Aaron44126

Yes, it will run, but obviously you can’t move it around…

Battery error. Fully disconnect the battery until the replacement arrives.

That, I believe, is the dGPU power cable. He'd only have to unplug that because it is "in the way". The two small ones would have been the fan power/sensor cables.

I think re-seating the battery cable is the correct first choice. Replacing the heatsink assembly and bottom cover should not have messed anything else up... the problem likely came from messing with the battery cable. The battery cable can indeed be detached from both ends but I have never messed with anything other than the motherboard end. I would start with that end first and leave the other one alone unless the problem is still occurring after a re-seat. Holding the power button down for a long time will reset the BIOS settings to default. (It also might not work at all if you have a BIOS password set.) I don't see why this would be necessary if things were working fine before the parts replacements.

If there is not a heat pipe in the way, you could probably just drill a hole where you need to put a screw ...? I don't know the details of this situation. I was under the impression that high-end NVIDIA GPU heatsinks were cross-compatible between the 7730 and 7740. If the two screws that line up happen to be opposite from each other, it might be fine; you just need "good enough" contact with the GPU die, these GPUs are so power-limited that it is pretty hard to overheat them. You can try calling Dell spare parts dept. in your region and see if they will sell you a heatsink. They sometimes have parts like this for very reasonable prices. When I got a M6700 heatsink "upgrade" (two-pipe version for the CPU vs. the one-pipe version that mine came with) the price I got from them was something low like $12. Granted, that was like ten years ago... I believe the phone number in the U.S. is 800-372-3355. Otherwise, you are looking at aftermarket sites like eBay, PartsPeople, TaoBao, ...

So I have been able to get nvlddmkm.sys to calm down before by going to NVIDIA control panel and setting the power management mode from "automatic" to "prefer maximum performance", not globally, but for a specific application that I was using. This would be under "Manage 3D Settings", then the "Program settings" tab. This would potentially slightly increase power draw (+ heat, noise), but I think the NVIDIA GPU constantly changing power states is part of what causes DPC latency from its driver. Wondering if you still have the same issue with an external monitor connected, because at least based on the screenshots that you shared above, that seems to cause wdf01000.sys to calm down a lot.

Yeah, I am/was a Dell Precision guy and I fought with this on and off for years, thinking I had it sorted only to have it crop up again after a driver or Windows update. Sad to see it seems to impact many high-end laptops. Also, I started typing "wdf0" into DuckDuckGo search and the first suggestion that popped back was "wdf01000.sys latency", so must not be a standalone issue! I ran across this thread, and on the second page someone suggests making sure that certain audio processes do not run on cores 0 or 1, maybe that will help. I myself have sworn off using Windows for my personal system about six months ago and I have never run into anything like this on Linux or macOS. (My work system is still Windows, but I do back-end dev work, audio doesn't really matter.)

Looks like they have now closed the loophole more fully, server-side. Windows 7 & 8 keys will not activate even if you can use them to install. https://www.theverge.com/2023/10/11/23913107/microsoft-windows-11-block-windows-7-8-keys-upgrade-activation

Classic DPC latency causing audio drop-outs... In a nutshell — Background processes running at the lowest level (mostly device drivers, but also perhaps OS processes or security software) can grab hold of the system to do work, causing everything to freeze. They should "release" control quickly enough that you do not notice that anything happened (1ms or less). You might notice something visually if such a process keeps the system held for, say, 2-5ms (since a 60 Hz monitor needs frames generated at 16ms), but that is enough time to run the audio buffer dry and create a noticeable audio blip, especially if you are trying to do real-time audio work. The whole system freezing issue might be a more extreme version of this. To get a hint as to what is causing this, download "LatencyMon" and run it while you work. (The text at the top of the linked page sort of explains this issue.) Just hit the "Play" button and leave it running in the background. After some time has passed (or you observe one of your audio blips), hit "Stop" and then go take a look at the Drivers tab. They should automatically be sorted with the worst offenders on top. For a system running smoothly we would not want to see any "highest execution time" values higher than 2ms or so. Next step would be to take a look at this list (which will be a list of .sys files), figure out what the worst offenders correspond to, and then see if there is anything you can do to fix it. Example — ndis.sys is part of the Windows network stack, and if it is showing high on your list then it could be an issue with your network driver. Try using Ethernet instead of Wi-Fi, or vice versa, to see if that makes the problem go away. If so, you know which device driver you need to fight with.

