[Support] Hacked usbport.sys (500 Hz)

Discuss & post Update Pack addons here.
Locked
ViVa

[Support] Hacked usbport.sys (500 Hz)

Post by ViVa » Tue May 16, 2006 11:03 pm

File is version 5.1.2600.2846
http://addons.ryanvm.net/viva/ViVaAddonsUSB_1.3.cab (61 KB)
MD5: CDED81601D95639053FAA5F6EA329051

This hack is not recommended for wireless mice!
A big thanks goes to the founder of the hack, RaZieL. I would also like to thank Zacam, who discovered how to patch newer versions. :)

PS: I will release 1.4 next month (file version 5.1.2600.2891), when KB918005 is included in Ryan's next Update Pack, stay tuned.
Last edited by ViVa on Tue May 23, 2006 6:34 pm, edited 10 times in total.

User avatar
Zacam
Moderator
Posts: 615
Joined: Tue Sep 13, 2005 7:46 pm
Location: Portland, OR

Post by Zacam » Wed May 17, 2006 3:57 am

Woah......I had _completely_ forgotten about that nifty! Okay, I'm now going to have to re-aquaint myself with more system files again. :)

Much appreciated ViVa!

armond
Posts: 263
Joined: Tue Sep 13, 2005 11:43 am
Location: Glendale, CA USA

Post by armond » Wed May 17, 2006 4:48 am

ViVa, What the hacked file actually do?

User avatar
Aserone
Posts: 306
Joined: Sat Mar 11, 2006 8:46 pm
Location: Opioid dependent head.

Post by Aserone » Wed May 17, 2006 4:50 am

Oh, darn it Zacam, we missed one hacked file. :P ;)

btw really great work ViVa!

I'm always on the lookout for ways to hack windows files to get more freedom. If you can post how it was hacked/more info it would be mucho appreciated! (nLite way?)

@armond: it will increase the hz(frequency info is sent) of your usb devices. Wich means smother mouse movements etc. max hz with ps/2 is 200hz in windos default. If you for example use this file with your mouse instead you get 500hz(it's locked @ 125 default, less than ps/2!) using the usb port. Great for games if you need precision.
//Aserone-Foxy | EULA: Everything I say/post is a lie and/or forwarded from swim.

User avatar
Siginet
Site Admin
Posts: 2894
Joined: Fri May 27, 2005 1:07 pm
Location: Planet Earth
Contact:

Re: [Release] Hacked usbport.sys (500 Hz)

Post by Siginet » Wed May 17, 2006 5:44 am

ViVa wrote: PS: This isn't for wireless mice!
Is there an issue with wireless mice? Or it just doesn't do anything for wireless mice?

ThX ViVa!
Image
--Siginet--

Techware
Your Virtual Technician
Computer Management Software

User avatar
Zacam
Moderator
Posts: 615
Joined: Tue Sep 13, 2005 7:46 pm
Location: Portland, OR

Post by Zacam » Wed May 17, 2006 5:48 am

So, why is it a direct HEX compare of this file vs the one in RVM.2.0.6 shows no difference?

User avatar
Zacam
Moderator
Posts: 615
Joined: Tue Sep 13, 2005 7:46 pm
Location: Portland, OR

Post by Zacam » Wed May 17, 2006 6:01 am

Okay, this is bizzare. I booted to safe mode on a test system and replaced it's usbport.sys file with the one from RVM 2.0.6. I then ran a patcher utility to put it @ 500mhz (after first verifying it as being at 125). Patch _supposedly_ worked as the detection program now reads up to 500mhz, but there is NO difference between the original 2.0.6 SYS file and the _supposedly_ patched file nor is there any difference between the patched file and this file any more than there is a difference between this file and the one in 2.0.6. (test computer is running 2.0.5 and the same test applied with those older files also shows no difference)

Whiskey-Tango-Foxtrot, over?

Further, all posts that I've found anywhere that pupport to changing the usbport.sys file list a specific key string to search for. This hex does not exist in the 2.0.3, 2.0.4, 2.0.5 or 2.0.6 usbport.sys files, so now I'm really curious.

(EDIT: Trying to look at the files that the Logitech MX510 Mouseware installs for it's rampant acceleration doesn't help any because it uses it's own files without replacing any windows files. I also haven't noted and registry key's being changed via the purported _patching_ applications)

User avatar
Aserone
Posts: 306
Joined: Sat Mar 11, 2006 8:46 pm
Location: Opioid dependent head.

Post by Aserone » Wed May 17, 2006 6:43 am

Zacam wrote:So, why is it a direct HEX compare of this file vs the one in RVM.2.0.6 shows no difference?
Yepp, different hash, but the NT header checksum is the same 2F850 and correct, size the same on the byte.

