geekhack
geekhack Community => Keyboards => Topic started by: genegold on Tue, 16 June 2009, 16:27:27
-
6/23: For current state of the problem, see post #40
-------------------
While awaiting a blue cube, I got a Radio Shack adapter that worked right off with a used 1391401 that I had picked up on ebay. Then the cube came, plugged it in and the keyboad Caps Lock and Scroll Lock flashed on and off, which seems to be this keyboard's way of saying, not working (can't start the computer). So I tried a different USB port and it started right up. Strange, wonder what caused that?
Today, I shut down the computer, unplug my regular Kensington board from the PS/2 port, plug in the cube and it doesn't work, just the flashing lights. Switch USB ports - there are 8 - and no go on any of them. Try the RS one, same deal. Then I discovered something: If I start up with the Kensington board, shut down, then plug in the cube or the Radio Shack adapter and power up within 10 seconds or so the computer boots fine. Wait much longer, or try consecutive adapter start ups, no go.
So, what's going on here?
-
weird. i'd say its a current flow issue but i'm not sure exactly how. But for instance i've had usb drives that worked on the back usb ports but not the front ports and the explanation was that the front ports get less current and the slight difference was enough to cause some usb devices to not work.
Here it sounds almost like once the juice has started flowing on the port, its able to keep going for the (heavy current drawing) model M, but if you cold start it with the M, no go. I dont know how the current is controlled on the ports, but it almost sounds like a hysteresis issue maybe.
Anyway I'd blame the current draw that the blue cube (as an active controller) is adding, maybe. Thats my initial superficial guess.
-
Bad cube. Send it back.
I've given up recommending dongles. Next time someone asks I'll just tell them to buy a new PC.
Try reading my post again (is it not clear?). It can't be the cube, or at least the cube alone, since the same pattern occurs with the Radio Shack one. Though before the cube arrived, the RS one fired right up each time I used it.
-
weird. i'd say its a current flow issue..
Anyway I'd blame the current draw that the blue cube (as an active controller) is adding, maybe. Thats my initial superficial guess.
The Radio Shack one appears to be active, too: http://www.radioshack.com/product/index.jsp?productId=2103747. Both support USB 1.1 and the motherboard manual (a new Gigabyte GA-MA790X-UD4P) says USB 2.0/1.1. I'll have to try each of these after the computer has been sitting off for awhile.
-
IF I read that correctly the RS adapter worked until you plugged in the blue cube... I'm wondering if the high power draw of the M (100mA) + power draw of the cube caused some sort of overcurrent that fried your USB controller...
-
another thing to try would be to get a powered external usb hub and try it thru that. (assuming the usb controller isnt fried in your computer).
-
another thing to try would be to get a powered external usb hub and try it thru that. (assuming the usb controller isnt fried in your computer).
Same thought came to mind, but right now I've got the RS adapter hooked into the same USB port as the cube was last in. Maybe there is some other kind of interaction between the two and the mb. Could it by any chance be the keyboard's own board? I've got another kb on the way, but have to make a decision about this one by tomorrow.
By powered hub, do you mean powered off the bus or from the wall plug, presumably with a transformer? Looking at hubs online, what they mean by "self powered isn't always clearly stated.
-
It could be the keyboard. Controllers are like any other electrical equipment and 20 year old ones do go bad. That's why I recommend plugging it into another computer. If the lights start flashing you better send it back.
Is there harm in plugging in one of these hot? I don't have access to a computer where I can control the power.
-
By powered hub, do you mean powered off the bus or from the wall plug, presumably with a transformer? Looking at hubs online, what they mean by "self powered isn't always clearly stated.
powered off wall plug with a transformer. that way you can be sure its getting enough current and see if something else is the problem.
-
No problem if you're talking about hotplugging a USB port. Don't hotplug a PS/2 port - your computer lab (library, whatever) won't be happy.
Ideally a PS/2 port computer would be ideal but let's see what happens.
..... the Computer Lab's lights start dimming....
Does the PS/2 port keyboard have to be unplugged? I'm thinking about somebody not being happy if I or they have to reboot a computer on the PS/2 reattachment.
-
On a reattachment you don't need to reboot. At least I don't with an Asus P4B533-VM and windows xp.
