I-Appliance BBS The Official Source for Internet Appliance Upgrades and Mods BBS Main List | Sign In | Sign Up | Search | Help | Linux-Hacker.net Reply to Thread | Printer |

Home / I-Opener Areas / I-Opener Technical Stuff USB fix: final version. Cures all USB transfer problems. Works in any OS. USB works at last......

USB fix: final version. Cures all USB transfer problems. Works in any OS. (modified 0 times) Georgie Profile So is there anyone intereseted in a working USB fix? This fix will enable YOU to use USB hard drives, floppy drives, CDRW drives, USB NICs with any network protocols - not only TCP/IP -, any kind of bidirectional USB device, under any operating system. Which has been impossible so far. I will release it here, in the coming days. Stay tuned. Georgie 04-13-2002 21:38:44

RE:USB fix: final version. Cures all USB transfer problems. Works in any OS. (modified 0 times) YouBecha Profile You do like teasing us don't you Georgie. ;) http://www.geocities.com/mr_bubba_zanetti/ 04-14-2002 08:19:43

RE:USB fix: final version. Cures all USB transfer problems. Works in any OS. (modified 0 times) redwood Profile well Georgie, let me be first to Thank you.. last time, was a bit of a tease.. but, we Know, you have the answers.. just not the time to post.. so, lets keep this thread at the top! :) gawd, I hope its something a dummy like me can figger out! :) 04-14-2002 08:30:34

Just one or two more days. (modified 0 times) Georgie Profile Yes, but you do like being teased with some really good news, don't you? ;)) I am sorry to make you wait, I need one or two more days for "beta testing". I simply want to know that it works on someone else's IOpener without modification. Then, I will take the time to post... Thanks for your patience. Keep the faith! Georgie 04-14-2002 21:55:17

Done..... (modified 0 times) Georgie Profile I am happy to let you know that you can read the USB fix in the Technical Reference area, here. Any questions and/or comments are welcome here. I wish a flawless USB port for Everybody! Later, Georgie 04-18-2002 00:16:07

RE:USB fix: final version. Cures all USB transfer problems. Works in any OS. (modified 0 times) alcuin Profile Hi Georgie, Good job on finding the proper usb settings; you may want to edit your last post and re-put the url links to John Fine's PCI program, so other people don't get lost trying to find it; a few questions though. The pci setting should have to be reinitialized in autoexec.bat every time I boot up? John Fine doesn't mention anything about PCI being run in W2k or WXP; Can it be done? If this works in any OS, how is the patch applied in linux? Thanks! -e 04-18-2002 03:43:19

RE:USB fix: final version. Cures all USB transfer problems. Works in any OS. (modified 0 times) alcuin Profile Nevermind about previous post; I suppose if i'm good, WP will make us a fixed bios =D -e 04-18-2002 03:48:32

RE:USB fix: final version. Cures all USB transfer problems. Works in any OS. (modified 0 times) redwood Profile ok, I have pci.exe in c:\ in a dos window I paste, pci r75=a3 r40,7,2=2 That, should do it? correct? .. but... I am testing, using, Labtec USB speakers, playing MP3's .. and, I'm sorry to say, it still stalls, when I move the mouse.. now, shouldn't this Fix my speakers also? I hope there is a way... I bought several sets of these speakers Long ago.. and, theyve been nearly worthless... Thanks for any tips! 04-18-2002 11:33:36

RE:USB fix: final version. Cures all USB transfer problems. Works in any OS. (modified 0 times) hotrod_nut Profile Georgie or anyone else, Thanks for the time you put into the USB fix. I am sorry but it is not working for me. First of all I had to search for where to get this "PCI.EXE" program and at first I downloaded this 1.3MB file and tired it. Then I relised it was the wrong one. I found out that I could get it here (correct me if I am wrong): http://www.execpc.com/~geezer/johnfine/ Next I "assumed" that I was to put the command in the autoexec.bat file and stays there for every boot. I have a USB network adapter that you can get at bestbuy, you know the really tiny one (USB100M) found here: http://www.linksys.com/products/product.asp?prid=402&grid=26 Now here is the funny thing. I was not able to get it to work at all on one of my IO systems (call it IO1). So a friend tried it on his and it worked and still works today. I am running 98lite from Win98SE and he is running full Win98SE, maybe that is it. Anywho, it Works great for him. So I tried it on my second system (IO2), 98lite again, and it worked great. I downlaoded 5GBs of MP3s over that night that I installed the OS and had no problems, so grabbed an image of it. I even tried it just now out in the car and it GREAT, I was even surfin via internet connection sharing. I put that image from IO2 on IO1 (was not working right) and it WORKED! ... until the next day. It is funny because it is intermitten, those are a B*@#H. I just tired it again and it work but then eventually it quits. So I saw your patch and tried it and it dose nothing. Now I cannot get it to work again, it still doesn't work. :( Any ideas? What could I have done to it? Cheers, Patrick 04-18-2002 19:55:03

