Leaderboard

This post is just to summarize what I learned from the very detailed and involved DRM discussion about 80 posts back, along with similar posts from other threads: The NPAPI version of Google's Widevine CDM plugin will run on Vista but not XP, so it's no help. Stephen Fox has a PPAPI version of Widevine that will run on XP, but you'll need Chrome, not FF or one of its variants, to run that. The Adobe Primetime CDM plugin will install and run on Basilisk 52 just as with FF 52 ESR (although you need to toggle media.ffvpx.enabled to false in about:config), but few providers of DRM-protected content will use it. (I suppose you could try anyway, by spoofing an older FF version with those sites.) One last possible fallback is Microsoft's Silverlight. But as with #2, you'd generally need to spoof an older FF version that didn't support Widevine, in order to get sites to try to use it. (With some sites, such as Amazon Prime, you also need to spoof Windows 7 in your user agent string.) Do I have that about right?

Yeah, I'm fine, as well as my web server. I'm just busy in $DAYJOB because of aftermath of typhoon.

We are supposed to all want Microsoft's patches without thinking. They work hard to create this mentality, in order to herd users ever more effectively. Fortunately (they would say unfortunately) I actually DO think, as do you (or you wouldn't be here reading). I think about these so-called "vulnerabilities" - some of which have never been seen in the wild - and what I can do about them. Knowing how things work is better than not knowing. I am most certainly NOT helpless in the digital world. I don't run software "from the wild" without vetting it. I have surrounded myself with a network environment that practically and substantially reduces the risk I'll visit a web site that will try to infect me or take data from me, while at the same time shunning the "run-of-the-mill" approaches that provide only marginal security. I resist "cloud" software that wants to update itself all the time. I don't want the "latest", I want the "most stable" that does what I need. My systems run for months 24/7 without faults, however hard I use them. Unlike most folks, I actually measure performance objectively, and can tell when the OS or application efficiency changes. I always realize there are tradeoffs - security is never a purely "more security is better" thing. I haven't had malware turn up on a MalwareBytes scan - ever, so I guess what I'm doing is effective. That said, I always look for ways to improve. Possibly most importantly: I never allow myself to get a false sense of security. If I do something stupid, and one of the things I've set up protects me, I still beat myself up over doing something stupid and strive not to do it again. If you look at the pricing of high-end systems - say, workstation prices at Dell - depending on how close to the top end the hardware is, you can see that computer systems delivering even just 10% more compute performance can cost literally thousands of dollars more. Why would I want to intentionally turn my system now, to mitigate vulnerabilities for which there are no known exploits yet, into a system that performs as poorly as the ones I passed up when I chose to pay top dollar? I simply don't subscribe to the sentiment "you WILL become infected if you don't patch to the very latest OS code", because Microsoft is not the company they used to be. It is no longer as high on their priority list to deliver a good, serious computing experience. There is no guarantee that a patch from Microsoft delivers better code than what it's replacing. Time has shown that they can (and do) deliver instability and even new vulnerabilities. I always try to gauge the tradeoffs. And let's not forget that they've let much of their testing organization go. It's clear Microsoft wants to bring everyone under their control (which invariably involves updating to their latest software) and they're using every trick in the book to get you off your old system where they do not yet have that control. They do not care whether your existing system/hardware works worse for what you need it for; you're not paying them to keep it. You WOULD, however, be paying them if you replace it. Microsoft software, even the very mature versions, is nowhere near optimized as well as it could be. There's no reason we have to expect it to get less efficient as newer versions are released. It should steadily be going the other way. For example, I've seen with my own eyes that their latest compilers are delivering faster and faster instruction sequences for the very same source code. Why aren't their OS patches/releases speeding up the system? It appears for every 10% improvement they make, they layer on 20% more junk. How many processes does Windows 10 have to run just to host an empty desktop for you nowadays (hint: well over 100)? Hosts? Brokers? Medics? Bleh. I have some systems from which I don't demand the utmost in performance, and for which security is a greater concern, and I've got them completely up to date. Others I have stopped at the December 2017 patch level, because there are significant disadvantages, while at the same time Microsoft just hasn't delivered any improvements that matter. Your mileage may vary. Just make sure to know what it is. -Noel

I've later found out that there are even newer hal.dlls (v. 5.1.2600.5687), but just for multicore processors (and not needed unless one's machne has more than 8 logical processors). They're found in KB958244 and, as above, the KB article is gone, but the hotfix remains available by request from MS. Long live, XP!

lol. The first reply is a bot, but the followin ones are not. I mean, there's a real operator there. I contacted them other times and some of them were funny and very polite. Besides, a bot wouldn't ask me to fill a report (a feedback), saying how was my experience with the operator. They basically have a set of pre-configured answers (like the one you send with your mobile when someone tries to call you, but you are busy), but they can write things down themselves as well.