-
I checked it over at a university library with the Radio Shack adapter - there wasn't room for the cube. Although it didn't like a hot plug in - same port, keys not the same - on restart and cold boot it did work ok (though not quite as quick a response as the Dell keyboard). That was a slow server system vs. my 1 minute boot, but sufficient to establish that it's probably not the keyboard itself.
-
I checked it over at a university library with the Radio Shack adapter - there wasn't room for the cube. Although it didn't like a hot plug in - same port, keys not the same - on restart and cold boot it did work ok (though not quite as quick a response as the Dell keyboard). That was a slow server system vs. my 1 minute boot, but sufficient to establish that it's probably not the keyboard itself.
i still vote for wonky current issues. Have you tried a powered usb hub? If its a current issue basically that combination of adapter/keyboard may just not work on that particular laptop (plugged in directly anyway). On other hand, if all it is is a current supply issue, a powered usb hub may solve it. (Again assuming nothing was fried inside the computer itself).
-
Welly, you and I may be the two worst guys to be debugging Genegold's problems. You of course have had Ultranav problems and are going USB.
I was sitting here trying out the Das again and "be-bomp" the USB connection dropped in the middle of a post. It had been working fine for 15 minutes!! I had it plugged into my Dell monitor so now have moved it to my Cables-To-Go powered hub to see if it is power related.
We're jinxed!
lol. when in doubt, upgrade! I always say ;)
I vaguely remember someone complaining about a Das dropping the connex, though maybe its another board i'm thinking of. (Hows that for helpful diagnosis ;) )
-
Welly, you and I may be the two worst guys to be debugging Genegold's problems. You of course have had Ultranav problems and are going USB.
I was sitting here trying out the Das again and "be-bomp" the USB connection dropped in the middle of a post. It had been working fine for 15 minutes!! I had it plugged into my Dell monitor so now have moved it to my Cables-To-Go powered hub to see if it is power related.
We're jinxed!
I bring good luck. Cheers!
Have to find a powered hub around my small town today.
-
i still vote for wonky current issues. Have you tried a powered usb hub? If its a current issue basically that combination of adapter/keyboard may just not work on that particular laptop (plugged in directly anyway). On other hand, if all it is is a current supply issue, a powered usb hub may solve it. (Again assuming nothing was fried inside the computer itself).
Just tried the Belkin wall-powered 4 seater. Worked at first and seemed to solve the problem. But just turned off the computer, then tried restart - nothing. Switched adapters, still nothing. Unplugged the IBM keyboard from the hub and plugged in Kensington PS/2 - still nothing, first time! Unplugged the Belkin from the computer's USB port, starts right up.
I keep wondering about the new motherboard and PSU, but what do I know.
-
i bet VIA is behind this somehow!
buy a new cube, or try an adaptor from radio shack or a computer store.
-
i bet VIA is behind this somehow!
buy a new cube, or try an adaptor from radio shack or a computer store.
You must have skipped the beginning. Cube and Radio Shack involved. But what are thinking re VIA? This is a two-week old Gigabyte GA-MA790X-UD4P board.
-
But just turned off the computer, then tried restart - nothing.
you mean the computer wont start? Or do you mean the keyboard isnt recognized?
if you mean the former, then you're saying with the hub plugged in the computer wont start, but without the hub plugged in, the computer starts?
Have you tried different usb ports on the computer? Is it the same result for the other ports? (It could be just one bum port you have, or a couple of bum ports if they're connected together internally. I did have a laptop once where one of the usb controllers was fried so two ports didnt work right but the other two worked fine).
-
you mean the computer wont start? Or do you mean the keyboard isnt recognized?
if you mean the former, then you're saying with the hub plugged in the computer wont start, but without the hub plugged in, the computer starts?
Have you tried different usb ports on the computer? Is it the same result for the other ports? (It could be just one bum port you have, or a couple of bum ports if they're connected together internally. I did have a laptop once where one of the usb controllers was fried so two ports didnt work right but the other two worked fine).
It seems that the original post needs to be replayed:
While awaiting a blue cube, I got a Radio Shack adapter that worked right off with a used 1391401 that I had picked up on ebay. Then the cube came, plugged it in and the keyboad Caps Lock and Scroll Lock flashed on and off, which seems to be this keyboard's way of saying, not working (can't start the computer). So I tried a different USB port and it started right up. Strange, wonder what caused that?
