xrayer

I installed nUSB 3.5 and checked again that all files from the package are at the right place. Only a few of files are localized in CZ windows so I put them here: http://rayer.g6.cz/1tmp/nusb35cz.zip please can you make installer package of it? Unfortunatelly it still doesn't work. Here's my devmgr: I tried Refresh button after plug in but any new device didn't appeared in any section... Please upload a sshot of your devmgr on X58 to have better idea how it should look like. Is there some automatic tool that will carefully clean up any trace of previous nUSB from system, just to be sure if it doesn't interfere with some previous install. Probably I would need to try fresh install with your chipset infs and then nusb to prove it if the problem is not caused by some old config... BTW my USB stick has a LED that lights up when system detect it and communication is established. The LED blinks when system R/W the media. But in W98 it only blink for very short time when plugged in and doesn't light up. So it seems like EHCI/hub is not initialized.

I checked USBD.SYS and it's on a place: C:\WINDOWS\SYSTEM32\DRIVERS version 4.10.0.2226, size 18944B I tried to get nUSB 3.5 from MSFN thread but the link on zshare seems so be dead. After some googling I finally founf eng. version here http://www.tmeeco.eu/9X4EVER/GOODIES/(there are also a lot of drivers that may be useful) I extracted the files and going to find matching cz counterparts. Is it possible somehow to run installation proces from directory with extracted files? There seems nothing like setup.exe or else to run (I extracted via Winrar)...

Thanks, I'll check if USBD.SYS is on place. BTW I could upload CZ version of USB drivers and DLLs from CZ WinMe used by nUSB 3.6 but I don't have skills how to repack it as installer executable.

I used nUSB 3.3d CZ before and then I tried nUSB 3.6 EN. Does the difference of USB 1.1 driver matter? I think not since I have no any UHCI. Is new nUSB still based on W2K USBSTOR.SYS? I think it is and it should matter for USB 2.0 systems. But I don't know Win98 USB and nUSB internals. I don't know clear how nUSB is dependent on Windows drivers and how much does it replace them. Maybe the problem is that device detection wizard and systray watcher is somehow linked to USB1.1 and if this subsystem is missing it doesn't launch. But how R.Loew did it? So there must be a way to link it with USB2.0 controller...

Hi, I booted my Win98SE to safe mode, deleted old P3x inf files and put there your new INFs. I went to devmgr and deleted all USB controllers and devices, IDE controllers and some sytem devices linked to old P31 chipset. I rebooted and let the system redetect new HW. It automatically recognized and installed C200 USB 2.0 controllers and hubs. I have no "!" mark on any USB device but it still doesn't work. System doesn't detect any plugged USB device. I also reinstalled nUSB but same result.

>LoneCrusader Thanks for the INF update, I'll try during weekend I'd like to put INFs into my existing installation so I will remove conflicting older INF for P3x chipset first... EDIT: INF thread is here

I had a problem with setup.exe from Win98SE CZ install. It siletly exit - did nothing. I replaced it by setup.exe from WinME (or Win98FE) and installation process started as expected (of course I had setup himemx memory limit before).

And what you exactly mean by "preload" ? You make your own installation CD that install this drivers during setup phase?

How do you preload this driver? Is it done by nUSB or needed to do manually or via other inf? I have USBD.SYS present and it's referenced by other USB*.SYS drivers and USB.INF, IntelICH7USB.INF - maybe I need to update IntelICH7USB.INF to match PCI device IDs to my P67 PCH?

Yes, I was learnt that USB 2.0 hub contains also speed translator so it can connect USB 1.1 devices. The main difference is that now all ports (and connected devices) goes through the hub and not directly to EHCI. BTW DOSUSB driver should handle this, it also detected mass storage device but failed later on acessing the drive letter. OK, this gives me a hope. I'll try to clean old nUSB and reinstall again. I'm just confused that there's no "!" mark in devmgr on any USB controller but it silently ignore plugged devices. I see only a short flash from LED on USB stick when plugged in. No new device detected, neither unknown device in devmgr... The question is if the unofficial nvidia driver core really contains some new code for 8xxx GPUs or if the lines in INF files was only added for future use/some testing but never released. Anyway 8xxx cards are now far obsolete too. Maybe it would be possible to compile some simple 2D accelerated driver based on Nouveau opensource driver for Linux but it would be still a lot of hardwork...