I have both those entries in my Add/Remove Programs. Searching for the winusb.dll file brings up the same version (6.1.7600.16385) in system32, but there are copies of the file in two other folders relating to connectivity software for mobile phones, both my old Nokia phone and my current Samsung phone. The copy in the Samsung folder is older (6.0.6000.16386), the one in the Nokia folder is the same as the one in system32. If you have similar software installed that may well be where it came from.

Yep , I read the "no longer" as implying that the change was recent enough, while in my opinion it dates back many, many years, Bill Gates left MS (as CEO) in 2008, so it is roughly 10 years, that puts it in the right timeframe. jaclaz

I'm not talking short term! I think I began to sense the change when Bill Gates stepped down from active management. That was a while ago, for sure. They spent a lot of years throwing things together then tidying it up later, but if you followed the "wait until Service Pack 1" (at least) philosophy, Windows has been a pretty good workhorse. I still remember when it went from something that had "reboot fairly often" built into its design to "runs virtually as long as you want" - which for me happened around SP2 (I think it was) of Vista. I got work out of Windows 3.1 for Workgroups and all the versions since, and to this day I still get a LOT of work out of Windows 8.1 - of course after taming its desire to be something it's not. That taming just doesn't work as well with each subsequent new version. Inertia took Microsoft a long way and is still carrying it along, but they're losing sight of the fact that the world needs Windows to be the serious, no nonsense business system that actually facilitates people's work on inexpensive hardware. Even though they made a lot of money being that, now it just seems like they want to facilitate their own updates and bloat, and have all but forgotten people don't run Windows just to run Windows, but to actually DO things. There is a helluva lot more a lot of folks need to do besides check Facebook and Twitter! You have to admit, the no-nonsense Windows 7 update philosophy - putting YOU in charge - was way more "good, serious computing experience" than this modern "Windows as a Service" BS where they take over whenever they choose to. And who said it was okay to change our settings, or delete our files? That was the realm of MALWARE before they started trying to social engineer people to change the way they look at computing. Even now - in the very latest versions of Windows 10 - we see them adding things like "WaaS Medic" and being ever more aggressive in taking over, while of course they say they're not. Just try to disable certain services and see how long that lasts. So yeah, they're not the company they used to be, not even last year. -Noel

@Dave-H You can tell them if something it's wrong: as long as it's just a feedback, MS support doesn't check the licence. I do it all the time pretending to be on POSReady. As to this one, I'm gonna report it as well as it's happening on my machine as well. Anyway, I just disabled the Update Root Certificates from the Windows Components Wizard, as @mixit suggested. Hopefully, this is gonna solve the problem, as I already update my certificates once a month using the Cert_Updater_v1.3.exe

My \Windows\System32 folder has a winusb.dll, size 24136 bytes created/modified 02 Nov 2006 7:00:10AM, version 6.0.6000.16386, with valid MS signature. This is the only version of this file that I have anywhere on the system, no second copy in dllcache, drivers, any of the windows update uninstall folders, etc. I can't find any reference to such a file in my notes about updates that I added so I'm guessing it came preinstalled, I don't know exactly what it does. The properties sheet description says Windows USB Driver User Library"Windows USB Driver User Library". The version number would suggest mine came from a package released for Vista (despite the date being too early), while yours looks like something released for W7rtm. That file is NOT mentioned in dencorso's thread about usb updates at https://msfn.org/board/topic/162733-usb-functional-patches-for-win-xp-after-sp3/ , nor at https://blogs.msdn.microsoft.com/usbcoreblog/2009/11/13/list-of-the-latest-usb-hotfixes-for-each-windows-version/ I do NOT have the item in Add/Remove Windows Components, but Add/Remove Programs does have "Microsoft WinUsb 1.0". Unlike the other program entries there is no information about size, frequency of use, last used date, etc.

IE9 (32-bit) on Vista SP2 Home Premium 32-bit here The youtube.com/html5 test is correct when it comes to IE9 on Vista (no VP8 support here): IE9 has no support for MSE; after Google ceased supporting Adobe Flash Player (all 3 variants of it, i.e. activeX, NPAPI, PPAPI) on youtube, the only thing left for IE9 to handle was progressive download of mp4 files (encoded in h264/aac) over HTML5, and I recollect the limitation of not being able to go Fullscreen on the embedded player (a message about unsupported browser was displayed ) ... It was at the start of summer that I accidentally noticed yt was unplayable on IE9, as I'm hardly using IE9 for anything this day and age... I did not lament the loss of yt on IE9, in any case it had been an unsupported browser many moons ago already... But it was this thread that made me want to investigate things... So: Well, youtube claim they don't officially support any IE version anymore, yet tests show that IE10/IE11 still work, while (obviously) previous versions don't... Loading Introducing iPhone XS, iPhone XS Max, and iPhone XR in IE9 (after you declined to upgrade your browser, that is...), you get the empty embedded player, as per OP: For troubleshooting, open IE9's developer tools (F12) and load (browser) Console; refresh youtube page: To load their embedded player, yt use, among other things, a JS script named base.js; URI for my locale is: https://www.youtube.com/yts/jsbin/player-vflkUTZn2/el_GR/base.js That script uses in 19 instances the window.matchMedia code snippet; a Google search on matchMedia yields the following two results (among others): https://developer.mozilla.org/en-US/docs/Web/API/Window/matchMedia#Browser_compatibility https://caniuse.com/#feat=matchmedia where one can see that the needed API was introduced only in IE10+, i.e. IE9 does not understand window.matchMedia and chokes, failing to load the HTML5 youtube player: Mozilla Firefox, OTOH, supports the function from an early version 6 - and Chrome from version 9... Not strange anymore, is it?