Today, I shut down the computer, unplug my regular Kensington board from the PS/2 port, plug in the cube and it doesn't work, just the flashing lights. Switch USB ports - there are 8 - and no go on any of them. Try the RS one, same deal. Then I discovered something: If I start up with the Kensington board, shut down, then plug in the cube or the Radio Shack adapter and power up within 10 seconds or so the computer boots fine. Wait much longer, or try consecutive adapter start ups, no go.
Add:
- No dead USB ports; it's worked and not worked off the same ports
- The hub worked at first through several startups this afternoon, then the problem this thread is about returned. Had to unplug the hub to get the computer to start even with a regular modern keyboard.
- Since using the IBM keyboard - intermittently since Friday - I've been losing the PCI parallel card port to the printer. When the hub was unplugged, the printer started churning pages again. But since I had just unplugged/replugged it, I don't know what the difference was (yet).
-
if your scrolllock/capslock lights are blinking, can you look up and see if those are diagnostic codes that its blinking? If they're blinking in a particular pattern they might be a code that you can look up. (Dell desktops for example work that way with four led's on the back panel). There must be somewhere where you can look up what that blinking means (whether its a pattern or not), probably on the computer manufacturer's support page.
Its definitely odd that the computer intermittently wont start with the hub plugged in. (have you googled that?). If you're using a powered hub, then it shouldnt be a power supply issue anymore, I think.
-
if your scrolllock/capslock lights are blinking, can you look up and see if those are diagnostic codes that its blinking? If they're blinking in a particular pattern they might be a code that you can look up. (Dell desktops for example work that way with four led's on the back panel). There must be somewhere where you can look up what that blinking means (whether its a pattern or not), probably on the computer manufacturer's support page.
Its definitely odd that the computer intermittently wont start with the hub plugged in. (have you googled that?). If you're using a powered hub, then it shouldnt be a power supply issue anymore, I think.
Seems the latter was an occasional problem with Win98, but little then and nothing more.
Found IBM error codes at http://forums.techarena.in/guides-tutorials/1045252.htm, but I assume those require at least a start into bios. One person reported similar flashing and it turned out to be a bad memory module. That doesn't make sense here, because the flashing started when the IBM keyboard was first hooked into the PS/2 port, pre adapter.
-
another question - if you leave the M keyboard out of the equation altogether, and only use the hub and the kensington, through several reboots and etc, does the problem occur? Or does it ONLY occur shortly after the M has been introduced into the sequence of events?
-
It seems that the original post needs to be replayed:
I'm afraid this isn't 100% clear. "Nothing happens" can mean many things:
- PC doesn't show any sign of life;
- Some lights and fans come on but drives don't spin up and there's no video signal;
- Drives spin up, but no video signal and no actual disk activity;
- You get no video but lots of disk activity as if the OS is trying to boot;
- You get video but the OS doesn't try to boot.
You get the idea, these details are needed to diagnose a fault.
Let's get back to basics. Firstly, stop hot-plugging PS/2 connectors if you are doing so. Even the die-hard "I've done it hundreds of times without any problems" people will accept it's not a good thing to do when diagnosing a fault like this.
Next, do a series of tests. (Ideally repeat each stage a few times.)
Start with no keyboards or PS/2-to-USB adapters plugged in at all. Power up. How far does it get?
Power off. Wait 15s. Plug in a PS/2 keyboard straight into the PS/2 keyboard port. Power up. How far does it get?
Power off. Wait 15s. Unplug the keyboard. Plug in a PS/2-to-USB adapter only - no keyboard. Power up. How far does it get?
Power off. Wait 15s. Plug a keyboard into the PS/2-to-USB adapter. Power up. How far does it get?
The sooner you see a problem, the more you've isolated a fault. (There might be more than one.)
One question occurs to me: is the BIOS set up to boot from USB devices?
-
I'm afraid this isn't 100% clear.
Let's get back to basics. Firstly, stop hot-plugging PS/2 connectors if you are doing so. Even the die-hard "I've done it hundreds of times without any problems" people will accept it's not a good thing to do when diagnosing a fault like this.