RE:USB fix: final version. Cures all USB transfer problems. Works in any OS. (modified 0 times) vailr Profile Wasn't part of the problem, that USB & video share IRQ's? Even though other IRQ's are unused, IRQ 15 has 2 VIA USB devices plus the Trident video device. WinXP. And the bios won't let you assign to specific IRQ's, to correct this. 04-18-2002 22:54:38

Updates (modified 0 times) Georgie Profile I have just updated my post in the technical reference area. It should be complete now. alcuin, Yes you should put pci.exe in your autoexec.bat. Yes, it will be part of the BIOS. redwood, Your speakers may have another, although related problem. After applying my USB fix your USB port transfers data out without corruption. But I think that your problem was not your USB port in the first place. In order to learn what your problem actually is I would need you to answer a few questions: What is your IO config: IO version, CPU speed, HDD, RAM? Are you using animated mouse cursors? What software are you using to play MP3s? Are there any differences between what you can hear through the USB speakers and what you can hear through the built-in speakers? In other words, if you uninstall the USB speakers do you experience the same kind of stalls through the built-in speakers? hotrod_nut, I am sorry for the lack of the "how to" paragraph. It is updated and hopefully complete now. On another note, that your USB NIC worked on your IO prior to my USB fix means that you use TCP/IP as your network protocol. TCP/IP has built-in error correction. Try NetBEUI and you will understand, that this USB fix works. Most probably your LAN will be twice as fast with NetBEUI and the fix than with TCP/IP and without the fix. It sounds to me that your problem of not being able to use your USB NIC on one of your IOs is not related to this fix, or not directly. What differences are there between the two IOs in terms of hardware? vailr, IRQ15 has the PCI holder for IRQ steering too. But this is how it is supposed to be. And it works flawlessly for me. No problem there. Later, Georgie 04-19-2002 08:06:24

RE:USB fix: final version. Cures all USB transfer problems. Works in any OS. (modified 1 times) hotrod_nut Profile Georgie, Looks like you have your hands full, I hope you knew what you were getting into. Again thanks for all your time. I don't think it is a network problem but I could be wrong. I will try installing NetBEUI tonight. Otherwise my other IO2 still works great and IO1 still works every so often. What I do see happen is that the LEDs on the adapter are off. It seems like the USB driver gets locked up. The LEDs are on for a moment during boot of windows then they turn off. Windows also will not shut down when the adapter is not working. Sometimes when the adapter is plugged in Windows wont even boot, it locks at the splash screen. As for hardware differences there is only one, the other IO has two MAX205 chips to convert the RS232 signals on both comm ports. Otherwise everything else is the same, even the same HD part number. This includes: -HD -Heat sink mod -IDE cable (badflash.com) -Sound mod -CF mod -Modem cards have been removed -disabled Sandisk (cut two traces on the board) -32mb ram (using seperate partition HD for swap file) -the OS install was made from the same image -BIOS Version (badflash.com) Any other ideas would be apreciated. What could I have different in the BIOS settings to cause this? Kind of off subject in the BIOS, why does the BIOS still need an ESC key in order to exit one submenu and go back to the main menu? Cheers, Patrick 04-19-2002 11:26:16

RE:USB fix: final version. Cures all USB transfer problems. Works in any OS. (modified 0 times) Wild_Pencil Profile | Email How frustrating! A great fix that deserves to be part of the BIOS, and all my stuff is in transit! Since it's no longer a Bus-Speed Specific fix, I can safely add this to the base BIOS. It'll be out as soon as I can get my new Lab set up. I'm still a little confused about the overall solution -- does this new fix replace the old Georgie USB-Tweak fix, or is it intended to augment the old Georgie USB-Tweak fix? -WP 04-19-2002 13:40:11

