jaclaz

I don't understand what you mean . It seems to me like a plain, nice confirmation that XP 32 bit has no issues "in itself" with 4 k sectored disks, and that the theory that up to 16 Tb is possible is perfectly sound, but that there may be issues with the controller (actually with the controller drivers) . And you also found your 3xB disk make/model example! About the pre-partitioning/pre-formatting, you can IMHO simply forget that any manufacture will EVER ship a 4k "native" disk drive with "MBR" style. It's a long time that they all got on the 64 bit/UEFI/GPT bandwagon, and I suspect - given the fast turnover in the industry - that their junior (soon to become seniors) technicians either never heard of MBR or know about it just like - say - we know that steam engines exist(ed) but never actually used one. jaclaz

Basically the 32 bit limitation is in the number of sectors that can be stored in a MBR style partition table "slot". Set aside the CHS part (and Partition ID and Active status) there are two LBA fields, "Sectors Before" and "Sectors in partition" or "Size in Sectors", each being 32 bit, thus with values from 0 to 2^32-1=4294967295, the limit is the "Sectors in Partition". If you have a device which exposes 512 bytes sector the largest possible accessible partition is 4,294,967,295*512=2,199,023,255,040 bytes, i.e. the (infamous) 2.2 Tb. If the device exposes 4096 bytes sector the largest possible accessible partition becomes 8 times that, 4,294,967,295*4,096=17,592,186,040,320, i.e. around 17.5 Tb (much more than any existing disk) the related topic is here: http://www.msfn.org/board/topic/176480-2-tib-limit-size-in-mbr-hard-drives/ The "over 2.2TB limit, maxing out at 4 TB in practice" was related to using a particular partitioning pattern on 512 bytes/sector devices, the thesis is here: http://www.msfn.org/board/topic/176480-2-tib-limit-size-in-mbr-hard-drives/?tab=comments#comment-1135913 it has been confirmed as working on Windows 7 32 bit MBR by Tripredacus, starting from here: http://www.msfn.org/board/topic/176480-2-tib-limit-size-in-mbr-hard-drives/?do=findComment&comment=1142727 until here: http://www.msfn.org/board/topic/176480-2-tib-limit-size-in-mbr-hard-drives/?do=findComment&comment=1143244 and later tested thoroughfully, this is enough to prove that the theory is sound and works, but later some issues prevented Tripredacus from replicating the behaviour on XP (and there could be possibly some other roadblocks on XP, but not related to the 32 bit size of the LBA MBR entries). @dencorso There are no such things as a xMx or xGx disk on the market, the M or the G uniquely depends on the "style" with which you partition the thingy. A 3xB (which is at the same time a 3MB and a 3GB depending on the style that you use) is *any* "native" 4Kb sector disk larger than 2.2 TB, I believe they exist alright, but I'll check and let you know. jaclaz