Next, do a series of tests. (Ideally repeat each stage a few times.)
Start with no keyboards or PS/2-to-USB adapters plugged in at all. Power up. How far does it get?
Power off. Wait 15s. Plug in a PS/2 keyboard straight into the PS/2 keyboard port. Power up. How far does it get?
Power off. Wait 15s. Unplug the keyboard. Plug in a PS/2-to-USB adapter only - no keyboard. Power up. How far does it get?
Power off. Wait 15s. Plug a keyboard into the PS/2-to-USB adapter. Power up. How far does it get?
The sooner you see a problem, the more you've isolated a fault. (There might be more than one.)
One question occurs to me: is the BIOS set up to boot from USB devices?
Second sample of 1391401 keyboard (5/89), same problem.
In Rajagra's 4-step test, it boots all the way until step 4, at which point the Caps Lock and Scroll Locks lights flash and the computer does not power up at all. That says it's the keyboard - or the keyboard + adapter - but we already knew that. There is no USB only setting in the BIOS, and it doesn't seem relevant given what I've reported. So, here's a summary of the three scenarios that occur:
- Computer cold with IBM keyboard attached (USB adapter) - computer powers right up
- Computer shut down with IBM attached and then restarted - no power up, only flashing Caps Lock and Scroll Lock lights
- Computer started with other (Kensington) keyboard, shut down, then keyboard switched to IBM and restarted within 5-10 seconds - powers up - if longer wait, no power up
I thought it might be a capacitor discharge issue, since I wasn't waiting the recommended 45-60 seconds between power ups, but I tried at one-minute intervals for six minutes and still no go. Two hours, no problem.
Seems like there's some kind of interaction going on between keyboard, USB port and motherboard/PSU. If the problem was that the keyboard doesn't work with the mb, then the cold boot shouldn't work. Guess I need an electrical/electronics engineer or a different keyboard.
-
Or a different computer.
I've got another case in the corner. ;-)
There's got to be something about the motherboard, normal or malfunction, that starts cold but not warm when an clickity kb is attached.
-
Gigabyte board, eh?
This does sound like some kind of obscure cold boot problem. Current Gigabyte boards are somewhat renowned for having boot problems. You might want to try to contact Gigabyte and ask for advice (update your BIOS etc.)
-huha
-
Gigabyte board, eh?
This does sound like some kind of obscure cold boot problem. Current Gigabyte boards are somewhat renowned for having boot problems. You might want to try to contact Gigabyte and ask for advice (update your BIOS etc.)
-huha
Yeah, a new one, maybe worth a return. BIOS is up to date. Just emailed Gigabyte.
-
I think it's a power problem. When you push the power button, the PSU supplies power to some of the power wires but not all. The motherboard needs to return a "power good" signal back to the PSU, which then switches on all the power lines.
I think the IBM is taking too much current and maybe overheating a component. From a stone-cold boot it might be OK because the component doesn't have time to overheat. But if you power off/on the hot component plus the high current draw causes the motherboard to tell the PSU it is not safe to power up.
When you power off and back on within 10s it overrides the "power good" check - this explains why you can change from the Kensington to the IBM if you are quick.
You say you need to wait up to 2 hours to get the system to cold boot with the IBM. Try this: pull the mains lead out and then hold the PC's power button in for 15s. This discharges residual power from the PSU and motherboard. Plug the mains lead back in and try to power up. It should be the same as if you'd waited 2 hours.
I would've expected the powered USB hub to fix all this, but it only helped a bit. And I noticed about the parallel port failing with the USB hub on.
I say assume this is a power problem until proven otherwise. It's amazing how many wierd problems disappear when you put in a quality PSU.
What make PSU is fitted?
-
The PSU is a Corsair VX450, considered very conservatively rated and one of the better ones around.
Rajagra, your suggestion seems a good one but didn't work. However, in the process I found a couple of new scenarios with the IBM keyboard:
- Once the computer is warm, power down and then try restarting within 5-10 seconds, the computer starts up - wait longer and it doesn't (before that only happened by switching keyboards)
- In the latter case (longer wait), switch USB banks and the the computer powers up
(Note: I discovered, at least with this keyboard sample, that flashing lights don't always mean no start; and I ran this test with the Radio Shack adapter - blue cube next time)
What does that tell you? I can say that it's not a bad USB port per se, because I went back and forth twice (four boots) between the same two on different banks.
