True or False? USB2 + Win 98 + Intel 915 = Impossible

[deomsh]

3 hours ago, waltah said:

Okay, so I have no idea how to add a picture to this post.  I attached it but that meant you'd have to download it to see ... not something anyone wants.  ???

 

My method (using Windows 10): select on screen with your mouse after Windows-key+Shift+S, paste in Irfanview, save as *.png, than drag the file into your post where the cursor is currently. 

Edited February 25 by deomsh
Missing +Shift

[NotHereToPlayGames]

You do not need Irfanview.

Just click the "print screen" button on your keyboard then paste (ctrl-V) with the cursor in the reply box (as in below [screen on right is maximized System Informer just to intentionally hide items on that screen before posting to the public).  The "print screen" button will not appear to do anything, it's not until you PASTE what the "print screen" button COPIED.

Or, just paste into something like LEGACY PAINT, edit, crop, add lines and arrows and boxes and highlights, you name it.  Then copy from legacy paint and paste that into the reply box.

The Win-Key-S won't work for "everybody" (that's a "Snipping Tool" shortcut that people like me UNINSTALL (because I have better third-party software for that function)).

Everything needed is already on *any* Windows system.  No need for Irfanview (unless you ALREADY use it, point is, you don't need to ADD anything to your Windows system if you don't want to).

 

 

Edit - but be aware that you can only paste so many images here at MSFN before you RUN OUT OF ROOM allocated for you.  You'll need to go to your profile and delete past images in order to paste new images once you hit your profile limit.

Edited February 25 by NotHereToPlayGames

[deomsh]

 

5 hours ago, NotHereToPlayGames said:

Edit - but be aware that you can only paste so many images here at MSFN before you RUN OUT OF ROOM allocated for you.  You'll need to go to your profile and delete past images in order to paste new images once you hit your profile limit.

True you can directly paste (see 'afbeelding' below), but filesize with Irfanview is more economical with highest compression (see Clipboard_xxx.png below).

 

About using Windows-Key+Shift+S you are missing the point: pasting a full print-screen is most of the time unreadable if containing chars, with Windows-Key+Shift+S you can select without using some extra tool like Paint (even without Irfanview).

In my opinion this is a good and easy Windows 10 tool for daily use. Sadly Windows 11 gives some dialog after the mouse-selection, with long waiting time going away to make next print-screen selection. If not included in the OS used, there will be numerous snipping tools available.

About IrfanView: of course this is not an advise to install such a tool, I just described my method to post pictures on MSFN. To be fully clear: for me 'Description' is not the same as 'Prescription'.

 

 

 

 

Edited February 26 by deomsh

[NotHereToPlayGames]

55 minutes ago, deomsh said:

'Description' is not the same as 'Prescription'

 

Yeah, the Win-key-s/Irfanview is actually similar to my default.  I don't generally paste "print screens" and my hotkey and program is different.  My primary point is that if a first-time image-poster is trying to figure out HOW, it just seems easier to direct him/her to something that should already be on his/her computer and not making him/her think Irfanview is "required".  That's all I was tryin' to say.  But hey, written communication is FAR from my strong suit.  

[LoneCrusader]

10 hours ago, waltah said:

Thanks for clarifying all that, Lone.   As I say, I'm no expert.  

One observation:  Operating systems and drivers for complete computers sold to a mass market are mainly paid for out of hardware sales.  The incentives are to make current sales successful but mostly not to support activities (such as continuing to use obsolte items) that would compte with future hardware sales.   The maker of a motherboard sold to an end user has different incentives because his customers are nearly all people like us and supporting non-current systems may be a competitive advantage.  

Which means an aftermarket board with a certain chipset is likely to have better support for earlier systems.

Now you begin to understand... haha. Yes.. Dell only has an incentive to support their systems as sold, whereas component manufacturers are more likely to provide wider compatibility testing and push more fixes for less-common or "backward compatibility" issues to the public. Dell is not "wrong" or "bad" for this, but it's not "good" for us.

Now, barring some weirdness like I described before, here are the only other ideas I have.

First - the SATA issue I mentioned before. You said you had problems with the SATA patch.. how so? What steps did you take? It's been a long time since I've had to work with it on a running system, but I know it has a very specific instruction set.

Second - I remember USB2 controllers used to be "switchable" - i.e. you could run them in USB2 mode or switch them to USB1.x. This is what "USB2STOP.VXD" does in my XUSBSUPP update for 95 - turns the USB2 controller off so that 95 can use the USB1.x controllers. I wonder if something similar is happening on your system - possibly the controllers are initially USB1.x by default and USB2 is not switching on (under 98) to take over? The problem is I have no idea how to investigate this, rloew wrote that driver for the package and he is no longer with us.