RE:USB fix: final version. Cures all USB transfer problems. Works in any OS. (modified 0 times) hotrod_nut Profile Georgie, I tried installing NetBEUI and no difference. If anyone has any other ideas I am willing to try them. Cheers, Patrick 04-20-2002 09:18:21

RE:USB fix: final version. Cures all USB transfer problems. Works in any OS. (modified 0 times) ckbone Profile Hotrod_nut any others reading this thread.....This latest USB fix from Georgie, so far as I understand, does only one thing: It corrects USB transfer problems over various LAN devices, and corrects outward USB transfers to disk drives, cdrws, etc. It does not correct device driver problems, configuration problems, OS compatability problems, and other VIA MVP4 chipset problems ....which there are many. I have tried many different USB devices on my three different I-Openers, and on a Compaq computer of mine which has the same MVP4 chipset as the I-Opener. Many different USB problems show up for me. The worst, I guess, and it happens with both computers, is the chipset occasionally losing a device which previously worked ok. You boot up..the device is gone. Maybe it shows up in device manager as an "unknown" device, maybe not. Sometimes it will start working if you reload the driver, sometimes not. Other problems....sometimes either the I-Opener or the Compaq will not boot if an externally powered USB device is attached and powered up. This happens with several USB hubs I have. Sometimes the computers will not shutdown properly if this USB device is attached. I have a Cannon scanner...not externally powered....My compaq will not shutdown if this device is attached. In short, VIA chipsets, not only MVP4s, have all sorts of USB problems. Georgie's latest fix will not cure every USB/I-Opener problem you may have. I suspect the only way to fix some of these problems is to trash the offending device, try a different piece of hardware. Georgie pulled off a great patch for I-Opener USB problems......but it will not fix every USB problem. By the way, if you can, try Windows XP. It completely fixed my USB problems with my Compaq, and my I-Opener.....so far anyway. 04-20-2002 15:14:01

RE:USB fix: final version. Cures all USB transfer problems. Works in any OS. (modified 0 times) ckbone Profile For USB help from VIA try here....... http://forums.viaarena.com/categories.cfm?catid=16 04-20-2002 17:34:59

RE:USB fix: final version. Cures all USB transfer problems. Works in any OS. (modified 0 times) laserfan Profile For the record, I've been using my IO as strictly an email (the LED is a joy!) and surfing machine (less joy but most sites are OK in 800x600) with only a USB network adapter to connect to my home LAN. The outbound USB issues have been a real pain because I could never trust whether an email attachment or website download could be off-loaded to my server or other networked PCs "accurately". So the fix is wonderful, and I thank George for persisting to figure this out for us! 04-21-2002 07:19:28

RE:USB fix: final version. Cures all USB transfer problems. Works in any OS. (modified 0 times) zonyl Profile I run the "other OS" (Linux) and really hadnt noticed the problem.. Running everthing on IP of course though so it doesnt surprise me that there is some corruption somewhere. Does someone have a some info on how to accomplish this change to the PCI register while in Linux? I boot from Flash so I dont have room for an FAT partition to loadlin it. 04-21-2002 08:38:19

RE:USB fix: final version. Cures all USB transfer problems. Works in any OS. (modified 1 times) milhouse Profile EDIT: Nevermind 04-21-2002 17:51:07

Answers? (modified 0 times) Georgie Profile hotrod_nut, You misunderstood what I wrote about NetBEUI. I just wanted to point out that you have been using TCP/IP as your network protocol, which is why you have actually never experienced the USB-out data corruption problem. TCP/IP has built-in error-correction, if a corrupted data packet arrives it will be retransmitted. Therefore, you never experience data corruption on TCP/IP, but all the retransmissions degrade network speed. NetBEUI has no error correction, thus it is perfect to demonstrate that the USB fix works: almost all your files will be corrupted without the fix. I just wanted to say that the fix does work and does what it is supposed to. It is your decision which protocol you use for your LAN. On another note, if you write about the green LEDs, it is a serious problem. Yes, your driver seems to have communication problems to your hardware (NIC)? Have you tried it with another NIC? The one that is known to work on the other IO? Yes, there may be a difference in the CMOS settings affecting your USB NIC. Make sure you have the exact same CMOS settings on the two IOs. One way to do it is with Charles Dye's cmos.com. Just a question: have you updated your uhcd.sys driver? You can read about it again in my post in the techref area. Georgie 04-22-2002 15:27:58

