Jump to content

LmnSour

Member
 View Profile  See their activity

  • Posts

    82

  • Joined

  • Last visited

Awards

This user doesn't have any awards

Recent Profile Visitors

The recent visitors block is disabled and is not being shown to other users.

LmnSour's Achievements

Apprentice

Apprentice (3/14)

31

Reputation

1

Community Answers

  1. LmnSour
    Performance investigation on the intel 13900HX using Warhammer II benchmarks. Theory: Due to limited power available, and with the GPU able to dynamically boost up to 175W, reducing CPU cores should enable more GPU power and higher boost clocks, not only directly, but indirectly by reducing the heat load and fan speeds. Laptop: Eluktronics Mech 17 GP2, 13900HX, RTX 4090, 32GB DDR5 5600 CL40 and 6400 CL38 Tests done: 6-P cores and 4-E cores with 5600 and 6400 Ram vs all cores (8p and 16e) with 5600 and 6500 ram. Ran the GPU undervolted at 2335 and +900 core. Only did one run, no averages; I'm not getting paid for this Ran the Warhammer 2 benchmarks Battle, Campaign, and Skaven, at Ultra settings, DX11, and 2k resolution with AA disabled an Gsync/vsync disabled. Max fans with the LPP watercooling and used Throttlestop to keep P-core clocks at 5200-5300 and E core clocks at 3900. P-core undervolted by .145v and e cores by .130v. First, ram benchmarks: 5600CL40 1.1v 6400CL38 1.35V --------------------------------------------------------------------------------------- For the results we have the individual results plus the averages of the 6e, 4p and the averages of the 5600 and 6400 ram speeds: Battle Benchmark: *Discussion: I was not expecting the 6400 memory to perform so poorly here. I think the issue with this benchmark is the power draw and the DDR5 6400 memory at 1.35V might be reducing the available power budget to the CPU/GPU? Cant be sure but I could try in the future to set the fans to auto and see how that affects performance in this benchmark. But we can clearly see that disabling the two p-cores and 12 e-cores netted a performance improvement. Campaign Benchmark: *Discussion: This is about what I was expecting; disabling 2 p-cores and 12 e-cores definitely saw an improvement in performance and the higher ram speed also saw a considerable improvement. Skaven Benchmark: *Discussion: Here we see the opposite of the Battle Benchmark where the ram speed saw an improvement but the extra cores helped more than the extra power budget. I think this benchmark is more CPU intensive than the other two (Skaven battles always throttle my CPU) and the extra cores and extra memory bandwidth certainly saw a large improvement. I might experiment with 8 p-cores and 4 e-cores. I know inevitably someone will ask to disable all e-cores but TBH, I like the idea of some e-cores for background applications. I'm trying to figure out how to overclock this laptop because there is a LOT of headroom in the boost clocks. For some reason, clocks seem to be capped at 5.3.
  2. LmnSour
    Quick GPU overlcock. I'm stuck at 2563 core and memory loses stability around +1350 - 1400. Fans never spun up the entire time testing but TBH, the run never lasted more than a minute. I don't know if this is game stable and haven't done a 10min run; I was just seeing how far I could push it with reckless abandonment.
  3. LmnSour
    Some R23 testing. Warm up loop conducted prior to each run and fans were allowed to spin down to make sure the heatsink was heat soaked and fans returned to normal operations. HWinfo max/mins reset prior to each run. A longer heat soak was given for the water. Its not exactly scientific but its a good indication of the performance the extra cooling gives. No overclock (AVX Offset enabled), 32GB DDR5 5600 CL40 ram, "Beat mode" (I think this is just all power limits removed), LM TIM replaced with PTM pads, and thermal pads replaced with putty. Auto Fans only: 30522 Max Fans: 31121 Water with Auto fans: 31699 Water with Max Fans: 31603 Looks like we hit a clock wall and more cooling wasn't doing much. Time to "overclock". For the Overclock, I basically disabled AVX offset (usually this results in thermal throttling for no benefit) using Throttlestop and enabled Speedshift but I don't think that helps in R23. Water with Fan on Auto: 34156 Water with Fans on Max: 33989 Curiously, I think the Max fans reduce the power available to the CPU, dropping the R23 scores. With the water and auto fans, the R23 run is over before fans reach 50%. I'll do some GPU overclock and gaming next.
  4. LmnSour
    The LPP comes with the laptop but its not needed: You can read up on the Eluktronics Mech-17 here: https://www.notebookcheck.net/Eluktronics-Mech-17-GP2-laptop-review-MSI-GT77-Titan-performance-without-the-size.699099.0.html
  5. LmnSour
    I travel a lot and use my laptop setup in my hotel. I'm a fan of desktop replacements. This works very nicely, can carry it in my backpack. I don't (yet) take the LPP on travel. At home the kids use it so keeping it quiet was a big win but admittingly it was a bit too much in the end.
  6. LmnSour
