jaclaz

There are still quitre a few BBS's accessible via Internet/Telnet, if you want to have the feelings of the good ol' times. http://www.bbscorner.com/ http://www.bbscorner.com/usersinfo/bbslists.htm And an IMHO interesting article, JFYI: https://www.theatlantic.com/technology/archive/2016/11/the-lost-civilization-of-dial-up-bulletin-board-systems/506465/ jaclaz

As always when it comes to those stupid chips, there may be an issue also with the "thermal pad" that is used to connect to the cooler, maybe (have no idea) there are better quality (with higher heat transmission capacity) than the standard ones (just like there is better and worse thermal paste). And of course taking care of the fan and of the duct to be clean and open usually helps, I seem to remember that those "burned" DELLs had *anything* in the duct (hairs, dust, grease, etc.). jaclaz

As said earlier, most of the point about boot optimization, at least the part connected with the sequence of loading files, is derived more or less linearly from the difference between "random seek/read" and "sequential seek/read" of the underlying storage media. So in the mentioned good ol' times (PE on CD) the difference was dramatic (even for greatly reducing the queer noises that came from the CD seeking up and down), the same technique applied to (USB 2.0) sticks roughly reduced boot time to 1/2 or 1/3 s (very, very noticeable), on internal hard disks (connected via SATA or however fast buses and with modern largish read ahead cache on the hd itself) it became not so noticeable, I don't expect that much difference on a SSD, but of course it is a nice experiment to perform. jaclaz

Once upon a time you set the explorer view as you liked it, then closed the windows clicking on the right top cross icon while holding the CTRL key and the view setting was saved. But I am not sure it applies to NT 4.00 Is that what you are asking? jaclaz

As a side note, I remember repairing a couple of laptops with those stupid Nvidia boards, and had to reflow them (it only worked once), but it happened on DELLs running (from factory) a plain enough XP (BTW used mostly for office work/web browsing, not gaming or other form of high speed/resources graphics), and they "burned" nonetheless, and I believe XP required less graphic resources than Vista (Aero or not Aero). jaclaz

Are you trying to join the old grumpy people club? I can put a good word for you, if you are interested . 8 bit processors? LUXURY! Why in my day .... https://tinyapps.org/blog/misc/200702250700_why_in_my_day.html ... and we LIKED it! jaclaz

We are not talking of "generic schools", we are taking of "computer classes", where the people that tomorrow (or the day after tomorrow) will be likely qualified to maintain your personal system (or your bank system, or the system that runs traffic lights in your city, whatever) learn the base knowledge on how to do that. I was hoping that they were better educated by their teachers. I was not expecting (I know how schools have often older systems) times expressed in seconds, but 5-10 minutes, come on . jaclaz

Well, allow me to disagree, students in a computer class should have been taught how to speed up their booting, I see it more as a failure of the course than wasted learning time, if the teachers/instructors didn't manage to teach (in the remaining minutes after the looooong time to boot) how to optimize the system for faster booting (and/or if the students didn't decide to study on their own how to solve this evident anomaly) there is a problem. jaclaz

Naah, it's ok , it's the FSB clock that can be raised to 1066 Mhz, I thought it was the RAM, as there is DDRAM2-1066 : https://en.wikipedia.org/wiki/CAS_latency#Memory_timing_examples jaclaz

Anyway it is 99% some "fake" news. It seems like it originated a couple years ago when someone saw the screen of a technician's laptop that (with some sense of humour) had an XP looking background. http://www.theregister.co.uk/2015/12/15/windows_xp_royal_navy/ https://www.theregister.co.uk/2015/12/18/windows_for_warships_not_on_queen_elizabeth_class_aircraft_carriers/ Hard to say how it re-surfaced. In any case some of HMS's do run a (obviously "special") version of Microsoft OS's: https://blogs.msdn.microsoft.com/ukgovernment/2008/12/17/windows-for-submarines/ jaclaz

Any source for this? Not really my field of experience but I remember that clocking/CAS/etc. is tricky business. But 1066 mhz should be DDR2, shouldn't it? Post a link to your exact motherboard (and manual). jaclaz

I usually find RAM (and most other things) at the other end of a web search. Like here (example only): http://www.ebay.it/itm/1GB-MDT-DDR1-RAM-PC3200U-400MHz-CL2-5-64Mx8-M924-400-16B-Desktop-/281914586060?hash=item41a36b33cc:g:LZIAAOSw0HVWA-8x More seriously, do you really need a CAS latency of 2.5? (CL3's are more common and cheaper) jaclaz