More answers.... (modified 0 times) Georgie Profile Wild_Pencil, Yes, the old "Georgie USB fix" is history. This one replaces that old one completely. Use this one alone. Thanks! zonyl, I am sure you can find a Linux program to do this. But you won't need it anyway once the fix is part of the BIOS. EVERYBODY, ckbone has a valid point. Please, make sure that you setup your hardware correctly using the latest drivers possible. The same goes for any OS updates (uhcd.sys !!). When everything is setup correctly and you are actually able to use your USB device, that is when you will need this fix to insure that outgoing data transfers will be accurate, uncorrupted. The absolute prerequisite for the application of this fix is that you are able to setup your USB device in your OS correctly, it shoes up and you can actually access/use it. Georgie 04-22-2002 16:02:49

Retry... (modified 0 times) Georgie Profile Ooooppps.... Let me try that again. The absolute prerequisite for the application of this fix is that you are able to setup your USB device in your OS correctly, it shows up and you can actually access/use it. Sorry fo the typo, Georgie 04-22-2002 16:08:03

RE:USB fix: final version. Cures all USB transfer problems. Works in any OS. (modified 0 times) laserfan Profile George, you said: >you never experience data corruption on TCP/IP I have been using TCP/IP, not NetBEUI, and have been experiencing USB-out data corruption. "What do you mean"????? 04-22-2002 20:17:18

RE:USB fix: final version. Cures all USB transfer problems. Works in any OS. (modified 0 times) hotrod_nut Profile Georgie, ckbone, etc ... Georgie thanks again for all your time. I don't mean to get on anyone's bad side, with all the bold and red letters and all ... am I the only one that thinks that statement was totally not necessary? I see how I misunderstood your NetBEUI thing. I was thinking this was a good spot to discuss USB issues. Maybe someone else might have ran into the same issue and maybe found it to be a corrupt file that I would have to delete and reinstall or so on. I can see that I should make a new thread and not crowd Georgies Fix. Yes, I did update my uhcd.sys as instructed on your install notes. I will try compairing CMOS settings and going to other message boards. laserfan What NIC are you using? Cheers, Patrick 04-22-2002 21:07:22

????? (modified 0 times) Georgie Profile Hi laserfan, Are you saying that the only network protocol installed in Network Properties is TCP/IP? Don't you also have NetBEUI installed? TCP/IP should correct any corruption by resending the offending packet. George 04-22-2002 21:12:40

RE:USB fix: final version. Cures all USB transfer problems. Works in any OS. (modified 0 times) Georgie Profile hotrod_nut, Please, don't take all the red and bold letters personally, they are intended to emphasize important stuff mostly for those just scanning through the posts. It wasn't meant for you. I just wanted to avoid getting too many posts about USB devices not willing to work at all (no bold here, ;)) ). Since that can be caused by many things, and almost certainly the cause is not the data-out corruption. It is a good place to initiate the discussion of other USB problems as well, nothing wrong with that, and I will devote whatever time I can to helping others out with those ones too. But it is important to differentiate between an otherwise working USB device which only has problems with data-out corruption (in which case this fix is the solution), and a given USB device not working at all for whatever reason. Last but not least, please understand that I am a really friendly person and I am not into offending anyone here. Even if you feel I tend to overuse bold and/or colour I only do so to emphasize what I feel is important, and never to offend somebody. Georgie 04-22-2002 21:40:52

RE:USB fix: final version. Cures all USB transfer problems. Works in any OS. (modified 0 times) laserfan Profile George: I uninstalled NetBEUI from my IO long ago--been running only TCP/IP (and had xsfer errors) Patrick: I'm using Netgear's EA101 USB network adapter 04-23-2002 05:06:09

Even TCP/IP corrupted??? (modified 0 times) Georgie Profile laserfan, WOW!! You have used TCP/IP and even that hadn't worked prior to the fix?! Surprise, surprise! I mean really! TCP/IP should be able to deal with data corruption gracefully, but apparently the USB-out data corruption can be so bad on the IO that even TCP/IP cannot salvage the data. That truly sucks... The interesting thing is that to my knowledge many people have used the IOpener in a LAN with a USB NIC and TCP/IP without serious problems. May the severity of the USB problem vary from IOpener version to version? George 04-23-2002 16:03:53