Personally i'm not using this one, i'm a ps/2 freak, and i'm to lazy to search, if someone can post info/files that patch usbport.sys please do!
//Aserone-Foxy | EULA: Everything I say/post is a lie and/or forwarded from swim.

User avatar
5eraph
Site Admin
Posts: 4621
Joined: Tue Jul 05, 2005 9:38 pm
Location: Riverview, MI USA

Post by 5eraph » Wed May 17, 2006 6:45 am

Do the file hashes match between hacked and unhacked versions? This is the defining factor I usually use.

EDIT: Nevermind. :)
Zacam wrote:Whiskey-Tango-Foxtrot, over?
LOL

ViVa

Post by ViVa » Wed May 17, 2006 7:10 am

Holy sh*t! Here's the story:

The founder of the hack is RaZieL, http://www.warleagues.com/viewnews.asp?view=45&page=1
That hack only works until 5.1.2600.2180 of usbport.sys since "3C 08 73 09 C6 86 0A 01" isn't in 5.1.2600.2846 at all!
I contacted RaZieL about this and I will get back to this matter asap.
This updated usbport.sys is something from KB913365 (April 25, 2006), which was introduced in RyanVM Update Pack 2.0.6 as well.

Lemonzest
Posts: 114
Joined: Tue Mar 07, 2006 8:19 am

Post by Lemonzest » Wed May 17, 2006 11:29 am

maybe you should try this

http://www.stud.ntnu.no/~skara/files/UsbRate.rar

its made by RaZieL and can patch ALL rates in real time, see if it works with that usbport.sys you have, if not email him.
AMD Athlon64 3000+ @2.4GHz, DFI Lanparty UT NF3 250Gb, 2x1024MB Geil Value Ram PC3200
ATI Radeon X800XT VIVO, SyncMaster 913N LCD, Logitech MX518, Creative Audigy2 ZS
Enermax Noisetaker 485W, Western Digital SATA 160GB, Windows Server 2003 R2 Standard

User avatar
Zacam
Moderator
Posts: 615
Joined: Tue Sep 13, 2005 7:46 pm
Location: Portland, OR

Post by Zacam » Wed May 17, 2006 4:20 pm

@Lemonzest: Using the file noted there and many others on the usbport.sys file from RVM 2.0.5 & 2.0.6 _say's_ that it works, this and the "MouseRate" program that's floating around _does_ register higher Mhz for the mouse after usage. However, checksum and direct hex compare of the file from before and after show _absolutely no changes_ made to the file, so I'm at a loss to explain how or why or where the "MouseRate" executable is managing to poll it's information from or how/why the results change.

@ViVa: Correct. The hex sequence also does not exist in SYS version 5.1.2600.2730 either. And what few files that I have found to _patch_ usbport.sys all derive from that information, but there is still the incongruity that something in the detection scheme changes even though the file affected apparently does not. I even hex searched for the string relevant to the 500Mhz speed and that code is not present either in a supposedly patched sys file.

Here are the links for the_supposedly_ working MouseRate Switcher for USB and Mouse Rate Checker that I used for testing. Note that any file labled as XP_USB_Patcher.zip works specifically on the provided HEX sequence and fails, the above MSRS and the utility linked by Lemonzest are the only ones that gives the apperance of working.

*edit* Followup: I'm looking into the ASM level to see what registers are being resourced by that hex sequence to determine if said registers exist in later versions of the file and if it's manipulatable at all. I should hopefully have some info soon as to wether or not it's still possible to effect this modification, or if the process of patching is doing something else entirely. I'm also going to rip apart my Logitech MouseWare install files to see how _they_ accomplish the rampant increase in acceleration and see if it's pointers can be used in a default XP sys file.

User avatar
Zacam
Moderator
Posts: 615
Joined: Tue Sep 13, 2005 7:46 pm
Location: Portland, OR

Post by Zacam » Mon May 22, 2006 1:56 am

Okay, so in looking at how and where the patching methods work on the .2180 files, it has something to do with the values for ntoskrnl.exe!InterlockedDecrement. How the patch actually manages to work properly by how it changes what it does, I still have no idea and there doesn't seem to be any way to translate it over to the newer DLL's. I haven't given up yet, though, but I figured someone would want an update.

*EDIT*
Fixed it! The problem is with this "3C 08 73 09 C6 86 0A 01". Namely, that post .2180 files, the endian is actually "0E 01". So, simply searching for "3C 08 73 09 C6 86" will find the new offset (in the same approximate location) as the old ofset (instead of 0x015D00 it's now 0x015D60).

Will patch the 2 post .2180 versions to 500mhz for hosting in addition to the .2180 file.