In the attached screenshot of the motherboard's layout, the two banks I switched between are labeled USB_1394 and USB_1394_1. I didn't try the others this time.
-
I posted this topic on what seems to be the closest thing to an official Gigabyte forum, http://forums.tweaktown.com/f69/. The resident electric guru posted the following reply:
----------
It sounds to me like the IBM keyboard is either drawing too much current for the adapter and/or the motherboards USB ports. Or through a combination of the above the adapter is sending corrupt Data via the USB Bus.
Without knowing what the max current draw is of the IBM keyboard in question and this article you linked too:- Model M keyboard - Wikipedia, the free encyclopedia states:-
"Sometimes it may be problematic to get them to communicate properly with a modern computer. A PS/2 to USB adapter would be necessary for computers without PS/2 ports, and the AT Model Ms additionally require an AT to PS/2 adapter. Some PS/2 to USB adapters are unreliable because the Model M tends to draw more power than some adapters expect. In response to these complaints, Unicomp added USB models to their lineup."
So as per my comments above, either the adapter is "falling over" due to the high current draw of this keyboard and or it could be exceeding (keyboard and adapter combination or even have a fault) the USB 2.0 current standard of a maximum of 500ma Universal Serial Bus - Wikipedia, the free encyclopedia
Either which way, i am slightly baffled why you wish to use a 20 year old keyboard that has been documentated as being troublesome when used with modern hardware and via an adapter.
Whilst it might be interesting to know what is specifically causing the "fault", the combination of a keyboard that wasn't designed for USB, using an adapter per-se, using an adapter with an adapter unfriendly keyboard makes it more likely that you've actually created a problem, rather than unexpectedly experiencing a fault.
__________________
Try other adapters, or another version in the line of these keyboards?
-
I love this quote:
Either which way, i am slightly baffled why you wish to use a 20 year old keyboard that has been documentated as being troublesome when used with modern hardware and via an adapter.
It's not troublesome, it's just that your crap can't handle 20 y/o technology. What is that?
Anyway, what you might want to do is use a powered USB hub as it will have an external power supply to supplement the anemic USB ports on your mobo. Hopefully, this helps. I only skimmed your post; did you try just using PS/2 yet?
-
I'm baffled why your computer doesn't play nice with the keyboard and two different dongles. Ours does.
Either:
1) Bad computer
2) Bad keyboard
3) You got 2 Bad dongles (unlikely)
Seriously, you need to do some swapping around of stuff until you isolate the problem. Get that keyboard to work with someone elses laptop first and then we can help you out more.
Don't listen to Tweaktown gurus that get their info from Wikipedia.
To be fair, in an earlier post by someone who questioned using a 20 year old keyboard, I sent them to Wikipedia. As much as I don't like or trust the latter, in this case it's the best all-around explanation.
Perhaps Rajagra will have something to add, but I'm suspecting an electrical incompatibility, at least using these two dongles. Unfortunately, I don't have any laptop available or other way to better isolate it that I know of, except for a PSU replacement.
-
I love this quote:
It's not troublesome, it's just that your crap can't handle 20 y/o technology. What is that?
Anyway, what you might want to do is use a powered USB hub as it will have an external power supply to supplement the anemic USB ports on your mobo. Hopefully, this helps. I only skimmed your post; did you try just using PS/2 yet?
No go with a powered (Belkin) USB hub. I've been assuming that's because the problem is at the USB port bank and the hub can't correct that. Not sure that explanation is correct, tho.
-
It sounds like some bad USB ports/controller on the mobo. Have you been able to try it with any other PCs or just straight PS2. I used to use a $4 POS (well, not really) adapter I got from Monoproce.com several years ago, and worked without a problem. I always had it plugged into a powered KVM switch, though. I can't imagine that there is anything wrong with the adapters or the keyboard itself. There is an outside chance that the PSU is bad (or has a bad/underpowered rail), but I would think you would see other problems if that were the case. Do you overclock your PC? I have heard in the past that it can cause some peculiarities with USB.