Yeah, I have a six-year-old phone here myself. I haven't really seen a need to replace it — until now... I am replacing it in the next 3-4 weeks. Apple has ended "latest OS" support at six years (though it did just get another security update earlier today and I expect those to continue for a while yet), and it is starting to occasionally have some strange long-delay / lock-up behaviors when doing trivial tasks; I think the flash storage might be starting to flake out. Though really, I see the mobile landscape as having matured to the point where, unless you really like having the latest-and-greatest, constant upgrades aren't really warranted and you should be able to be fine with a phone for 5+ years, as long as you start with a good one. Sort of like the PC landscape hit I'd say 10-12 years ago or so. (All the better if phone makers can keep the software updates coming for longer!) Though on the other side, there's the money factor; The Verge notes that Pixel 8 Pro has some software features that Pixel 8 doesn't for no apparent reason other than "just because", and there will always be an incentive to limit support for new features to new phones just to prop up sales. Another example would be Microsoft who is (seemingly) trying their darnedest to prop up falling PC sales by messing with OS support. Looking back to the Windows 11 launch in 2021, they didn't offer it to or officially support any PCs that were more than ≈3 years old at the time of the launch, with no good reason given on why a 6th-gen or 7th-gen system (with a TPM even!) couldn't run it just fine, what the heck. Rumors persist that Windows 12 is coming "next year", we'll see if they pull that again .....

Eh. Cool, but "security updates" isn't exactly the same as full OS upgrades.

Regarding Pixel 8‚ I saw that Google is committing to seven years of OS support and also parts for repairs, a tremendous bump up from what you typically see from an Android phone. Maybe things are shifting. (If only Samsung would follow suit.) https://www.theverge.com/2023/10/9/23910082/google-will-stock-pixel-8-spare-parts-for-seven-years

Does no one know about the "Files" app ...? These days, you can use the "Files" app on the iPhone, connect to a file share on your laptop via SMB, and copy/paste files from the file share to the local file system. This is what I usually do if I want to move a bunch of video files to the phone for offline viewing with an app like nPlayer. (The "Files" app can also be used to deal with files on external storage like a USB drive or an SD card, if you have the appropriate adapter hooked up to your phone.) It's still a little clunky because of the small screen and touch UI for dealing with files, but I'd consider it a step up from having to use the file transfer mechanism in iTunes.

Windows 12 "subscription fee" rumor seems to have been shot down. https://www.windowscentral.com/software-apps/windows-11/no-of-course-windows-12-wont-require-a-subscription-to-use ...Not to say that Microsoft won't look to take advantage of anything that they think that they can get away with, to make money from users running Windows. Ads, I mean "recommendations"... plus pushing people to Bing for more ad revenue, and pushing people to Edge so that they can monetize the "shopping features" and whatever other crud they are adding there. (Really, it makes sense, no one would go for a subscription fee. They can't even charge people for Windows upgrades like they used to because people just won't buy them, and then how would they start taking advantage of the new monetization features in their latest version of Windows?)

Eh. I think the number of people who actually use their phone charging port for data transfer is pretty low. I'm in an iPhone family and I know of no one who uses it for this except for me. The vast majority of people won't notice the "low data rate", and anyone who really cares about it will hopefully be aware enough to just get an iPhone 15 Pro.

iPhone 15 (non-Pro) uses the A16 SOC from last year, so that would make sense...

This looks fine. Fan replacement is what I would do. They don't send fans individually for these systems, since the 7000 series came out. You'll get a whole heatsink assembly with two fans attached to it. My Precision 7560 also has bad fan noises sometimes. When it is operating at a certain range of low speeds, it makes a kind of grindy/grumble sound. I usually hear it when it is "slowing down" from a busy load to an idle load. I will be having them replace the heatsink assembly before this thing goes out of warranty.