User avatar
Zacam
Moderator
Posts: 615
Joined: Tue Sep 13, 2005 7:46 pm
Location: Portland, OR

Post by Zacam » Mon May 22, 2006 4:00 am

*edit*

Replaced by this topic
Last edited by Zacam on Tue May 23, 2006 6:14 am, edited 2 times in total.

User avatar
RyanVM
Site Admin
Posts: 5189
Joined: Tue Nov 23, 2004 6:03 pm
Location: Pennsylvania
Contact:

Post by RyanVM » Mon May 22, 2006 8:36 am

heh, and there'll be another new one next month for KB918005 :P
Get up to $200 off on hosting from the same people who host this website!
http://www.ryanvm.net/forum/viewtopic.php?t=2357

ViVa

Post by ViVa » Mon May 22, 2006 10:08 am

Updated first post, thanks Zacam for the research!! :)

I have been able to reproduce. The differences are (build 5.1.2600.2846):

Code: Select all

         125  500

15D66:	3C	B0
15D67:	08	02
15D6A:	C6	52
15D6B:	86	8C

User avatar
Zacam
Moderator
Posts: 615
Joined: Tue Sep 13, 2005 7:46 pm
Location: Portland, OR

Post by Zacam » Mon May 22, 2006 5:08 pm

Heh. And since I'll also be getting a laptop myself soon, a useful file to have. I've already patched it, btw.
Last edited by Zacam on Mon May 22, 2006 5:29 pm, edited 2 times in total.

ViVa

Post by ViVa » Mon May 22, 2006 5:16 pm

You've already patched what?

User avatar
Zacam
Moderator
Posts: 615
Joined: Tue Sep 13, 2005 7:46 pm
Location: Portland, OR

Post by Zacam » Mon May 22, 2006 5:28 pm

5.1.2600.2891, of course. :)

User avatar
RyanVM
Site Admin
Posts: 5189
Joined: Tue Nov 23, 2004 6:03 pm
Location: Pennsylvania
Contact:

Post by RyanVM » Mon May 22, 2006 5:37 pm

You guys don't have to wait for me, you know ;)
Get up to $200 off on hosting from the same people who host this website!
http://www.ryanvm.net/forum/viewtopic.php?t=2357

User avatar
Zacam
Moderator
Posts: 615
Joined: Tue Sep 13, 2005 7:46 pm
Location: Portland, OR

Post by Zacam » Mon May 22, 2006 5:40 pm

*edit*

Replaced by this topic
Last edited by Zacam on Tue May 23, 2006 6:15 am, edited 1 time in total.

ViVa

Post by ViVa » Mon May 22, 2006 5:46 pm

Ah, that's what you meant. Well, I decided to wait for Ryan's next pack, since it most likely will cause problems with ppl that do not read further than the download link. ;)

User avatar
Stellvia12
Posts: 63
Joined: Mon Feb 27, 2006 11:38 pm
Location: Japan

Post by Stellvia12 » Tue May 23, 2006 12:30 am

Question, I know this patch makes my G7 mouse smooth at higher dpi, but I'm concerned about my flash drive. Does anyone have proof that this patch does not burn your flash drive?
FYI GUNDAM SEED Destiny DVD 3-4 ARE COMING OUT SOON!!! Scratch that its out....

User avatar
Zacam
Moderator
Posts: 615
Joined: Tue Sep 13, 2005 7:46 pm
Location: Portland, OR

Post by Zacam » Tue May 23, 2006 3:25 am

Considering that I just tested every version since .2180, I can assert that, patched to 500 hz, I have not had any problems (outside of whatever native problems existed for that version #) in regards to USB device usage.

To more specifically answer your question, 5.1.2600.2891 has had no detrimental side effects to my mouse, keyboard, joystick, scanner, thumb drive (including a flash-card to thumb drive converter), external hard-drive, external dvd-rw burner. However, not all usb hardware or motherboard manufactured USB ports are made the same. If you have any concerns regarding 5.1.2600.2891 specifically, test it in it's original vanilla state first.

As for the patch portion of it, if for your G7 mouse you can install the Mouseware drivers and enable acceleration in them without any side effects to your USB flash drive, then this file should pose no difficulties.

User avatar
Aserone
Posts: 306
Joined: Sat Mar 11, 2006 8:46 pm
Location: Opioid dependent head.

Post by Aserone » Tue May 23, 2006 5:06 am

Zacam wrote:Release: ZacamAddon_Patched_USB_5.1.2600.2891.cab
MD5: 4DF774CD787E3CC50267CC8D2B42110B
(For use with KB918005 (Validation Required) & the next RVMUpdatePack)
Thanks Zacam, will integrate this to my RVM 2.0.6 now...
//Aserone-Foxy | EULA: Everything I say/post is a lie and/or forwarded from swim.