    I got an Eluktronics Mech-17 13900HX, RTX 4090. After a ton of research, I settled for this laptop mostly for the Prema mod bios, allow full undervolt and overclock options, including memory and memory timing. But not going to lie, I was also excited about the Liquid cooler! For those not familiar: Water cooling a GeForce RTX 4090 laptop: Eluktronics Liquid Propulsion Package G2 review - NotebookCheck.net Reviews But when the LPP cooler arrived, I was less than impressed and honestly can't recommend this laptop. I don't NOT recommend it, but I don't quite recommend it either but that's another story. So to my modding adventures: First issue was the Aluminum radiator. Eluktronics says to run just distilled water but that's just begging for corrosion, that radiator needed to go. Taking that out wasn't hard but finding a copper radiator that could fit was not easy. It had to be exactly 120mm in width and not many 120mm radiators fit the bill. I was able to find the OEM company that made the Alu core radiator and to my luck, they made a copper core version: https://www.amazon.com/gp/product/B09CDCWRY4/ref=ppx_yo_dt_b_search_asin_image?ie=UTF8&psc=1 (its currently unavailable here) ------------------------------------------ Next issue was harder. The pump was loud and the software control sucked. It offered no control and regulated the pump and fan speed based off temp so the pump would ramp up and down and sounded worse than the laptop fans. Piling it on, I'm pretty sure the laptop would lower fan speeds and your custom fan curve no longer worked, so your overall temps were not as good as they could be unless you forced the fans on max. Solution: Alphacool DC-LT pump https://www.amazon.com/gp/product/B01H0KV6GO/ref=ppx_yo_dt_b_search_asin_title?ie=UTF8&psc=1 Alphacool Eisstation 40 DC-LT Pump Top/Reservoir https://www.amazon.com/gp/product/B081HGW89K/ref=ppx_yo_dt_b_search_asin_title?ie=UTF8&psc=1 I'm not going to belabor you the absolute, total PITA this was and will go ahead and recommend you NOT TRY THIS. I'm also not going to post how much it cost me, to include killing the control board on my original LPP and having to buy another one... @ $250... mostly because I don't even want to know the cost. Wasn't worth it, but once I started it, I couldn't stop. "In for a penny, in for a pound." Here it is: Original (from Notebookcheck.com): Test Fit (version 3 IIRC): All back together with fill/drain port: (Yes, the front plate JUST BARELY fits) Leak test: Results (I don't know exactly what it usually idles at but I think 50C; this is just a quick snap after initial boot; I just needed to make sure the water was flowing): --------------------------------------- So that's hardware but the software side of things was also a PITA. The laptop connected to the LPP and controlled it via Bluetooth and as I mentioned above, user control was not a thing. The original hardware has a flow indicator and the software uses the indicator to know the pump is working. If there is no flow, the software will fault and shut down. Now the issue was that the Alphacool pump needed more voltage than the original and when the laptop was idle, the software would slow down the pump, in this case, causing the Alphacool to shut off (or spin too slow for water circulation), ultimately causing the software to fault. But I found a fix that allowed me to remove the flow indicator, control the laptop fan speeds independent of the LPP pump and fan, run the pump and radiator fan at a constant speed, and it runs ABSOLUTELY QUIET to boot (the only issue is I forget to turn it off because it runs so quiet -bsitting 2 feet from me, I don't hear it). To run the LPP in manual mode, click the LPP power button to put it into pair mode then push and hold down the button for about 3-5 seconds which runs it in, I guess, override / manual mode at a constant speed. I think it's just running at 12V, so max speed. Thank goodness I also changed the fan to an Arctic P120. Phew! At least after almost losing my sanity and blowing more money than I care to admit I have something to show for it! More results pending. I have 5 kids and a needy wife so it might be a while before I can get around to posting results.
  7. LmnSour
    I like the 7.1 setting on my Soundblaster X3. Its not "perfect 7.1" but I've found it to accurately portray sound direction in Call of Duty Using Drop + Sennheiser Pc38x headset.
  8. LmnSour
    Ive been thinking about this for a while. This isn't the first time someone called out LTT for ethics or accuracy and I've watching LTT make mistakes for quite a while now. Why is Gamers Nexus, who defended Linus's integrity and accuracy before, now making this video? Was whole thing an objective piece of journalism and positive feedback from one professional to another, for the benefit of the community? Did we all end that video enlightened or did we end it thinking LTT = Scumbags? Steve did released the video with no financial benefit to him and his channel, right? Or were their other motives?
  9. LmnSour
    LmnSour LinusTech

    Linus, I know this is sucking.  Steve ego smeared you under the cover of "objectivity" and the credibility of his subjectivity and smearing was bolstered by his, well, facts...  Unfortunately, he's right. And worse, he attacked you and brought in a lot of your employees and put you, and them, in a very bad spot.  The knee jerk reaction is to lash out with emotion; which, unfortunately I think you already did.  And Steve pounced on that.  It's pretty despicable but unfortunately he's credible.