-
The PSU is a Corsair VX450, considered very conservatively rated and one of the better ones around.
They are a good make. Rating is not really relevant since you aren't even getting the PC to power up when the problem hits.
This is very strange:
- You've used a mains-powered hub. That should isolate the PC from any excessive power draw from a keyboard. (Unless there is a bad short, which doesn't seem to be the case.)
- The PC isn't being confused by "chattering" or other confusion from the keyboard, because it isn't even powered up!
I think this is a motherboard problem, either design fault or failure.
You might be able to confirm this by plugging in a bunch of other USB devices and testing.
The only thing I can think of trying now is a different mains-powered USB hub with more power rating. Seems like a long shot though.
EDIT> Well, two other suggestions. 1: Leave the keyboard in the PS/2 port, or 2: Plug the keyboard into USB after booting and unplug it before shutting down.
Not fixes, but maybe a workaround you can use.
-
itlnstln, I just ordered the Monoprice adapter. If that doesn't fix it, the mb goes back for replacement. No overclocking.
Rajagra, I don't quite understand your last suggestion, but I think just changing the USB port to the next bank over is the same amount of effort.
Corsair (PSU) tech also thought it likely the mb and suggested I check for grounding. I pulled the mb onto cardboard, but for some reason it wouldn't start. Then I switched cases, something I'd been procrastinating about, but the USB problem was still there.
Thanks to everyone who offered help. Will report back if and when there's some major change.
-
Corsair (PSU) tech also thought it likely the mb and suggested I check for grounding. I pulled the mb onto cardboard, but for some reason it wouldn't start. Then I switched cases, something I'd been procrastinating about, but the USB problem was still there.
The mobo needs the screws going into the case for ground; that's why it didn't work on the cardboard. I wouldn't send the keyboard back, I would try it on a few other PCs first. I think you have a bad motherboard.
-
It should also work on cardboard. Screws for grounding are nice to have for a variety or reasons (read: EMI), but they're not terribly neccessary. Ground is usually connected via the ATX plug.
-huha
-
It should also work on cardboard. Screws for grounding are nice to have for a variety or reasons (read: EMI), but they're not terribly neccessary. Ground is usually connected via the ATX plug.
-huha
If this is the case, then there is something definitely wrong with genegold's motherboard. I would imagine, then, that there is a power handling issue with the mobo causing the USB ports not to function properly.
-
If this is the case, then there is something definitely wrong with genegold's motherboard. I would imagine, then, that there is a power handling issue with the mobo causing the USB ports not to function properly.
I am hesitant to draw that conclusion from my cardboard test. When I first installed this mb and tried starting up a few weeks ago, the cpu and video card fans started to turn, then stopped, spinning freely a few times. I think that turned out to be the power switch connector needed reversing or something. Yesterday, at first, there wasn't anything, then the brief fan spin happened a couple of times. I checked and remade all the connections but couldn't find the problem, which is what led me to switch cases, now that the mb was out.
-
The Monoprice PS/2 to USB active adapter came today. Immediately prior to giving it a whirl, I'd been using the modern (Kensington) keyboard (USB) with a PS/2 adapter into the keyboard port. Upon plugging in the IBM/Monoprice the computer wouldn't power up, just the flashing kb lights like before. Then, I switched USB banks and it started right up. Same old song, except this time it did it without having been plugged in first. At that moment I'm wondering about the motherboard. However, ...
As a test, I swtiched back to the Kensington keyboard, this time using its USB connector. It worked no matter what, whether in the same or different USB ports. What I draw from that is the difference of electrical load - IBM vs. modern - is the kicker and that the motherboard itself is probably not at fault. I'll still probably RMA/replace it, just to be sure. To be continued.
-
Some motherboards abort power up if they don't detect certain fans spinning. Might have been worth disabling that setting in the BIOS.
Mobos shouldn't need to be grounded by the support posts (indeed some are plastic). But I'm thinking:
- When the mobo is out of the case you have no case fan connected - the sytem may have realised this;
- When you draw too much power from the USB ports maybe that is stopping the fans from spinning up fast enough.
- A poorly designed motherboard might well be dependent on additional grounding via the support posts.