ViVa

Post by ViVa » Tue May 23, 2006 6:51 am

Aserone: that's a mistake! You need KB918005, like I stated in my first post!

ViVa

Post by ViVa » Tue May 23, 2006 6:53 am

Zacam: I'm having doubts about the patch... Can it be that just patching usbport.sys isn't enough? Could there be other patched files for it to work too? When I use my patched file, http://www.stud.ntnu.no/~skara/files/UsbRate.rar still claims I have 125 Hz. When I patch it real time with the patcher, it says I have 500 Hz, but usbport.sys is completely unchanged (since it's already patched of course)...

User avatar
Aserone
Posts: 306
Joined: Sat Mar 11, 2006 8:46 pm
Location: Opioid dependent head.

Post by Aserone » Tue May 23, 2006 6:54 am

ViVa wrote:Aserone: that's a mistake! You need KB918005, like I stated in my first post!
Yeah I know, and I know how to integrate/replace the whole hotfix too. Don't worry.. ;)
Last edited by Aserone on Tue May 23, 2006 6:55 am, edited 1 time in total.
//Aserone-Foxy | EULA: Everything I say/post is a lie and/or forwarded from swim.

ViVa

Post by ViVa » Tue May 23, 2006 6:55 am

Aserone: but stating that you will integrate it to "your RVM 2.0.6" looked confusing... :)

User avatar
Aserone
Posts: 306
Joined: Sat Mar 11, 2006 8:46 pm
Location: Opioid dependent head.

Post by Aserone » Tue May 23, 2006 6:56 am

ViVa wrote:Aserone: but stating that you will integrate it to "your RVM 2.0.6" looked confusing... :)
Hehe, yeah but i'm building my own updatepack with ryans as a base all the time..
//Aserone-Foxy | EULA: Everything I say/post is a lie and/or forwarded from swim.

ViVa

Post by ViVa » Tue May 23, 2006 7:01 am

:)

User avatar
Zacam
Moderator
Posts: 615
Joined: Tue Sep 13, 2005 7:46 pm
Location: Portland, OR

Post by Zacam » Tue May 23, 2006 7:05 am

ViVa wrote:Zacam: I'm having doubts about the patch... Can it be that just patching usbport.sys isn't enough? Could there be other patched files for it to work too? When I use my patched file, http://www.stud.ntnu.no/~skara/files/UsbRate.rar still claims I have 125 Hz. When I patch it real time with the patcher, it says I have 500 Hz, but usbport.sys is completely unchanged (since it's already patched of course)...
I ran into a similiar "defeciency" of detection as well. However, observing the behavior of the UsbRate.rar when it does both realtime (and non-realtime) patching, it's not touching _anything_ else. I just think it's SET to default assume that if you're running it, that the value at the offset is going to be 125 unless/until it registers that IT has patched the file. Other mouserate detectors not built into that patcher do detect properly.

ViVa

Post by ViVa » Tue May 23, 2006 7:07 am

Hmm, you could be right. Are you sure it doesn't touch other files? Checked with some kind of file monitor?

User avatar
Zacam
Moderator
Posts: 615
Joined: Tue Sep 13, 2005 7:46 pm
Location: Portland, OR

Post by Zacam » Tue May 23, 2006 7:10 am

File access monitor and a registry access monitor as well as a real-time memory scanner. Just to be even more sure, I snap shot the hard drive and registry before and after.

ViVa

Post by ViVa » Tue May 23, 2006 7:11 am

So, the only thing possible is that the patcher has 125 Hz set as a default? And usbport.sys is the only file that needs to be patched?

User avatar
Zacam
Moderator
Posts: 615
Joined: Tue Sep 13, 2005 7:46 pm
Location: Portland, OR

Post by Zacam » Tue May 23, 2006 7:16 am

As far as I can determine. Without source code to prove contrary that would be my best guess. And what sourcecode I've seen for other "patchers" they default assume and/or can only scan for the "3C 08 73 09 C6 86 0A 01" instead of the shorter (more accurate cross version) "3C 08 73 09 C6 86" value.

Do a clean test with the patched files and Mouse Rate Checker 1.1b and see what it says. Then swap out an umodified version and double check, Hz will change accordingly.

User avatar
RyanVM
Site Admin
Posts: 5189
Joined: Tue Nov 23, 2004 6:03 pm
Location: Pennsylvania
Contact:

Post by RyanVM » Tue May 23, 2006 6:36 pm

Get up to $200 off on hosting from the same people who host this website!
http://www.ryanvm.net/forum/viewtopic.php?t=2357

Locked