One long shot - find a copy of the time-limited Orangeware USB2 driver mentioned by Petr here and see if it works for your controllers. This driver was specifically written for 9x, not 2K, and was used by Intel for the 865/875 chipsets. The version distributed by Intel for those chipsets is locked to their device ID's though, and would require a patch by someone with more know-how than me to bypass this.

[SweetLow]

1 hour ago, LoneCrusader said:

This is what "USB2STOP.VXD" does in my XUSBSUPP update for 95 - turns the USB2 controller off so that 95 can use the USB1.x controllers

I always assumed that turning off EHCI controllers in Device Manager gives the same effect.

1 hour ago, LoneCrusader said:

I wonder if something similar is happening on your system - possibly the controllers are initially USB1.x by default and USB2 is not switching on (under 98) to take over?

If you reread the statements of topic starter you will find this one: 150 MBytes @ 60 seconds == 2,5 MB/s. It is simply impossible on USB1 controllers (of course if it measured correctly because some OS options can highly distort usual copy procedure time).

Edited February 26 by SweetLow

[waltah]

The ideas help.  Let's try again, pasting a .jpg of my 'devices by connection':

That looks better.  The trick I was missing was .jpg instead of .bmp:  Thanks, folks. 

I don't see anything wrong with it

I have Everest (early version of AIDA64) on the machine:  There's a ton of information there but nothing that leaps out as a problem.  As I understand things, USB 1.1 is typically 1-10 Mb/s, USB 2 20-30.  With the PNY drives I move 150M in just about 40 seconds, = 3.8 Mb/S while using a SanDisk USB 3.2 drive (shows as 'USB' in the picture since '98 has no awareness of USB 3) the time is 32 seconds or 4.7Mb/S.  

I am having another discussion with AI as to whether anyone has ever gotten USB 2 speeds on Win 98 on a 4700.  So far he's repeated the claim for success on a 4600 -- irrelevant since a different chipset -- and some armwaving about the need for NUSB 3.6 or maybe one of the later packages containing .sys files from XP.   All of which is already done -- in order to get to the picture above.  

My wife uses a paid AI chatbot: It sounds like it's better than the Luxxle free chat  I'm using.  

In other news I got a Belkin 802.11g card working in this machine without trouble and Retrozilla 2.3 loads at least some modern sites adequately.  And I have an ATI x300 PCI-e card on order so that'll free a PCI slot for a game card.   (4700 has 2 PCI, 1 PCI-e x16, 1 PCI x1.)

I can't recall the details of my attempt to patch SATA; I probably got something wrong there.  I'll try again.  And once I've got the PCI-e video card for this machine I'll put the PCI one back in the 4600 and check out it's USB speed.  That's a much easier machine for running Win 98.  

I've got PowerQuest Second Chance on this machine so I can try things and roll them back just as with XP's System Restore.  Added advantage:  PQSC lets you build a boot disk so you can do a restore even if the machine won't boot.  

Edited February 26 by waltah

[schwups]

I also would like to remind you of HDTach 2.7 - small but nice and does what it's supposed to do . I always use this tool first for such a purpose.

https://web.archive.org/web/20031108132105/http://www.simplisoftware.com/Public/download.php?request=HdTach-2.7
https://web.archive.org/web/20031030124506/http://www.simplisoftware.com/Public/index.php?request=HdTach

[waltah]

I have installed Win 98 on my 4700 with ACPI on and the exact same NUSB3.6 and other .inf/.sys tweaks as formerly.  Makes NO DIFFERENCE in speed. 

Copy of a 150 Mb file from my various thumb drives to disk takes the same amount of time ACPI ON as ACPI OFF. The only difference noted is that with ACPI ON you can bring up or shut down the system without an hour or more delay (effectively a hang) if there's a thumb drive inserted.

Win XP times average around 7 seconds = 20 or so Mb/s -- low end of USB 2 but definitely not USB 1.1. And that's really too short a test for precision.

This is good news: It means that Win XP still knows a USB trick that I haven't yet taught Win 98. So the problem can be solved -- it is not buried in the bowels of the BIOS.

The post-NUSB3.6 files I'm using are Win 2000.  I can check the other packages I've found and see if anything later will work. 

Probably this will turn out to be like many other problems in my life -- User Error.   

[SweetLow]

$2 USB3.0 UHS-I SD Card Reader + UHS-I Samsung 32G SD Card, Windows 98SE, Nvidia NForce2 (the slowest EHCI controller I know)

The same on Windows XP.

[awkduck]

PNY USB 2.0 FD <-- is this the result of formatting the way I mentioned? Or, is this formatted elsewhere/differently?