And I didn't say exactly that, I meant that maybe a reboot in a VM may not be as "complete" as it is on real hardware and anyway (when compared to a direct access to a physical drive) there are added "layers" of *whatever* that may make a "cold start" very different from a "reboot". Of course comparing different OS's on a "same" VM makes a lot of sense, and of course comparing different VM's (to see which is faster) is nice, it is the comparing of Virtual vs Real that I see as not fully appropriate because a lot of things may be different in the way things work. jaclaz

Brought to you by the same guys that call "system" what everyone else calls "boot" and call "boot" what everyone calls "system", BTW . http://www.multibooters.co.uk/system.html jaclaz

If I may, comparing "real" machines with "virtual" ones is not really-really appropriate anyway. Maybe a virtual machine using direct physicaldrive access (as opposed to a - I believe that is what you use - a vhd/vdi/vmdk file) may give some more comparable results, but when using a virtual device such an image file it is likely that any kind of filesystem or OS or Virtual Machine cache may have a role in faster reboot times. jaclaz

It seems to me like a very good result . Warpdisk (AFAICT) is essentially a boot optimizer, as I read it (but I may well be wrong of course) all the stuff about defragmentation is the usual mumbo-jumbo and commercial bull§hit[1] but a read ahead cache AND making files sequential does make a big difference (the bigger the slower is random access times vs. random access ones). In the good ol'days of BartPE (and of real CD's) using a "sort.txt" with mkisofs did make miracles, by simply writing to the CDFS (where files are already contiguous by definition) the files in the exact same order they were requested by the booting OS. jaclaz [1]a file ( a single file) is either fragmented or it is contiguous, once every file in the boot sequence is made contiguous there is no fragmentation, the difference maybe in the contiguity (and right sequential order) of several files

Which was the reason why sometimes times I had to remind you how some of your perceptions, when using powerful systems with an array of SSD's (and/or a gazillion Gb of RAM), can be very different from those experienced by a large part of the users base. As you well know, history repeats itself, one of the reason why Vista at the time was such a huge delusion for most people (much more than what it deserved) was that it was installed by users (and even installed by OEM's on new machines) on too low power machines, and as such it was slow as molasses. jaclaz

It is one of the classical yes/no answers. The good guys that put together the EFI/UEFI "standard" (and then summed it up in some 2,200 - that is twothousandtwohundred - pages of specifications ) had something (actually pretty vague and fundamentally "wrong" when it comes to backwards compatibility) in any case each and every UEFI vendor may have read in those pages everything and the contrary of everything so *anything* may happen. In theory UEFI would provide an EFI "shell" which could be as powerful as a real mode OS, like - say - DOS, allowing to do almost anything in the phase before the OS is booted, in practice not only this shell has become a mere replica (with senseless graphical effects) of the settings you find in *any* BIOS, but noone (with a few exceptions that I can count on the fingers of one hand) develops any software for the EFI environment. But (with some few exceptions of "pure UEFI" systems) most firmware implementations have - inside or besides UEFI - a CSM (Compatibility Support Module) which is simply a BIOS, just like the good ol'one. So you shouldn't have any problem in booting in BIOS mode, and even in UEFI mode, there are anyway ways. Some UEFI are (poorly IMHO) coded in such a way that they automatically disable CSM if they detect a EFI loader, that is an issue (that can be also fixed with a workaround or two, still it is more complicated). The main "mistake" that people makes (BTW induced by the - I have to presume intentionallt UNclear communication from Intel, MS and friends) is that GPT "style" of partitioning (while part of the UEFI "standard") is totally independent from it. An OS may (quite a few are) not able to boot in UEFI mode but perfectly capable to access a GPT disk (and most with a few tricks can be made into also booting from a GPT disk, even if the OS is said to be not able to). Anyway the main thing that one must understand is that BIOS (or UEFI) are largely irrelevant once an OS like NT based systems are booted. Typical boot sequences: BIOS: BIOS->MBR of first disk->PBR of activepartition on first disk->OS loader->rescan of system, HAL, drivers->booted OS UEFI: UEFI->OS loader listed in NVRAM, but residing on some device->rescan of system, HAL, drivers->booted OS or: UEFI->OS loader in FAT EFI partition on disk->rescan of system, HAL, drivers->booted OS So, once booted, the system will be essentially (like 99.99999%) the same, of course ALL OS that can be booted in UEFI mode surely support GPT "style" disks, while systems that only can boot in BIOS mode usually do not support GPT disks for booting (but as said there are workarounds) but most support GPT for "data only" disks. So, if you have a CSM in your UEFI you can ignore the UEFI and continue booting through BIOS (and from MBR), if you have not a CSM module you can most probably (but it may need a tweak or two) boot from MBR, or you can have your boot disk as GPT (besides being in some regards a "stupid" standard it has some advantages over MBR, namely a replicated partition table that may come of use in case of disk corruption for recovery). If you are going to multiboot different systems the BIOS (CSM) is still the best choice IMHO, one of the (senseless and evidently imposed to "push" something on users) is that with UEFI the OS must have the same bitness of the hardware, i.e. you cannot use a 32 bit Os on 64 bit hardware. If you are going to use only one OS or a couple of new ones that support UEFI and GPT you may also use GPT only devices without issues. jaclaz

