All Activity

I'm happy to hear that you are making progress but sad that you are having to go through it.

I think my RAM sticks are maxed out at 2667.

I want it in GPU-Z. This is me from one of the very best days I have had the last two months. Today going out and eat (outside my preferred place). I don't know if I will be the same as before. It will take time. Loads of time before I will/start feel well with myself. I feel broken🙁 They say it takes at least a year before you can feel more confident. The first 3 months after the stroke is critical. After one year you can start feel safer and try come back to yourself. The progress goes slow but steady forwards. Thanks for asking my good friend🙂

Timespy Dev wrote that the 5080 causes CPU overhead issues due to their code. (so it is not ideal to test with) Just tried 48x @ -0.120V and came in similar. Not sure if because I have steam also running since that's how I got the demo. CPU at 85.6* Fans are set to full speed.

just a follow up to my question - apparently you can order 2026 Grams with US ANSI keyboard from amazon.com, they do ship to Europe

As it turns out B&H had an open box 9955HX3D + 5090 model for $3375. Supposedly that config was discontinued by MSI as they ran out of Hynix 32GB modules (Micron and Samsung have compatibility issues with Fire Range).

Makes sense. I would love to compare/contrast the much cheaper V-color CL26 sticks but they're both pricey endevours. Still mulling it over, but those cores aren't great on second glance. Very wide range of voltages and pretty beefy on some of them. IMC is fantastic though overall. It really is a sample focused on fine tuning memory as the primary goal. I wouldn't put anything past Nvidia especially after they purposely killed off MXM and removed SLI as an option on consumer level cards. We're seeing them in true boss form right now in the worse way. It's a combination of being a reviewer, work (albeit that is mainly a lot of Mac hardware nobody cares about in these threads), and sometimes being asked my true assessment of particular hardware from a few NDA sources who apparently value my opinion....who whudda thunk it? 🤣 I will say this, expect a lot and I do mean a lot of cut down options (16GB + 512GB) on models that normally would come with 32GB + 1TB standard on some mid tier models wise and expect a lot of 2024/2025 rehashing of CPUs (14th gen Intel, 8000 series AMD) paired with low to mid (5050->5070) 5000 series GPUs for grossly inflated pricing in the land of laptops. We're already seeing refreshed 290hx models on the top end going for $5-6k which is absolutely mind boggling but even worse we'll be seeing rehashed 14th gen models with mid tier or lower 5000 series GPUs (5060 even) going for $2k to start. If I was in the market right now for a decent laptop, I would look to last year's inventory clear out and then look for open box or used on top of that to save a major bit of coin. Same 5000 series and for gamers, the CPU refreshes aren't really bringing anything major to the table. Right now, with things in the pipeline as they are, the 9955HX3D Raider 18 is a good option if you can lock down a sub $3k variant. The 5090 version is quickly entering unicorn status it seems. Hopefully not a situation of Nvidia influencing them to not add it as a readout. Now we need MSI AB to add it in for their OSD. How ya feeling @Papusan? Hopefully things are moving in a positive direction for ya.

Not sure why GPU-Z is so slow NVIDIA RTX 50 Series Hotspot Temperature Readings Are Back Through HWMonitor Utility https://wccftech.com/nvidia-rtx-50-series-hotspot-temperature-readings-are-back-through-hwmonitor-utility/

I still use my laptop every day at home and taking it with me to work or when traveling. My wife has now switched to either using the desktop connected to our TV or the Lunar Lake MSI Claw 8. Its a shame that Thin and Light dominates the market. I have heard Nvidia prefers those types of designs and could even refuse to supply their chips into something that they deem would go against their marketing direction like the old X170 or GT75. Then we have the BIOS lockdown cartel where even if you have the right combination of hardware you may not be able to tune it. It is cool that they sent those laptops to you for free. Is that a deal through your work? The pricing is getting pretty crazy, especially when you get into these preconfigured models where there is no option for less RAM or SSD. 9955HX3D Raider 18 is definitely a good option if you can find a deal. B&H sometimes has good pricing on open box MSI models if they also have a sale active on the new model. I still think Fire Range is superior to Arrow Lake when it applies to mobile platforms. Not even 290HX Plus can really boost 1% lows up with how difficult it is to run tuned SODIMMs stability.

@Will Short answer : Yes and yes Yes, the dGPU will be used a a video accelerator selectable in windows. This allows even GPUs with no display outputs at all such as MXM gpus like the quadro tesla M6, P6, P106m, etc, to work in MXM based laptops Yes, I had put a P3000m in my zbook 17 g3 and I would get no dipslay out if I forced dGPU in the vbios, the reason being that zbook had eDP on DP_C and the card had eDP on DP_D. What I ended up doing was bridging the 3.3v and PNL_ON and PWM of the MXM slot and then forced the graphics mod to dGPU only in the vbios to get a video out. I then had a picture on the screen with no way to adjust brightness, it was stuck at 100%. Using an eDP cable mod v1 or v2 will be much simpler (and less stressful) than soldering direclty to the pins. HOWEVER, nvidia drivers will decide "screw that" after they install and will disconnect the dGPU from the screen, usually ~10 seconds after windows loads and the driver starts. SuperMG found a workaround in windows to force the screen to stay on after the drivers install luckily, it is detailed in his "eDP cable mod" PDF instrutions.