As an iPhone user, I think this is absolutely the right move, and it's sad that the EU had to drag them into doing it. (Though I think they would have done before too long, anyway; they've already been transitioning the iPad from Lightning to USB-C.) iPhone will join the rest of the world with being able to plug in any old USB-C thing and have it mostly just work. The worst thing about it will be the transition period. In my house, my wife and I both have iPhones, and my kids both have iPads, and they are all still Lightning devices. We have charging cords scattered around and it is nice that you can just plug one in to charge whatever device. My wife will be getting an iPhone 15 Pro in a few weeks, so we'll start the period of having both USB-C and Lightning cables hanging around until we get all of the devices transitioned (which I am in no particular hurry to do just because of the connector change). And it's not just chargers. We plug the phone into the car to enable CarPlay, and now I will have to have both a USB-C and Lightning cable in the car which will need to be swapped around depending on whose phone needs to be connected, until I also get a phone with USB-C. I have a collection of Lightning accessories (HDMI out, USB in, 3.5mm hookup for speakers, SD card reader) that will eventually need to be replaced as well.

SBF trial started today with jury selection; expected to last about six weeks. https://www.theverge.com/23894366/ftx-sam-bankman-fried-trial-updates-news His parents are also facing some legal issues. https://www.npr.org/2023/10/02/1200764160/sam-bankman-fried-sbf-parents-ftx-crypto-collapse-trial-stanford-law-school

Pretty sure you can't use a non-eval key with eval media at all. Doesn't matter where you get the key from.

Microsoft is going to stop allow people to use Windows 7, Windows 8, and Windows 8.1 keys to activate Windows 10/11. ...Or so they say. https://arstechnica.com/?p=1972170 It appears that this only applies to new versions of Windows 11. If you have an older version, you can still activate with an old Windows 7/8 key (for now). So, presumably you could save a Windows 11 22H2 ISO, use it to activate with an old key, and then upgrade to the current Windows 11 build with Windows Update.

Funny because the issue occurs with Optimus turned on... if the GPU upgrade was causing an issue with the display panel, I would expect it to be more problematic with Optimus off. (With my M6700 + M5000M, I basically had to keep Optimus on because of odd behaviors and stability issues that would occur if I did not.)

First off, I'd reseat the display cable (probably won't fix anything but easy enough to try and there is the small potential that it is a stupid simple fix like that). Another thing that you could try is to switch the setting for NVIDIA Optimus / graphics switching on or off and see if that changes the behavior, could give a hint as the display panel would be "attached" to the iGPU or the dGPU depending on the configuration there. But assuming that changes nothing, next, I'd be worried about the display panel itself and start thinking about replacing it.

One day later... CrossOver 23.5 is out! With D3DMetal integrated, tons more games now work. https://www.codeweavers.com/blog/mjohnson/2023/9/27/crossover-235-is-a-real-game-changer