    Please try to make a professional and objective video to apologize (if it needs to be done) and present a way to fix some of the obvious process issues but do it without ever mentioning Steve or Gamer's Nexus.  Just my humble opinion.

    1.   Show previous replies  12 more
    2. SamKirk

      SamKirk

      2 minutes ago, NCSGeek said:

      I can't see how the video comes across that way to you. The data is misleading to consumers and therefore is a bit of a problem. I've watched GN for years and he's always seemed very genuine and not money/fame hungry.

       

      As for the cancelling/woke statement, this definitely isn't that. I even more so cannot see how this is like that. I hate woke stuff and cancel culture, but this is the same overuse of the terms that discredit them as problems.

       

      All of this is blown out of proportion (except for the Madison stuff if it's true) because people want drama or to fight for their "team".

      Fair response I guess, however, I will point out that Steve is in fact very brand aware and is afterall a business person with overheads. He has quite successfully carved himself out a niche in tech reporting with a fantastic focus on details and accuracy however he also uses this borderline OCD level of detail to invoke his "cancel culture" powers. Lets take a look at his biggest money earners from YT shall we...

      Slamming of ASUS
      Slamming of NewEgg
      Slamming of LTT

      Look, he has found that going after "the man" is earning him some serious views and spending the time he has to launch a campaign against LTT is not done out of care for you, its done out of a double win for him, he gains more followers at the same time as damaging LTTs following and it being such a hot topic his views are through the roof.

      And before you jump in and say "he didnt monetize the video" did you notice he did for the second one? If it was done out of pure care then why do it.

    3. NCSGeek

      NCSGeek

      12 minutes ago, SamKirk said:

      Fair response I guess, however, I will point out that Steve is in fact very brand aware and is afterall a business person with overheads. He has quite successfully carved himself out a niche in tech reporting with a fantastic focus on details and accuracy however he also uses this borderline OCD level of detail to invoke his "cancel culture" powers. Lets take a look at his biggest money earners from YT shall we...

      Slamming of ASUS
      Slamming of NewEgg
      Slamming of LTT

      Look, he has found that going after "the man" is earning him some serious views and spending the time he has to launch a campaign against LTT is not done out of care for you, its done out of a double win for him, he gains more followers at the same time as damaging LTTs following and it being such a hot topic his views are through the roof.

      We'll probably have to agree to disagree. The Newegg thing was definitely big, in that case what initiated it was they personally wronged him. They then talked about it and ultimately got a really insightful talk with some of the company heads.

       

      Point being I don't see GN as living for this kind of "drama".

       

      17 minutes ago, SamKirk said:

      And before you jump in and say "he didnt monetize the video" did you notice he did for the second one? If it was done out of pure care then why do it.

      Sure. Though that wasn't a dedicated video, just a segment in the usual news series. Feels more like a "we saw the forum post, here are some points we'd like to address from it before we drop the topic and move on".

       

      I feel like even if he didn't monetize that news video then people will say it doesn't matter since he "gained followers and will monetize future content so he still benefited from this".

    4. LmnSour

      LmnSour

      4 hours ago, NCSGeek said:

      I can't see how the video comes across that way to you. The data is misleading to consumers and therefore is a bit of a problem. I've watched GN for years and he's always seemed very genuine and not money/fame hungry.

       

      As for the cancelling/woke statement, this definitely isn't that. I even more so cannot see how this is like that. I hate woke stuff and cancel culture, but this is the same overuse of the terms that discredit them as problems.

       

      All of this is blown out of proportion (except for the Madison stuff if it's true) because people want drama or to fight for their "team".

      You are absolutely right.  There are issues with LTT that need to be corrected but there are many ways GN could have done this.  He ignored the issues for a long time, even defended LTT from similar accusations in the past, and only when LTT launches "Labs" does he make this video... He also invested $250k into his own labs; because he see's new competition.  He's creating a facade of "no monetary gain to GN" (he started the video with that), helping the community, helping the billet startup, etc... all to put him on a moral and ethical soap box so that he could smear Linus. 

       

      Look at all the out cry, the 200k+ subscribers LTT lost in two days.  GN's video was not an objective video of "helping LTT do right."  Everyone that watched that had came away with the same conclusion: Linus is a scumbag.  