The thing that is even more stupid than Samsung doing this is the reality that some people will fall in line, like good little sheeple, and do it. The will rationalize the decision because they "have nothing to hide" or because it is "for the greater good" but at the end of the day everyone else knows they are idiots that are a danger to themselves and everyone around them.

Ready to send over your data to help Samsung bro @Mr. Fox🙄 Me buying an Samsung smartphone? Nope. Not in hell. Hell will freeze first. Samsung is holding your health data hostage to train its AI Samsung Health now requires users to consent to AI training on their health data or lose access to it, raising serious privacy concerns

Regarding the GPU, it's known that the ZBook 15/17 from G1 to G3 (and definitely the G4 as well) have eDP on DP_C. Therefore, if a graphics card with eDP on DP_D were installed - let's say an RTX 4050, for example - I wouldn't get a display on the internal monitor if the SBIOS were configured in discrete/dedicated graphics mode, correct? But if the option were changed to hybrid graphics, would I be able to use it normally with the video passing through the iGPU? And another question: if the GPU has standard DP on DP_C and I use the backlight cable mod, I would get a display on the internal monitor and could use it in discrete graphics mode, correct? @JadeRover@SuperMG can you guys give me some insight on this please?

EDIT: here are some updates after opening the keyboard and accessing the primary memory modules: - Update 1: the laptop had a 32GB SODIMM in DIMM_A. I installed another 32GB SODIMM into the other primary slot - DIMM_C. The laptop then booted just fine (as expected) with total of 96GB. - Update 2: I then relocated the other SODIMM from DIMM_B to DIMM_D, and the laptop still boots fine with 96GB seen. I just acquired a used Dell Precision 7760 laptop from eBay. It's working fine, so I thought I would upgrade RAM. The laptop came with a SODIMM module installed in "slot 3" (as per CPU-Z) which I don't know whether that's DIMM_A or DIMM_C, but it's definitely one of the 2 primary module slots under the keyboard. I have not gotten to opening up the keyboard to access the primary module area, so I thought I'd put an extra 32GB matching SODIMM in DIMM_D (which is on the much more conveniently accessible bottom section of the laptop). The laptop did not boot, with 2 amber lights and then some white lights (sorry, forgot to count precisely). I then switched the new SODIMM into DIMM_B on the bottom side, and the laptop boots and I see 64GB installed. BIOS did report only 1 memory channel active so my guess is that the primary slot corresponding to DIMM_D is not populated, which may hopefully explain the non-boot situation. I have studied the service manual as well as the spec / setup PDFs, but am unable to get clarity on 2 things, so would love if anyone can help with these questions: 1. How do the DIMM_[A/B/C/D] slots correspond to "Slot 1/2/3/4" in CPU-Z? EDIT: DIMM_A is "slot 3" in CPU-Z per my later updates 2. More importantly, is it a known requirement that both the primary modules should be populated first before their corresponding secondary module slots are filled? EDIT: empirically, this does seem to be proven that you need a SODIMM in the primary slot corresponding to the secondary slot you are trying to fill, else the laptop does not boot. Thanks in advance!

He is using the G.SKILL Trident Neo 6000 C26 2*16 1.450V magic sticks. It is unusual to find any others that can do so well, and part of the reason for that is the effort ASUS has put into making them be able to.

I don't blame ya. I'm a 2x24GB+ user now. Sweet spot between 32GB and 64GB of ram and the prices are insane. Priorities first. Scary thing is they're supposed to continue to go up as corporations push for AI centers everywhere to the detriment of the world and citizens as a whole on a local level. I'm trying to see the downsize to the 270k as time rolls on. Yes, X3D in pockets bests it and 13/14th gen pretty handily, but overall? 270k is a price:performance banger across the entire spectrum especially at 4k gaming. I saw this last night. I asked a few questions in PM. Nice fellow. Makes you wonder what his other 9950X3D2 can do......he's on a chiller so you have to figure that in too to some degree. Has a fantastic 1:1 IMC 2200/6600 is super nice Has an above average 2:1 IMC capping out at 8400 Static chiplet testing is.....average. It's an SP118. Then you get samples like Sugi's doing 8800 and 5700+ but everybody has their own levels of "personal use stability." This one *might* have some of that under water and delidded....$980 + S&H + fees = ~$1050. I'm wondering what memory he was running at 2200/6600 with those timings......that's pretty damn good. I'd pick it up for it's IMC primarily at 1:1. That is pure beast mode.