As said the confusion is between different "features" of the drive. A disk can be: 1) "traditional" 512 bytes/sector 2) "Advanced Format" 4 Kb/sector BUT exposing 512 bytes/sector 3) "Native 4 Kb", exposing 4 Kb/sector M) Partitioned MBR Style G) Partitioned GPT Style S) Smaller than 2.2 Tb B) Bigger than 2.2 Tb XP 32 bit (without added "special drivers") cannot "understand" ANY disk drive partitioned with the "GPT style". XP 32 bit (without added "special drivers") CAN "understand" ALL disk drives partitioned with the "MBR style" BUT it won't be able (normally) to boot from "native" 4 Kb sectored disks AND it won't be able to accesss a larger than 2.2 Tb partition on 512 bytes sectored ones. Even if the experiments carried by Tripredacus were not finalized, I am still convinced that is possible to access "MBR style" disks (512 Bytes/sector) up to 4 Tb IF (and ONLY IF) they are partitioned with two partitions, of which the first is slightly smaller than 2.2 Tb. (Tripredacus succeeded without issues with Windows 7 32 bit on such a scheme but had issues with disk drivers on XP, so the expereiment is still to be finalized) So, disks compatible with XP 32 bit AND with later OS's (without using particular tricks or third party drivers/whatever): 1MS 2MS 3MS <- these don't really exist 3MB <- usable for storage BUT NOT for booting from it [1][2] With the Paragon GPT loader (and possibly/maybe also with some transplants from Server 2003), still for storage only (not bootable) and limited (if the Paragon GPT loader is used) to internal disks ONLY: 1GS 2GS 3GS 1GB 2GB 3GB Possibly with the special partitioning, limited to two partitions (or however with a single partition starting immediately before the 2.2 Tb limit and spanning over the rest of the disk, given that the rest does not exceed 2.2 Tb, in practice limited to 3 Tb or 4 Tb disks, bootable, both internal and external: 1MB <- these don't really exist 2MB It's a mess , I know . @Tomcat76 Don't worry, the "GPT style" scheme includes what is called a "protective MBR", any OS that has not been coded to recognize the "GPT style" scheme will simply see an unmountable disk and won't access it, unless you intentionally (DiskPart/Disk Management and similar) delete such protection. jaclaz [1] a number of external USB enclosures designed for "Big" disks may expose 4 Kb/sector even if the disk drive inside is 512 bytes/sector, so they can be used in XP 32 bit (for storage not for booting). [2] I wouldn't totally exclude that with a number of tricks these can be made also bootable, but no senseful experiments have been made about it that I know of.

Which operations do you expect to be destructive? Naah, don't worry, you can use GPT disks just fine, if you install the appropriate third-party driver (not free). What you cannot do is to boot from such a disk. And without the driver you won't be able to access the partitions on the disk so no damage can be done. (of course using Diskpart or Disk Management or similar tools and fiddling with the partition table is NOT a good idea). @dencorso Sorry, but it is "the opposite" (the mixup is between MBR vs. GPT and 512 bytes/sector vs. 4 Kb/sector. You can have 4 Tb MBR disks (4 Kb sectored only) both as internal and external, but you cannot have GPT disks (no matter if 512 bytes or 4 Kb sectored) if not with the GPT driver (which is only for internal disks): https://www.paragon-software.com/technologies/components/gpt-loader/ jaclaz

@dencorso As said it depends on programs used and on specific use of the programs, @Tomcat76 Photoshop is often mentioned as an example of one that *requires* the pagefile, but AFAIK/AFAICR it is only a (stupid) "cold" requirement, i.e. you *feed* it with a 100 Mb pagefile and it will start just fine, it will simply refuse to run without a pagefile, which does not automatically mean that it *needs* a pagefile, the difference is very slight, but it is there alright. The general point that I was trying to make is that if one runs the "normally" installed system in his/her "normal" way, using the programs "normally" used in the "normal" way he/she uses it and monitors pagefile usage for a few days, the decision on size (and *need*) of a pagefile can be taken correctly, based on actual data. And I will repeat myself - if needed - we are not talking of an alien OS just landed on Earth from an outer galaxy. We are talking of a 16 years old OS, that has been consistently run for years by millions of people with "automatic pagefile setting", which means - according to Mark Russinovich: https://blogs.technet.microsoft.com/markrussinovich/2008/11/17/pushing-the-limits-of-windows-virtual-memory/ Now, it is evident (at least to me) how the algorithm used is - to say the least - approximated, only a little more sophisticated than a generic (and "safe") rule of the thumb, and since the minimum size is required only by the *need* for enough size for a full crashdump the lower limit is in itself m00t. If we assume that the formula is correct (reductio ad absurdum) it plainly means that a 4 GB RAM system with a pagefile of - say - 512 Mb has (roughly because of the known issue with reserved unused memory under 4 Gb) the same total capacity as a 1 GB RAM system with a fully committed/used 3 GB pagefile or as a 2 GB RAM system with a pagefile expanded up to 2 Gb or DOUBLE that of a 512 Mb machine with a fully committed/used 1.5 GB pagefile . The above (machines with 512 Mb or 1 Gb or 2 Gb of RAM running XP) represent probably 70% (or more) of the hundreds of millions machines that ran XP in the last 5-10 years of life of the OS. Yet the reported crashes are not so common, it would reasonably mean that the provision is "big enough". What I personally suggest is - when the "full" 4 Gb of RAM are present to make a variable sized pagefile, range 100 Mb to 4 Gb, then run the system normally for a few days, checking how much the pagefile expanded. Once determined this expanded size, make a fixed size pagefile of corresponding size and be done with it. (and of course, you can well move it to the "past 4 Gb" Ramdisk if the size is suitable). If the size did not grow, try running without a pagefile for a few more days, if everything works as before then you don't *need* a pagefile. As said IF there is the need of a pagefile, AND the resulting size is in the range 1-3 times the 4 Gb RAM, it means more or less that you would be better served by a 64 bit OS. jaclaz