       

      And one more thing while Im on my own soap box.  I utterly reject Steve and all the other elitist tech reviewers that think they are God's gift to the community.  As if I can't think on my own and utterly rely on his God blessed talents.  I appreciate his work but I reject the premise that the proverbial "community" is so easily duped by bad tech influencers.  That we need to be saved from ourselves.  Steve, I hope you read this.  I enjoyed your past work on data collection and reviews but I don't need "saving" from LTT or any other Tech reviewer... 

  10. LmnSour
    I've been noticing errors on a lot of videos and it was only time for 60 Minutes, uh GamersNexus to post a video. But to say Steve's video was "purely objective" is questionable. It came off as an ego filled smear video like all of his "objective" reporting videos. Self righteous smearing. Ending the video with *shrug* "that's my thoughts on this..." was an attempt to deflect responsibility of the video by Steve. He knew what he was saying... But I think what LTT need to be mindful of is this delusion that permeates society of "you are good at your job by virtue of being hired to it." LTT employees need to understand what they are reporting and saying and need to be a bit less judgemental in their videos IMHO. They should be more mindful of objectivity, professionalism, and just getting things right. And I agree with Steve. If its wrong, it probably should corrected in the manner he prescribes.
  11. LmnSour
    The issue is the IHS. I'm running a supercool computers direct die block, a 420, x2 360's, and a 140mm radiator, and x2 D5's but also have an RTX 3090 STRIX with the 1000W bios which draws 600+ watts at load so... 13900k with SP Performance core score of 107 (I think 100 overall) Overclocked to 60, 60, 60, 60, 59, 58, 57, 57. I could have gone further but my PSU (Seasonic 1300 PRIME) was running out of watts... At this overclock, I got random hard crashes in games (PSU trip) until I undervolted the GPU. I run it stock with an undervolt. At 5.7 all core, 300 max watts. The loop isn't heat soaked but I generally get temps to level out in the lower to mid 70's in R23 after looping it. If you want, this is a 12th gen block and I just received a 13th gen. I can send you the 12th gen for your own testing and you won't have to wait for shipping from Asia. As you can see, works just fine on a 13th gen CPU. Btw, for those wondering, this is all stock, ASUS Multicore Enhancement disabled, and undervolted. Not a huge difference between the two except for 100+ additional watts.
  12. LmnSour
    With the high RTX memory temperatures there has been a ton of discussion on thermal pads but also there is a new (to me at least) and not very well known thermal putty that claims 10 w/mk (TG-PP10) sold by digikey (TG-PP10-50 t-Global Technology | Fans, Thermal Management | DigiKey) A thermal comparison between the leading thermal pads, thermal putty, and a pros/cons of both would be great. I've grown to appreciate the ability of the putty to be used with laptops particularly as higher performance thermal pads tend to be too dense and either don't conform well enough to allow consistent proper contact while also not impeding the GPU/CPU contact with laptop coolers (and full cover waterblocks). I would be very interesting to see a comparison and weather or not the 10w/mk claim adds up.
  13. LmnSour
    Going back through the post, maybe the SMD's you soldered back on are not properly soldered and they are only intermittently working. Maybe getting a proper repair done would fix it.
  14. LmnSour
    NO WORRIES BROTHER! We're all here to help; I've received my share of online help and support and its no problem to pass it on! The fact that it ran eliminates a bad fuse so that's good. We're making progress! A few questions: when it worked, how long did it run, how long was it able to maintain a continuous load, what were the temps, and how well did it run overall? If it ran just fine, its probably not a GPU issue. Maybe its not getting proper voltage at initial startup? When you turn off your computer, do you keep your power supply on and plugged in? Are you getting clean power to your house? Is you PS good quality? Are all your power connectors properly seated into the GPU and powersupply? Is your PCI-E connector fully and properly seated? Just FYI, on power, Tripp Lite makes power conditioners and are highly recommended whether you are having issues or not (Amazon.com: Tripp Lite 1800W Line Conditioner, AVR Surge Protection, 120V, 15A, 60Hz, 6 Outlet, 6 ft. Cord, 2 Year Warranty & $25,000 Insurance (LC1800): Home Audio & Theater) The other issue could be your GPU. Maybe a bad cap on the GPU that's not providing proper voltage? Maybe there is some dust buildup, liquid metal, or some other foreign object that is bridging/shorting something on your card? But if it ran just fine but its just not getting properly recognized at initial boot, I smell a bad power supply, power supply connection issues, or dirty power from your house. One thing you can do if you have a second computer is try the GPU on the other computer to see if the problem persists. If so we know its a GPU issue. If not, we know the issue is with something else in your main rig (probably power supply).
  15. LmnSour
    Could be the integrated GPU as mentioned above.
×