The fact you and the other user reports a manual profile reducing GPU power limits completely solves the issue points to a faulty PSU failing under higher loads or a board level power delivery issue. If you had a wall power meter to see how much is actually pulled, I think you will see your shutdowns are occurring at a certain overall draw level. First thing, and easiest thing to do, is to order a 380w PSU from Amazon with a 30 day return policy and see if that fixes your problem. If it doesn't, at least you can rule out the PSU and move onto a board level power delivery issue somewhere.

That's the problem, isn't it? There isn't really one. Theoretically the Clevo NH57xx series, like the NH57VR with the AM4 5000 series support but soldered 30 series gpu. And also the NH57JNNQ with the LGA 1700 socket and soldered 30 series gpu. Beyond that there's nothing... 😞

Dark Delta Storm & Heavy Rebel Soul

New Fallen Empire - Wastelander

I switched out the heatsink on my p775 laptop so I no longer need these. you can find it here https://www.ebay.co.uk/itm/117295603313?_skw=p775+tm1g+heatsink&itmmeta=01KXE1FA24JZQ7AQKH6T8G4MFJ&hash=item1b4f5cde71:g:RKYAAeSw4VVqUSKe&itmprp=enc%3AAQALAAAA8GfYFPkwiKCW4ZNSs2u11xA9UrrOJJvfQJLZv8zv%2BaPiNTrHDtf1BoscAGHupPDwzGnbvQX0szVwbwsOjbh0G4haJ1J2KF4i2IovltnojtT%2FWuEQXB%2BTF3wHbdWSFrdpP%2BPPMh996z2AUsVO3rkaRBRlDJoqNSOJ6KLGPJFRVFQP4IjTaVSKT5Q2uwBttPepsbfteTzf3BJcPsXik9sB4d4UzO3vEETqvPnyJjgMuUdwRb16WxcRADKalw84blU3%2FR80HNw%2FQfW43rI8sS73Sn5tffEKZIw6ZX7PI1HFW9Q2g6N0P36P3Yv6iM1shAAang%3D%3D|tkp%3ABk9SR66hvcHrZw

Hi everyone, I'm trying to determine whether this is an issue affecting my specific unit or if other owners of the ASUS ROG Strix G18 G815LW are experiencing the same behavior. System specifications ASUS ROG Strix G18 G815LW-S9076W Intel Core Ultra 9 275HX NVIDIA RTX 5080 Laptop GPU (175 W) 32 GB DDR5 RAM Original ASUS 380 W power adapter Windows 11 24H2 BIOS 338 Latest Intel chipset/ME drivers Latest NVIDIA Game Ready driver Latest Armoury Crate and ASUS updates Problem The laptop performs an instant hard power-off while gaming. There is: no BSOD no application crash no NVIDIA driver reset no WHEA errors no thermal warning Event Viewer only records Kernel-Power Event ID 41 after reboot. The shutdown is identical to pulling the power plug. Games where it happened inZOI Starfield Subnautica 2 Benchmark Space Marine 2 The crashes are completely random and not tied to a specific amount of play time. What I've already tested BIOS updated to 338 Latest NVIDIA driver Latest Windows updates Latest Intel drivers MyASUS hardware diagnostics (no errors) Event Viewer analysis (nothing except Kernel-Power 41) Temperatures appear normal before shutdown. Interesting observation I normally use an Armoury Crate Manual profile with: custom fan curve GPU Core Offset: 0 MHz GPU Memory Offset: +200 MHz CPU PL1: 75 W CPU PL2: 90 W While using this Manual profile I don't remember experiencing any shutdowns. The most recent shutdown happened after an Armoury Crate update. After reboot I noticed Armoury Crate had switched my laptop to Turbo mode, making me suspect the system may have been running Turbo when the shutdown occurred. Another G815LW owner I recently found another owner with the exact same model reporting almost identical symptoms: instant hard power-off no BSOD only Kernel-Power 41 BIOS 338 latest NVIDIA driver original 380 W adapter clean Windows installation normal temperatures, voltages and power readings He can reliably reproduce the shutdown with Hashcat benchmark in Performance mode, while Silent mode passes. He also reports that a Manual profile with reduced GPU power limits completely solves the issue. Interestingly, Prime95 + FurMark together run without problems on his system, suggesting this is not simply a thermal issue. Questions Does anyone else with a G815LW (Core Ultra 9 275HX + RTX 5080) experience similar hard shutdowns? Have you noticed any difference between Silent, Performance, Turbo and Manual profiles? Has anyone identified whether this is related to GPU boost clocks, Dynamic Boost, power delivery or firmware? Has ASUS acknowledged this issue or provided any engineering feedback? Any additional data or experiences would be greatly appreciated. Thank you!

@tps3443 did you see this? https://www.overclock.net/posts/29599251/

In 2000 i think went to a summer 2 day festival at Zambujeira d Mar in my country and they were one of the best live acts There were more famous bands but the live energy from them was whole. It just flowed.