It "could" make sence, if this is the result of formatting the way I suggested. Floppy disk are formatted in a similar layout.

But if not, there is a chance that you are having a "Geometry Translation" issue. I'm pretty sure the "FD" is floppy disk/drive.

Edit: Suppose it could mean Flash Drive. It is just an ID associated text descriptor, and probably not checking anything beyond that. I'm just over fishing for a simple fix.

Just curious. I can't examine this on my own, right now. 

Edited February 27 by awkduck

[waltah]

Impressive stats, SweetLow.   I'll run the same if I get USB 2 going at correct speed on the Win 98 system.

After another afternoon of being led around by AI without measureable progress, I have ... no measureable progress.  It would be nice to have a way to log USB actifities starting with negotiation to at least narrow the territory in which the problem lies but after assuring me that such a tool exists (and how to start it) AI later told me there is no such thing.  If it does exist someone here probably knows ... ???

Other than that I've swapped USBEHCI.SYS from a current WinXP SP3 system -- it perhaps improved speed slightly but not close to USB 2.  In order to do much else I'd have to start over without NUSB3.6  since that mods some modules.  

This is a difficult problem because of the number of different chipsets, the varying goals (is it working USB 2 if it only goes at 1.1 speed?), and the lack of specific instructions for most packages.  I'm not sure whether trying to get help from AI gave any net benefit.  The only appealing thing about it is it always gives an answer, but most of them are either wrong or useless.    

I don't think anyone has gotten USB 2 to go full speed on an ICH 6 machine under Win 98 without using an add-in card.  Since I can unplug the Win 98 HDD and plug in the one with XP and get about a 6x speed boost it has to be possible but either it hasn't been done or the result is very well hidden.  

Same argument applies to format/geometry questions:  It's the same drive on XP at 20+ Mb/s as under Win98 at 3-4 Mb/s. 

I'd be happy to be told I'm wrong.  

[awkduck]

1 hour ago, waltah said:

Same argument applies to format/geometry questions:  It's the same drive on XP at 20+ Mb/s as under Win98 at 3-4 Mb/s

You are still missing something; but I probably haven't explained it well enough. May not fix the issue anyway.

This geometry issue can be due to your BIOS' design. Earlier I mentioned initializing devices, with Grub2, before booting Win98. Keep this in mind.

Win2k+ and Linux "more" fully break the ties with BIOS's device initiation. Win98 does "not". Please note: This is not a basic Grub2 boot. I had to use a specific Grub2 command, to initialize Grub's USB2 drivers; otherwise Grub will rely on your BIOS' support.

So the geometry, from your BIOS, could be detecting the drive as some kind of zip drive or supper floppy (this does happen). I would think this would not be an issue, if you are not booting with the drive connected to that machine; but I have seen evidence that BIOS can still interfere. It seems similar to a USB handover issue, but happens even if the machine was not booted with the device in. Maybe just the right control register getting poked, or something.

Edit: On other machines, with similar hardware issues, I've sometimes had to look at MSR settings for that particular machine. It would be interesting to see how your system was configured different, between XP and 98.

Or, if indeed it was a geometry issue (only affecting Win98 [potentially bios related]), there may be a FAT32 format layout that causes the flash translation/emulation less work.

Edited February 28 by awkduck

[SweetLow]

9 hours ago, waltah said:

 I'll run the same if I get USB 2 going at correct speed on the Win 98 system.

I waited that you running the same in ANY case as I asked you this for your problem insulation, not for my curiosity. But no luck.

[waltah]

4 hours ago, SweetLow said:

I waited that you running the same in ANY case as I asked you this for your problem insulation, not for my curiosity. But no luck.

This isn't a misunderstanding:  I simply don't see the benefit.   Benchmarks are valuable for comparisons across different platforms and situations, particularly when small differences may be important -- is this an improved model, or not?    My situation is a (roughly)  6x difference in performance between two operating systems on the exact same hardware (I just plug one sysres volume or the other) and copying the same data from the same device.   

It is true that a benchmark might tell me the  truth is one system is only 4.9x faster (or 6.9, or 7.2 or ...) but how would that matter?  As far as I can tell, it wouldn't matter at all.  To the extent that I have a real issue at all (rather than just a 'ship in a bottle' exercise) being able to copy a 150 meg file is the bottom line.  

What would be useful now is a tool that fingers some specific part of the process and thus tells us which module and function aren't going right.   I haven't found anything that'll do that, though AI told me that it was ebedded in Win 98 and gave specific directions for turning it on. (Which did not work ... And when I specifically questioned that, it couldn't find anything.)

If I'm wrong about all this -- just misunderstanding something -- please enlighten.