[back to collection]

Virgin Webplayer

(last updated: 04-Jan-2016)


Operational Status

Configuration

I have swapped the previously hacked-in hard disk for a CompactFlash 'disk' using IDE-to-CF adapter.

Major Events

Still To Do


Description

Acquisition

15-Jul-2009

Cruising around eBay recently I was pleasantly surprised to see this Webplayer on auction with an opening bid of just $9.99 with 0 bids and just over a day to go. The seller clearly stated that this was a hacked system, had been running Windows 98, but now would no longer boot up. The offer was for the main unit, keyboard, and AC power supply, all AS-IS.

I setup an esniper job to put in my bid, and lo, I was the only bidder! So I snagged this unit for only $10 plus $12 shipping. Very, very cool! The unit arrived today, well packed and as described. The seller even included a CD-ROM with Windows drivers and a printout of one online hacker how-to.

Cleanup

15-Jul-2009

The unit was in pretty good shape, other than a bit of accumulated dust from being (no doubt) in storage for some time. I performed my usual 'baby-wipe' cleanup job and other than a few nicks and dings in the case, it looks great!

The unit has already been hacked, with two major chassis modifications. They are the installation of a CPU heatsink/fan and the replacement of the original power switch with a big On/Off rocker switch. Internally, both the modem and modem jack panel have been removed, the Disk-On-Chip (DOC) has been removed, and an 802.11b networking mini-PCI card has been installed.

My guess on the power switch is that the prior owner/hacker wanted to shut the system completely off, once there was a spinning harddisk installed. I don't know yet, but I suspect that the original power switch was a sort of 'soft-switch' that just put the unit into 'sleep' mode when pushed, rather than completely powered off.

Also, without the DOC, I don't have the original Virgin software which would have been interesting to experince. Still, it would have been no good without the modem anyway, I guess.

I've come across a number of resources:

http://www.mambaza.com/projects/webplayer/
http://www.larwe.com/technical/webplayer_main.html
http://www.dougpile.com/computer/webplayer/webplayer.html
http://en.wikipedia.org/wiki/Virgin_Webplayer
http://www.webplayer.0catch.com/

After reading up on a few of these sites (particularly the last) it would seem to confirm my guess that the power switch is really a soft-switch after all. My Webplayer's previous owner cut the two power switch wires going to CN9 (pins 1,2) when he installed the massive power switch. I think I'd like to try resurrecting that old power switch functionality.

Initial power-up

15-Jul-2009

I put batteries (AA x2) into the keyboard and powered up the system. It showed the POST screen and BIOS configuration screen, then tried to boot, but the hard disk kept spinning up, clicking, powering down, then cycling again and again. It looks like the hard disk is bad. But the system itself seems to work.

The keyboard keys are bit mushy and 'stick' a bit when the plungers don't slide very easily into the stem on an individual key, like an old IBMjr 'chiclet' keyboard. But it works.

The display contrast is really high. Everything is washed out, just inside the 'almost okay' range when I turn the contrast all the way down using the front panel wheel.

Mass storage

17-Jul-2009

Using the printed hacking document included with this unit, I cracked it open and pulled the hard disk out. I connected the drive to a separate desktop system using my 2-1/2" to 3-1/2" IDE adapter and fired it up. The drive worked fine! I put the drive back into the Webplayer and experienced the same spinup-click-spindown cycle again. I'm starting to think that the drive isn't getting enough current to power the spindle motor.

Since I would prefer to have the unit operate with no noise whatsoever, without spinning disks and fans, etc., I decided to go ahead at this stage and install a 4GB CompactFlash card using a 44-pin CF-to-IDE adapter that I had on hand. Booting the system showed that the CF card was recognized as a disk just fine, and all I had to do now was create a file system and move all of the necessary installation files onto it.

Windows 98SE installation

17-Jul-2009

After a little bit of fussing around to make the 4GB CF card bootable (Win95 DOS) on a seperate system I copied over the installation files for Windows 98SE, 98Lite, and the hardware drivers provided by the seller. Now ready to go for it, I moved the CF card back into the unit and fired it up.

I started by running the 98Lite installation which prepared the 98SETUP directory with all of the windows installation files. I choose to use the 98Lite 'sleek' option which uses the faster and smaller desktop and windows explorer files from Windows 95. With this all prepared, I started the Win98SE setup.