-
Some motherboards abort power up if they don't detect certain fans spinning. Might have been worth disabling that setting in the BIOS.
Which setting is that? manual: http://www.giga-byte.com/Support/Motherboard/Manual_DownloadFile.aspx?FileType=Manual&FileID=19157
XP/SP3. All the USB ports work, just - looking at them as banks of ports - not necessarily each one at any specific time. It depends on whether the machine is cold (all work) or has been just shut down, and how long it's been since the latter (<~10 secs ok, >10 seconds have to change USB bank).
Update: A new wrinkle. I have a master console. That ~10 sec. figure applies if I turn off the console power to the computer; if left on, then it seems to restart ok on the same port at >10 secs. I've tried up to 90 secs. so far (Monoprice adapter).
-
Which setting is that? manual: http://www.giga-byte.com/Support/Motherboard/Manual_DownloadFile.aspx?FileType=Manual&FileID=19157
Don't worry, it doesn't have the option.
-
After the behavior described at the end of my last post, I RMA'd the motherboard for replacement and then hooked up my three year old K8 board (MSI K8NGM2-FID). The IBM keyboard now works without any problem so far off the PS/2 port(!), as well as, with adapter, the USB ports. That suggests to me it was either a bad Gigabyte motherboard or poor engineering. Hopefully, the first.
-
Good luck with the new motherboard.
I liked how the Gigabyte forum guy blamed the keyboard. The Power Supply guy blamed the motherboard.
For a while I even doubted the Blue Cube!!! Still da Winnah!!
A winner. The Radio Shack and Monoprice work just as well too.
-
Maybe - nobody has put those through the paces like da Blue Cube. Plus, I'd NEVER trust anything from Radio Shack. I even had a simple phono to submini jack coverter that was bad once. And don't get me started on their electronics...
Blue Cube - why risk anything else for a $6 dongle?
I've put the Monoproce one through its paces. It's always a go.
-
Should add to the wiki what worked for you.
Hmm... Will do. I need to get a copy of that software used in the main article.
-
Hmm... Will do. I need to get a copy of that software used in the main article.
There's a link in the article pointing you to the place to download the software.
If you're not keen on adding it yourself, just send me the data and I'll add it.
-huha
-
There's a link in the article pointing you to the place to download the software.
If you're not keen on adding it yourself, just send me the data and I'll add it.
-huha
No prob. I just have to find the time at work to do it. I know it works for every keyboard in my sig (even the USB ones, minus the ABSs, using the PS/2 adapter). I just need to find an ISO 'board to test with. They're not easy to find around here.
-
I've put the Monoproce one through its paces. It's always a go.
Has anyone else tried it with the SGI AT101? CX reported that it didn't work so it'd be interesting to know whether that's a one-off or not.
-
Da Blue Cube sez:
Can't you handle a SGI in da face!
My Monoprice adpater handles my Dells just fine (same 'board).
-
My Monoprice adpater handles my Dells just fine (same 'board).
I know they've got the same shell but are you sure they've got the same controller as well as that'll affect whether they work with the adapter - and it was designed for SGI systems rather than for PCs generally so it's a good chance it may not do.
-
I know they've got the same shell but are you sure they've got the same controller as well as that'll affect whether they work with the adapter - and it was designed for SGI systems rather than for PCs generally so it's a good chance it may not do.
As far as I know, they share the same internals the only difference being the dampers on the switches. If someone has both 'boards, preferrably of close vintage to help eliminate time-based hardware revisions, maybe they can look into it (no pun intended).
-
I own both keyboards. Are you asking to test them with the "blue cube" or with the Monoprice adapter? I do have one of the blue cubes available as well as one or two other PS/2->USB adapters, but none from Monoprice.
-
I own both keyboards. Are you asking to test them with the "blue cube" or with the Monoprice adapter? I do have one of the blue cubes available as well as one or two other PS/2->USB adapters, but none from Monoprice.
We are wondering if the SGI and the Dell have the same internals (Controller, PCB, etc.). We know the Monoprice adapter works with the Dell, but there was a report that it didn't work with the SGI. We don't know if this a regular occurance or if it was just one person's experience. If the internals are the same, then it might just be a one-off that it didn't work.
