic in Desktop Hardware
If I know I'm doing temp test mounts when testing various CPU stuff on my X170SM-G I will leave the K5 Pro as is, but when it is my final remount I will use a thermal paste applicator and clean up the surrounding squeeze out and reapply it. Takes about 10 min tops. -
17400 is definitely with e-cores off as in an earlier post he said he got 23400 at full tilt. It definitely is a touch better than the NH55 atm but it is using slightly higher power limits and on paper and visual inspection has a much better cooling system so we will see what happens as tweaks are made. There is also the random luck of silicon too which can't be user addressable on BGA laptops.
-
I'm lean more towards real application testing like the software you're using now. Have you tested your software on a well cooled desktop vs Prime95? On a laptop it could be a case of Prime95 forcing a more brutal sustained throttle scenario via a combination of heat or pull (or both) vs your software giving the impression your software is hitting it harder and hotter when in actuality it is not. I've seen this happen before as Prime95 overwhelms a laptop. What is the name of your software? Which Clevo do you use?
-
WTH............. That is an insult to the Xx70 line if that is their successor.....
-
Prime95 is a near waste on desktops and a super waste on laptops unless the idea is to smash it into the ground with unrealistic loads. Ranks right up there with running Furmark to stress test your GPU.
-
Zen and the art of binning aka why silicon quality counts in constrained laptops 5 10900k's. 2x off the shelf 1x ZtecPC binned 1x LTX Golden Limited Edition SP96 1x Silicon Lottery SP101 After much testing on the desktop to quantify and qualify them: EVGA CLC 360mm AIO + KPx TIM used for all desktop testing. The two garbage chips off the shelf had their stock lids. They weren't worthy of a delid The ZtecPC Binned was re-delidded and re-lidded by yours truly so real world temps may have been a touch higher if stock lid. The LTX was already delidded so I cleaned it up very professionally and relidded it with a BartX so real world temps may have been a touch higher if stock lid. The Silicon Lottery CPU still has its stock lid and has never been modified On the desktop, the Ztec, LTX and SL all could reach a full -0.130 uV and pass a 10 min CB23 (water equalized) run no problem. Desktop uV != Laptop uV since they run much hotter so laptop uV will always be lower. With NO delid, the SL ran cooler and pulled the least even with the Ztec and LTX being delidded. Testing in the X170SM-G: Measured center dollop of Nanogrease Extreme (using gram scale) for equal weight. Randos: I knew what the results were going to be with the off the shelf randos in my X170SM-G but popped them in there anyhow. Both immediately thermal throttled and couldn't handle > -0.055 uV without Bsod during a 2 run CB23. Poor silicon will have you falling behind right out of the gate in DTRs. ZtecPC binned 10900k: The established baseline binning criteria set by @Mr. Fox for ZtecPC shows why as with a min requirement of 5.1/1.35v or 5.0/1.31v V/F curve and -0.090 uV requirement they could pass CB23 at stock but temps did creep into the 90s. I'm sure if you went LM you could bring that down even more (as ZtecPC did sell it with this option). LTX 10900k: A clear step above the ZtecPC binned 10900k and closer to the SL as temps were in the high 80s during runs and could be done with ~ -0.100 uV SL SP101: The king but the LTX was right on its heels. Able to sit in the mid 80s with a ~0.100uv for CB23 run. I'm sure with a delid it would run even cooler/better but it runs so well as is and it is a rare 10900k that right from the factor core temp differentials are 5c max which is rare. I don't feel a need to delid it. Conclusion: LTX 10900k is a worthy competitor and clearly bests all the other chips but that SL SP101 is magic. If I didn't have the SL chip and just the LTX I would be just as happy with it they are just that close enough. On the desktop one thing I noticed is the LTX really stabilizes under Vdroop when benching and didn't fluctuate as much as the other chips. I'm so used to staring at the Vcore under load and watching the fluctuations that to see it sit there and barely fluctuate but lock in much better was odd. Maybe that was part of the Intel binning process for those chips? (less transients). @jaybee83 @Clamibot
-
Looking forward to a full head to head Scar vs Raptor....
-
Awesome! I'll go take a looksy. Looking forward to the results and higher level breakdown from you vs the standard YT affair.
-
Score another one for ZtecPC. While breaking down my SM to test my Silicon Lottery 10900k vs an LTX Golden Sample 10900k I decided to remove the tape/gunk and it was already removed and cleaned. They really did their homework looking to optimize the SM as much as possible.
-
Welcome Nico and thank you so much for adding more data and BIOS to the "maximize this mini beast" party! It is greatly appreciated. Agreed. So far, we have your PCS version and now Nico's Clevo-Computer version. Still need Eurocom's. Hopefully you and Prema can figure out how to get the Unicorn running free in the field! I keep loading up Eurocom's configurator page fighting the urge to order one. I wish they would offer a barebones version. I'll wait for a used model to hopefully pop up at a good price on eBay or similar. 🙂