>rloew I played with RTLNIC.SYS and SPLIT8MB a bit more (I used SPLIT8MB from before on my P31 system because I cannot boot without it). But in this case I found it goes a bit better when I commented SPLIT8MB out. Then I was able to boot with disabled NIC and enable it later after boot. It freezed computer for about 10-20s but then it started and NIC became operational, I did a few pings out. But after reboot it hanged again. So I would need to enable it on demand after boot and don't forget to disable it before shutdown. I also asked Realtek support if they will upgrade the W9x driver, just for fun But then I found another working approach. I observed in list of NIC drivers that Windows provide that I could use a 16-bit mode ODI or NDIS drivers. So I installed the RTGND.DOS via OEMSETUP.INF (the last version 1.52 from year 2015), it also added a line "net start" to my autoexec.bat that invooke RTGND.DOS loading before Windows start (it reports using 1Gbit link). Sure it would have worse performance than native 32bit driver but it works stable and speed is far as good for common browsing. I just tested DL from my hosting FTP and got about 750kB/s (not kb/s) so I'm happy with that. Only disadvantege seems to be little slower boot because it waits some additional seconds to initialize DOS driver. BTW I wonder why Windows are waiting so long if network is enabled. A clean install without net can boot in 10s but with net it maybe 30s or more... The bigger issue may be with USB. I read that intel dropped UHCI since it's 5-series chipsets so it wouldn't be possible to get even USB 1.1 slow speed. It also have some USB2.0 hub between the ports and chipset that behaves like a wall that any DOS USB driver can pass - I tried my favourite USBASPI.SYS - it detected EHCI and hub but failed to detect target device. This intel suxx ASPI Manager for USB mass-storage Version 2.27 (C)Copyright Panasonic Communications Co., Ltd. 2000-2008 Controller : 00-26-0 VID=8086h PID=1C2Dh (1458h-5006h) EHCI : MEM=FBFFE000h-FBFFE3FFh(1024Bytes) Controller : 00-29-0 VID=8086h PID=1C26h (1458h-5006h) EHCI : MEM=FBFFD000h-FBFFD3FFh(1024Bytes) ... Initializing Host Controller 1/2 ... ... Initializing Host Controller 2/2 ... ... Scanning USB Devices ... USB Device : VID=8087h PID=0024h HS(HUB) ... Scanning USB Devices ... : VID=8087h PID=0024h HS(HUB) ... Scanning USB Devices ... ERROR : Target USB device not found.1 char device installed.Driver loaded.I just tired to install nUSB 3.6 and it detected 6-series USB root hubs and I don't have any "!" mark in devmgr but any USB device plugged in is not recognized by system. Do you have working USB on your Z87 system? About SATA, yes I use my SMB utility to access HDD directly. I can force it to talk via 1F0 legacy port and it works. You can read various HDD info and SMART status with it. It's freely available at http://rayer.g6.cz/programm/programe.htm#SMB I did some upgrades but not released yet... >LoneCrusader Yes I read your separate thread about installation mod INFs. I wonder there's so stupid 64k limit from RM DOS era. Is there some #inclue directive like in C to include other files? I don't have any problem to select manually right INF for my system and copy it manually. But I would prefer to be able to use it also after install (I'd like to preserve my program setting that I tuned for years than throw it and do clean instal). I think I could copy selected filed to my INF directory before I move system from old PC to new PC and let it redetect new HW. I'm just waiting for better CPU and DDR3 to complete my new setup before I will defintely replace the MB then I'd really like to try your INFs About the patch - so it means that now with new memory patch it is possible to use any 7xxx PCI-E VGA with 512M? Wouldn't it also help with 8xxx cards? Seems that nobody succeed with them. Fortunately I have some spare 7600 and 6600 with 256 or 128M if my 7900 fails...

OK, I found the Tenda driver RTLNIC.SYS of version 5.621.304.2005, size 73984B, I uncommented RTL8111 lines and installed it without problems but after reboot PC hangs. I have to go to safe mode and disable it in device manager. When I enable it in normal mode it hangs immediatelly. Is this the version that you tried? And on what HW did you tried it, D or E? How should I test if SATA is in legacy not native mode? What utility? I can see it can be accessed via legacy ports 1F0 like std. IDE...