I trust this more: https://www.forensicfocus.com/Forums/viewtopic/t=15588/ https://www.forensicfocus.com/Forums/viewtopic/p=6589197/#6589197 Just creating a %windir%\perfc file is enough, but for added safety (you never know) also creatre a file %windir%\perfc.dat AND set it as Read Only. jaclaz

Easier is to use contig to make the $MFT file contiguous (and of course UltraDefrag is an exceptionally good tool IMHO): https://en.wikipedia.org/wiki/Contig_(defragmentation_utility) but the issue you are worried about is not about defragmentation of the $MFT, it is about unused entries in it, and about its size. You want to "compact" the $MFT or "shrink" its size. I am not so sure that a free tool exists with this exact feature, surely there are a couple (Commercial) tolls from Paragon that can do that, and also Ultradefrag has a "MFT Optimization" that does more or less the same: https://sourceforge.net/p/ultradefrag/discussion/709672/thread/a7a67d93/ jaclaz

Not so surprising, since it pre-dates most of them by years. We are talking 1993/1994 here. Use this Nirsoft little tool, adding psapi.dll should work on NT just fine: http://www.nirsoft.net/utils/driverview.html jaclaz

There wasn't even internet (as you know it) at the time: https://en.wikipedia.org/wiki/History_of_the_web_browser https://en.wikipedia.org/wiki/Tim_Berners-Lee jaclaz

Start by reading the guide: http://diddy.boot-land.net/grub4dos/Grub4dos.htm Although now more than a bit dated (more recent grub4dos 0.4.5c and 0.46a series have countless added features) it still cover the basics nicely. I see however that there is still a communication problem between us. Our conversations are becoming pretty much pointless as you (I have to presume deliberately) continue to largely ignore the information I try to share with you and attempt - at every interaction - to shift the topic at hand to something else, I won't even try to comment on these As I tried to tell you before it is YEARS that several different methods to run XP entirely in RAM exist, there is nothing to be invented or added to them: Personally I believe that using either Winvblock or Firadisk is easier and more handy than using the good ol' Euhenio's way (making use of the MS Ramdisk.sys coming from server 2003 SP1) but of course beauty is in the eye of the beholder. jaclaz

I don't know. On one side this kind of lists are a handy way yo know about existing applications, on the other a "full review" that omits benchmarks (when we are talking of tools that are - or should be - intended to speed up things) and focuses on lists of (often theoretical only) features (present or missing) is pretty much useless. Besides the 15 defraggers list (that very likely almost entirely use the same plain MS defragging API), there is another list with 42 (fortytwo) "Free Data Destruction Software Programs": https://www.lifewire.com/free-data-destruction-software-programs-2626174 with *any*, really *any* kind of tool mixed up together, good, extremely good, bad and pure crap, without any sensible evaluation of WHAT they actually do and how much time it takes to do the ONLY kind of wipe that is needed is appalling to me. There are only 3 tools among the 42 listed that do "the right thing" of using Secure Erase, and the Author perfectly knows about the differences: https://www.lifewire.com/what-is-secure-erase-2626004 still he puts all together in the same basket. I fear that by putting out too many options it will only confuse the reader. On the other hand (you may be less lucky than me ) I appreciated the image on the ad for "Hot Russian Ladies". jaclaz

Hey peeps, do we really need yet another burst of posts on this matter? And Dybia, you know how I appreciate your enthusiasm :), and believe me, I have a lot of sympathy for young people and for there eagerness to learn, their sometimes brilliant ideas, etc, but you shouldn't really make statements like "When back in 2013 I bought a new desktop" or however pose yourself as a long time professional user of any OS that was ever run in the world (unless you actually are one of those). How old were you in 2013? jaclaz