Sure, but ... A pagefile that will likely never be hit, so a largely otiose use anyway. Seriously, if the programs you use (or your particular usage of the system) consistently hits the pagefile, then you will be much better served by an operating system allowing more "direct" RAM, even if Imdisk or Gavotte are fast (faster than a hard disk and possibly even faster than a SSD) they are much slower than direct RAM access. On the other hand, if the programs you use (or your particular use of the system) never *need* a pagefile you are introducing a (remote possibility as over the years both Imdisk and Gavotte's has been demonstated as being very stable) an unneeded possible point of failure. jaclaz

Or - possibly much-much better - don't use IE at all. Recent news, JFYI: https://hotforsecurity.bitdefender.com/blog/internet-explorer-bug-can-reveal-the-contents-of-your-address-bar-19020.html https://www.brokenbrowser.com/revealing-the-content-of-the-address-bar-ie/ jaclaz

Yes, surely. JFYI (some related links): jaclaz

Ok , I'll try again. There are two distinct problems. One is accessing RAM beyond the 32 bit limit (4 Gb) and another one is accessing all memory within that limit. What was the difficult part in: http://www.msfn.org/board/topic/130001-32bit-windows-not-usingseeing-all-4gb-ram/ More specifically read this (it is the first link among those listed in point 2.): http://www.msfn.org/board/WindowsXP-Pro-won-t-recognize-4GB-of-mem-t92568.html that you won't be able to access unless you use my little address translator or translate manually to: http://www.msfn.org/board/index.php?showtopic=92568 which will land you here: http://www.msfn.org/board/topic/92568-windowsxp-pro-wont-recognize-4gb-of-memory/ Now, the first one may (or may not, given the instability of the patch on some hardware) be solved by enabling PAE, the second may (or may not) be solvable by changing (if available) some settings in the BIOS (or using a modded BIOS if available), in any case you won't ever see all 4 Gb if all the RAM you have is 4 Gb, values actually available range from 3 Gb (or even below in some "edge" cases) up to 3.75 Gb (most motherboards with normal settings/peripherals will have 3.25 or 3.50 Gb available). jaclaz

Which likely has nothing to do with PAE but rather with the specific mapping of RAM done by the BIOS, devices and drivers. See: AND provided links. jaclaz

@Rloew The Grub4dos hide command does change the partition ID's in the partition table in the MBR (i.e. adds 0x10 to the "normal" partition ID's), but of course it does so to the actual volumes slots, i.e. primaries and to the various EMBR's, not to the extended partition in itself, so it is possible that this triggers *something*. @ppgrainbow The duplicate ghost drives come from *something* else, not from merely hiding them, as said grub4dos hides them in the "normal" way. Maybe the issue here? (or something loosely related to it). Anyway, this has become (actually it was since the start, but I failed to recognize it at first sight, I am getting older, beside grumpier) clearly a S.E.P. Have fun. jaclaz