>LoneCrusader Plese could you send me your intel INF 9.4.0.1017 mod for personal testing as you don't have time to finish it? What "patches for the 7xxx" are developed? To fix the windows shutdown issue? I don't have any other problem with that latest unofficial drivers on my 7900GT. >rloew Did you investigate the differences between 8111B and E? I belive it would't differ much. I used older NDIS driver RTEND.DOS for DOS TCP/IP from 2011 and it works fine with 8111B and 8111E too. I tried to edit the INF file of latest Win9x driver (30.10.2008) and imported there 8111E specifics (chip revision 6 section, registry entries..) from the INF file of XP drivers. Win98 detected my 8111E but I got ! mark in device manager. I got also error 1F6 during installation of my modded INF. Then I tried to boot with RTENIC.SYS from older 2k/XP drivers (year 2009 - supp. 8011D and 2010 - supp. 8011E) it booted but still with ! mark. There are no HW resources conflicts. Maybe the Win9x RTENIC.SYS could be patched - if there is some chip revision check inside to bypass or so... I also got idea about HW chip downgrade (I work with QFN chips at work quite often but I can see that 8111E has different package and pinout that 8111B so direct replacement is not possible. Would you reveal what other driver you find to work? Limitation to 100Mbit absolutely don't bother me, the warning is problem but I'd like to test it. Yes I know that Win9x don't recognize AHCI (this should work now for NT4 with UniATA driver) and INT13h extension take place so Win98 can run via BIOS access - compatability mode. It means performance degradation but not so ugly, even when put on SSD... I belived that SATA3 (6 Gb/s) should be 100% compatible with SATA2 (3 Gb/s). The BIOS should provide SATA IDE legacy mode and SATA IDE native mode same way for both class of ports but it's not 100% well done. While NT4 or DOS doesn't see difference between ports Win9x see. I think that this may be problem of BIOS that doesn't setup SATA3 ports 100% correctly to legacy mode (maybe SMI IO trapping is involved) or it may be HW issue that SATA3 (at least intel's implementation) controller has some differences that cannot be configured by BIOS to 100% compatability mode. This is probaby also the reason why my previous experiment with Win98 on B75 board ~2 years ago failed - I also had connected the SSD on SATA3 port (of course)... It would be good to highlight this information someway here on forum so another experimentators wouldn't fall into the same trap (and save a day of endless reinstallations trials

Holy f*ck! I just plugged my SATA HDD to another port and it works! Instalation process continued, detected devices, another reboot and wow, I have a basic working desktop. It boots within 11s (80GB older seagate, no SSD). Even I was able to boot my previous installation - it redetected the HW but there are left some issuses. USB seems not work at all and Realtek RTL8111E has conflict mark (it worked with older RTL8111B - I though it's just minor difference and drivers would work.). Anaway I'm happy with this limited functionality. What is the difference between SATA ports? Before the change I had it plugged to SATA3 port 0. And for some obscure reason, Win9x doesn't like SATA3 ports even if set to IDE legacy mode (for ports 0-3). Other 2 SATA2 ports works fine (the rest 2 are unavailable in IDE mode). What's interesting - it works in AHCI mode. If AHCI mode is selected, then AHCI PCI option ROM is activated and it take control of INT13h and I can then use any SATA port (yes, also SATA3) and Win9x boots from it. Here's summary: SATA3_0 = CH0, master (cannot boot Win98 in SATA native/IDE mode but can boot in AHCI mode) SATA3_1 = CH1, master (cannot boot Win98 in SATA native/IDE mode but can boot in AHCI mode) SATA2_2 = CH0, slave (can boot Win98 in SATA native/IDE and AHCI mode) SATA2_3 = CH1, slave (can boot Win98 in SATA native/IDE and AHCI mode) SATA2_4 = RAID/AHCI port 4 (can boot Win98 in AHCI mode) SATA2_5 = RAID/AHCI port 5 (can boot Win98 in AHCI mode) On current machine I have W98+DOS+NT4 on SSD primary partition 0, WinXP on SSD primary partition 1, linux on primary partition 2 + data HDD and I'd like to profit from SATA3 connection. So I would need to move W98 on data HDD or try AHCI way (but AHCI is problematic for direct access tools)...