Exactly right. There are tons of choices with various pros and cons. Only you can decide what is best the best option for you. What can I say ..... I got fed up! Fed up with Windows. Fed up with the state of high-end laptop offerings (specifically from Dell but many of the issues are "global"), compromises involved with them, thermal issues, power management issues (specifically random GPU power drops because of it trying to balance CPU/GPU power and not doing it right), and general lack of attention to detail all around. I've posted about these issues elsewhere on the forum. (Never mind the issues I have with the direction that Microsoft is going with Windows 11, I tried switching to Linux for about two months and posted in that thread about some frustrations I was having with Windows; and later my rationale for switching from Linux to macOS, and I've also written about my frustrations with the Dell Precision 7770 which I was initially so excited about.) So I went to a Mac because it seemed like the only place to go that made any sense. (Apple's recent attention to the gaming space did help push me over the edge.) What I have found that I didn't fully expect going in is how %@#$ good of a laptop a MacBook Pro is. I'm not talking about "as a PC", but "as a laptop" specifically. It offers a balance between high performance when needed and cool/quiet/long battery life the rest of the time that you simply can't get from anyone else, where you have to pick between one or the other when deciding which model to get. And it offers other things that you'd want a laptop to have; quick battery charging, a really good display panel, good speakers as well, and a solid mostly-metal build. And yes, there is definitely a compromise when it comes to both "tweakability"/"upgradeability" and "absolute top performance". For the latter, I am settling for "good enough" performance (which is pretty darn good) because, to me at least, the benefits from this system more than make up for it. And, I mean, I played through Shadow of the Tomb Raider at 1510p resolution and "highest" graphics settings preset, it had no trouble maintaining a stable 60 FPS, and that was running through the Rosetta 2 x64->ARM CPU emulator to boot, so I have nothing to complain about with regards to gaming performance. I might not be able to run at 4K/120 or use the latest ray tracing glitz, but I can play games fine and they look nice enough for me. With regards to the former, it's sort of something that I've just accepted that I have to deal with. But, laptops have been become less upgradeable all around so if I'm going to be pushed down that road no matter which manufacturer that I go with, I might as well get a something that works as a good laptop rather than something that tries to be a mini desktop and, as a result, is on the bulky side and runs noisy and hot. (I'm really at a point where I can't have my computer stuck in one place, so actual desktops are not under consideration for me right now.) Still, while I don't mind having to pay up front to max out the CPU/GPU/RAM, the lack of at least replaceable/upgradeable storage is indeed ridiculous. To be clear, I have found multi-threaded performance to be not quite as good as a 12th gen Core i9 laptop CPU (never mind a 13th gen which is notably faster). The M2 Max doesn't top the chart in CPU performance, but it is up there near the top, and certainly can trounce anything more than a couple of years old. (Intel sure does have throw a lot of power at the CPU to beat the M2 Max, though; M2 Max probably wins in terms of power efficiency.) Something I've been mulling over lately. What's the difference between a translation layer and "native support", anyway? Modern games are largely just bits of computer code that target a set of APIs, and if you swap out those API implementations with something else that works the same... is it a "translation layer" or just an "alternate execution environment"? Could a "translation layer" still offer "full performance"? ...Leaving Mac aside for a moment — obviously it is running a whole different CPU architecture so it must be translated if you want to run x86/x64 games. Let's look at trying to run Windows games on Linux, on a PC with a typical x64 CPU. Modern Windows games generally just x64 binaries (with supporting x64 libraries & data files) that make calls to various Windows APIs to handle various things that they need (open files, play sounds, network chatter, etc.) and use the DirectX 11 or 12 APIs for graphics. On Linux, running on an x64 CPU, running the code from the game binary should not be any slower than it is on Windows. What you have to worry about are the external API calls that the game makes, which flat out don't exist on stock Linux. Valve, CodeWeavers, and others have basically cooked up replacements for the APIs (Wine, plus DXVK & VKD3D for DirectX/graphics). The game doesn't care if it is making calls to "real" Windows+DirectX or to Wine+DXVK, as long as the implementation is good and the right stuff happens, it just executes the same code and the game chugs along. And, if the "alternate API implementations" are good enough, there is no real reason why it should be "slower" than running a game on Windows directly. After all, it's still just the same sort of CPU churning through the same x64 game code, in the end, just the parts where it calls out to the OS to do something have been swapped out. And, you'll find that under Proton, many games run just as well as they do on Windows ... and sometimes, even better. There are issues obviously when the behavior of the alternate implementation doesn't match up with Windows's behavior, which is why some games are broken (DirectX 12 in particular since that is newer), but if you compare Proton today to just like three years ago you will see that they have made incredibly rapid progress in fixing that up. On the Mac side, to pull this off you need to both have a replacement backend for the OS APIs (something like CrossOver with D3DMetal) and you have to translate the CPU instructions from x64 to ARM (Rosetta 2). The biggest performance hit seems to come from translating graphics API calls right now (D3DMetal or MoltenVK), but that has been doing nothing but getting better over time. D3DMetal performance has actually gone up like 20%-80% (depending on the game) over just the past three months or so. On the CPU translation side, Rosetta 2 performance actually seems to be really good, again looking at my experience playing Shadow of the Tomb Raider which has not been compiled for ARM. It does have a one-time performance penalty hit when you first launch the game, as it basically trawls through the executable and translates all of the x64 code that it can find to ARM code. It might happen again when you hit new bits of code (i.e. when getting from the menus to the game proper for the first time). The results are cached, so you don't have to wait for the same translation a second time (unless the game is updated). This is probably why I originally complained that loading times seemed longer than they should be early on, but that problem quickly went away. So, is there performance overhead? ......... In some cases, yes, though it is shrinking as these technologies mature, and there are groups of people that are working really hard on "unlocking" games from the environment that they were written for. Honestly, the whole approach of making alternate implementations of backends that games need (and ideally, open source ones) is critical for long-term preservation. There's been chatter recently following this study that, unlike other media like music and films where translating from one format to another is relatively easy, the vast majority of games ever made are currently unavailable to general users. If you are like me and like playing older games as well as new ones and don't want to have to maintain a collection of old hardware, then you've gotta get used to making your own backups and playing in an emulator or translation environment of some sort. Heh. That's all of my off-topic ranting for now!