RE:USB fix: final version. Cures all USB transfer problems. Works in any OS. (modified 0 times) zonyl Profile I worked a little on the pegasus driver for Linux back in the IO hay-day, and there was an awfull lot of corruption of ethernet packets. It would trigger a breakage in the driver, as the maintainer did not have broken hardware (an IO), I was relaying information and pressing him on handling the error's better. My guess is you are seeing dynamic results on whether IP worked at all with a USB adapter in Windows, would be due to specific nic vendors drivers that were not handling this correctly either to varying degree. To this day with the latest and greatest Linux pegasus driver, I am still seeing quite a bit of "babbling" from the driver on my IOs and I see my error count's climb as I use them, but for a while now are quite stable, albiet slow. It was quite frustrating for me in the beginning as all of my IOs remote boot and would get frequent lock-ups from the driver getting hosed. 04-24-2002 06:27:26

RE:USB fix: final version. Cures all USB transfer problems. Works in any OS. (modified 0 times) milhouse Profile Hi, It's been a while since I messed with my iO. I am running win2k and don't understand how to use hOda's apps to set the registers for proper USB function. I have used the WPCREDIT to tweak the memory settings on an old Via chipset but I seem to be lost here. Help/input from the guru's will be much appreciated. thank you :) 04-25-2002 19:58:59

How to apply the fix using WPCREDIT ? (modified 0 times) Georgie Profile milhouse, There are two registers to change. The first is register 75 in the Northbridge. When you start WPCREDIT, you will be looking at the 256 PCI registers of the Northbridge. Just go to line 7, column 5 and you have it. Click it, and begin editing the bit pattern in the lower right. These are the bit values to achieve "A3" hex: 10100011 . For the other register you have to open the 256 registers of the USB controller. Click Edit/Device, and select Bus 0, Device 7, Function 2, the first USB controller you will see in the list. In the matrix click on register 40, and change the bit pattern from "00000000" to "00000010" in the lower right. You should prepare a WPCRSET config file for it, and have it autoloaded by Windows. Georgie 04-26-2002 19:57:44

RE:USB fix: final version. Cures all USB transfer problems. Works in any OS. (modified 0 times) bflem82 Profile hey, i was wondering:i recently purchased a BF-110N pc-pc network cable. i have been having some problems, and was wondering if this is a usb fix issue. on my Desktop, i have a cable modem hooked up to the ethernet card. by sharing this connection with the USB network bridge in my Local Area Network, i am sometimes able to get an internet connection on my iopener v5, but not always. i cant seem to figure out the problem. i'll have an internet connection for 20minutes and then its gone. the same is the case for my file sharing. sometimes i can connect a local network between the two to share files and such but not always. and when i am able to, the network doesnt last very long, as the "packets" that i see being sent from my desktop to my laptop suddenly stop being received by the iopener.if this is a usb fix problem, where can i find the fix and once i find it, what do i do with it? i read something about puting it in the autoexec.bat and pasting pci r75=a3 r40,7,2=2 but this isnt really clear. thanks 05-06-2002 18:27:07

RE:USB network cable (modified 0 times) Georgie Profile bflem82, Yes, it awfully sounds like the USB data corruption problem. And as such, I feel that this USB fix will solve your problems. I've just written a clean how-to at the end of the USB fix thread in the technical reference area. Please, read that whole thread again, and follow the detailed instructions in the last post. In case you have any problems, feel free to ask again. In any case, let us know about your results. Thanks. Georgie 05-06-2002 22:50:46

RE:USB fix: final version. Cures all USB transfer problems. Works in any OS. (modified 0 times) rscott Profile <continuation from thread in moderated area - sorry!> Yup, completely confused on the address. What threw me off is that pcitweak didn't like register 75 - it wants word aligned register accesses. So, armed with my new knowledge, I've made progress in the register setting part of the fix. Here's the result: pcitweak -r 0:0:0 0x74 0x00d2a10e pcitweak -w 0:0:0 0x74 0x00d2a30e pcitweak -r 0:0:0 0x74 0x00d2a30e pcitweak -r 0:7:2 0x40 0x00010000 pcitweak -w 0:7:2 0x00010002 pcitweak -r 0:7:2 0x40 0x00010002 So, as you can see, I've successfully changed the registers. However, I still can't do any NFS write operations, I can do NFS read, rsh, etc, into other boxes on my network. The write operations just timeout with "nfs: server xxxx not responding", followed by "nfs: task xxx can't get a request slot" I'm running redhat 7.3, AMD K6-2 @233 MHz, have used both a Netgear EA101, and a 3com 3c19250, with the same results. Am I doomed to never write to my network?? Thanks! 05-17-2002 22:15:55