>Drugwash Yes, I already have GB P31 MB so I know CTRL+F1 trick. On this MB it onlu unhide one option - the PCI latency, thats's unrelated to this issue. I already set the BIOS much fail safe as possible, disables USB, HDA, NX, VTX, HT, set 1 core, sata IDE mode... >MrMateczko Well, it's positive that Win98 works on your i7 Haswell so newer CPU and chipset is no crucial problem (drivers will be but it's not fatal for bare system) >jaclaz Thanks for the documentation. I just tried both /p i and /p j options (2 installations) but seems no difference, always hangs after 1st boot. Also the bootlog looks very similar - the last line is a binary garbage all the time... See here: http://rayer.g6.cz/1tmp/GA-P67-DS3-B3/boot95.txt http://rayer.g6.cz/1tmp/GA-P67-DS3-B3/bootme.txt http://rayer.g6.cz/1tmp/GA-P67-DS3-B3/boot98pi.txt Anyone have a clue? It seems to die quite on early stage of boot, I doubt PnP/ACPI stuff plays role here...

Yes, I limit the RAM via XMS and system.ini as standard process. In my BIOS I don't have any option to disable ACPI, you mean just use /p i switch? I didn't remember it and it's not listed by /? - a hidden feature? I can try again. Isn't ACPI stuff loaded later in 2nd phase of installation where PnP devices are detected (after reboot)? BTW I found this video (if not a fake) proving W98 run on core i7 3960X (AMI BIOS) https://www.youtube.com/watch?v=YOWzorOD-II

Hi, anyone else did try some newer intel 6x/7x/8x board since my last try? Now I have GIGABYTE GA-P67-DS3-B3 with Award preUEFI BIOS + Sandybridge intel core i3 CPU. I set my SETUP options to max. conservative and legacy (IDE mode and so on) I limit the XMS with himemx /MAX=512M My current installation of W98SE is unable to boot neither in safe mode, don't display any message. For my curiosity I ran istallations of all following systems: Win95, Win95 OSR2, Win98, Win98SE, WinME On all systems I was able to pass 1st installation phase that ended with request for reboot (youi know that window reboot in 15, 14, 13... s) But after reboot it never comes up. It always hanged after files are update, loading windows message. Except WinME that restarted instead hang but it's same effect. I tried safe mode and various options of win.com but without any success. I didn't find anything usefull in bootlog.txt - there was any error and sometimes the last line was unfinished and rest of file filled by some binary garbage. In WinME log last lines are: [00011171] Loading Vxd = msmouse.vxd [00011171] LoadSuccess = msmouse.vxd [00011171] Loading Vxd = dynapage [00011171] LoadSuccess = dynapage [00011171] Loading Vxd = vcd [00011171] LoadSuccess = vcd [00011171] Loading Vxd = vpd [00011171] LoadSuccess = vpd [00011171] Loading Vxd = int13 [00011171] LoadSuccess = int13 [00011171] Loading Vxd = VDD [00011171] LoadSuccess = VDD [00011171] Loading Vxd = VKD [00011171] LoadSuccess = VKD I wonder that I was unable to run even poor safe mode on all of W9x systems I'd like to know if someone else succeed on new intel core with different chipset or different BIOS. BTW Windows NT 4.0 with UniATA and VBEMP run fine on this MB, my DOS programs including network too. So why are Win9x so poor? Waht BIOS API is missing or what HW component cause the incompatability?

I belived that windows (and other OSes) use non-paged pool memory for critical things like ISRs and drivers to prevent be paged out. What do you exactly mean by "disable the PagingExecutive"? You mean disable page file in control panels?

Yes, may be an option. For me, the advantage of 4GB PAE is far much less than time spent by fresh instal of OS, apps, drivers, etc. I'm running my current XP installation for about 8 years while I'm patching and improving it, I never needed make clear install during that time bacuse of my backup and testing system. And because I didn't radically changed my HW setup - I did just some upgrades that doesn't affected comaptability like CPU, RAM, SSD that boosted system a lot. Still giving enough power for common work this days...

Thanks but this are the same files included in XP64G patch (from w2k3 server) that I have already replaced... I have no idea what usb releated drivers should I replace more. Maybe I could install W2k3 server to a VM and then copy all *.sys drivers to XP to try if something will change...