It depends on HOW (exactly) you did WHAT. Choose one (and one only) given link/tutorial and follow it EXACTLY. If it still doesn't work, post again DETAILing which one you chose to follow and the EXACT behaviour it has, AND the hardware you are using it on. jaclaz

BUMP! (#1) jaclaz

Yep , but without knowing the actual setup it is hard to say, it is also possible that *somehow* there is a "mixed boot", with *some other* io.sys loaded, however the normal search order for grub4dos is; (hd0,0) (hd0,1) (hd0,2) ... (hd1,0) (hd1,1) ... but grub4dos (I believe it depends on versions) can also look on hidden volumes, so it has to be seen what really happens. I inserted the: root command in the suggested commands (chich is otherwise unneeded and to be removed when inserting in a menu.lst) exactly to verify that the "right" root is found, but of course that was implying that some feedback would have been provided. jaclaz

Whatever floats your boat is good But you cannot misuse the tool and post an overcomplex workaround (actually you can, of course). If you detail how EXACTLY your disks are setup, I can probably help you, but you must not misrepresent what you did. Let's see together the set of commands (to be issued on command line) I posted. I assumed that the IO.SYS that you want to boot (and the corresponding Windows 95) bothe reside on (hd2,0). map (hd0) (hd2) map (hd2) (hd0) map --hook At this point (hd0) is (hd2) and (hd2) is (hd0). find --set-root /io.sys this will find the first io.sys it can find and set root to that disk partition. If the io.sys is on (hd2,0) (which is now (hd0,0), the following: root will output (hd0,0), so you have the "right" root. At this point you have two (actually three) possible way to boot the Windows 95 (DOS 7.x): 1) chainloader +1 (this will chainload the bootsector of the current root partition) 2) chainloader /io.sys (this will chainload the io.sys directly, bypassing the bootsector) 3) chainloader (hd0)+1 (this will chainload the MBR of the disk, if there is normal boot code in it, and if the first partition on that disk is active, it will chainload the bootsector, like #1 above). It is possible (and it may depend on the exact grub4dos version you are using) that the one or the other doesn't work, however. jaclaz

Sure, if you just added the map --hook command in that menu.lst, you tried booting from the "wrong disk", you set root BEFORE exchanging disks. Moreover you first experiment in command line, and then you put it in a menu.lst entry, like: map (hd0) (hd2) map (hd2) (hd0) map --hook find --set-root /io.sys root chainloader /io.sys boot The above is not "booting from third disk" (which is AFAIK impossible, as DOS wants to be on first disk, actually on first active partition of first disk, or however whatever gets drive letter C:\, not negotiable), it is booting from third disk remapped to first disk. What you are probably doing (and that is philosophically "wrong") is to hide all partitions on first and second disk so that the first partition on third disk becomes C:\. But without knowing how the disks are partitioned/setup it is hard to say. jaclaz

As I tried telling you here: http://reboot.pro/topic/21586-installing-os2-on-a-second-hard-drive-with-grub4dos/ you are NOT exchanging disks if you do not hook the mapping. If you want it more bluntly, try using the grub4dos commands properly, before overcomplicating the matter. Dos/Windows 95 want to be booted form the First hard disk, just exchenge (properly) the disks drives. jaclaz

Well, everyone has to start somewhere/sometimes , that (being new) seems not a particular issue. Which is good, and creates a nice, symmetrical situation, as I am also not an English guy. jaclaz

Seriously, it seems like you want *someone else* to help you fix your issue BUT you do not provide the basic info needed and asked for. Not necessarily the info asked for will actually be enough to get this kind of assistance, let alone guarantee a success, but surely without this info NOONE (among those actually capable of providing meaningful assistance) will be able to provide that. All you will get this way will be either nothing or half-botched suggestions by clueless people. Standard Litany, please: https://web.archive.org/web/20160604095422/http://homepage.ntlworld.com/jonathan.deboynepollard/FGA/problem-report-standard-litany.html jaclaz

Ah well, that explains the issue. I also suspect that your XP was NOT originally installed on a full moon by a virgin[1]. Anyway, since my crystal ball is in the shop (again) for tuning and Tarots are so imprecise, you can have an I-Ching answer just now (much earlier than 2000 years): #19 Lin: http://www.psychic-revelation.com/reference/i_l/i_ching/hexagram19.html jaclaz [1] this not among the requirements by MS, but usually helps.