-
Here are two photos of the PCB/controller immediately above the function keys on my SGI AT101 (part # 9500900). I may be able to get similar photos from a Dell AT101W tonight or tomorrow.
-
Here's the Dell AT101W. It's nothing like the SGI AT101:
-
Skriefal, thanks for opening those up. When were both of those produced? Looks like I might need to do a test on the SGI with the Monoprice adpater.
-
I own both keyboards. Are you asking to test them with the "blue cube" or with the Monoprice adapter? I do have one of the blue cubes available as well as one or two other PS/2->USB adapters, but none from Monoprice.
Ideally we want everyone to test every keyboard they've got with every adapter they've got and report the results so we can come up with one that works with everything. In this case though testing the SGI AT101 with both adapters would answer the point at hand.
-
Skriefal, thanks for opening those up. When were both of those produced? Looks like I might need to do a test on the SGI with the Monoprice adpater.
This SGI AT101 has a May 1995 build date. I forgot to note the build date of the AT101W.
-
After the behavior described at the end of my last post, I RMA'd the motherboard for replacement and then hooked up my three year old K8 board (MSI K8NGM2-FID). The IBM keyboard now works without any problem so far off the PS/2 port(!), as well as, with adapter, the USB ports. That suggests to me it was either a bad Gigabyte motherboard or poor engineering. Hopefully, the first.
Back again. Motherboard replacement arrived. Plug the IBM 1391401 keyboard directly into the PS/2 keyboard port and it works fine from cold computer. Try starting from ~30 seconds to a minute or two after shutdown and computer won't power up, like before (adapter/USB no different). Wait 25 minutes, as I just did, and it starts fine. Guess that's cold enough. That leaves either motherboard design/testing (hardware, BIOS, ..?) or PSU. Since the keyboard worked with a three year old K8 board no problem, I would guess that rules out the PSU, but what do I know.
The key hardware is:
- Gigabyte GA-MA790X-UD4P
- AMD Athlon 64 X2 7750 Kuma 2.7 GHz Dual Core Black Edition
- Corsair VX450 PSU
(all new)
-
Putting aside the keyboard for a second, the other idea is to really check the system power. It's clearly only an issue under load so maybe try running SuperPi or Prime95 and then take a voltmeter to a power connector and check for whether you're getting true 5V and 12V.
Only other idea is that the particular Model M you have is really corroded or something and is drawing more power than usual. Once again, I don't believe it's a common problem or a whole bunch of people around here would be complaining about their boot ups.
Since it's the second M keyboard that's had a problem with this mb/PSU combo, I just decided to RMA the PSU. Should have a replacement within a week. I've also got a new monitor with USB ports coming and wonder if that would have any effect or just be like the USB hub weirdness last time.
-
Replacement PSU today. Good news is that the problem seems to have disappeared, at least so far (1391401 into PS/2 port). I'll qualify the PSU as the solution by saying at the same time I moved the mb from cardboard to case and hooked up all the case switches and fan. Don't know if having those around changes the whole kb-mb interaction vis a vis residual current/capacitance. The bad news is the wiring inside the PSU hits its fan. I moved it around to kill the tapping, but I'm not sure that's a viable long-term solution yet.
-
OK, a couple of ways to fix that power supply. Be safe - the caps in there can knock you on your butt or worse.
Thanks, I'll bookmark that. The problem seems to be the placement of the cables inside relative to the fan, since moving it around a bit from the outside got rid of the tap. Have to look in daylight. Certainly don't want to give up a PSU that's got the keyboard working right.
-
I may be wrong, but in the pic of the SGI AT101 PCB, it looks like capacitor C5 is bad or damaged. I wouldn't expect to see the "inner cylinder" sticking out as it is on the right side.
-
I see. I've never seen one that's like that before that's by design; I've seen it on radials where physical stress on the cap has pulled a section out the bottom and it looks like that.
-
Whew, that was a lot of work - we're getting close to 100 posts on this. Kudos to the unnamed Geekhack trio:
Hope you're enjoying those Model M's now!
Thanks to everyone for suggestions, comments and patience. It feels anticlimatic after all this to find that the problem was the PSU. In any case, I have been enjoying the Model M, and now without having to wait between restarts.