RE:USB fix: final version. Cures all USB transfer problems. Works in any OS. (modified 0 times) Georgie Profile rscott, Use the "-b" switch for 1-byte register accesses. Try it this way. It might be an issue. Anyway, are you sure your NFS setup is correct? Have you tried any other USB transfer? Georgie 05-21-2002 01:26:46

RE:USB fix: final version. Cures all USB transfer problems. Works in any OS. (modified 0 times) rscott Profile I'll try the -b switch, next time I'm home (sometime next week), thanks! The nfs server is in use by all of the other computers, without a hitch. (It also does the usual SMB, DHCP, HTTP, mail, X-10, etc, services. Linux is great!) As far as other USB operations, no, I haven't tried any. I do have a compact flash USB reader/writer I could try. (And an old 8MB card I don't mind loosing rscott 05-21-2002 16:48:50

need help with WPCREDIT (modified 0 times) nascarnights Profile sorry - I posted this in the moderated section by accident, so I reposted it here -------------------------------------------- I'm trying to translate the PCI.EXE command into something I can use with H. Oda's tools. (I'm a pci register newbie) I'm guessing that the northbridge is zero and it's left out of "pci r75=a3" cause it's implied? The USB interface is device 7, but what is the extra 2 in "r40,7,2=2" I still have usb problems with my SMC ethernet adapter even with XP, so I want to try these tools but I can't quite figure them out. Any help would be appreciated. thanks. 05-28-2002 20:08:38

Help with WPCREDIT (modified 0 times) Georgie Profile nascarnights, Sorry for being slow to answer. > I'm guessing that the northbridge is zero and it's left out of "pci r75=a3" cause it's implied? Yes, zeroes - "0,0" in this case - can be just left out. > The USB interface is device 7, but what is the extra 2 in "r40,7,2=2" Function. The possible digits mean: r"register number","PCI device number","PCI device function number" I hope this helps. Georgie 06-07-2002 16:11:47

RE:USB fix: final version. Cures all USB transfer problems. Works in any OS. (modified 0 times) rscott Profile Another chapter in the linux USB network story... Well, I finally got a chance to try the -b switch, but still no joy Also tried writing to a CF card via my USB-to-CF adaptor - worked as fast as on my other systems, no corruption that I could detect. So, perhaps it's my choice of USB ethernet adaptors? The're both based on the Kawasaki chipset. Thanks, rscott 06-07-2002 18:37:38

RE:USB fix: final version. Cures all USB transfer problems. Works in any OS. (modified 0 times) spikeorama Profile Georgie, Thanks for the fix! It cured my Linux TX problems with NFS. Prior to this, RX would work fine but TX seemed to take twice as long and was very intermittent. I just used: pcitweak -w 00:07:02 0x40 0x00010002 for the usb controller. I did notice some interesting benchmarks. Using FTP to and from a server, I get about 470 kB/s, both with and without the tweak. FTP has always been stable on usb for me. However, NFS is much slower, around 250 kB/s - I'm guessing the NFS protocol has more overhead than FTP? In case anyone is curious, I'm using RH 7.3 but with a custom kernel (2.4.18). I was unable to get RH's patched source of 2.4.18 to work on the IO properly, although I can't remember anymore what the problem was. 06-18-2002 18:08:18

RE:USB fix: final version. Cures all USB transfer problems. Works in any OS. (modified 0 times) Jimmerman Profile See my post in the moderated thread. Setting 0:7:2 register 40 may not be a good idea. Doing so may force 8DW burst mode, not 16DW burst mode. 07-23-2002 20:35:44

AAAhhhhh!!!! (modified 0 times) Georgie Profile Jimmerman, Welcome to the board, you are obviously new here. The user ID of the message author is on the right hand side, not on the left. Nobody calls himself Profile here.... As for your comments everywhere, you should have been a bit more circumspect before posting them. You are dead wrong about what you say: we are talking about the VT82C686A Southbridge, not the VT82C586A/B chip, as you thought we were. No problem, but next time don't think everybody else is stupid, it probably never is the case. Take care, Georgie 07-24-2002 22:33:47