So the BBC spotted a large trove of running XP's: http://www.bbc.co.uk/news/technology-41306321 (of course the rest of the article has the usual FUD about XP's security). This time the new entry is a Dr Steven Murdoch "a cyber-security expert at University College London" that provides the usual boilerplate scaremongering, even if his field of expertise is (according to Wikipedia) in other matters (contiguous but not exactly the same as OS security). jaclaz

It is extremely difficult to suggest a theme, as anyone will have his/her own preferences, and - generally speaking - themes are "excessive". Check if any among the themes here: http://customize.org/teschio/gallery?page=2 might do. The QNX/XP might be similar to what you asked for: http://customize.org/xpthemes/19645 though cannot say how much functional is it. jaclaz

Replies to what? These patches are known to work perfectly (or at least good enough) on some setups and to not work or work unreliably on some others. You posted that it doesn't work on your machine. There could be tens of reasons why the patches don't work, including among them an improper install/setup, some of these may be solvable, some may well not. Without any detail (EXACT hardware involved, EXACT patch/files used, DETAILED description of the way they were installed/setup, etc., etc.) apart: what did you expect? jaclaz

Maybe, or maybe he thought of a possibly common incident involving high-end Aston Martins : (though most probably somethign similar may happen to Bentley's as well) JFYI, we (dencorso and me) had recently a PM exchange on the opportunity to have a BfB (Bang for the Bucks) unit of measure when comparing performance of things that can be bought. Personally (being besides grumpy also cheap) I rate cars using the PEV (Panda Equivalent Value) which is an abstract and arbitrary unit of measure roughly valued at the moment Eur 13,000 (i.e. the average "list" price of a Panda) but expressed in thousands, 13k. Max Speed (comfortable/sustainable) of the Panda is assumed 130 km/h, so the S ratio is a nice, round 130/(13*1)=10 As an example a Renault Clio Turbodiesel has a PEV of 19k/13k=1.46 so S is 150/(13*1.46)=7.89 An Aston Martin Vanquish Coupè has a Pev of 273k/13k=21 so S is 220/(13*21)=0.81 The PEV is useful because you can say "wait a minute, I can buy more than twenty Pandas with that kind of money". @dencorso Naah, you are missing a point, if you have a Lagonda, you also have a chauffeur, and you can afford another one when his points are lost. jaclaz

Ok, XP sucks (when compared with Windows 2000). As a matter of fact Windows XP is Windows 2000 with a number of mostly unneeded bells and whistles bolted on it. Again the base issue is the "one OS to rule them all". Until XP we had Windows NT and 2000 for "professionals" and Windows 9x/Me for "home users". Both professionals and home users were pretty much happy about their respective OSes as each did what they expected. XP re-unified the two categories, adding multi-user logon (and permissions, and quotas) to the innocent home users, making the setup and mantainance of the system much more complex, while adding (largely unneeded) plug 'n play extensions, multimedia features etc, to the professionals. The peak of abomination in XP came with Service Pack 2. Since the OS was deployed (because of the "home users" target) on scarce quality hardware with unstable drivers that were conflicting with managing large amounts of memory the good MS guys decided to remove PAE support to more than 4 Gb RAM from XP, thus damaging the professionals that actually had more than 4 Gb. On the other hand (thanks to the sheer quantity of "home users") the professionals benefitted from a number of software tools and new hardware (good enough for professional use) that would probably had never been developed if the user base had the same numbers of Windows 2000. And with some care, and once removed the most intruding bells and whistles, XP is as good as Windows 2000, actually it is Windows 2000, only a tad bit more bloated than needed. jaclaz