The setup went without a hitch, as normal as could be. Once the system rebooted, it showed that drivers were missing for four devices:

- PCI bridge
- Audio device
- Video device
- Network device

Using the drivers from the seller, I was able to update the audio, video, and network device drivers, but was unable to find anything to deal with the PCI bridge device. The seller's set of drivers included a UDMA driver that I tried, but that wasn't a proper match. For now, I just left it as Unknown, and the system seems to work anyway.

One issue, though, was with the audio. I could get no output until I went into the audio controls and muted the Line-In channel. Then everything started working just fine.

So, I'm now a happy camper, as far as the system's software configuration goes anyway. Next steps are to decide if I want to try to return the system to a more 'virgin' (ha ha) state, without the CPU fan.

Disassembly

19-Jul-2009

Now that I have confirmed that the system is operational and I know how it generally works, I decided to take the system completely apart to see what's what inside, and to review my options for returning the system to a more 'virgin' state, and to see if it would be reasonable to disable the fan, making the system completely quiet now that the hard disk is gone.

Yep, I confirmed that the original power switch was of the soft-switch variety, and that it should have two leads going to CN9. I removed the two pins from the connector and I'll try to find the correct sized wire to go to a momentary soft-switch that I'll eventually have to find. I guess I can at least try out the functionality of these two pins by just shorting them together for the time being, just to confirm my understanding.

Also, I can see that the previous owner has modified the power-in socket so that he could tap it for the fan, as well as fit his hard on/off switch, which cuts all power to the system when switched off.

I removed the heatsink/fan assembly from the CPU and confirmed that the CPU is the original Geode. The seller thought that he might have upgraded the CPU, but this is not the case. So now I have to ask, what's with the big heatsink/fan on the CPU? If it's the original CPU, a standard heatsink would have been enough. He needn't have gone to all the trouble of cutting a giant hole in the case for the fan. Maybe it has to do with Windows 98, which he was running. Win98 is too old to have the smarts to know when it's in an idle loop and run slower. I've noticed that in my Netpliance i-opener that the CPU got really hot when running Windows, but when running Linux or NetBSD, it was cool, but then it's running a 'regular' type of CPU, not one designed for low power.

Perhaps the case of running Win98 is the reason the previous hacker installed the fan. Still, when I was running my installation of Win98SE on this unit, the heatsink stayed absolutely cool to the touch, even though I was playing Space Quest I which runs the CPU 100%. Hmmm.... In any case, I do plan to disable the fan and see what happens. I just want this thing to be quiet! I guess, though, that I'll go ahead and leave this same heatsink in place, with the fan disabled, since the case has already been hacked up (literally!) and it wouldn't look good if I messed with trying to close the fan cutout.

As for mass storage, there is an unpopulated CF card socket location on the motherboard. It is immediately adjacent to the 44-pin IDE header, and I'll bet a CF socket could be soldered onto the board and it would 'just work'. The chassis has clearance for the card already 'built in' so that with the black side-panel removed a CF card could be inserted or removed without further disassembly! I'm going to try this once I get a CF socket ordered.

So, what I need, to clean up the system so to speak, is a momentary push button to replace the power switch rocker, and a CF-card socket. I'd be pretty happy with the system at that point.

Assembly with CF pseudo-disk

26-Sep-2011

At last, after many months I have returned to this system. I have finally received another IDE-to-CF adapter and can now return this sytem to working status with a solid-state disk of sorts. Using a 1MB CompactFlash card as the psuedo-disk, I installed the adapter and CF card to the end of the IDE cable, then bagged that to prevent electrical shorts.

With the mass storage safely tucked away inside the system, I buttoned it all up.

Power connector re-soldered

24-Sep-2015

The power connector on the PCB inside has come free. This system had arrived with a weekly soldered power jack in the first place, and I had previously resoldered it. But, this system is also one of those "no lead" RoHS units, and I recall having a difficult time getting the no-lead solder hot enough to do a good job. Clearly, it wasn't a good job. Today, using a newer and hotter soldering iron, I resoldered the connector, and it is firm. I reassembled the system, and it runs a champ!


[back to collection] [top]

Copyright © 2006-2025 Jared Blaser. All rights reserved.