RE:USB fix: final version. Cures all USB transfer problems. Works in any OS. (modified 0 times) Jimmerman Profile Yeah. I realized the username is on the right after I sent that first message. Anyway, I wasn't implying that you were incorrect or stupid. I just wanted to let people know that your fix doesn't apply to the VT82C586A/B. I don't think you mentioned what chipset you were talking about ... tho as you noticed I am new to this board and I may have missed it. My I-Opener most certainly does have the VT82C586A/B (at least that is what the chip ID is). That is why I asked you which one your's had. I didn't say anything about your fix being wrong on the 686. Also I did and still do want to know your source for the VT8501 registers since I cannot find that info anywhere. I am one of those people who likes to know what I am mucking with before I go blindly writing to config registers. Like your fix for HD to tripple speeds doesn't mention that register pci0:7:1 4B (X) is really two settings (each half byte) for recovery time and pulse width, however you gave very good choices even if you did not explain them. For a "Technical" board, your post just seemed a bit light on the details and I followed up on some. As for the BIOS. I certainly don't want to install a BIOS that is going to set 8DW burst mode ... which that one will on the VT82C586A/B, and neither should anyone else with the 586A/B. I have everything I need to write my own BIOS, but prefer to stick things in the bootloader, because it's a lot easier to update that than the BIOS. Sorry if you were offended. I was just trying to make more information available, so other people with the 586A/B wouldn't tweak those registers. 07-25-2002 11:57:47

Please, open up your IOpener, and check the SB chip..... (modified 0 times) Georgie Profile Jimmerman, I did not have to specify which Southbridge I was talking about. The IOpener chipset is VIA MVP4, implying that the SB chip is VT82C686A. To my knowledge, no other version of MVP4 was ever done. > My I-Opener most certainly does have the VT82C586A/B (at least that is the chip ID That is an ID some software gave you. Either the software is inaccurate, or you took a piece of information incorrectly. Please, open up your IOpener, do have a look at the SouthBridge chip, and confirm that you also have a 686A SB. Please, do it, so that we can avoid the totally unnecessary confusion here. Or you can opt to just trust me on this one: you have a VT82C686A SouthBridge. > I did and still do want to know your source for the VT8501 registers Surprise: the documentation. It can be found, you just have to look for it. Or ask. > For a "Technical" board, your post just seemed a bit light on the details This is supposedly a BBS for hardware hackers. A true hacker would try to understand what the meaning of all that stuff is. Great, do so. But I don't think I was ever supposed to write a manual documenting all the settings the HDD speed fix does, actually it changes many things. I did try to give a good hint on what the USB fix does, though. It was short enough. All in all, I gave you the solutions to two major problems. After having done that let me at least expect that whoever is interested enough in understanding what does exactly what, will make the necessary steps themselves. Again: no offense was taken, but it was an unnecessary nuisance. Take care, Georgie 07-27-2002 21:49:29

Reply to Thread | Printer | All times are PST Powered by UltraBoard v1.62

I-Appliance BBS General - NEWS, Breaking NEWS - linux-hacker.net BBS on CD Questions - User Groups in Your Town - E-Bay, Yahoo Auction LINKS ONLY (MODERATED) - Jokes & Humor - usdtv box I-Opener Areas - I-Opener General Posts - I-Opener Technical Stuff - I-Opener Technical Info Reference (MODERATED) WebSurfer Areas - WebSurfer General Posts - WebSurfer Technical Stuff - WebSurfer Technical Info Reference (MODERATED) Other I-Appliances - Virgin Webplayer - WebPal - MailStation - 3Com Audrey - ThinkNIC / Oracle NIC - Gateway/AOL Connected Touch Pad - Philips/AOL-TV - ACER NT-150 STB - MSN Companion - Compaq IA - Emachines IA - IBM NetVista IA - Intel dot.station - MSNTV2 Network Computers - Fujitsu-Siemens Scovery Linux Net Computers - GallantWeb Server Consoles - Dreamcast - xbox hacking Web Pads - Epods One General Post - Panasonic CF-01 - Fujitsu Stylistic 1200 - Cybiko Thin Clients - Thin Clients DigitalVideoRecorders - DTV Recorders, Replay, TiVo, Etc... - Tivo Hacks Linux - Midori Linux General Board - ITX MotherBoards Single Board Computers - Earth Computers mARMalade Elite Member Area MISC Areas - The Swap Meet WE TAKE NO RESPONSIBILITY FOR ANY TRANSACTION HERE AND NO COMMERCIAL POST PERIOD !!! - Wireless Net - Cameras - Robotics - Routers/Gateways - Mattel JuiceBox