One of my friend using PAE patch reported me another problem with HP Scannjet 3800 scanner - when command to scanning is sent it doesn't start and locks the scanner application that cannot be closed any way. Also it cause some corruption to Windows that cannot be restarted/shutdown normal way but freeze at last turning off screen. Again, when booted with original unpatched kernel it works fine. Seems similar issue like with my logic analyzer. So XP64G patch has only limited usage and it would need some improvement that could make some workaround for that faulty drivers to not load them above 4GB and make sure they don't need handle any address >4GB (using some buffers in "low" mem), etc. simply so they can run same as on nonPAE system. I really don't know how complex modification would be needed...

But in this case no 3rd party driver is involved (I don't count the plain inf file), winusb.sys is from microsoft same as low level usb*.sys drivers. So they give to other driver writers a bad example... I tried with hidusb.sys from w2k3 but no change, as I expected this driver is used for HID devices, log. analyser is not a HID class.

Hi, I found another compatability issue with XP PAE. I have USB logic analyser and using driver and software from Saleae. Their's driver is based on winusb.sys that is some new universal MS filter/API for various usb classes. It adds top layer over low-level USB drivers. User can call functions via winusb.dll The behavior is weird because I don't get any BSOD but instead if I select higher sample rate than 4MS/s the application freeze and sampling progress bar even doesn't move. But this application freeze also freeze entire OS, I cannot launch task manager and cannot turn off system normal way so reset buton is the only possibility. Sometimes it even feeze the mouse. When I select 4MS/s or lower everything works fine. When I disable XP64G and boot unpatched kernel it works fine for all sample rates up to 24MS/s. It is 100% predictable behavior, I tried many times. I also tried to update winusb.sys from win7 sp1 (6.1.7601.17514, previous was 6.1.7600.16385) but any change. I also tried newer one from Win8.1 but it doesn't work at all. But most probably winusb.sys is not the root of problem, it relies on low level drivers. I have replaced all usb*.sys drivers from win2003 so I don't have idea where else the problem can be burried... Hm there's some depencies diagram I found that HIDUSB.SYS in my system is original from XP (5.1.2600.5512) so maybe replacement will help..

Hi, for many years I was happily using Mozilla SeaMonkey 2.0.14 - the latest one that is fully working under Win98+KeX4.5.2 also under WinXP which is my primary OS. So one browser - one profile under both OSes, no problem. But currently there are more and more websites that doesn't correclty (or even at all) with this ancient browser. I was forced to download Firefox portable and use it for such specific cases. Along time the count of such messy websites rise so I finally decided to upgrade to latest Mozilla SeaMonkey 2.31 but the profile and addons are not fully compatible. After some fidling and patching seamonkey.exe I finally run both versions together sharing emails, stored session except browsing history. This is because SeaMonkey 2.x.x use new DB layout for places.sqlite which now stores visited URLs and bookmarks but old version stored only visited URLs there. So I you run newer seamonkey over old profile the DB gets updated but when you run old version DB gets corrupted. I was searching for this crap for a long time last night because it was an UNICODE string I missed it first time (seamonkey.exe uses mix of both for filenames). So if someone is interested here are some technical details: Following files from old profiles are renamed by replacing last char. of filename by number 2 seamonkey.exe strings replacement: prefs.js -> pref2.js extensions.ini -> extension2.ini places.sqlite -> place2.sqlite (UNICODE string, search for: 70 00 6C 00 61 00 63 00 65 00) places.sqlite.corrupt -> place2.sqlite.corrupt (UNICODE string, search for: 70 00 6C 00 61 00 63 00 65 00) then recalc image checksum in PE header nsExtensionManager.js strings replacement: extensions.* - > extension2.* Older incompatible add-ons are placed locally in: C:\Program Files\MOZILLA\Extensions\ Newer incompatible add-ons are placed locally in: L:\Program Files\MOZILLA\Extensions\ Universal add-ons are placed in shared profile: C:\Program Files\MOZILLA\PROFILES\User\Extensions\ BTW I had some issues with font and UI size in new Seamonkey 2.31 os I strongly reccomend to install Theme Font & Size Changer addon that allows easily to fix it.

I have a newer one, 5.2.3790.4555 , from some later hotfix update, props: 5.2.3790.4555 (srv03_sp2_qfe.090718-1230) I got second time restart when launching BlazeDTV app for DVB so it's not just a coincidence. I googled hard and found a newer FlyDVB hybrid PCI tuner driver (2 years after current) that is unified for WinXP-Win8.1 so I could hope that it will help.