Since you're able to log in normally and can do everything, there is probably no real corruption. Run regedit and go to: HKEY_LOCAL_MACHINE/SOFTWARE/Microsoft/Windows NT/Current Version/ProfileList Highlight each entry and you will be able to see the temporary profiles that were created and are being used. The correct profiles are still there but have had ".bak" extensions added to them. Delete the temporary ones and rename the correct profiles by getting rid of the .bak extensions, which will probably fix it. However, the last time it happened to me, Windows kept recreating the temporary profiles, so I just deleted the temporary profiles and the ones with the .bak extensions, and rebooted, which fixed it.

The last time it crashed my profile is a few years back. For me it always started with various error messages like Event ID Error 1517 and 1524 until the profile could not be loaded anymore. Since then, I have been using the User Profile Hive Cleanup Service v1.6g, where the functionality since Windows Vista is part of Windows and has not had any problems since then.

No, you ain't! No way! Win 10 has actually an insane interface and is really perplexing to use. And you're not alone, I, for one, do abhor Win 10 with a passion. I don't think I can be of much help to you with Win 10, sorry, but rest assured you're my hero for even giving it a try! You have all my sympathy.

Well, I've seen that happen, maybe some half-a-dozen times with my own XP installations (nowadays I have four bare-metal and one virtual XP SP3 boot-partitions in three physical machnes, but they've been seven bare-metal plus one virtual XP SP3 boot-partitions up to when I decommissioned my two beloved A7V600-X Athlon XP machines -- because they are non-SSE2, to limit how many minorities I belong to -- and gave the wife one Phenom II quadricore machine I assembled but used very little). Observe that all my XP installs, but the virtual one, are on FAT-32, not NTFS, so that permissions were sure not the issue, then (none of the instances happened with the virtual XP installation). My solution, in all cases was to redeploy the previous backup image of the system partition involved, since those never were older than one month, and usually were even more recent. IMO, there actually *is* some quirk, possibly a race-condition or some other problem arrising from a complex interaction among processes in XP that does cause it to clobber the profile info (most commonly the user.dat) once-upon-a-blue-moon. And in every case, after having redeployed the previous system image, the problem did not happen again anytime soon. I think it actually did happen twice on my oldest install, but the two events were two or three years apart, so I deem them unrelated. Sorry for the long-winded post.

Well, you will need to provide a timeline for this change, I don't think they had "deliver a good, serious computing experience" anywhere on their list, most probably in the last 10 (ten) years, surely not in the last 5 (five) years [1]. jaclaz [1] to give some context, Windows 8 was RTM on August 2012, i.e. roughly 6 (six) years ago and soon Windows 8.1 will be 5 years old.

latest drivers for the Intel I219-V ethernet LAN adapter from Intel web site: https://downloadcenter.intel.com/product/82186/Intel-Ethernet-Connection-I219-V also apply latest available cumulative update for Windows 10 v1803

These updates are not yet ready for use by most users. They are not compatible with USP3, SE2ME, or UPXed files. Updating a previous KernelEx 4.5.x installation with just the new DLLs misses the registry and Core.ini changes that go with them. I am currently testing nine new files. (I am having no problems with .db files in any of many browsers.) Please do not create any more packages until I post these new files. This should be in the next few days. Then we can coordinate a full installation for limited (MSFN.org) release. I already have more API improvements for the following update. If I can also resolve the aforementioned compatibility issues, I'd like us to have a Release Candidate by the end of September and a Final for general distribution the first week of October. Until then, remember that currently this is not an open beta for just anyone to access and try. These update files are only for testing by registered members of MSFN.org who are interesting in helping me improve KernelEx. As previously discussed, no file attachments from this thread should be redistributed in any form on outside servers.

Thanks, I'm safe as usual. The only thing I worried is the possible power outage caused by typhoon.

official build has --disable-eme while my UXP builds has --enable-eme, so DRM may work (I haven't tested)

Useful link with more info: https://www.anandtech.com/show/13201/intel-preps-h310-revision-with-win7-coffee-lake-support alacran